No title

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm Investigat...

Author: Emerald

267 downloads 925 Views 6MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

Investigation into wrinkle behavior of woven fabrics in a cylindrical form by measuring their tangential force S. Shaikhzadeh Najar Department of Textile Engineering, Amirkabir University of Technology, Tehran, Iran

Wrinkle behavior of woven fabrics 7 Received 27 May 2008 Revised 21 July 2008 Accepted 21 July 2008

E. Hezavehi Department of Textile Engineering, Research and Science Center, Azad University, Tehran, Iran

Sh. Hoseini Hashemi Department of Mechanical Engineering, Iran University of Science and Technology, Tehran, Iran, and

A. Rashidi Department of Textile Engineering, Research and Science Center, Azad University, Tehran, Iran Abstract Purpose – The purpose of this paper is to describe a unique approach to investigate the wrinkle force of textile structures in a cylindrical model. Design/methodology/approach – In this research, an apparatus was designed and constructed in order to investigate the torsional and wrinkle behavior of textile structures in a cylindrical model under a different rotational level using data acquisition and micro-controller systems. Findings – In the light of research results, the fiber and fabric type, fabric physical and mechanical properties and imposed rotational level significantly contributed to wrinkle characteristics of worsted fabrics. It was noticed that with increase of rotational level, the wrinkle force, and energy increased along weft and warp directions. Wrinkle characteristics along warp direction exhibited greater values than in weft direction. Originality/value – The study is aimed at determining wrinkle behavior of worsted fabrics under the combined influences of compression and torsional strains. Keywords Data collection, Wool fabric, Textile technology, Fabric testing, Deformation, Bagging Paper type Research paper

The authors wish to thank Dr R. Ghazi Saeidi for his significant technical support of Lab-view software and data acquisition instruments. The authors also wish to express their gratitude to the manager of Iran-Merinous Textile Co., and in particular Mr Emami and Mr Dokhanchi, for providing worsted fabric samples and also for testing fabric physical and mechanical properties by SIROFAST tester.

International Journal of Clothing Science and Technology Vol. 21 No. 1, 2009 pp. 7-30 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910923728

IJCST 21,1

8

1. Introduction Wrinkle is a form of three dimensional fabric deformation that usually appears at the back of the knee in a trouser leg. In fact, whenever, bending occurs in more than one direction, so that the fabric is subjected to double curvature (Hearle et al., 1969; Amirbayat and Hearle, 1986) and as a result, some degree of in-plane and out-of-plane permanent deformations will be resulted (Amirbayat and Alagha, 1996). When a wearer bends his leg, the top part of the trouser leg is extended resulted to the bagging deformation into the fabric. It was shown that the in-plane fabric tensile properties contributed to bagging behavior of woven fabrics (Abghari et al., 2004). On the other hand, during bending the trouser leg, the effective length of the inside of the trouser leg is reduced leading to buckling and hence wrinkling of the fabrics (Postle et al., 1988). It is shown that when a thin cylinder shell compressed in its axial direction, the wave length of buckling is dependent almost on shell diameter and thickness and is independent on mechanical properties of the material of the shell (Timoshenko and Gere, 1961). Skelton and Freeston (1971) investigated the shear deformation of a rectangular specimen formed into a cylinder subjected to an axial load. They related theoretically the shear stiffness of the fabric to the cylinder axial load, various geometrical parameters, angular rotating, and the tangential force. The authors designed an apparatus to measure the tangential force against shear deformation for loom state, heat-set and coated fabrics. Shinohara et al. (1991a) analyzed the garment wrinkling by deforming a fabric cylinder in axial compression. They found that the buckling pattern generated in knitted fabrics is different from that in woven fabrics. In their further theoretical work, they proposed a mathematical model to describe the buckling deformation of a woven fabric cylinder in axial compression (Shinohara et al., 1991b). Basset et al. (1999) reviewed the experimental methods for measuring fabric mechanical properties and discussed the proposed test methods for cylindrical specimens of permeable fabrics. In AATCC wrinkle recovery tester (1994), the fabric is formed as a cylindrical shell compressed and rotated under constants of axial load and rotational angle. However, there is little information available to study wrinkle force of worsted fabrics under the combined influences of compression and torsional strains. Therefore, the aims of this work are to investigate the wrinkle behavior of worsted fabrics by measuring their tangential force using a new developed test method and compare the obtained results with different fabric types and torsional strains. 2. Experimental 2.1 Material and methods In this research, 13 different woven twill worsted fabrics were used. Table I shows the general fabric specifications. The mechanical properties of the specimens including fabric thickness, bending rigidity, extensibility, formability, and shear rigidity were measured using SIROFAST tester as shown in Table II. The tensile elastic properties were measured using Tensolab Mesdan tensile tester. In this test, the fabric width and length and test speed were adjusted at 5, 15 cm, and 300 mm/min values, respectively. In addition, the Poisson ratio along warp and weft directions (V1 and V2) were measured using a photography technique at strain levels of 6.66, 13.3, 20.00, and 26.6 percent, respectively, and then the average value was calculated.

Fabric density Fabric A B C D E F G H I J K L M

Construction

Yarn count (Nm)

T2/2 T2/1 T2/1 T2/1 T2/1 T2/1 T2/1 T2/1 T2/1 T2/1 T2/1 T3/1 T2/2

40/2 40/2 40/2 40/2 48/2 48/2 48/2 48/2 60/2 60/2 60/2 60/2 60/2

Fiber content (percent) 45w 45w 20w 7w 45w 30w 38w 20w 35v 7w 20w 45w 45w

55p 55p 80p 93p 55p 70p 55p,7v 80p 65p 64p,29v 80p 55p 55p

Weight (g/m2)

Picks/cm

Ends/cm

Thickness (mm)

268 256 260 233 222 229 231 221 192 195 213 225 195

18 18 18 19 21 19 20 19.5 24 25.5 22 26 22

29 29 28.5 32 29 31 29 28.8 27 31 36 34.5 34

0.686 0.664 0.586 0.67 0.57 0.6 0.58 0.67 0.475 0.57 0.543 0.59 0.48

Notes: w, wool; p, polyester; v, viscose; T, twill

All experiments were performed under the standard conditions of 22 ^ 28C and 65 ^ 2 percent r.h. 2.2 Fabric wrinkle tester To investigate the effects of torsional and compression strains on woven fabric wrinkle force, a unique test method is developed. A photograph of the wrinkle force tester is shown in Figure 1. A schematic diagram of wrinkle force tester is also shown in Figure 2. The wrinkle tester includes of two main electrical and mechanical parts. The electrical part includes of a load cell (Model BONGSHIN, Type DBBP-S-Beam, 20 kg), stepper motor (Model Sanyo Denki, Type 103H 89222-6341, 22 kg cm) with its driver and intermediate electrical board and also a PC. The speed of the stepper motor is kept constant at a rate of four step/s and its rotational direction is clockwise. The intermediate board receives analog signals from load cell and then converted to digital signals using a A/D board. Lab-view software Ver.6 (National Instrument Co., Austin, TX) is used in order to run, control the apparatus and register and monitor the data. The mechanical part contains one frame, two gears (118 and 68), two circular rings in 90 mm diameter and a spiral shaft. The bottom ring is connected to the load cell and the upper ring can be freely rotated and vertically displaced over a spiral shaft through the stepper motor. In this research, three different spiral shafts are used. The specification of spiral shafts is shown in Table III. Before positioning the fabric sample in the wrinkle tester, the two circular rings are covered by a double-sided sticky tape. Then, the fabric sample with dimensions of 290 £ 160 mm is mounted in a cylindrical form between the two circular rings while a constant tension force of 2 N is applied to the fabric. The fabric sample is tightly fastened over the circular rings in order to prevent from any slippage during the test. When the top ring is rotated clockwise and moved downwards, the load-cell measures

Wrinkle behavior of woven fabrics 9

Table I. Fabric general specifications

A B C D E F G H I J K L M

11.4 11.3 11.6 11.2 11.1 19.3 10.3 8.7 10.8 12.4 16 14.3 5.4

B2 (mN m) 17.5 21.4 19.6 19.5 17.2 24.3 14.8 19.4 14.9 14.9 22.2 18 9.1

B1 (mN m) 48.6 37.9 41.5 43.7 47 46.6 31.8 44.3 42.2 42.3 37.11 41.2 27.1

Shear rigidity (N/m) 2.38 1.81 3.89 3.37 1.86 2.09 0.82 1.54 1.24 5.25 1.13 3.56 0.68

E2 (N/mm) 1.83 1.89 1.75 5.52 1.5 1.78 0.53 3.09 0.25 3.9 3.27 1.2 2.08

E1 (N/mm)

0.55 0.7 0.56 0.69 0.62 0.78 0.68 0.6 0.6 0.78 0.74 0.67 0.72

V2

0.6 0.75 0.59 1.17 0.8 0.9 0.78 1.22 1.02 0.83 0.85 0.75 1.12

V1

10 2.74 2.30 2.40 2.50 1.10 1.10 1.40 1.00 1.10 1.10 1.10 2.00 1.10

2.40 2.40 2.40 2.90 2.00 1.10 2.30 2.90 2.90 2.00 1.10 1.30 2.90

Extensibility (percent) E100-1 E100-2

Notes: V1, Poisson ratio (warp direction); V2, poisson ratio (weft direction); E1, tensile elastic modulus (warp direction); E2, tensile elastic modulus (weft direction); B1, B2 (bending rigidity in warp and weft direction)

0.36 0.43 0.32 0.75 0.40 0.49 0.30 0.42 0.25 0.45 0.18 0.29 0.21

0.87 0.83 0.80 0.60 0.47 0.24 0.3 0.35 0.25 0.27 0.35 0.53 0.27

Fabric

Table II. The fabric mechanical properties

Formability (mm2) F-1 F-2

IJCST 21,1

Wrinkle behavior of woven fabrics

Stepper Motor

11 Gear

Driven

Driving Gear Load cell

Figure 1. A typical picture of fabric wrinkle force tester

Sample

Stepper Motor

Computer

Upper Ring

Driver

Spiral Shaft Load Cell Microcontroller

Shaft specification Total effective length (mm) Number of rotation turns throughout the effective length External diameter (mm) Pitch (mm) Spiral angle (degree) Rotational level (turn/m)

Figure 2. A schematic diagram of wrinkle force tester

Bottom Ring

1

Number of shafts 2

110

110

0.75 22 146.6 25.24 6.81

1 22 110 32.14 9.10

3 110 2.75 22 40 60 25

Table III. Spiral shafts specification

IJCST 21,1

the tangential resistance force against bottom ring rotation. This force represents as an indication of wrinkle force of the specimen. The friction of circular ring around spiral shaft can influence the detected force. This effect is assumed to be small which is ignored in this work. The output data in real time is transferred to a PC computer using the data acquisition system. To store and analyze the data, a MATLAB software is also used. 5 tests were performed for each fabric type.

12

13 12 11 10 9 8 7 6 5 4 3 2 1 0

N/Deg. (1st cycle)

Torsional Force (N)

Torsional Force (N)

2.3 Wrinkle parameters investigated A typical diagram of the wrinkle force against torsional strain for one cycle loading is shown in Figure 3. In order to evaluate the wrinkle behavior of fabrics, different parameters including the maximum wrinkle force Fw, wrinkle energy Ww, the wrinkle recovery energy Wwr, the dissipated wrinkle energy Ew (Ew ¼ Ww 2 Wwr), the resilience of the fabric wrinkle Rw (Rw ¼ Wwr/Ww) and the percentage of hysteresis Hw (Hw ¼ Ew/Ww) were calculated. The experimental results of fabric wrinkle test are shown in Tables IV-VI. The experimental results of fabric wrinkle characteristics values were statistically analyzed using ANOVA and Multiple Range test methods.

MAX(F) = 4.95N ← Ww = 99.2919N.Deg

Wwr = 20.9002N.Deg

Hw = 78.9507%

0

45

90 135 180 225 270 315 360 405 450 Torsional Strain (Degree)

N/Deg. (1st cycle) 13 12 MAX(F) = 13.2N ← 11 10 9 8 7 6 5 4 Ww = 388.5152N.Deg Wwr = 182.3777N.Deg Hw = 53.0578% 3 2 1 0 0 45 90 135 180 225 270 315 360 405 450 Torsional Strain (Degree)

(a)

(b)

Figure 3. Typical diagrams of wrinkle force against torsional strain in one cycle loading for fabric type D in warp direction (a ¼ 6.81, b ¼ 9.10, c ¼ 25 turn/m)

Torsional Force (N)

N/Deg. (1st cycle) 18 MAX(F) = 17.44N ← 17 16 15 14 13 12 11 10 9 8 7 6 5 Ww = 1,335.7002N.Deg Wwr = 897.2177N.Deg 4 Hw = 32.8279% 3 2 1 0 0 45 90 135 180 225 270 315 360 405 450 Torsional Strain (Degree)

(c)

Fabric A B C D E F G H I J K L M

6.81 turn/m Warp Weft 3.6 (0.09) 1.56 (0.11) 4.81 (0.29) 4.86 (0.17) 2.64 (0.17) 1.77 (0.12) 1.81 (0.09) 2.73 (0.05) 0.65 (0.06) 3.44 (0.07) 2.23 (0.04) 1.13 (0.07) 1.10 (0.06)

2.37 1.41 1.74 2.76 1.98 1.13 1.42 1.93 0.46 2.63 1.95 0.75 0.78

(0.05) (0.07) (0.11) (0.09) (0.16) (0.08) (0.05) (0.05) (0.09) (0.07) (0.08) (0.10) (0.63)

9.10 turn/m Warp Weft 17.36 (0.21) 14.30 (0.12) 13.90 (0.61) 12.95 (0.26) 10.94 (0.3) 12.53 (0.23) 13.32 (0.31) 15.22 (0.19) 2.75 (0.04) 8.61 (0.17) 9.73 (0.3) 10.31 (0.30) 7.7 (0.12)

8.52 (0.056) 5.63 (0.1) 8.59 (0.17) 8.82 (0.11) 7.22 (0.21) 10.28 (0.07) 4.8 (0.13) 5.12 (0.1) 1.76 (0.05) 6.78 (0.16) 6.63 (0.11) 7.28 (0.19) 4.01 (0.184)

25 turn/m Warp 17.85 (0.09) 16.9 (0.04) 16.95 (0.04) 17.13 (0.09) 16.91 (0.05) 16.92 (0.04) 16.91 (0.04) 16.84 (0.03) 16.96 (0.04) 16.63 (0.03) 16.89 (0.05) 16.94 (0.04) 16.92 (0.02)

Weft 16.59 16.76 16.84 17.07 16.82 16.87 16.85 16.99 16.88 16.83 16.76 16.88 16.86

(0.05) (0.05) (0.03) (0.02) (0.13) (0.05) (0.03) (0.04) (0.05) (0.03) (0.08) (0.05) (0.07)

Note: The data in brackets are SD values

3. Result and discussion 3.1 Wrinkle mechanism The wrinkle generation process with wrinkle tester is shown in Figure 4. It may be considered that the wrinkle mechanism of worsted woven structure consists of four zones. As upper ring gradually rotated and moved downwards, the fabric shell layers start to buckle and hence torsional and compression buckling would be created. In this zone, the wrinkle force is almost zero since the structural instability prevents fabric layers to be completely twisted and also the load cell can only measures the torsional force rather than compression force. In addition, the low rate of rotational strain can also be responsible for occurrence of this event. With further twisting the fabric structure, and at low wrinkle force, the fabric layers are partially twisted while they are compressed and wrinkle characteristic in this second region is presumed to be elastic. Increasing the torsional strain and hence the wrinkle force overcomes the internal fiber and yarn friction and fiber slippages takes place. Thus, in this third region, the fabric wrinkle force increases nonlinearly with increasing rotational angles. Further increasing the rotational angles, the fabric layers is laterally compressed and twisted and in this forth region, a highly packed and twisted compressed fiber assembly can be considered. The friction of fabric around spiral shaft may affect the results and this manifests partially itself as an increase of energy loss or hysteresis. It may be considered that for the spiral shaft of 25 turn/m (Figure 3), the wrinkle force variation is different from in other spiral shafts. It is shown that the fabric will be rapidly reached towards the non-linear region and hence a high twisting level will be obtained. 3.2 The effect of fabric type on wrinkle characteristics along warp and weft directions A summary of ANOVA statistical analysis result and Duncan test result of wrinkle force, energy and hysteresis in 5 percent confidence limit for different worsted fabrics along two warp and weft directions at different rotational levels are shown in Tables VII-XIII. The variation of these wrinkle characteristics against fabric type along two warp and weft directions are also depicted in Figures 5-10. As shown in

Wrinkle behavior of woven fabrics 13

Table IV. The experimental results of the wrinkle tester (maximum wrinkle force value, N) Fw

2,005.87 (0.3) 1,786.62 (1.93) 1,508.77 (3.28) 1,343.01 (7.52) 1,469.97 (8.13) 1,754.30 (1.95) 879.31 (2.30) 1,313.01 (2.71) 728.24 (6.51) 920.20 (0.55) 897.99 (3.48) 966.34 (2.01) 1,155.6 (0.3)

Note: The data in brackets are SD values

683.21 (8.83) 439.37 (0.6) 449.59 (0.16) 409.21 (1.5) 320.75 (0.12) 483.34 (0.32) 228.65 (0.32) 573.95 (0.35) 53.50 (0.12) 240.74 (0.22) 327.92 (0.47) 343.7 (0.29) 243.41 (0.21)

81.32 (0.1) 19.75 (0.09) 71.31 (0.13) 98.88 (0.11) 20.40 (0.09) 56.88 (0.07) 27.02 (0.07) 49.75 (0.06) 4.19 (0.06) 75.80 (0.05) 42.05 (0.1) 12.36 (0.08) 15 (0.09)

A B C D E F G H I J K L M

Table V. The experimental results of the wrinkle tester (wrinkle energy and hystersis), Ww, Hw

Ww (warp) (N Degree) 6.81 turn/m 9.10 turn/m 25 turn/m 61.01 (0.39) 18.36 (0.1) 45.32 (0.16) 69.66 (0.23) 15.28 (0.17) 47.58 (0.32) 12.27 (0.09) 45.34 (0.15) 3.16 (0.10) 5.09 (0.1) 40.19 (0.15) 4.39 (0.07) 8.06 (0.13)

263.61 (0.41) 144.15 (0.32) 320.49 (0.46) 408.82 (0.38) 224.07 (0.44) 338.38 (0.38) 80.58 (0.32) 123.50 (0.36) 20.5 (0.28) 154.59 (0.34) 221.81 (0.17) 234.64 (0.21) 88.14 (0.18)

1,891.20 1,591.82 919.71 1,029.81 1,245.17 680.92 752.02 1,175.20 714.51 523.91 729.36 826.24 820.46

(1.80) (1.66) (3.02) (2.73) (1.18) (0.85) (0.75) (0.5) (1.55) (0.69) (2.60) (1.23) (1.90)

Ww (weft) (N Degree) 6.81 turn/m 9.10 turn/m 25 turn/m 76.19 (0.03) 75.57 (0.66) 75.10 (0.35) 79.26 (0.09) 83.23 (0.08) 55.23 (0.12) 87.14 (0.1) 70.73 (0.78) 80.78 (0.02) 73.27 (0.23) 76.46 (0.23) 84.01 (0.38) 63.26 (0.31)

54.64 54.92 59.63 55.18 48.66 62 51.87 54.02 64.83 54.02 58.05 56.79 55.94

(0.3) (0.09) (0.19) (0.11) (0.24) (0.08) (0.24) (0.15) (0.28) (0.12) (0.07) (0.18) (0.15)

54.74 (0.08) 43.9 (0.61) 72.34 (0.09) 32.82 (0.08) 50.22 (0.04) 27.42 (0.09) 52.24 (0.06) 35.81 (0.02) 33.49 (0.02) 34.31 (0.02) 44.07 (1.01) 47.97 (0.01) 44.87 (0.1)

Hw (warp) percent 6.81 turn/m 9.10 turn/m 25 turn/m

71.25 85.15 80.82 80.06 70.06 63.84 86.97 80.83 82.57 70.49 76.64 59.58 59.10

(0.07) (0.09) (0.05) (0.04) (0.06) (0.13) (0.07) (0.04) (0.04) (0.06) (0.04) (0.07) (0.09)

48.16 47.94 60.92 77.44 40.46 42.39 55.62 74.52 66.76 62.42 73.50 43.16 46.44

(0.1) (0.04) (0.02) (0.05) (0.05) (0.17) (0.04) (0.04) (0.04) (0.02) (0.06) (0.06) (0.04)

50.96 (0.01) 41.74 (0.02) 55.11 (0.06) 32.82 (0.09) 28.46 (0.06) 25.18 (0.01) 28.28 (0.07) 52.57 (0.02) 35.84 (0.01) 36.3 (0.02) 54.52 (0.03) 27.26 (0.03) 37.47 (0.06)

Hw (weft) percent 6.81 turn/m 9.10 turn/m 25 turn/m

14

Fabric

IJCST 21,1

Fabric

Rw (warp) (percent) 6.81 turn/m 9.10 turn/m 25 turn/m

Rw (weft) (percent) 6.81 turn/m 9.10 turn/m 25 turn/m

A B C D E F G H I J K L M

22.89 23 21 30.36 19.11 44.09 12 29.88 18.89 27.10 23 11.98 35.97

28.11 14 12.1 20.39 28.97 36 11.98 19 16.94 19.05 28.95 15.3 40

(0.19) (0.03) (0.05) (0.4) (0.19) (0.08) (0.01) (0.11) (0.15) (0.03) (0.05) (0.05) (0.07)

44.96 45.09 40.08 47.05 47.98 31.04 42.02 46.03 35.04 45.03 42.03 42.96 44.11

(0.03) (0.11) (0.06) (0.04) (0.02) (0.04) (0.03) (0.03) (0.03) (0.02) (0.05) (0.02) (0.03)

44.94 (0.05) 70.98 (0.07) 52.1 (0.09) 67 (0.05) 41.25 (0.05) 71.98 (0.03) 54.10 (0.05) 71.25 (0.03) 64.25 (0.05) 65.05 (0.05) 45.11 (0.03) 71.97 (0.04) 64.10 (0.02)

(0.06) (0.04) (2.01) (0.32) (0.03) (0.04) (0.01) (0.05) (0.04) (0.04) (0.04) (0.03) (0.08)

51.25 32.11 29.25 44.25 59.11 59.21 44.12 25.05 34.36 37.16 30.11 56.31 44.11

(0.08) (0.03) (0.02) (0.05) (0.04) (0.06) (0.10) (0.04) (0.03) (0.03) (0.04) (0.03) (0.04)

54 55.03 44.08 47 49.93 81.11 69.12 47.08 66.98 45 56.95 4915 62.08

(0.03) (0.03) (0.08) (0.01) (0.06) (0.06) (0.16) (0.05) (0.01) (0.02) (0.03) (0.05) (0.05)

Note: The data in brackets are SD values

Table VII, fabric type has significantly influenced on wrinkle force, energy and hysteresis along two warp and weft directions. Tables VIII-XIII compare and classify the wrinkle characteristics along warp and weft directions according to fabric type. It is shown that the differences in wrinkle force of fabric types I, M, E, F, G, B, C, A are statistically significant at the 5 percent level along warp direction. However, the differences in wrinkle force of fabric types L, J, K and also fabric types D, H are statistically in-significant at the 5 percent level along warp direction. In addition, it is represented that the differences in wrinkle force of fabric types A, B, C, D, F, G, H, I, K, L, M are statistically significant at the 5 percent level along weft direction. However, the differences in wrinkle force of fabric types E, J are statistically in-significant at the 5 percent level along weft direction. The results of experiments show that the wrinkle force in warp direction is significantly higher than in weft direction. This is because the fabric samples tested along warp direction exhibited almost a higher formability and bending rigidity than in weft direction. It is found that the fabric samples A and I exhibited the highest and lowest wrinkle force in warp direction, respectively. It is shown that the fabric type A with a twill construction of 2/2 and fiber contents of 55/45 polyester/wool has the highest thickness, weight, shear rigidity, extensibility and formability in warp direction compared with other samples. However, the fabric type I with a fiber contents of 65/35 polyester/viscose has the lowest thickness, weight, and formability in warp direction compared with other samples. It is deduced that this extreme difference in fabric physical and mechanical properties would lead to a significant difference between the wrinkle force of these fabrics. It is indicated that the fabric samples with a higher polyester fiber content exhibited a higher wrinkle force than other samples provided that the fabric weight and thickness to be almost similar. This is attributed to the higher stiffness of polyester fibers compared with wool and viscose fibers. It is also indicated that the fabric type D with a twill construction of 2/1 and fiber contents of 93/7 polyester/wool has the highest wrinkle force in weft direction compared with other samples. This result is attributed to the high value of formability of this fabric in weft direction. Similar to previous findings, the fabric type I with a fiber

Wrinkle behavior of woven fabrics 15

Table VI. The experimental results of the wrinkle tester (wrinkle resilience), Rw

IJCST 21,1

16

Figure 4. Wrinkle generation process for fabric type K in warp direction, with 9.10 turn/m rotational level. (a) u ¼ 08; (b) u ¼ 1508; (c) u ¼ 3008; (d) u ¼ 3508

(a)

(b)

(c)

(d)

Wrinkle Maximum Wrinkle Maximum wrinkle force wrinkle force energy (Ww) energy (Ww) in weft in warp in warp in warp direction direction direction direction Table VII. A summary of ANOVA statistical analysis results for worsted fabric wrinkling properties

Fabric type Rotational level Fabric type * rotational level

Wrinkle hysteresis in warp direction

Wrinkle hysteresis in warp direction

þ þ

þ þ

þ þ

þ þ

þ þ

þ þ

þ

þ

þ

þ

þ

þ

6.7826

15 15 15 15 15 15 15 15 15 15 15 15 15 1.000

1

N

1.000

8.5772

2

0.109

9.5242 9.5654 9.6203

3

1.000

10.0010

4

1.000

10.4117

5

Warp max force (N) 6

1.000

10.6844

Subset

1.000

10.9257

7

0.385

11.6020 11.6510

8

1.000

11.8910

9

12.9433 1.000

10

Notes: Means for groups in homogeneous subsets are displayed. Based on type III sum of squares. The error term is mean square (error) ¼ 0.024; auses harmonic mean sample size ¼ 15.000; ba ¼ 0.05

I M L J K E F G B H D C A Sig.

Duncana,b Type of fabric

Wrinkle behavior of woven fabrics 17

Table VIII. Duncan Homogeneous Subsets test of maximum wrinkle force (Fw) for worsted fabrics in warp direction

Table IX. Duncan Homogeneous Subsets test of maximum wrinkling force (Fw) for worsted fabrics in weft direction

15 15 15 15 15 15 15 15 15 15 15 15 15

N

1.000

6.3700

1

1.000

7.7767

2

1.000

7.6943

3

1.000

7.9389

4

1.000

8.0180

1.000

8.3091

1.000

8.4516

Weft max force (N) Subset 5 6 7

0.054

8.6779 8.7512

8

1.000

9.0605

9

1.000

9.1638

10

18 1.000

9.4294

11

9.5562 1.000

12

Notes: Means for groups in homogeneous subsets are displayed. Based on type III sum of squares. The error term is mean square (error) ¼ 0.011; auses harmonic mean sample size ¼ 15.000; ba ¼ 0.05

I M G B H L K E J C A F D Sig.

Duncana,b Type of fabric

IJCST 21,1

4

5

6

7

8

9

10

11

12

13

Notes: Means for groups in homogeneous subsets are displayed. Based on type III sum of squares. The error term is mean square (error) ¼ 145.650; auses harmonic mean sample size ¼ 15.000; ba ¼ 0.05

3

15 261.9600 15 378.3287 15 412.2480 15 422.6573 15 440.8047 15 453.0120 15 603.7113 15 617.0380 15 645.5727 15 676.5580 15 748.5840 15 764.8433 15 923.4720 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000

2

I G J K L M E D H C B F A Sig.

1

N

W.Warp (N Degree) Subset

Duncana,b Type of fabric

Wrinkle behavior of woven fabrics 19

Table X. Duncan Homogeneous Subsets test of maximum wrinke energy (Ww) for worsted fabrics in warp direction

15 15 15 15 15 15 15 15 15 15 15 15 15

1.000

227.8607

1

1.000

246.0593

2

1.000

281.6307

3

1.000

305.5560

4

1.000

330.4547

0.161

355.0927 355.6273

1.000

428.517

W.Weft (N Degree) Subset 5 6 7

1.000

448.0180

8

1.000

494.8407

9

1.000

502.7667

10

11

1.000

584.7887

20 738.6087 1.000

12

Notes: Means for groups in homogeneous subsets are displayed. Based on type III sum of squares. The error term is mean square (error) ¼ 1.083; auses harmonic mean sample size ¼ 15.000; ba ¼ 0.05

J I G M K L F C H E D B A Sig.

N

Table XI. Duncan Homogeneous Subsets test of maximum wrinkle energy (Ww) for worsted fabrics in weft direction

Duncana,b Type of fabric

IJCST 21,1

48.2200

15 15 15 15 15 15 15 15 15 15 15 15 15 1.000

1

N

1.000

53.5227

2

1.000

53.8693

3

1.000

54.7287

4

1.000

55.7560

1.000

58.1327

0.116

59.5313 59.7047

H.Warp (percent) Subset 5 6 7

1.000

60.7080

8

1.000

61.7680

9

1.000

62.9287

10

1.000

63.7533

11

69.0287 1.000

12

Notes: Means for groups in homogeneous subsets are displayed. Based on type III sum of squares. The error term is mean square (error) ¼ 0.090; auses harmonic mean sample size ¼ 15.000; ba ¼ 0.05

F H J M D B K I E A L G C Sig.

Duncana,b Type of fabric

Wrinkle behavior of woven fabrics 21

Table XII. Duncan Homogeneous Subsets test of wrinkle hysteresis (Hw) on worsted fabrics in warp direction

15 15 15 15 15 15 15 15 15 15 15 15 15

1.000

43.3380

1

1.000

43.8080

2

1.000

46.3307

3

1.000

47.6767

4

1.000

56.4087

1.000

56.7920

1.000

56.9587

H.Weft (percent) Subset 5 6 7

1.000

58.2667

8

1.000

61.7273

9

1.000

63.4467

10

1.000

65.6220

11

22 1.000

68.2247

12

69.3127 1.000

13

Notes: Means for groups in homogeneous subsets are displayed. Based on type III sum of squares. The error term is mean square (error) ¼ 0.004; auses harmonic mean sample size ¼ 15.000; ba ¼ 0.05

L F E M J A G B I D C K H Sig.

N

Table XIII. Duncan Homogeneous Subsets test of wrinkle hysteresis (Hw) on worsted fabrics in weft direction

Duncana,b Type of fabric

IJCST 21,1

Wrinkle behavior of woven fabrics

Maximum Force(N) Warp Direction

20 18 16 14 12

23

6.81 9.1 25(turn/m)

10 8 6 4 2 0 A

B

C

D

E

F

G

H

I

J

K

M

L

Type of Fabric

Figure 5. Comparison of maximum wrinkle force in worsted fabrics (warp direction)

Maximum Force(N) Weft Direction

18 16 14 6.81 9.1 25(turn/m)

12 10 8 6 4 2 0 A

B

C

D

E

F

G

H

I

J

K

L

M

Ww(N.Deg) Warp Direction

Type of Fabric

Figure 6. Comparison of maximum wrinkle force in worsted fabrics (weft direction)

2,500 6.81 9.1 25(turn/m)

2,000 1,500 1,000 500 0 A

B

C

D

E

F

G

H

Type of Fabric

I

J

K

L

M

Figure 7. Comparison of wrinkle energy in worsted fabrics (warp direction)

Figure 9. Comparison of wrinkle hysteresis in worsted fabrics (warp direction)

Figure 10. Comparison of wrinkle hysteresis in worsted fabrics (weft direction)

Ww(N.Deg) Weft Direction

Figure 8. Comparison of wrinkle energy in worsted fabrics (weft direction)

2,000 6.81 9.1 25(turn/m)

1,500 1,000 500 0 A

B

C

D

E

F

G

H

I

J

K

L

M

Type of Fabric

Hw(%) Warp Direction

24

contents of 65/35 polyester/viscose exhibited however the wrinkle force value along weft direction. It is also shown that the fabric samples with a higher polyester fiber content represented a higher wrinkle force than other samples provided that the fabric weight and thickness to be almost similar. It is also shown that the differences in wrinkle energy of all fabric types are statistically significant at the 5 percent level along warp direction (Table VII). Similar to previous findings for wrinkle force, fabric samples A and I exhibited the highest and lowest wrinkle energy in warp direction, respectively. However, fabric samples A and J exhibited the highest and lowest wrinkle energy in weft direction, respectively. The fabric types G and J in which their fiber compositions are viscose, polyester, and wool also exhibited lower wrinkle energy along warp direction. In addition, it is represented that the differences in wrinkle energy of all fabric types are almost statistically

100 80 60 40 20

6.81

9.1

25(turn/m)

0 A

B

C

D

E

F

G

H

I

J

K

L

M

I

J

K

L

M

Type of Fabric

Hw(%) Weft Direction

IJCST 21,1

100 80 60 40 20 6.81

9.1

25(turn/m)

0 A

B

C

D

E

F

G

H

Type of Fabric

significant at the 5 percent level along weft direction. However, the differences in wrinkle energy of fabric types F, and L are statistically in-significant at the 5 percent level along weft direction. This result is attributed to the similar thickness and weight of these fabrics. The results of experiments show that the wrinkle hysteresis is almost higher in warp direction than in weft direction. However, for fabric types D, H, I, J, and K this result is reverse presumably due to the interaction effect of fabric type and torsional angle. In general, the wrinkle hysteresis varies from 48 to 69 percent and from 43 to 69 percent along warp and weft directions, respectively.

Wrinkle behavior of woven fabrics 25

3.3 The effect of rotational level on wrinkle characteristics along warp and weft directions The ANOVA and Duncan statistical analysis results of wrinkle energy, force, and hysteresis in 5 percent confidence limit for different worsted fabrics tested at three different rotational levels are also shown in Tables XIV-XIX. The variation of these wrinkle characteristics against rotational level along two warp and weft directions are also depicted in Figures 11-16. As shown in Table VII, rotational level has significantly influenced on wrinkle characteristics along two warp and weft directions. Tables XIV-XIX compare and classify the wrinkle energy, force and hysteresis along warp and weft directions according to rotational level parameter. It is shown that the differences in wrinkle characteristics of fabric samples tested at rotational levels of 6.81, 9.10, and 25 turn/m are statistically significant at the 5 percent level along warp and weft directions. Similar to previous findings, the results of experiments show that

Duncana,b Rotational level 6.81 turn/m 9.10 turn/m 25 turn/m Sig.

Weft max force (N) N

1

65 65 65

1.6445

Subset 2

3

6.5762 1.000

1.000

16.8519 1.000

Notes: Means for groups in homogeneous subsets are displayed. Based on type III sum of squares. The error term is mean square (error) ¼ 0.011; auses harmonic mean sample size ¼ 65.000; ba ¼ 0.05

Duncana,b Rotational level 6.81 turn/m 9.10 turn/m 25 turn/m Sig.

Table XIV. Duncan Homogeneous Subsets test of rotational level for maximum wrinkle force (Fw) in weft direction

Warp max force (N) N

1

65 65 65

2.5065

Subset 2

3

11.4732 1.000

1.000

16.9849 1.000

Notes: Means for groups in homogeneous subsets are displayed. Based on type III sum of squares. The error term is mean square (error) ¼ 0.024; auses harmonic mean sample size ¼ 65.000; ba ¼ 0.05

Table XV. Duncan Homogeneous Subsets test of rotational level for maximum wrinkle force (Fw) in warp direction

IJCST 21,1

26

the wrinkle force and energy of fabric samples tested in warp direction are significantly higher than in weft direction. However, the wrinkle hysteresis of fabric samples tested along warp and weft directions are almost similar. As shown in Figures 11-14, with increasing the rotational level or torsional angle, the wrinkle force and energy of fabric is increased. This is because more torsional strain imposed into the fabric and hence leads to an increase in torsional stress of the fabric. The trend of this increase depends on the direction of strain along the warp and weft direction. It is shown that with increasing of rotational level, the wrinkle force and energy are nonlinearly increased almost for all fabric samples tested along warp direction. However, this trend is partially linear for fabric samples tested along

Duncana,b Rotational level

Table XVI. Duncan Homogeneous Subsets test of rotational level for maximum wrinkle energy (Ww) in warp direction

6.81 turn/m 9.10 turn/m 25 turn/m Sig.

1

65 65 65

44.2118

3

369.298 1.000

1.000

1,282.6375 1.000

Notes: Means for groups in homogeneous subsets are displayed. Based on type III sum of squares. The error term is mean square (error) ¼ 145.650; auses harmonic mean sample size ¼ 65.000; a ¼ 0.05

6.81 turn/m 9.10 turn/m 25 turn/m Sig.

W.Weft (N Degree) Subset 2

N

1

65 65 65

28.9048

3

200.2842 1.000

1.000

992.3366 1.000

Notes: Means for groups in homogeneous subsets are displayed. Based on type III sum of squares. The error term is mean square (error) ¼ 1.094; auses harmonic mean sample size ¼ 65.000; ba ¼ 0.05

Duncana,b Rotational level Table XVIII. Duncan Homogeneous Subsets test of rotational level for wrinkle hysteresis (Hw) in warp direction

Subset 2

N

b

Duncana,b Rotational level Table XVII. Duncan Homogeneous Subsets test of rotational level for maximum wrinkle energy (Ww) in weft direction

W.Warp (N Degree)

25 turn/m 9.10 turn/m 6.81 turn/m Sig.

H.Warp (percent) N

1

65 65 65

44.1531

Subset 2

3

56.1997 1.000

1.000

75.4131 1.000

Notes: Means for groups in homogeneous subsets are displayed. Based on type III sum of squares. The error term is mean square (error) ¼ 0.090; auses harmonic mean sample size ¼ 65.000; ba ¼ 0.05

weft direction. The results also indicates that at rotational level of 25 turn/m value, the wrinkle force of all fabric samples are almost similar particularly for fabric samples tested along weft direction. On the other hand, with increasing the rotational level, the wrinkle hysteresis is significantly decreased both along warp and weft directions. It is implicated that at rotational level of 25 turn/m, the fabric samples tend to be purely twisted while in other torsional levels the fabric samples are simultaneously compressed and twisted. In addition, as shown in Figure 3(c), the behavior of fabric in wrinkle recovery for rotational level of 25 turn/m is similar to wrinkling process which in turn leads to a lower energy loss and hence a lower wrinkle hysteresis during wrinkling process.

Wrinkle behavior of woven fabrics 27

H.Weft (percent)

Duncana,b Rotational level 25 turn/m 9.10 turn/m 6.81 turn/m Sig.

Subset 2

N

1

65 65 65

38.9654

3

56.9051 1.000

74.4171 1.000

1.000

Notes: Means for groups in homogeneous subsets are displayed. Based on type III sum of squares. The error term is mean square (error) ¼ 0.004; auses harmonic mean sample size ¼ 65.000; ba ¼ 0.05

Table XIX. Duncan Homogeneous Subsets test of rotational level for wrinkle hysteresis (Hw) in weft direction

20 Maximum Force(N) Warp Direction

18 16 14 12 10 8 6 4 2 0 6.81

9.1

25

Rotational Level (turn/m) A

B

C

D

J

K

L

M

E

F

G

H

I

Figure 11. Comparison of maximum wrinkle force of worsted fabrics along warp direction at different rotational levels

28

18 Maximum Force(N) Weft Direction

IJCST 21,1

16 14 12 10 8 6 4 2 0

Figure 12. Comparison of maximum wrinkle force of worsted fabrics along weft direction at different rotational levels

6.81

9.1

25

Rotational Level (turn/m) A

B

C

D

E

J

K

L

M

F

G

H

I

Figure 14. Comparison of maximum wrinkle energy of worsted fabrics along weft direction at different rotational levels

2,000 1,500 1,000 500 0 6.81

9.1

25

Rotational Level (turn/m) A

Ww (N.Degree) Weft Direction

Figure 13. Comparison of maximum wrinkle energy of worsted fabrics along warp direction at different rotational levels

Ww(N.Degree) Warp Direction

2,500

B

C

D

E

F

G

H

I

J

K

L

M

2,000 1,500 1,000 500 0 6.81

9.1

25

Rotational Level (turn/m) A

B

C

D

E

F

G

H

I

J

K

L

M

Hw(%) Warp Direction

4. Conclusion The aim of this paper was to study woven fabric wrinkle behavior at different rotational levels. An apparatus was designed and constructed in order to investigate the torsional and wrinkle behavior of worsted fabrics in a cylindrical model under different rotational levels using data acquisition and micro-controller systems. Thirteen different worsted fabric samples with different wool and polyester fiber compositions and fabric designs were prepared and then torsional force was continuously measured along warp and weft directions while constant rotational levels (6.81, 9.10, and 25 turn/m) imposed on the specimen. The results showed that the wrinkle force, energy and hysteresis of worsted fabrics is significantly influenced by fabric physical and mechanical properties and imposed rotational levels. It is shown that with increase of rotational level, the wrinkle force, and energy is increased along weft and warp directions. However, with increasing the rotational levels, the wrinkle hysteresis is significantly decreased both along warp and weft directions. The results indicates that at rotational levels of 25 turn/m value, the wrinkle force of all fabric samples are almost similar particularly for fabric samples tested along weft direction. The results of this research revealed that worsted fabric samples with higher polyester fiber content exhibited higher wrinkle force values.

29

100 80 60 40 20 0 6.81

9.1

25

Rotational Level (turn/m)

Hw(%) Weft Direction

Wrinkle behavior of woven fabrics

A

B

C

K

L

M

D

E

F

G

H

I

J

Figure 15. Comparison of maximum wrinkle hysteresis of worsted fabrics along warp direction at different rotational levels

100 80 60 40 20 0 6.81

9.1

25

Rotational Level (turn/m) A

B

C

K

L

M

D

E

F

G

H

I

J

Figure 16. Comparison of maximum wrinkle hysteresis of worsted fabrics along weft direction at different rotational levels

IJCST 21,1

The results of this research also suggested that the wrinkle force and energy of worsted fabrics in warp direction is significantly higher than in weft direction. Further experimental works are needed to investigate and measure the wrinkle force of different textile and technical structures. In addition, the cyclic wrinkle behavior of textile structures at higher torsional strain rates are warrant further studies.

30

References AATCC Test Method 128-1989 (1994), AATCC Technical Manual, American Association of Textile Chemists and Colorists, Triangle Park, NC, Vol. 69, pp. 217-18. Abghari, R., Shaikhzadeh Najar, S., Haghpanahi, M. and Latifi, M. (2004), “Contributions of in-plane fabric tensile properties in woven fabric bagging behavior using a new developed test method”, International Journal of Clothing Science and Technology, Vol. 16, pp. 418-33. Amirbayat, J. and Alagha, M.J. (1996), “Objective assessment of wrinkle recovery by means of laser triangulation”, Journal of the Textile Institute, Vol. 2, pp. 349-55. Amirbayat, J. and Hearle, J.W.S. (1986), “The complex buckling of flexible sheet materials. Part I: theoretical approach”, International Journal of Mechanical Sciences, Vol. 28, pp. 339-58. Basset, R.J., Postle, R. and Pan, N. (1999), “Experimental methods for measuring fabric mechanical properties: a review and analysis”, Textile Research Journal, Vol. 69, pp. 866-75. Hearle, J.W.S., Grosberg, P. and Backer, S. (Eds) (1969), Structural Mechanics of Fibers, Yarns, and Fabrics, Vol. 1, Wiley-Interscience, New York, NY. Postle, R., Carnaby, G.A. and de-Jong, S. (1988), The Mechanics of Wool Structures, Chapter 14, Ellis Horwood, Chichester, pp. 369-86. Shinohara, A., Ni, Q. and Takatera, M. (1991a), “Geometry and mechanics of the buckling wrinkle in fabrics. Part 1: characteristics of the buckling wrinkle”, Textile Research Journal, Vol. 91, pp. 94-9. Shinohara, A., Ni, Q. and Takatera, M. (1991b), “Geometry and mechanics of the buckling wrinkle in fabrics. Part 2: buckling model of a woven fabric cylinder in axial compression”, Textile Research Journal, Vol. 91, pp. 100-5. Skelton, J. and Freeston, W.D. (1971), “The shear behavior of fabric under biaxial loads”, Textile Research Journal, Vol. 41, pp. 871-9. Timoshenko, S.P. and Gere, J.M. (1961), Theory of Elastic Stability, McGraw-Hill, New York, NY, p. 457. Corresponding author S. Shaikhzadeh Najar can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

The single-row machine layout problem in apparel manufacturing by hierarchical order-based genetic algorithm

Single-row machine layout problem

Miao-Tzu Lin

Received 2 February 2008 Revised 28 May 2008 Accepted 28 May 2008

Department of Fashion Design and Management, Tainan University of Technology, Taiwan

31

Abstract Purpose – The purpose of this paper is to address the topic of minimizing the moving distance among cutting pieces during apparel manufacturing. Change machine layout is often required for small quantity and diversified orders in the apparel manufacturing industry. The paper seeks to describe a hierarchical order-based genetic algorithm to quickly identify an optimal layout that effectively shortens the distance among cutting pieces, thereby reducing production costs. Design/methodology/approach – The chromosomes of the hierarchical order-based genetic algorithm consist of the control genes and the modular genes to acquire the parametric genes, a precedence matrix and a from-to matrix to calculate the distance among cutting pieces. Findings – The paper used men’s shirt manufacturing as an example for testing the results of a U-shaped single-row machine layout to quickly determine an optimal layout and improve effectiveness by approximately 21.4 per cent. Research limitations/implications – The manufacturing order is known. The machine layout is in a linear single-row flow path. The machine layout of the sewing department is independently planned. Originality/value – The advantage of the hierarchical order-based genetic algorithm proposed is that it is able to make random and global searches to determine the optimal solution for multiple sites simultaneously and also to increase algorithm efficiency and shorten the distance among cutting pieces effectively, according to manufacturing order and limited conditions. Keywords Garment industry, Textile machinery and accessories, Programming and algorithm theory, Process planning Paper type Research paper

Introduction Good machine layout and shorten moving distance among materials are important for reducing production costs. Tompkins et al. (1996) indicated that moving non-value added material often takes up 20-50 per cent of the total manufacturing costs, and an efficient layout can save 10-30 per cent of the total manufacturing costs implying that an optimal layout can improve manufacturing schedules and therefore efficiency. Apparel manufacturing involves small quantities of diversified items that often require changes in machine layouts according to different materials, specifications, and manufacturing processes and methods. If the machine layout is able to be re-arranged quickly, then the change time, labor required, and moving distance can be reduced.

International Journal of Clothing Science and Technology Vol. 21 No. 1, 2009 pp. 31-43 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910923737

IJCST 21,1

32

To solve problems involving longer search times and difficulties in operations resulting from manufacturing large quantities of product series, this study adopted a hierarchical order-based genetic algorithm to modularize sub-assembly lines and reduce the order of the search. The chromosomes of the hierarchical order-based genetic algorithm include the control genes and the modular genes that are used to acquire the manufacturing order of the parametric genes; later a from-to matrix and precedence matrix were adopted to calculate the moving distance of cutting pieces. Therefore, the proposed method has the advantages of hierarchical structure and modularization and real-time random and global searches for an optimal solution that also improves algorithm effectiveness. The paper used a men’s shirt manufacturing as an example to quickly determine the optimal machine layout, shorten moving distance among cutting pieces, and improve production efficiency. Literature review A machine layout with a linear single-row flow path exists in different configuration, such as a straight line, U-shaped, serpentine line, and loop. Although modern material handing systems often allow for complex flow path configurations, linear single row flow paths are still popular in industry. Due to its ease of construction and control, the linear single-row machine layout is the most commonly used layout (Ho and Moodie, 1998). The machine layout at the plants is one of the issues of this equipment layout study. Optimal algorithms are used to acquire the optimal equipment layout but more time consuming. Another application is the sub-optimal heuristic algorithm that was categorized by Kusiak and Heragu (1987) into four types: construction algorithms, improvement algorithms, hybrid algorithms, and graph theoretic algorithms. In recent years, due to the development of improved computer algorithms, many researchers have proposed meta-heuristic algorithms such as simulated annealing, Tabu search, and genetic algorithms similar to optimal algorithms to determine an acceptable optimal solution within a reasonable time. The optimal algorithms are time consuming while the sub-optimal heuristic algorithm takes solution quality into consideration within the limited time and algorithm’s ability. The linear single-row machine layout problem is identified as a NP-complete problem (Suresh and Sahu, 1993). Accurate mathematical solutions do not exist for such problems. The complexity of such problems increases exponentially with the number of devices. For instance, a system consisting of N machines will comprise a solution space with the size N. For arranging devices in the systems, the number of possible solutions is equal to the number of permutation of N elements (Ficko et al., 2004). Hollier (1963) presented four flow-line analysis (FLA) methods for four single-machine-type problems where only one machine of any type is allowed in a flow line. Bragalia (1996) proposed a combination of simulated annealing and genetic algorithms to minimize total backtracking in the linear ordering machines. It is assumed that the machine locations and facilities are equally space. A two-phase layout procedure combining flow-line analysis and simulated annealing is suggested by Ho and Moodie (1998). Phase one modified Hollier’s four FLA methods (methods 1, 2, 3 and 4) and presented two new FLA methods (methods 5 and 6). Method 5 is suitable for bidirectional flow lines and method 6 is suitable for unidirectional

flow lines. Ponnambalam and Ramkumar (2001) proposed the two best methods, FLA method 5 and FLA method 6, combining an order-based genetic algorithm to reduce the material handling cost by efficient layout design. Ficko et al. (2004) presented a model of the flexible manufacturing systems (FMS) in single or multiple rows with the order-based genetic algorithm. In studies of machine layouts, the selection of the objective function has been focused on minimal moving time, minimal total moving distance, minimal moving distance among parts, minimal moving costs of materials, and minimal return times (Heragu and Kusiak, 1988; Kumar et al., 1995; Ho and Moodie, 1998; and Sarker et al., 1998). The application of the genetic algorithm can convert an objective function into a fitness function to satisfy the survival of the fittest principle.

Single-row machine layout problem 33

Mathematical model From-to matrix The commonly used linear single-row machine layout in the manufacturing industry is shown in Figure 1. The distance among stations as the from-to matrix is shown in equation 1: 2

T I £I ¼ tij I £I

t11

6 . 6 . 6 . 6 6 t ¼6 6 i1 6 . 6 . 6 . 4 tI 1

···

t1j

···

..

.

.. .

..

···

t ij

···

..

.

.. .

..

···

t Ij

···

t 1I

3

.. 7 7 . 7 7 7 tiI 7: 7 .. 7 7 . 7 5 tII

.

.

ð1Þ

I is the number of stations; tij is the distance from station i to station j (unit: meter). 2

1

.

3

.

i 1

2

3

.

.

Conveyor .

I

.

j

.

.

. j

3

1

.

j

2

.

.

. 3

AG/

.

i

I

1

I . (c)

.

.

I

.

.

2

j

(b) .

i

.

AG/

(a)

.

.

i

. (d)

Figure 1. The linear single-row machine layout with respect to different types: (a) loop layout; (b) straight-line layout; (c) U-shaped layout; (d) serpentine layout

IJCST 21,1

34

Precedence diagram and matrix The precedence diagram with circles and arrows to indicate the relationship of the stations is shown in Figure 2. The relationship among stations as the precedence matrix is shown in equation 2: 3 2 p11 · · · p1j · · · p1I 7 6 6 .. . . .. 7 .. . . 7 6 . . . . . 7 6 7 6 6 pi1 · · · pij · · · piI 7 P I £I ¼ pij I £I ¼ 6 7 7 6 7 6 . . . . . . . .. 7 . . .. 6 .. ð2Þ 7 6 5 4 pI 1 · · · pIj · · · pII ( pij : pij ¼

1

from station i to station j

0

other

The moving distance algorithm among cutting pieces Based on the from-to matrix and the precedence matrix, we are able to acquire the total distance matrix shown as equation 3 and the total distance D shown as equation 4: M I £I ¼ ½mij I £I ¼ ½t ij £ pij I £I

D¼

I X I X

ð3Þ

ð4Þ

mij :

i¼1 j¼1

Hierarchical order-based genetic algorithm Chromosome As shown in the left part of the precedence diagram of Figure 3, according to the manufacturing order of the sub-assembly line, adjacent machine layout has a minimal

i

j

2

3

I 1

Figure 2. Precedence diagram

1

3

2

4

Single-row machine layout problem

1 2

5

35 6

3

7

4

8

5

Precedence diagram

Modular diagram

Figure 3. Precedence relationship of modular order

distance that modularizes the sub-assembly line in the new sequence and order as shown in the right part of the modular diagram of Figure 3. The hierarchical order-based genetic algorithm chromosomes are consisted of the control genes and the modular genes that are used to acquire the parametric genes as shown in Figure 4. 0 1 1 1 0 3 1 4 5 2

Chromosome A

Control genes

Modular genes

Control genes

0

1

1

1

0

Modular genes

3

1

4

5

2

Parametric genes

6

7

8

2

1

3

4

5

XA = (6,2,1,7,8,3,4,5)

1 0 1 1 1 3 1 4 5 2

Chromosome B

Control genes

Modular genes

Control genes

1

0

1

1

1

Modular genes

3

1

4

5

2

Parametric genes

6

7

8

1

2

5

4

XB = (6,1,2,7,8,5,4,3)

3

Figure 4. Chromosomes of the hierarchical order-based genetic algorithm

IJCST 21,1

36

Initial population A population pool of chromosomes can be randomly set initially. A conventional binary genetic algorithm was applied to the control genes. The modular genes adopted an order-based genetic algorithm to identify an order-based list such as one, two, three, . . . , I. The positions of the chromosomes were selected in accordance with random number from the order-based list and then deleted it from the list. The procedure was repeated until the order-based list was emptied. Fitness function This study convert the total distance into a fitness function by employing a positive number times the reciprocal of the objective function as shown in equation 5; the one with the larger fitness function is considered optimal: FitðDÞ ¼ 1000 £

1 : D

ð5Þ

Roulette wheel selection Roulette wheel selection is one of the most common techniques being used for such a proportionate selection mechanism. We put the fitness function in order and divided the value of the individual function by the total function value. The ratio acquired was illustrated as a roulette wheel and the fittest chromosome and occupies the largest interval. The numbering of the larger interval was more likely to be selected (Man et al., 2000). Crossover Not all chromosomes were put into crossover, resulting in the definition of Pc, probability of crossover, with a typical value between 0.6 , 1 (Man et al., 2000). This generates offspring from the parents, based on a randomly generated crossover mask. The control genes adopted the uniform crossover. The operation is demonstrated in the left part of Figure 5. The modular genes adopted the uniform order-based crossover. The operation is demonstrated in the right part of Figure 5. Mutation The definition of Pm, probability of mutation, at a typical value less than 0.1, is used to escape from a local optimum (Man et al., 2000). The control genes adopted the conventional binary genetic algorithm mutation. The modular genes used a scramble sub-list mutation as shown in Figure 6. We randomly selected two bits from the bit string and the chromosomes in between were randomly listed (Chan et al., 1998). Termination condition Normally, the termination condition is shown in three aspects. When more operating times are required than evolution times, the algorithm of the generation is set to terminate. The termination also initiates when the objective function reaches its target value. When the group is shown with homogeneity, convergence is presented and it is close to local optimal solution. This study adopted the generation termination algorithm.

0

1 2 3 4 5

0 1 1 1

3 5 1 4 2

0 1 1 0 1

0 1 1 1 0

Parent 1

0 1 1 1

Parent 2

1

Mask

Single-row machine layout problem

3 5 1 4 2 0 1 1 1

2 3 4

0

37

1 5

Crossover operations 1

3

0 1 1 1

2

5 1 4 1 2 3 4 5

Offspring 1

0 0 1 1 1

5 2 3 4 1

Offspring 2

1 1 1 1

3 1 4 5 2

0

Uniform crossover

1 2

3

4

Uniform order- based crossover

5

Figure 5. Crossover of the hierarchical order-based genetic algorithm

6 7 8 Original Chromosome

Choose two random positions 1 2

3

4

5

6 7 8 New Chromosome

Permute sublist randomly Remark: with the beginning and the end of the selected sublist marked by "|" Source: Chan et al. (1998)

Flow chart Flow chart of the hierarchical order-based genetic algorithm is shown in Figure 7. Result and discussion To test the effectiveness of this study, we used the same men’s shirt manufacturing in the genetic algorithm study of Chan et al. (1998). Figure 8 shows the precedence relationship diagram of the example. The chromosomes and their structure are shown in Figure 9. The task of each station is tabulated in Table I. The condition of the example is shown as follows: . Population size: 50. . Probability of crossover (Pc): 0.6.

Figure 6. Scramble sub-list mutation

IJCST 21,1

START

Create initial random population

38

Parametric genes evaluate from chromosomes

Apply fitness

Termination criterion satisfied?

Y

END

N

Roulette wheel parent selection and reproduction

Crossover (Pc)

Figure 7. Flow chart of the hierarchical order-based genetic algorithm

Mutation (Pm)

. .

Probability of mutation (Pm): 0.008. Termination condition: Termination until generation 200.

According to the manufacturing order, the machine layout is listed as (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41). However, before applying the proposed algorithm and put order in the equation, we acquired the total distance among cutting pieces, D1 ¼ 70 and fitness value, Fit (D1) ¼ 14.29. By applying the order-based genetic algorithm until generation 200, we acquired the machine layout as (1, 41, 40,39, 3, 4, 2, 5, 7, 8, 9, 10, 25, 26, 15, 16, 17, 21, 22, 23, 20, 19, 18, 24, 32, 11, 12, 13, 14, 27, 28, 29, 31, 6, 36, 38, 37, 35, 34, 33, 30), resulting in a total distance among cutting pieces, D2 ¼ 68 and fitness value, Fit (D2) ¼ 14.7. After the application of the hierarchical order-based genetic algorithm, the chromosomes of the control genes in Figure 10 are shown as (0, 1, 0, 1, 1, 0, 0, 0, 1, 0, 0, 1, 1); the modular genes as (1, 3, 2, 7, 4, 9, 5, 6, 8, 10, 12, 13, 11). From the control genes and the modular genes we acquired the improved order of machine layout as (1, 2, 3, 4, 5, 6, 12, 11, 10, 9, 8, 7, 23, 22, 21, 17, 16, 15, 14, 13, 25, 26, 27, 28, 29, 18, 19, 20, 24, 30, 31,

1

13

18

21

25

34

2

14

19

22

26

35

3

15

23

27

36

Single-row machine layout problem

5 7 4

9

1 4

16

28

37

5

17

29

38

6

7

11

39 2

39

20

8

6

24

8

9 30

10

3

31

11

10 32

12

33 12

40

Figure 8. The precedence relationship diagram of men’s shirts manufacturing

13

41 Precedence diagram

Modular diagram

Source: Chan et al. (1998)

Chromosome

0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 Control genes

Modular genes

Control genes

0

0

0

0

0

0

0

0

0

0

0

0 0

Modular genes

1

2

3

4

5

6

7

8

9

10

11

12 13

Parametric genes 1 2 3 4 5 6 7 8 9 10 1112 1314 1516171819 202122 23 24 25 26 27 28 29 30 313233 34 35363738 39 40 41

32, 33, 40, 41, 39, 38, 37, 36, 35, 34), resulting in a total distance among cutting pieces, D3 ¼ 55, and fitness value, Fit (D3) ¼ 18.18. The total distance and fitness value of the generation algorithm acquired by the two methods are shown as in Figure 11 and the assessment of improved effectiveness is shown as in Table II. Discussion Comparison of the hierarchical order-based genetic algorithm and the order-based genetic algorithm When comparing two algorithm methods, we found the convergence of the order-based algorithm in generation 178 and the hierarchical order-based genetic algorithm in

Figure 9. Chromosomes of the hierarchical order-based genetic algorithm in men’s shirts manufacturing

IJCST 21,1

40

Table I. Tasks of each work station in men’s shirts manufacturing

Task no.

Task name

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Spot fuse collar fall Top fuse collar fall Sew collar stay pocket Runstitch collar fall Trim, turn and press collar fall Topstitch collar fall Hem collar band Attach collar band Turn and press collar band Topstitch collar band Sew collar band buttonhole Sew collar band button Set centre front placket Hem right front edge Trim neckline Sew centre front buttonhole Sew right front button Hem pocket mouth Crease pocket Set pocket Sew yoke pleats

Task name

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

Set yoke label Set yoke Join shoulder Set sleeve under placket Set sleeve top placket Finish sleeve placket Sew sleeve placket buttonhole Sew sleeve placket button Set sleeve Topstitch armhole Join side seam Hem bottom Hem cuff Runstitch cuff Turn and press cuff Topstitch cuff Sew cuff buttonhole Sew cuff button Set cuff Set and close collar

Source: Chan et al. (1998)

Chromosome

0 1 0 1 1 0 0 0 1 0 0 1 1 1 3 2 7 4 9 5 6 8 10121311 Control genes Modular genes

Control genes

0

1

0

1

1

0

0

0 1

0

0 1

1

Modular genes

1

3

2

7

4

9

5

6 8

10

1213

11

Parametric genes

35

36

1 2 3 4 5 6 121110 9 8 7 2322211716151413252627282918192024303132334041393837363534

37

38

39

41

40

33

32

31

30

24

20

19

18

29

28

27

26

1

34

Figure 10. U-shaped improved machine layout

Task no.

25 1

2

3

4

5

6

12

11

10

9

8

7

23

22

21

17

16

15

14

13

generation 39 indicating the shorter moving distance, better fitness value, quicker convergence, and the optimal solution acquired by the hierarchical order-based genetic algorithm. Advantage of the hierarchical order-based genetic algorithm Advantage of the modular genes: after modularizing the sub-assembly line, the number of groups in search is reduced. From the first to the sixth manufacturing steps in Figure 3 as the example, there are 6! (¼ 720) types of layouts. After modularization as

Single-row machine layout problem

Hierarchical order-based genetic algorithm fitness value Order-based genetic algorithm fitness value Order-based genetic algorithm moving distance value

1

11 21 31 41 51 61 71 81 91 101 111 121 131 141 151 161 171 181 191

250 240 230 220 210 200 190 180 170 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0

41

Moving distance value

Fitness value

Hierarchical order-based genetic algorithm moving distance value 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

Generation

U-shaped machine layout

Total distance D (meter)

1. Machine layout according to manufacturing order 2. Order-based genetic algorithm

D1 ¼ 70

3. Hierarchical order-based genetic algorithm

D3 ¼ 55

D2 ¼ 68

Assessment of improve effectiveness

Figure 11. The fitness value and the moving distance value of each generation

Remarks (fitness value) Fit (D1) ¼ 14.29

Improved by 2.9 per cent than D1 Improved by 21.4 per cent than D1; Improved by 19.1 per cent than D2

Fit (D2) ¼ 14.7 Fit (D3) ¼ 18.18

shown in Figure 12(b), there are only 3! (¼ 6) types of layouts that effectively reduce quantities the search. Advantage of the control genes: as shown in Figure 12 (c), when “control genes” is 0, it has the increase order (1-2) and when “control genes” is 1, it has the decrease order (5-4-3). Then these two sub-assembly lines can be moved to the main line to shorten the distance.

Table II. The assessment of improved effectiveness of U-shaped machine layout

IJCST 21,1

D = 13 1

4

3

6

5

1

2

4

3

6

5

2

(a) Order-based genetic algorithm

42

D=7 1

2

3

4

5

1

1 D=6 1

2

3

6

4

2

3

6 2

3

1

4

5

3

6 2

5 1

2

3

6

4

5

(b) Modular order 0

0

1

1

3

2

D=5

Figure 12. Machine layout order analysis

1

2

3

6

5

(c) Hierarchical order-based genetic algorithm

4

1

2

3

6

5

4

Conclusion This study addressed the topic of minimizing the moving distance among cutting pieces during apparel manufacturing and made the following conclusions. Based on the layout composition theory, modularization decreases the number of layout compositions and provides a corresponding increased or decreased order for the control genes to enable these two sub-assembly lines to change direction to the main line in order to shorten the moving distance among cutting pieces. From the examples of the U-shaped machine layouts, the proposed hierarchical order-based genetic algorithm has proven to increase effectiveness by 21.4 per cent thereby matching the limits of manufacturing order possibilities and improving algorithm effectiveness. References Bragalia, M. (1996), “Optimization of a simulated-annealing-based heuristic for single row machine layout problem by genetic algorithm”, International Transactions in Operation Research, Vol. 3 No. 1, pp. 37-49. Chan, K.C.C., Hui, P.C.L., Yeung, K.W. and Ng, F.S.F. (1998), “Handling the assembly line balancing problem in the clothing industry using a genetic algorithm”, International Journal of Clothing Science and Technology, Vol. 10 No. 1, pp. 21-8.

Ficko, M., Brezocnik, M. and Balic, J. (2004), “Designing the layout of single- and multiple-rows flexible manufacturing system by genetic algorithm”, Journal of Materials Processing Technology, Vol. 157-158, pp. 150-8. Heragu, S.S. and Kusiak, A. (1988), “Machine layout problem in flexible manufacturing systems”, Operations Research, Vol. 36 No. 2, pp. 258-68. Ho, Y.C. and Moodie, C.L. (1998), “Machine layout with a linear single-row flow path in an automated manufacturing system”, Journal of Manufacturing Systems, Vol. 17 No. 1, pp. 1-22. Hollier, R.H. (1963), “The layout of multi-product lines”, International Journal of Production Research, Vol. 2, pp. 47-57. Kumar, K.R., Hadjinicola, G.C. and Lin, T.L. (1995), “A heuristic procedure for the single-row facility layout problem”, European Journal of Operational Research, Vol. 87 No. 1, pp. 65-73. Kusiak, A. and Heragu, S.S. (1987), “The facility layout problem”, European Journal of Operational Research, Vol. 29, pp. 229-51. Man, K.F., Tang, K.S. and Kwong, S. (2000), Genetic Algorithms, Springer, Hong Kong. Ponnambalam, S.G. and Ramkumar, V. (2001), “A genetic algorithm for the design of a single-row layout in automated manufacturing systems”, International Journal of Advanced Manufacturing Technology, Vol. 18 No. 7, pp. 512-19. Sarker, B.M., Wilhelm, W.E. and Hogg, G.L. (1998), “Locating sets of identical machines in a linear layout”, Annals of Operations Research, Vol. 77, pp. 183-207. Suresh, G. and Sahu, S. (1993), “Multiobjective facility layout using simulated annealing”, International Journal of Production Economics, Vol. 32, pp. 239-54. Tompkins, J.A., White, J.A., Bozer, Y.A., Frazelle, E.H., Tanchoco, J.M.A. and Trevino, J. (1996), Facilities Planning, 2nd ed., Wiley, New York, NY. Corresponding author Miao-Tzu Lin can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

Single-row machine layout problem 43

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

IJCST 21,1

Effect of laundering on fabric drape, bending and shear

44

Department of Fashion and Apparel Studies, University of Delaware, Newark, Delaware, USA

Belinda T. Orzada Mary Ann Moore and Billie J. Collier

Received 19 July 2007 Revised 1 March 2008 Accepted 1 March 2008

Florida State University, Tallahassee, Florida, USA, and

Jonathan Yan Chen Louisiana State University, Baton Rouge, Louisiana, USA Abstract Purpose – The purpose of this paper is to investigate the effect of laundering on the drape, shear, and bending properties of bottom weight fabrics. Design/methodology/approach – Six bottom-weight 100 percent cotton fabrics were included. Collier’s Drape Tester was utilized to obtain drape values. Bending and shear values were measured on the KES-F Shear Tester and the Pure Bending Tester. Three laundering cycles (unlaundered, one and five home launderings) following AATCC methods were explored. Findings – Laundry cycle did not have a significant effect on fabric drape, shear or bending properties. However, drape values increased overall, while shear and bending modulus and hysteresis decreased, resulting in a more drapable, pliable fabric after five laundry cycles. Research limitations/implications – Future research examining a wider variety of fabrics and conducting a greater number of laundry cycles to approximate an average yearly number of laundry cycles is recommended. An expansion of this preliminary study should give more conclusive evidence of the trends observed. Originality/value – Objective measurement of drape and fabric mechanical properties related to drape after laundry treatments would assist the apparel manufacturer in developing laundry recommendations based on the fabric’s performance and in selecting fabrics which maintain their drape characteristics, mechanical properties, and dimensional stability with use. Higher quality garments with increased consumer satisfaction would result. Keywords Fabric testing, Cotton Paper type Research paper

International Journal of Clothing Science and Technology Vol. 21 No. 1, 2009 pp. 44-55 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910923746

Introduction Defined as the arrangement of fabric in graceful folds as a result of gravity (Mehta, 1985), drape is an important component of the aesthetic appeal of fabrics for both apparel and home furnishings. Assessing and analyzing the complex behavior of draped fabrics has been a concern of researchers for a number of years. Much of the initial work on fabric drape has focused on development of direct quantitative evaluation methods for drape and the correlation of these drape values with measured fabric properties thought to influence drape (Chu et al., 1950; Collier, 1991; Cusick, 1965, 1968; Morooka and Niwa, 1976; Sudnik, 1972). More recently, researchers have sought to predict drape in various configurations using numerical modeling techniques and measured fabric mechanical properties (Collier et al., 1991; Postle and Postle, 1998; Wu et al., 2003).

Although the prediction of fabric drape has important implications in apparel production and in garment use, one aspect of drape, changes in drape over time due to laundering, has been given little focus. A review of the literature revealed little research utilizing objective means of evaluation to examine effects of laundering on fabric mechanical properties, or aesthetic ones such as drape. Morris and Prato (1981) concluded that laboratory tests that include laundering are better indicators of performance properties than laboratory tests without laundering. Recognizing the need for research in this area, the main purpose of this exploratory study was to investigate the effect of laundering on fabric drape. Given this purpose, the specific objectives of this study were to: (1) determine the effect of repeated launderings on drape values of bottom weight fabrics; (2) determine the effect of repeated launderings on the fabric mechanical properties of shear stiffness, shear hysteresis at 0.258, shear hysteresis at 2.58, bending modulus and bending hysteresis on bottom-weight fabrics; and (3) to identify any significant correlations between these variables. Objective evaluation of fabric drape Understanding of drape has progressed gradually from the initial work by Peirce (1930). Since his study of properties affecting fabric hand, much progress has been made in the development of instruments and quantitative evaluation methods of fabric drape and hand. In the early 1960s, researchers evaluating low stress mechanical properties of apparel fabrics related bending, shear, buckling, and compression to garment tailorability, formability and drape (Shishoo, 1995). To evaluate drape quantitatively, an instrument capable of distorting a fabric sample in all three dimensions was developed by Chu et al. (1950). This drapemeter used an optical system to determine the projected area of a draped fabric specimen. In the test method developed for quantitative drape measurement (Chu et al., 1950), a 10-in. circular fabric specimen was placed between two smaller discs and was raised until the specimen edge did not touch the tester base. The image of the draped fabric area was then recorded using an integrator mechanism. A drape coefficient value (DC) was determined by dividing the area of the draped sample by the area of its annular ring (Chu et al., 1950). Cusick (1965) made further design improvements in the drapemeter, developing a vertical projection method to trace the outline of the draped fabric on paper. The drape coefficient was calculated using a planimeter to measure the shadow area. This instrument was found to be sensitive to specimen size for both high and low DC values (Cusick, 1968; Sudnik, 1972). Cusick (1968) found that a larger specimen size, which provided more fabric area to drape over the supporting disk, was needed to distinguish differences in the draping behavior of stiffer fabrics. Cusick (1968) also developed an easier method of calculating DC, which involved weighing paper tracings of draped and undraped images instead of determining areas mathematically. Collier et al. (1988) developed a Drape Tester that utilizes a bottom surface of photovoltaic cells to determine the amount of light blocked by a fabric specimen draped on a pedestal. A digital display gives the amount of light being absorbed by the photovoltaic cells, which is related to the amount of drape of the fabric specimen.

Effect of laundering

45

IJCST 21,1

46

When the fabric specimen is draped over the plate, the amount of light blocked is determined by the draping conformation of the specimen and normalized on a scale of 1-100 percent (Collier, 1991). Collier found a high degree of fabric drape resulted in high voltage readings while those fabrics with low drapability had low readings. Collier et al. (1988) drape tester was further validated in a study of the effects of interfacing type on shear and drape behavior (Collier et al., 1989), and in a study examining the relationship between fabric drape, fabric mechanical properties and subjective drape evaluation (Collier, 1991). Previous work using this Drape Tester indicated that drape values obtained on this instrument correlated well with fabric weight and shear hysteresis (Collier et al., 1988). Fabric mechanical properties related to drape The amount of drape a fabric undergoes has been correlated with fabric bending and shearing properties as well as with weight and thickness. Cusick (1965) found that both bending stiffness and shear stiffness influenced drape. He determined shear stiffness by measuring the shear angle at which a fabric began to buckle; fabrics that are resistant to shear deformation will buckle at a lower shear angle than those with lower shear stiffness. Sudnik (1978) confirmed the importance of shear stiffness in predicting fabric drape. Fabric bending is a major mode of deformation in draping (Hearle, 1969). For a fabric to form graceful folds, a low resistance to bending is essential (Collier, 1990). The series of instruments developed by Kawabata (1980) for the objective evaluation of fabric hand have been used to determine fabric mechanical properties related to drape (Morooka and Niwa, 1976). The Kawabata Evaluation System for Fabrics (KES-F) measures fabric mechanical properties under low loads, providing useful information for the assessment of fabric hand in the apparel manufacturing process and in the development of new fabrics. The KES-F system consists of four instruments developed to measure bending, shear, tensile, surface, and compressional properties (Kawabata, 1980). These instruments have been used to objectively measure fabric properties related to drape. Two instruments from the KES-F system have proven most useful in determining mechanical properties important to drape, the Kawabata Tensile and Shear Tester and the Pure Bending Tester. Early work using the Kawabata instruments to predict drape found bending modulus and fabric weight to be the most important factors (Morooka and Niwa, 1976). Collier et al. (1988), however, research employing the Kawabata shearing and bending instruments found that shearing properties were more significant predictors of fabric drape than were bending properties. Later, Collier (1991) used multiple regression analysis to determine shear hysteresis at 5.08 and bending modulus to be the most important drape predictors. Collier (1991) reported that a higher drape value obtained on the Drape Tester corresponds to a lower shear value, giving a negative correlation. Orzada et al. (1997) found negative correlations between drape values obtained on Collier’s Drape Tester and shear and bending properties determined on the Kawabata equipment. Hu and Chan (1998) conducted a comprehensive study of the relationship between drape coefficient and all the mechanical properties measured by the KES-F system. Bending, shear, surface roughness and weight all revealed significant correlations with drape coefficient.

Fabric properties related to laundering Laundering plays a significant role on the physical, mechanical, and aesthetic properties of a fabric. Research has concluded that laundering changes the appearance as well as the performance properties of fabrics. Laundering is a key cause of degradation in washable textiles (Goynes and Rollins, 1971; Handy et al., 1968; Breese et al., 1994). Research conducted by Breese et al. (1994) concluded that laundered fabric appears alike microscopically to fabrics in actual wear studies. In addition, detergents and the agitation process during laundering also affects fabric properties; detergents decrease the surface tension of the water and its ability to spread over and wet textiles (AATCC, 2004). Several studies have examined the influence of laundering on fabric physical and mechanical properties. In 1991, Postle reported that fabric mechanical properties are influenced by laundering and dry cleaning. Handu et al. (1976) concluded that both mechanical and chemical deterioration of fabrics occurs during laundering. Chu et al. (1950) reported griege cotton fabrics had poorer drapability than corresponding fabrics in bleached, and bleached and starched states. Likewise, Dhingra et al. (1989) found unfinished wool fabrics had improved shear and bending after finishing. Dhingra et al. (1989) investigated the effect of dry cleaning and steam pressing on fabric mechanical properties. Shear values of the wool fabrics tested were significantly improved by steam pressing; however, the degree of fabric set applied during finishing determined the effect of dry-cleaning on shear and bending. Weedall et al. (1995) utilized the KES-F system to assess fabric mechanical properties and obtain Kawabata Hand Values. The researchers found shear resistance was reduced considerably with increased washing and wearing, while other KES-F properties and primary hand values showed little effect from laundering. Response of fabric hand to laundering has been subjectively evaluated in several studies. Tinsley et al. (1991) investigated the effect of detergent products on fabric handle by means of consumer responses. The researchers concluded that detergent did influence fabric handle.

Effect of laundering

47

Methodology This exploratory study examining the effect of laundering on drape, bending and shear of bottom weight cotton fabrics was designed as a randomized factorial experiment with two independent variables: six fabrics and three laundering cycles (unlaundered, one and five home launderings). Drape, shear, and bending were the dependent variables and were evaluated after each of the three laundering cycles. Owing to the preliminary nature of the study, six 100 percent cotton fabrics representative of bottom weight fabrics were selected. Fiber type and fabric weight categories were controls. Physical fabric property measurements shown in Table I were determined according to American Society for Testing and Materials (ASTM) Methods. Fabrics 1 and 2 were Fabric 1 2 3 4 5 6

Weight (g/m2)

Thickness (mm)

Weave

Count (W £ F/25 mm)

273.97 287.65 413.33 432.61 315.69 271.26

0.575 0.543 0.785 0.853 0.620 0.585

Twill Twill Twill Twill Twill Plain

116 £ 56 114 £ 56 66 £ 38 57 £ 38 105 £ 52 72 £ 38

Table I. Fabric descriptive properties

IJCST 21,1

48

gabardines; Fabrics 3 and 4 were both heavy denims; Fabric 5 was a lightweight denim; and Fabric 6 was a plain weave. There was no significant variance in the weight of the fabrics; all were bottom weight with a mean value of 332.42 (g/m2). There was no significant variance within the six fabrics with respect to thickness; mean value was 0.66 mm. Fabric specimens were randomly cut from the six fabric samples and were randomly assigned to one of the three laundering cycles; unlaundered or control, one laundering, or five launderings. Specimens assigned to the laundering cycles were cut larger than required by the specific test methods; after laundering, all specimens were cut to the size required by the specific test method prior to measurement of drape, bending and shear properties (a 10 in. (254 mm) diameter circle for drape analysis, and a 20 £ 20 cm pattern for the bending and shear analyses). There were 108 specimens (18 from each fabric and 6 for each laundering cycle) for each of the dependent variables (drape, shear and bending) for a total of 324 specimens. Each specimen was laundered and dried according to AATCC standard home washer and dryer methods. Normal cotton/sturdy wash cycle with a water temperature of 1208 ^ 58F was used all for launderings. Detergent conditions were prepared according to test conditions; 66 g of AATCC Standard Reference Detergent was added to the automatic washing machine. Fabric specimens and enough 100 percent cotton Wash Load Ballast Type 1 were added to make a 4.00 ^ 0.25 lb. load. At the end of each of wash cycle, each specimen and ballast were immediately removed and placed in an automatic tumble dryer for drying. After drying, the specimens were conditioned according to ASTM D 1776, in an atmosphere of 70 ^ 28F (20 ^ 1 8C) and 65 ^ 2 percent RH before evaluation. Collier’s Drape Tester was used to evaluate drape quantitatively. Drape values were obtained by draping a 10 in. circular fabric specimen over a 5 in. center pedestal plate. Drape values were obtained by draping specimens face up, and then with the reverse side up, with duplicate measurements taken for each configuration. Selected fabric properties related to drape were measured objectively using two instruments from the Kawabata KES-F system. The Kawabata Pure Bending Tester was used to measure bending resistance to obtain bending modulus and bending hysteresis values. Shear resistance and hysteresis were determined on the Kawabata Tensile and Shear Tester (Kawabata, 1980). Specimens were sheared 48 in one direction and 48 in the opposite direction. Eight degrees is the standard shear deformation angle; however, the upper limit of shear strain range was adjusted to 48 because of the buckling caused by these stiffer bottom weight fabrics. Shear stiffness was calculated from the slopes of the forward and backward curves between 0.258 and 2.58. Shear hysteresis was calculated from the mean value of the hysteresis width at 0.258, and at 2.58 for the second hysteresis value. Measurements in warp and weft directions on three specimens of each fabric were made on the Kawabata instruments. Analysis of variance (ANOVA) was performed using the Statistical Package for the Social Sciences (SPSS) for each dependent variable; drape, shear, and bending properties. Pearson Product Moment Correlation Coefficients were determined to identify any significant correlations between the variables. Findings are presented according to the effects of fabrics and laundering cycles on each of the three dependent variables. Results and discussion Laundering had little effect on the physical characteristics of the fabric specimens. The laundering cycles did not change the weight (sign. of F ¼ 1.00) or the thickness of

the fabrics (sign. of F ¼ 1.00). Further, there were no significant differences (sign. of F ¼ 0.92) in the fabric count by laundering cycles; however, it is interesting to note that the fabric count increased after laundering. The mean count value of the unlaundered samples was 135, while the mean count rose to 140 after one laundering, and 143 after five launderings. The increase may be the result of fabric shrinkage after laundering.

Effect of laundering

Drape values Table II summarizes the drape values for each fabric after the laundering cycles. Although up to five (5) launderings did not make a significant difference in the drape values of the individual fabrics, it is worthy to note that overall the drape values, as well as the back drape values, increased for three of the fabrics (Fabrics 3-5) as the laundering cycles increased. Further, summed drape values across fabrics indicate a trend of increased drapeability with laundering cycles, as may be observed in Figure 1. The mean drape value for all unlaundered fabrics was 32.41; for five (5) laundry cycles, the mean drape value was 34.03. The increase in fabric drape may be related to the increased fabric count. It is known that the mechanical agitation during laundering enhances fabric shrinkage in cotton therefore increasing the fabric count. In addition, laundering also causes relaxation by mechanical agitation that may explain the increase as the laundering cycles increased. An ANOVA was performed to determine if there were any significant differences in draping configurations in the fabrics and to determine if there were significant differences in draping configurations due to face and back sides of the fabrics after laundering. (Table III). Fabric was not a significant source of variation in drape values ( p ¼ 0.43), face drape values ( p ¼ 0.06) or back drape values ( p ¼ 0.72). Drape values ranged from 26.43 to 41.33 percent. The lack of significant difference due to fabric is due to the similarity in the fabrics selected for this preliminary study; all fabrics were bottom weight fabrics with low drape, were identical in fiber type, and did not vary

49

Fabric

Laundering cycle

1

Zero One Five Zero One Five Zero One Five Zero One Five Zero One Five Zero One Five

2 3 4 5 6

Drape value

Drape values Face drape

Back drape

32.49 32.89 30.32 30.78 30.20 29.71 33.52 35.80 39.28 28.35 27.00 35.82 28.92 37.81 35.29 34.42 33.03 33.73

30.92 32.72 28.33 38.75 31.07 30.37 38.57 36.07 39.45 30.27 27.02 34.05 30.67 34.28 31.03 35.53 32.78 33.80

34.07 33.07 32.30 34.82 29.33 29.05 28.47 35.53 39.12 26.43 26.98 37.58 27.17 41.33 39.55 33.30 33.28 33.67

Table II. Mean values for fabric drape values after laundering

IJCST 21,1

34.5 34 33.5

Drape (%)

33

50

32.5 32 31.5 31 30.5

Figure 1. Mean values for fabric drape

30 Unlaundered

Source of variation

Table III. Analysis of variance for drape values

Laundering 1

Laundering 5

Laudering Cycles

Drape values Fabric Laundering cycle Face drape Fabric Laundering cycle Back drape Fabric Laundering cycle

Sum of squares

DF

Mean square

F

Sign. of F

61.0997 8.5396

5 2

12.2199 4.2698

1.0475 0.3326

0.4345 0.7222

118.5012 10.2394

5 2

3.7002 5.1197

2.9715 0.3765

0.0567 0.6926

66.0187 61.1751

5 2

13.2037 30.5875

0.5667 1.6130

0.7243 0.2320

widely in their weight, thickness and count properties. Laundering had no significant effect on drape values ( p ¼ 0.72), face drape values ( p ¼ 0.69), or back drape values ( p ¼ 0.23). The drapeability of these stiff, bottom weight fabrics were not significantly affected by laundering. Their original draping characteristics were retained after five laundry cycles. Therefore, there is a need to analyze the effect of an increased number of launderings, such as 10, 20, or 30 cycles, on drapeability since most bottom outerwear are expected to be serviceable for approximately one year. As report by El-Bayoumi (1980), garments that lose their drapeability would be discarded before they lose too much of their tensile properties to be functional.

Shear properties Summed shear values after laundering are presented in Table IV. Shear stiffness values for the fabric specimens ranged from 2.60 gf/(cm · deg) for Fabric 2 to 11.88 gf/(cm deg) for Fabric 4, the heaviest and thickest fabric. For Fabrics 3-5, the heaviest and thickest, experienced sharp decreases in shear due to laundering making them more pliable.

Fabric

Laundering cycle

1

Zero One Five Zero One Five Zero One Five Zero One Five Zero One Five Zero One Five

2 3 4 5 6

Shear stiffness (gf/cm deg) 2.92 3.92 4.25 2.60 3.77 4.10 10.25 4.92 5.06 11.88 6.54 5.67 5.94 4.00 4.25 3.73 3.85 4.04

Shear properties Shear hysteresis at 0.258 (gf/cm) Shear hysteresis at 2.508 7.17 9.38 10.56 7.00 10.98 11.83 20.75 7.23 9.69 21.38 7.06 10.58 21.21 7.73 9.69 5.93 6.71 7.75

8.15 10.00 11.13 8.33 11.69 11.92 22.92 9.00 11.17 23.13 10.50 12.71 17.10 9.08 11.33 8.19 9.04 9.98

Effect of laundering

51

Table IV. Mean values for shear properties by fabric for laundering cycles

Although changes in shear values due to laundering are inconsistent across fabrics, when summed, decreases in shear values are evident (Figure 2). Mean values for shear stiffness were 6.22 for all unlaundered specimens with mean shear stiffness dropping to 4.56 for specimens laundered five times. Likewise, shear hysteresis values dropped after five laundry cycles. These findings are consistent with findings by Weedall et al. (1995) who found initial inconsistencies in shear values based on number of launderings, but, in general, an overall increase in shear values across fabrics due to laundering. 16 14

Unlaundered One laundering

12 Five Launderings 10 8 6 4 2 0 Shear Stiffness (gf/cm. deg)

Shear Hysteresis at 0.25 degrees (gf/cm)

Shear Hysteresis at 2.50 degrees (gf/cm)

Figure 2. Mean values for shear properties by laundering cycles

IJCST 21,1

They used the Kawabata system to examine changes in mechanical properties due to laundering. An ANOVA (Table V) revealed no significant differences in shear stiffness for the main effects of fabrics ( p ¼ 0.07) and laundering ( p ¼ 0.40). In addition, there were no significant differences in shear hysteresis at 0.258 and at 2.508 due to fabrics and laundering ( p . 0.05).

52 Bending properties Table VI presents the ANOVA results for bending properties of both modulus and hysteresis. The data indicates that the main effects of fabric was significant ( p ¼ 0.000) for both bending modulus and bending hysteresis. However, there were no significant differences in the bending values of the fabrics due to laundering for either bending modulus ( p ¼ 0.85) or bending hysteresis ( p ¼ 0.72), respectively. It is worthy to note that increased launderings resulted in lower bending resistance and hysteresis of the fabrics as revealed in Figure 3. These lowered bending parameters result in increased fabric flexibility and elastic recovery from bending as is necessary for fabrics to drape, tailor and wear well (Dhingra et al., 1989). El-Bayoumi (1980) reported a decrease in bending modulus for woven cotton fabrics at all phases of the laundering cycle due to degradation that was emphasized by the mechanical damage of laundering. Bending modulus values ranged from 0.21 (gf cm2/cm) for Fabric 1 to 0.79 for Fabric 4. Bending hysteresis values were also highest for Fabric 4 and lowest for Fabric 1. Summed across fabrics, bending modulus decreased from 0.5 for the unlaundered specimens to 0.41 for those laundered five times. Likewise, the bending

Source of variation

Table V. Analysis of variance for shear properties

Shear stiffness Fabric Laundering cycle Shear hysteresis at 0.258 Fabric Laundering cycle Shear hysteresis at 2.508 Fabric Laundering cycle

Source of variation

Sum of squares

DF

Mean square

F

Sign. of F

52.4018 11.4023

5 2

10.4804 5.7012

2.7630 0.9884

0.0693 0.3951

96.2783 102.5084

5 2

19.2557 51.2542

0.6757 2.2898

0.6500 0.1356

99.4246 70.8480

5 2

19.8849 36.4240

0.9489 1.8973

0.4850 0.1843

DF

Mean square

F

Sign. of F

5 2

0.2026 0.0124

18.2212 0.1655

0.0001 0.8490

5 2

0.2057 0.0328

5.1746 0.342

0.0022 0.7157

Sum of squares 2

Table VI. Analysis of variance for bending properties

Bending modulus (gf/cm /cm) Fabric 1.0130 Laundering cycle 0.0248 Bending hysteresis (gf/cm/cm) Fabric 1.0287 Laundering cycle 0.0657

Effect of laundering

0.8 0.7 0.6

Unlaundered One Laundering Five Launderings

0.5

53

0.4 0.3 0.2 0.1 0 Bending Modulus

Bending Hysteresis

hysteresis mean dropped from 0.71 for the unlaundered specimens to 0.57 for those laundered five times. Figure 3 reveals the trend of decreased bending values due to laundering (Table VI). Conclusion For a fabric to form beautiful folds, a low resistance to bending is essential (Collier, 1990). Likewise, high shear values make draping and three-dimensional forming difficult, therefore lower shear values are desirable. Although the six bottom-weight fabrics used in this preliminary study were not significantly affected by the number of laundering cycles examined in this study, increased laundry treatments did increase overall drape values and reduce each of the shear and bending parameters examined, resulting in more pliable, drapable fabrics. While no significant differences in draping behavior were found among the laundered and unlaundered specimens, results confirm findings from other studies which reveal that washing relaxes the yarns in woven fabrics which decreases the frictional pressure between warp and filling yarns at the crossover points (Grosberg and Swani, 1966). Therefore, laundering allows the yarns to shear and bend more easily to assume the curvature that is necessary for draping. Trends of decreasing shear values with increased laundry cycles observed in this study are consistent with results obtained by Weedall et al. (1995). This exploratory study of six bottom weight fabrics has important implications for textile and apparel manufacturers. The continuing satisfactory performance of a textile fabric throughout its lifespan is of concern to consumers, and should be of concern to manufacturers as well. Objective measurement of fabric properties related to drape after laundry treatments would assist the manufacturer in developing laundering recommendations based on the fabrics’ performance and in selecting appropriate fabrics which maintain their drape characteristics, mechanical properties, and dimensional stability with use. Future research should examine a wider variety of fiber types, weave structures, and weights and expand the number of laundry cycles to confirm the trends observed in the present study.

Figure 3. Mean values for bending properties

IJCST 21,1

54

References AATCC (2004), AATCC Technical Manual, American Association of Textile Chemists and Colorists, Research Triangle Park, NC. Bresee, R.R., Annis, P.A. and Warnock, M.M. (1994), “Comparing actual fabric wear with laboratory abrasion and laundering”, Textile Chemist and Colorist, Vol. 26 No. 1, pp. 17-23. Chu, C.C., Cummings, C.L. and Teixeira, N.A. (1950), “Mechanics of elastic performance of textile materials. Part V: A study of the factors affecting the drape of fabrics – the development of a drape meter”, Textile Research Journal, Vol. 20 No. 8, pp. 539-48. Collier, B. (1990), “Assessment of fabric drape”, The FIT Review, Vol. 6 No. 2, pp. 40-3. Collier, B. (1991), “Measurement of fabric drape and its relation to fabric mechanical properties and subjective evaluation”, Clothing and Textile Research Journal, Vol. 10 No. 1, pp. 46-52. Collier, B., Paulins, V.A. and Collier, J.R. (1989), “Effects of interfacing type on shear and drape behavior of apparel fabrics”, Clothing and Textiles Research Journal, Vol. 7 No. 3, pp. 51-6. Collier, B., Collier, J.R., Scarberry, H. and Swearingen, A. (1988), “Development of a digital drape tester”, ACPTC Combined Proceedings, p. 35. Collier, J.R., Collier, B., O’Toole, G. and Sargand, S. (1991), “Drape prediction by means of finite-element analysis”, Journal of the Textile Institute, Vol. 52 No. 9, pp. 395-406. Cusick, G.E. (1965), “The dependence of fabric drape on bending and shear stiffness”, Journal of the Textile Institute, Vol. 56, pp. T596-T606. Cusick, G.E. (1968), “The measurement of fabric drape”, Journal of the Textile Institute, Vol. 59 No. 6, pp. 253-60. Dhingra, R.C., Liu, D. and Postle, R. (1989), “Measuring and interpreting low-stress fabric mechanical and surface properties. Part 2: Application of finishing, drycleaning, and photodegradation of wool fabrics”, Textile Research Journal, Vol. 59 No. 6, pp. 357-67. El-Bayouni, A.A. (1980), “An analytical approach to the effect of laundering processes on the drape and stiffness properties of cotton woven fabrics”, Journal of Engineering for Industry, Vol. 102, pp. 342-6. Goynes, W.R. and Rollins, M.L. (1971), “A scanning electron-microscope study of washer-dryer abrasion in cotton fibers”, Textile Research Journal, Vol. 41 No. 3, pp. 226-41. Grosberg, P. and Swani, N.M. (1966), “The mechanical properties of woven fabrics. Part IV: The determination of the bending rigidity and frictional restraint in woven fabrics”, Textile Research Journal, Vol. 36, pp. 338-45. Handu, J.L., Sreenivas, K. and Ranganathan, S.R. (1976), “Calcium-phosphorus deposition during laundering”, Textile Research Journal, Vol. 38, pp. 735-43. Handy, C.T., Arnold, H.W., Reitz, D.C. and Wilkinson, P.R. (1968), “Predicting the durability of dress shirts in home laundering”, American Dyestuff Reporter, Vol. 57 No. 14. Hearle, J.W.S. (1969), “Theory of the mechanics of staple fibre yarns”, in Hearle, J.W.S., Grosberg, P. and Backer, S. (Eds), Structural Mechanics of Fibres, Yarns and Fabrics, Wiley-Interscience, New York, NY. Hu, J. and Chan, Y.F. (1998), “Effect of fabric mechanical properties on drape”, Textile Research Journal, Vol. 68 No. 1, pp. 57-64. Kawabata, S. (1980), The Standardization and Analysis of Hand Evaluation, 2nd ed., The Textile Machinery Society of Japan, Osaka. Mehta, P. (1985), An Introduction to Quality Control for the Apparel Industry, J.S.N. International, Inc., Tokyo.

Morooka, H. and Niwa, M. (1976), “Relation between drape coefficients and mechanical properties of fabrics”, Journal of the Textile Machinery Society of Japan, Vol. 22, pp. 67-73. Morris, M.A. and Prato, H.H. (1981), “Consumer perception of comfort, fit and tactile characteristics of denim jeans”, Textile Chemist and Colorist, Vol. 13 No. 3, pp. 24-30. Orzada, B.T., Moore, M.A. and Collier, B.J. (1997), “Grain alignment: effects on fabric and garment drape”, International Journal of Clothing Science and Technology, Vol. 9 No. 4, pp. 272-84. Peirce, F.T. (1930), “The handle of cloth as a measurable quantity”, Journal of the Textile Institute, Vol. 21, pp. T377-T417. Postle, R. and Postle, J.R. (1998), “The dynamics of fabric drape”, International Journal of Clothing Science & Technology, Vol. 10 Nos 3/4, pp. 305-12. Shishoo, R. (1995), “Importance of mechanical and physical properties of fabrics in the clothing manufacturing process”, International Journal of Clothing Science and Technology, Vol. 7 Nos 2/3, pp. 35-42. Sudnik, M.P. (1978), “Rapid assessments of fabric stiffness and associated fabric aesthetics”, Textile Institute and Industry, Vol. 16 No. 6, pp. 155-9. Sudnik, Z.M. (1972), “Objective measurement of fabric drape: practical experience in the laboratory”, Textile Institute and Industry, Vol. 10 No. 1, pp. 14-18. Tinsley, A., Byrne, M. and Fritz, A. (1991), “The effect of detergents on fabric handle”, Journal of Consumer Studies and Home Economics, Vol. 15, pp. 223-9. Weedall, P.J., Harwood, R.J. and Shaw, N. (1995), “An assessment of the Kawabata transformation equations for primary-hand values”, Journal of the Textiles Institute, Vol. 86 No. 3, pp. 470-5. Wu, Z., Au, C.K. and Yuen, M. (2003), “Mechanical properties of fabric materials for draping simulation”, International Journal of Clothing Science and Technology, Vol. 15 No. 1, pp. 56-68. Further reading American Society for Testing and Materials (1985), Annual Book of ASTM Standards, 7.01, Easton, MD. Postle, R. (1991), “Fabric objective measurement technology: present status and future potential”, in Stylios, G. (Ed.), Textile Objective Measurement and Automation in Garment Manufacture, Ellis Horwood, Chichester, pp. 27-45. Corresponding author Belinda T. Orzada can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

Effect of laundering

55

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

IJCST 21,1

The abrasion resistance of socks ¨ zdıˆl, Arzu Marmarali and Nida Og˘lakciog˘lu Nilgu¨n O

Department of Textile Engineering, Ege University, ˆIzmir, Turkey

56 Abstract Received 22 April 2007 Revised 31 January 2008 Accepted 31 January 2008

Purpose – The purpose of this paper is to explain the yarn parameters and some finishing process that can affect the abrasion resistance of socks in detail. Design/methodology/approach – The abrasion tests were made on socks produced from the most popular fibers (cotton, wool, PAC, PES, PA, and blends of these) by the Modificated Martindale method. The effects of fiber type, yarn count (for single and ply yarn), combing process, softness process with silicone and mercerization process to the abrasion resistance were investigated. Findings – It was found that the use of coarse yarns, addition of polyester, polyamide fibers or elastane filaments to the structure and application of the mercerization process increase the abrasion resistance of the socks. However, the silicone softeners decrease this value. The resistance of wool socks is higher than acrylics. Originality/value – Socks, which are a necessary item of clothing, need to be comfortable, affordable and retain their quality throughout their life. The most significant problem is abrasion which can greatly reduce the material’s life. To determine the parameters affecting the sock abrasion will be useful both for producer and for consumer. Keywords Abrasion, Wear resistance, Yarn testing, Garment industry Paper type Research paper

International Journal of Clothing Science and Technology Vol. 21 No. 1, 2009 pp. 56-63 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910923755

Introduction Abrasion, which is an unavoidable problem, usually occurs on the on the heel, sole and toe of the socks. The socks rub within the shoes, slippers or even the ground. The life of sock becomes shorter with the changing in mechanical properties and decreasing in quality due to abrasion. The first stage of abrasion is small balls entanglement because of the loose fibers unravels from the fabric surface during usage and washing. Eventually the fibers which bind the balls to the surface breakdown and a hole occur. If the sock consists of synthetic fibers with natural fibers, during rubbing action natural fibers, which give the desirable properties of the sock, move away, only synthetic fibers remain. This gives the sock undesirable appearance and decrease the overall fabric thickness. Until now some research related with the abrasion resistance socks has been completed. Wisniak and Krzeminska (1987) made both laboratory tests and usage tests to research abrasion resistance of seven different types of socks consisting of different rates of Co-PA. For abrasion resistance they used pilling test device. They used the abrasion time of the sample as an assessment. They found that the results from the laboratory and the usage tests were different. Miajewska and Kazmierczak (1983) searched the abrasion resistance of terry socks. They used PA yarns for ground, wool and wool blends (wool þ PA, PAC þ PA, wool þ PAC) also cotton and cotton blends (Co þ PA, Co þ viscone þ PA) for pile. They found that the result of the wool and wool bends are evenly matched to each other. Abrasion resistance of the cotton socks was better than wool socks. They used different yarns for piles to increase the abrasion resistance of the socks and found that

the yarns spun with wool -PES blends gave the best results. The results of the PAC-wool yarns and PAC-linen yarns followed it, respectively. They observed that if the yarn twist was increased the abrasion resistance of the socks also increased. They also found that Pa and folded PAC yarns increase the abrasion resistance. The Sock Testing Consortium researched socks abrasion resistance and they used the Stoll Abrasion Tester (2001). When 3lb compression weight was used the average result of the tests was 1,500 tours with a variation coefficient of 7.6 percent. For 3 lb the average result was 2,706 with a variation coefficient of 13.5 percent. Isgoren et al. (2000) researched the physical properties of military socks. They used different six type of socks produced from Co-PA, wool-PES-Co and Co-PAC-wool blended yarns. They tested dimensional stability, weight loss after abrasion, pilling properties, color, perspiratory, and crocking fastness of the socks. They found that the weight loss was more than 5 percent except Co-PA socks. The aim of this study is to explain the effects of fiber type, yarn count (for single and ply yarn), combing process, softness process with silicone and mercerization process to the abrasion resistance of socks.

The abrasion resistance of socks 57

Materials and methods Materials Cotton, acrylic, wool, polyamide, and polyester fibers are mostly used in socks. So in this research, the effect of commonly used fibers to abrasion resistance was investigated. For this purpose the yarns in Tables I-VI were used. The socks were knitted in a single jersey structure using suitable hosiery machines with optimum adjustments. During the knitting process all machine settings were kept exactly same. Test methods In this study, abrasion resistance values are measured on modified Martindale instrument according to EN 13770:2001 (2001) (Figure 1). Material

Yarn count (Ne)

100 percent cotton

20/1 20/1 £ 2 30/1 30/1 £ 2 Carded 20/1 Carded 30/1 Combed 20/1 Combed 30/1

Material

Twist coefficient (ae)

Number of rubs at yarn breakdown

4.3 4.3 3.8 3.8 3.5 3.7 3.6 3.7

2,169 2,973 1,013 2,917 2,250 1,025 2,167 917

Yarn count

Percent 100 Co Ne 20/1 Percent 75 Co-percent 25 PA Ne 20/1 þ Td 150 Percent 50 Co-percent 50 PA Ne 20/1 þ Td 150 £ 2 Percent100 PA Td 150 £ 2

Table I. The abrasion resistance results of different yarn parameters

Twist coefficent Number of of cotton yarn rubs at yarn Total weight (ae) breakdown loss (percent) 4.3 4.2 4.0 –

2,413 16,750 66,000 110,000

1.65 13.29 12.11 4.07

Table II. The abrasion resistance results of PA and Co socks

IJCST 21,1

58

For each sock, four samples (two from heel-two from sole) are taken and the average abrasion resistance values were calculated. The endpoint was determined according to the following two criteria: (1) A hole which usually develops when one thread is broken causing a hole to appear. At this point the number of rubs which represent abrasion resistance is recorded. This assessment criteria was used all types of socks. (2) Thinning when the knitted footwear garment is constructed of filament and spun staple yarns plated in knitting, e.g. nylon/cotton, and spun staple yarn wear away leaving a base of the filament. In this sock test the weight loss in certain intervals, the number of cycles and total weight loss that the “one break” caused a hole are recorded. This assessment criteria was used only the socks constructed with Co/PA yarns.

Material Table III. The abrasion resistance results for Co-PES blends in socks

Table IV. The abrasion resistance results of Wo/PAC socks

Table V. The abrasion resistance results of socks with and without silicone softener

Percent 100 Co Percent 70 Co-percent 30 PES Percent 50 Co-percent 50 PES

20/1 20/1 20/1

4.0 3.9 4.0

2,000 2,500 3,000

Material

Yarn count (Nm)

Twist coefficient (am)

Number of rubs at yarn breakdown

Percent Percent Percent Percent

16/1 16/1 40/1 40/1

104.9 119.1 102.5 112.6

4,313 4,450 1,250 1,750

100 acrylic 50 acrylic-percent 50 wool 100 acrylic 100 wool

Material

Silicone

Yarn count (Ne)

Twist coefficient (ae)

Percent 85 Co-percent 15 PA Percent 85 Co-percent 15 PA

Without With

Ne 20/1 þ Td 75 Ne 20/1 þ Td 75

4.4 4.4

Material Table VI. The effect of mercerization on abrasion resistance results

Yarn count Twist coefficient (Ne) (ae) Number of rubs at yarn breakdown

Combed yarn Mercerized combed yarn

Number of rubs at yarn breakdown 34,500 23,000

Yarn count (Ne)

Twist coefficient (ae)

Number of rubs at yarn breakdown

56/2 36/2 56/2 36/2

3.9 4.4 4.0 4.5

930 2,280 1,050 2,700

The abrasion resistance of socks 59

Figure 1. Setting of sock kit on Martindale apparatus

SPSS 10.0 for Windows statistical software was used for evaluating the test outcome. To determine the statistical importance of the variations, ANOVA and correlation tests were applied. To deduce whether the parameters were significant or not, p values were examined. Ergun (1995) emphasized that if the p-value of a parameter is greater than 0.05 ( p . 0.05), the parameter would not be important and should be ignored. Results and discussion Effects of yarn parameters on abrasion resistance Table I displays the abrasion resistance of socks knitted with 100 percent cotton yarns that use a different spinning system, count, twist coefficient, and ply. The results indicate that the number of rubs for a hole and the abrasion resistance values increase when coarser yarns are used. With using plied yarns, these values increase also, because plied yarns are coarser. In this group, the differences are statistically significant. According to the results, the effect of the combing process in yarn production is statistically insignificant on abrasion resistance of socks. Effects of Co/PA ratio on abrasion resistance Filament yarns are commonly used in the sock industry because of the high tensile strength, which increases the abrasion resistant and usage life of socks. Throughout the life of socks, natural fibers move away from the surface and only the filament yarns remain. This event is referred to as thinning. In this study, socks were knitted using PA(filament) and Co yarns in different amounts as seen in Table II. Filament and cotton yarns were fed from separate yarn guides to the needles in sock machines. To determine the effect of elastic yarn the abrasion results of the socks knitted with percent 98 PA-percent 2 elastane yarn and percent 100 PA were compared also. Figure 2 displays the thinning appearance at the heel part of percent 75 Co-percent 25 PA sock during abrasion resistance tests. According to results and statistical evaluation, the abrasion resistance increases as the polyamide filament ratio increases and this relation is statistically significant (Figure 3). The results show that by using elastic yarn the abrasion resistance of the

IJCST 21,1

60

socks tested increased by almost 80 percent. The analysis of variance indicated that the effect of elastic yarn on abrasion values is statistically significant. Figure 4 displays the percentage of weight loss in grams for a number of revolutions. In pure cotton socks breaking occurs in the first 5,000 rubs; in Co-PA socks because of the affect of cotton yarn, loss weight in grams was obtained within the first 10,000 rubs. The weight loss with each rub decreases in proportion to the amount of PA used. The weight loss is minimum in PA-elastane socks and maximum in percent 75 Co-percent 25 PA socks.

Figure 2. The appearances of sock at 0-5,000-10,000 rubs during abrasion test

Figure 3. The abrasion resistance of Co-PA socks

Number of rubs at yarn break down

200,000 150,000 100,000 50,000 0 %100Co

%75Co%25PA

%50Co%50PA

%100PA

%98PA%2Elastane

Weight loss %

16

Figure 4. The results of weight loss in grams in cotton-PA socks

14

%98PA-%2Elastane

12

%100PA

10

%50-50 CO/PA

8

75-25 CO/PA

6

% 100 CO

4 2 0 0

10

20

30

40

50

60

70

80

90

100 110 120 130 140 150 160 170 180 190

Number of rubs (×103)

Effects of blend ratio of Co/PES on abrasion resistance In this study, different blend ratios of Co-PES yarns, which are mostly used in the sock industry, were also tested and the results are given in Table III. PES fibers were blended with cotton during the spinning process. So during the tests thinning was not seen. The number of rubs where one thread is broken causing a hole, was taken as the end point of the test. The results showed that the abrasion resistance significantly increases when the PES ratio increases (Figure 5) because of the higher tensile strength and bending resistance of PES fibers.

The abrasion resistance of socks 61

Effects of blend ratio of Wo/PAC on abrasion resistance The Wo-PAC blends in socks are generally preferred in winter. Table IV displays the abrasion results of socks from various Wo/PAC yarns. The comparisons of samples knitted with 100 percent Wo and 100 percent PAC yarn revealed that the abrasion resistance of wool socks is significantly higher than acrylic ones (Figure 6). However, the differences of abrasion values between 50 percent Wo-50 PAC and 100 percent PAC samples are not significant. Effects of softeners on abrasion resistance To give a softer feel, silicon softeners are applied to the socks. In order to determine the effects of softeners to the abrasion resistance, the samples with and without softeners were compared. It can be seen in Table V, the abrasion resistance significantly decreases with the usage of softeners. Because when silicone softener is applied, the surface of the sock becomes smoother. So fibers will easily slide over each other and move away from the surface.

Number of rubs at yarn breakdown

Effect of mercerization process on abrasion resistance A brighter appearance in textile products is achieved by applying mercerization process. To determine the effects of this process to the abrasion resistance, the socks knitted from combed and mercerized combed yarns with different counts were tested (Table VI). The results indicate that the socks knitted from mercerized yarns have greater abrasion resistance values for both yarn count and differences are statistically significant.

3,000 2,500 2,000 1,500 1,000 500 0 %100 Co

%70-30 Co/PES

%50-50 Co/PES

Figure 5. The abrasion resistance of socks knitted with Co-PES blends

IJCST 21,1

Figure 6. The abrasion resistance of different wool-acrylic blends

Number of rubs at yarn breakdown

62

5,000 4,000 3,000 2,000 1,000 0 Nm 40/1 %100 PAC

Nm 40/1 %100 Wo

Nm 16 %100 PAC

Nm 16 %50 Wo%50 PAC

This situation can be explained by the change of fiber morphology during mercerization process. The settling of the fibers in mercerized yarn will be more uniform, close to each other and parallel to the yarn axis. Therefore, the number of contact points increases and the friction between them becomes higher. The result is fibers which resist being moved from the structure giving a higher resistance to abrasion. Furthermore, after mercerization fiber tensile strength also increases and this creates a positive influence on the abrasion resistance. Conclusions The study showed that the abrasion resistance value of socks can be increased by a number of measures; use of thicker yarns, adding PA to the structure, adding elastic yarns to the structure, increasing the PES ratio in Co-PES yarns. The effect of the spinning process on abrasion resistance of socks is not significant statistically. The resistance of wool socks is higher than acrylic ones and the wool ratio in wool-acrylic samples has a positive influence on abrasion resistance. The abrasion resistance increases during mercerization process and decreases with the use of silicone softeners. References EN (2001), 13770:2001, “Abrasion resistance for hosiery”, Operator’s Guide, James H. Heal, Halifax. Ergun, M. (1995), “Bilimsel Aras¸tırmalarda Bilgisayarla Istatistik Uygulamaları”, SPSS for Windows, Vol. 107. Isgoren, E., Isgoren, N. and Agırgan, O¨. (2000), “Asker C¸oraplarının Fiziksel ac¸ıdan ˙Incelenmesi”, ¨ ru¨nlerindeki Teknolojik Gelis¸meler ile Temizlik KKK, LEMAS 2000Askeri Amac¸lı Tekstil U ve Bakım yo¨ntemleri Sempozyumu, pp. 86-8. Miajewska, M. and Kazmierczak, U. (1983), “Improving the end use properties of hosiery with a loop construction by using various raw materials”, Technik Wlokienniczy, Vol. 32 No. 2, pp. 47-50. Sock Testing Consortium (2001), “Abrasion study results”, available at: www.legsource.com/ Hosiery_Consortium_Testing/abrasion_study_result.htm

Wisniak, E. and Krzeminska, F. (1987), “Testing the abrasion resistance of men’s socks”, Przegland Wlokienniczy, Vol. 41 No. 10, pp. 408-10. Further reading Abrasion Resistance for Hosiery Modified Martindale (1999), Operator’s Guide, James H. Heal & Co. Ltd, Halifax. Corresponding author Nilgu¨n O¨zdıˆl can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The abrasion resistance of socks 63

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

IJCST 21,1

Factors affecting recovery of PTT shape memory fabric to its initial shape

64

Lihuan Zhao, Li Qin and Fumei Wang College of Textiles, Dong Hua University, Shanghai, People’s Republic of China, and

Received 8 June 2008 Accepted 8 June 2008

Hoe Hin Chuah Shell Global Solutions, Houston, Texas, USA Abstract Purpose – The purpose of this paper is to understand the recovery mechanism of poly(trimethylene terephthalate) (PTT) shape memory fabrics. Design/methodology/approach – Tests were designed to study the effects of force, temperature and their combinations on the fabrics’ crease recoveries. In the test a cantilever device and an ironing force which simulated people ironing their clothes were used, respectively. Findings – Temperature was found to have little effect on the recovery of both the warp and filling of the fabrics. Crease recoveries did not improve significantly when the temperature was increased to above the polymer’s glass transition. However, forces, applied in primarily compressive and tensile modes to simulate ironing and hand stroking actions, were found to be very effective in the fabrics’ crease recoveries. Recoveries were 81-87 per cent even when the applied force was very small, at 5 N/cm2. When forces were applied at elevated temperatures, just below and above the polymer’s glass transition, there were no significant improvements in crease recoveries. Therefore, force was the main factor in PTT shape memory fabrics’ recovery mechanism for the fabrics to return to their initial shapes. Originality/value – The results suggest that PTT shape memory fabric has excellent shape recoverability and easy care property and it has large application potentiality. Keywords Textile technology, Temperature, Textile manufacturing process, Fabric testing Paper type Research paper

International Journal of Clothing Science and Technology Vol. 21 No. 1, 2009 pp. 64-73 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910923764

1. Introduction Polytrimethylene terephthalate (PTT) shape memory fabric is a newly developed advanced fabric. Although it appeared in the market only within the last three years, it quickly gained consumer acceptance and had significant market growth. As such, PTT shape memory fabric caught the attentions of several major companies, both domestic and overseas, which invested significant amount of money in its study and development. PTT shape memory fabric is made from PTT filaments and has properties of a shape memory material. The fabric can be shaped at will; it readily returned to its initial flat shape and dimension by the application of external forces. Such property allows costume designers to design puffy-looking clothing, such as “umbrella” and “tulip” fashions. These clothes can be flattened by simply stroking them with hand and reshaped if the fashion is not to one’s liking. Because of this excellent recovery, clothes made with shape memory fabrics are easy to care, wrinkles

can be easily removed, and the fabrics can be flattened and returned to their initial shapes by simply stroking them with hands. Shape memory material is defined as a material which after being deformed and set into a new shape, could recover its original shape by the application of external means, such as thermal, chemical, mechanical, light, magnetic, electrical means, etc. Generally, shape memory material can be divided into three types: shape memory alloy, shape memory ceramic and shape memory polymer (SMP). Only shape memory alloys and shape memory polymers are used in textiles. Between these two, SMPs tend to have a larger recover-ability, lighter in weight, a superior molding property, lower in cost and have tunable thermal transitions in the appropriate temperature range (Wornyo et al., 2007) than shape memory alloys. Because of these advantages, SMPs are used as functional materials in various hi-tech applications (Zhang and Ni, 2007). Generally speaking, SMPs are polymers with networks acting as molecular switches which response to external stimulations (Behl and Lendlein, 2007). These networks consist of chain segments and net points The net points can be physical, such as phase aggregations or intermolecular interactions, or chemical, such as covalent bonding or crosslinking (Ji et al., 2007). They connect the switching (transition) segments and determine the permanent shape of the polymer, and are usually referred to as fixed phases (Choi and Lendlein, 2007). In thermally-induced SMPs, chain segments which are associated with domains having the second highest thermal transition temperature are called switching segments or reversible phases (Razzaq et al., 2007). Switching segments become flexible when the working temperature is higher than the transition temperature, which can be a glass transition (Tg) or a melting temperature (Tm) (Kaursoin and Agrawal, 2007). As soon as the switching temperature is reached, the polymer begins to recover its original shape (Khonakdar et al., 2007). When SMPs are spun into filaments and made into shape memory fabrics, the recovery mechanisms of the final fabric are more complex than those of the starting raw material. Besides, polymer structure, molecular weight (Chen et al., 2007), hard segment content (Jeong et al., 2000), cross-linking of hard segments (Chun et al., 2007), etc. which affect the recovery of the SMP’s fixed and reversible phases, frictional forces between fibers in the fabric, fiber surface properties, yarn structures and fabric textures are also involved. Therefore, whether a fabric could recover its original shape or not depends on whether the applied recovery force is greater than the sum of forces in the polymer’s reversible phases and the frictional forces between the fibers. Until now, there are very few reports studying PTT shape memory fabrics. As far as we are aware, factors responsible for a PTT shape memory fabric to readily return to its initial shape have not been studied. In this article, we carried out experiments to study those factors to understand PTT shape memory fabric’s recovery mechanisms. Based on customer feedbacks and consideration of several possible factors operating in a SMP, we focused on temperature and mechanical force in this study. 2. Sample preparation Three shape memory fabrics were used in this study; their specifications are listed in Table I. The tests were conducted in a constant temperature environment according to ISO 2313-1972. Six fabric strips, 4 cm £ 1.5 cm, were cut along the warp and fill directions

PTT shape memory fabric

65

IJCST 21,1

66

of the fabrics, respectively. The test strip was folded into two halves; a 2 kg weight was put on top, covering the entire folded strip, for 24 h. After removing the load, the test strip was allowed to recover for 5 min and tested for shape recovery. 3. Explorations of test methods 3.1 Effect of temperature The filaments used for making shape memory fabric were spun from PTT, which is a semi-crystalline thermoplastic polymer. Figure 1 shows PTT’s differential scanning calorimetry (DSC) scan. The glass transition temperature is 41.78C. From the DSC’s heats of fusion, crystallinities of the warp and weft yarns were measured and are shown in Table I. In a traditional SMP, the mechanism for shape memory recovery involves the polymer’s crystalline region to act as a fixed phase and the amorphous region as a reversible phase. Thus, when the temperature of a PTT filament reaches 41.78C, the amorphous reversible phase begins to have segmental chain mobility. If the elastic recovery stress of the fixed phase was large enough, PTT would begin to recover its original shape according to the traditional mechanism of a SMP.

Fabric number #1 Table I. Specification of shape memory fabrics

#2 #3

Sample source Shaoxing Silver Brige Spinning & Weaving Co,.Ltd Wujiang Yingxiang Chemical Fibre Co.,Ltd Shaoxing Silver Brige Spinning & Weaving Co,.Ltd

Yarn size (Tex)

Density (10 cm)

8.3 £ 8.3

820 £ 326

26.73

25.16

8.3 £ 8.3

928 £ 338

26.27

24.67

8.3 £ 8.3

895 £ 420

–

–

DSC/(mW/mg)

1.5

Crystallization point 1.0 0.5 0.0 –0.5

Glass transition: 41.7°C

–1.0

Figure 1. DSC of PTT shape memory filament

–1.5

Fusion point 0

50

100

Crystallinity (%) Warp Weft

150 200 Temperature/°C

250

300

We, therefore, carried out the following test to study the effect of temperature on PTT shape memory fabrics’ recoverability. Eighteen pieces of 4 cm £ 1.5 cm fabric strips from the warp and filling of the #3 fabric were cut, respectively, and were divided into three groups. They were folded into two-halves to form creases as described in the Experimental section. After the loads were removed, the specimens were allowed to recover for 5 min at 20, 40 and 608C. The crease recovery angle, (u, was then measured. An average of six specimens was reported. The crease recovery ratio was calculated as follows and the results are shown in Figure 2: Crease

recovery ratio ¼

PTT shape memory fabric

67

u £ 100%: 1808

We found temperature was not an effective mean for recovering PTT shape memory fabric’s creases in both the warp and fill directions. Crease recovery ratios were between 46-51 per cent and 35-39 per cent in the warp and fill directions, respectively, at temperature of 20, 40 and 608C. Since PTT’s glass transition falls between 40 and 608C, according to the fixed phase’s recovery stress theory, PTT’s reversible phase would have changed from glassy to elastomeric state, and the polymer would have returned to its initial shape at temperatures above its Tg. However, the results showed that the fabrics did not have high crease recovery ratios and they were independent of temperatures below and above the polymer’s Tg. Since we are dealing with a fabric, the recovery process is much more complicated than that of the starting SMP. Fabric is an assembly of fibers, deformation of a fabric involves not only deformation of the fibers’ configurations but also a change in their relative locations. The latter is highly dependent on the interactive frictional forces between the yarns and fibers. A possible reason that temperature did not have significant effect on the fabric’s crease recovery is that the yarn’s low crystallinities and, therefore, low-elastic recovery stress. The fabrics used in this study had dense constructions (either plain weave or twill), higher counts (Table I), and were made from either untwisted or low-twist yarns. 50

Crease recovery ratio/%

48 46 44

warp-wise

42

filling-wise

40 38 20

30

40 Temperature/°C

50

60

Figure 2. Effects of temperature on crease recovery ratio of PTT shape memory fabric in the warp and weft directions

IJCST 21,1

68

These factors resulted in high-frictional forces between the fibers; the total elastic recovery stress from the fixed phases was not large enough to overcome the frictional forces in such a fabric construction and, therefore, the fabric could not return to its initial shape even at temperature above its Tg. 3.2 Effect of force From customer feedbacks, we understood that wrinkles and creases of a PTT shape memory fabric could be easily removed by simply stroking the fabric with hand. We, therefore, experimented with different ways of applying forces to the creased fabric, and found the following two methods allowed the fabrics to recover its flat shapes quite remarkably and gave us insights to PTT shape memory fabric operating mechanism. 3.2.1 Method 1. Figure 3 is a schematic of the first method. Five levels of loads, 0.60, 1, 1.30, 2.60 and 4 N/cm, were applied to the fabric perpendicular to the specimen’s crease direction. The cantilever made smooth alternating motions and dragged the specimen over a drum, simulating the action of a hand stroking the fabric. The applied force included both compression and tension modes. After 1 min, the load was removed, the specimen was un-mounted gently and was allowed to recover for 5 min. Crease recovery angle was then measured. Figures 4 and 5 show the results. Crease recovery ratios for all five tensions fell between 80 and 90 per cent. Even at the lowest load of 0.6 N/cm, which was much less than a tensile force generated when a human hand rubs a fabric, crease recovery ratios in the warp and fill directions were . 83 per cent. Increasing the loads did not significant change the results. Values of the crease recovery angles were on par with those of good anti-crease wool and cotton fabrics rated at level 4 according to the American Association of Textile Chemists and Colorists anti-crease standard. In summary, crease recovery ratios in this test indicated that PTT shape memory fabrics could return to their initial shapes with the crease largely removed by the application of forces simulating rubbing the fabric by hand. 3.2.2 Method 2. The second test involved applying forces simulating ironing where the main force was in compression mode and tension mode was secondary. Three ironing forces, 5.0, 8.0 and 12.0 N/cm2, were applied to the fabrics at 208C. The iron Test sample

Crease

Fixed smooth axis

Figure 3. Schematic of cantilever device

f

Axis of rotation

PTT shape memory fabric

90

Crease recovery ratio/%

87 84 81 78

69

75 # 1 fabric

72

# 2 fabric

69 66 63 60

0.5

1.0

1.5

2.0 2.5 3.0 Tension/(N/cm)

3.5

4.0

4.5

Figure 4. Effects of forces on crease recovery ratios of the fabric in warp direction

96 Crease recovery ratio/%

90 84 78 72 66

# 1 fabric

60

# 2 fabric

54 48 42 0.5

1.0

1.5

2.0 2.5 3.0 Tension/(N/cm)

3.5

4.0

4.5

was moved against the fabric crease once at a speed comparable to when people ironing their clothes. The specimen was then allowed to recover for 5 min and the crease recovery angle was measured. Figures 6 and 7 show the results. Both warp and fill of PTT shape memory fabrics had . 81 per cent crease recovery ratios. There were virtually no differences in the crease recovery ratios at the three levels of applied force. The above results showed that there were no significant differences between the two test methods on the fabrics’ recoverability; both of them returned the fabrics to near their initial shapes with very high crease recovery ratios. In addition, the load and force required to achieve the recovery were very small, 0.6 N/cm and 5.0 N/cm2 in method 1 and 2, respectively. These tests showed that force is likely the main operating mechanism for PTT shape memory fabric to return to its initial shape. The fabrics used in this study had tight weave structures and, therefore, high inter-filament frictional forces. When the applied force was large enough to overcome frictional forces between the filaments, the fabric began to recover to its initial shape.

Figure 5. Effects of forces on crease recovery ratios of the fabric in weft direction

IJCST 21,1 Crease recovery ratio/%

70

90 87

Figure 6. Effects of ironing forces on crease recovery ratios of PTT fabrics in warp direction

84 81 78 75 72

# 1 fabric

69

# 2 fabric

66 63 60

4

5

6

7

8

9

10

11

12

13

11

12

13

Force/(N/cm2)

90

Crease recovery ratio/%

87

Figure 7. Effects of ironing forces on crease recovery ratios of PTT fabrics in weft direction

84 81 78 75 72

# 1 fabric

69

# 2 fabric

66 63 60

4

5

6

7

8

9

10

Force/(N/cm2)

Such a recovery process is akin to the process by which the “reversible phase is softened” and that the elastic recovery stress of the PTT shape memory “fixed phase” allowed the fabrics to recover their flat shapes in the traditional mechanism of a SMP. 3.3 Effect of force and temperature in combination The above results show that force is the main factor in PTT shape memory fabric recovery mechanism. Between the two force test methods, the ironing method is more convenient to use and a low-compressive force of 5.0 N/cm2 is sufficient for the fabric creases to recover. Since PTT is a thermoplastic, we expected temperature to also play an important role in the polymer’s recovery process, however, results in Section 2.1 did not show any significant temperature effect. We postulated that the filaments’ low crystallinity and, therefore, low-elastic recovery stress generated by the fixed phase, and the tight fabric constructions might be the causes. In order not to rule out the possibility that temperature might play a role in the recovery mechanism in addition to

force, we designed the following tests to check the combined effects of force and temperature on PTT shape memory fabric’s recoverability. Two criteria were considered when selecting the temperature: First, glass transition of a polymer typically occurred over a range of temperature with about a ten-degree spread around the reported DSC mid-point glass transition temperature. So PTT’s glass transition occurs over a range of temperature of 32-528C. Second, since PTT shape memory fabric could return to its initial shape by just stroking the fabric with hand, heat transfer from a human hand to fabric from stroking could raise the fabric temperature slightly but not by very much.We, therefore, chose 20 and 358C as the two temperatures in this test. The first temperature is below PTT’s Tg and the second temperature is close to a human body temperature and covers the lower spectrum of PTT’s Tg. The ironing force method was used by applying a 5.0 N/cm2 force at these two temperatures. The results are shown in Figures 8 and 9. At 208C, both fabrics’ crease recovery ratios were more than 81 per cent in the warp and fill directions. At 358C, warp crease recovery ratios increased slightly while the fillings did not. A possible reason is the warps were subjected to a higher tension force during weaving and resulted in higher inter-filament frictions.

PTT shape memory fabric

71

4. Conclusions PTT shape memory clothes wrinkled when they were being worn or stored, however, those wrinkles could be easily removed by just stroking the clothes with hand. The mechanism of such an excellent recovery is not well understood. Since stroking action combines both compressive and tensile forces, and possibly a moderate increase in temperature generated by human hand, the effects of temperature, force and their combinations on PTT shape memory fabric crease recovery were studied. The results showed that temperature, below and above the polymer’s glass transition, had little effect on crease recoveries of both the warp and weft of PTT shape memory fabrics. A possible reason for the low recoveries was PTT filaments’ low crystallinities; the elastic recovery force generated by the fixed crystalline phase was 90

Crease recovery ratio/%

87 84 81 78 75 72

# 1 fabric

69

# 2 fabric

66 63 60 18

20

22

24

26 28 30 Temperature/°C

32

34

36

Figure 8. Effects of force and temperature combinations on crease recovery ratios in warp direction

IJCST 21,1

Figure 9. Effects of force and temperature combinations on crease recovery ratios in weft direction

Crease recovery ratio/%

72

90 87 84 81 78 75

# 1 fabric

72

# 2 fabric

69 66 63 60 18

20

22

24

26 28 30 Temperature/°C

32

34

36

not high enough to initiate recovery in this tight construction fabric. However, . 81 per cent crease recoveries were obtained when the fabrics were subjected to forces applied either in compressive or tensile modes simulating ironing and hand stroking actions. Moreover, forces required for the excellent recovery were quite small, 5.0 N/cm2 in ironing mode. Combinations of force and temperature did not significantly improve crease recoveries than by force alone. Therefore, force is likely the main factor in the recovery mechanism for PTT shape memory fabric to return to its initial shape. When the applied force was large enough to overcome frictions between filaments of the tight woven structure, the fabric gained sufficient mobility, coupled with PTT filaments’ excellent elastic recovery properties, and gave excellent fabric crease recovery. References Behl, M. and Lendlein, A. (2007), “Actively moving polymers”, Soft Matter, Vol. 3 No. 1, pp. 58-67, available at: www.rsc.org/publishing/journals/SM/article.asp?doi ¼ b610611k Chen, S.J., Hu, J.L., Liu, Y.Q., Lien, H., Zhu, Y. and Meng, Q. (2007), “Effect of molecular weight on shape memory behavior in polyurethane films”, Polymer International, Vol. 56 No. 9, pp. 1128-34, available at: www3.interscience.wiley.com/journal/114114554/abstract Choi, N.Y. and Lendlein, A. (2007), “Degradable shape-memory polymer networks from oligo [(L -lactide)-ran-glycolide] dimethacrylates”, Soft Matter, Vol. 3 No. 7, pp. 901-9, available at: www.rsc.org/publishing/journals/SM/article.asp?doi ¼ b702515g Chun, B.C., Chong, M.H. and Chung, Y.C. (2007), “Effect of glycerol cross-linking and hard segment content on the shape memory property of polyurethane block copolymer”, Journal of Materials Science, Vol. 42, pp. 6524-31, available at: www.springerlink.com/content/ e370q5j0852887m4/ Jeong, H.M., Lee, S.Y. and Kim, B.K. (2000), “Shape memory polyurethane containing amorphous reversible phase”, Journal of Materials Science, Vol. 35 No. 7, pp. 1579-83, available at: http://apps.isiknowledge.com/full_record.do?product ¼ WOS&search_mode ¼ General Search &qid ¼ 3&SID ¼ 1Akl6cgHI1Dad5CdBfH&page ¼ 1& doc ¼ 1 Ji, F.L., Hu, J.L., Li, T.C. and Wong, Y.W. (2007), “Morphology and shape memory effect of segmented polyurethanes. Part I: With crystalline reversible phase”, Polymer, Vol. 48 No. 17, pp. 5133-45, available at: www.sciencedirect.com/

science?_ob ¼ ArticleURL&_udi ¼ B6TXW-4P1G9P3-4&_user ¼ 1314101&_rdoc ¼ 1 &_fmt ¼ &_orig ¼ search&_sort ¼ d&view ¼ c&_acct ¼ C000052297&_version ¼ 1 &_urlVersion ¼ 0&_userid ¼ 1314101&md5 ¼ 98bf837e49518cd07ecda1525eb0fb24 Kaursoin, J. and Agrawal, A.K. (2007), “Melt spun thermoresponsive shape memory fibers based on polyurethanes: effect of drawing and heat-setting on fiber morphology and properties”, Journal of Applied Polymer Science, Vol. 103 No. 4, pp. 2172-82, available at: www3. interscience.wiley.com/journal/113475155/abstract?CRETRY ¼ 1&SRETRY ¼ 0 Khonakdar, H.A., Jafari, S.H., Rasouli, S., Morshedian, J. and Abedini, H. (2007), “Investigation and modeling of temperature dependence recovery behavior of shape-memory crosslinked polyethylene”, Macromolecular Theory and Simulations, Vol. 16 No. 1, pp. 43-52, available at: www3.interscience.wiley.com/journal/114079355/abstract? CRETRY ¼ 1&SRETRY ¼ 0 Razzaq, M.Y., Anhalt, M., Frormann, L. and Weidenfeller, B. (2007), “Mechanical spectroscopy of magnetite filled polyurethane shape memory polymers”, Materials Science and Engineering, Vol. 471 Nos 1-2, pp. 57-62, available at: www.sciencedirect.com/ science?_ob ¼ ArticleURL&_udi ¼ B6TXD-4NB2WNM-8&_user ¼ 1314101 &_rdoc ¼ 1 &_fmt ¼ &_orig ¼ search&_sort ¼ d&view ¼ c&_acct ¼ C000052297&_version ¼ 1 &_urlVersion ¼ 0&_userid ¼ 1314101&md5 ¼ 83e5507abff62fa46993c66e71f464dc Wornyo, E., Gall, K., Yang, F.Z. and King, W. (2007), “Nanoindentation of shape memory polymer networks”, Polymer, Vol. 48 No. 11, pp. 3213-25, available at: www.sciencedirect. com/science?_ob ¼ ArticleURL&_udi ¼ B6TXW-4N85B80-7&_user ¼ 1314101&_rdoc ¼ 1&_fmt ¼ &_orig ¼ search&_sort ¼ d&view ¼ c&_version ¼ 1&_urlVersion ¼ 0&_userid ¼ 1314101&md5 ¼ 90ba6e9c093baa80becb0a9abcb7f37a Zhang, C.S. and Ni, Q.Q. (2007), “Bending behavior of shape memory polymer based laminates”, Composite Structures, Vol. 78 No. 2, pp. 153-61, available at: www.sciencedirect.com/ science?_ob ¼ ArticleURL&_udi ¼ B6TWP-4HD8BHG-1&_user ¼ 1314101&_rdoc ¼ 1 &_fmt ¼ &_orig ¼ search&_sort ¼ d&view ¼ c&_version ¼ 1&_urlVersion ¼ 0 &_userid ¼ 1314101&md5 ¼ 79e28dead4c5b15a12f299aa0faff99a Corresponding author Lihuan Zhao can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

PTT shape memory fabric

73

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

IJCST 21,2/3

82

Electro-conductive sensors and heating elements based on conductive polymer composites V. Koncar, C. Cochrane, M. Lewandowski and F. Boussu GEMTEX Laboratory, ENSAIT, Roubaix, France, and

C. Dufour IEMN, University of Lille 1, Villeneuve d’Ascq, France Abstract Purpose – The need for sensors and actuators is an important issue in the field of smart textiles and garments. Important developments in sensing and heating textile elements consist in using non-metallic yarns, for instance carbon containing fibres, directly in the textile fabric. Another solution is to use electro-conductive materials based on conductive polymer composites (CPCs) containing carbon or metallic particles. The purpose of this paper is to describe research based on the use of a carbon black polymer composite to design two electro-conductive elements: a strain sensor and a textile heating element. Design/methodology/approach – The composite is applied as a coating consisting of a solvent, a thermoplastic elastomer, and conductive carbon black nanoparticles. In both applications, the integration of the electrical wires for the voltage supply or signal recording is as discreet as possible. Findings – The CPC materials constitute a well-adapted solution for textile structures: they are very flexible, and thus do not modify the mechanical characteristics and general properties of the textile structure. Research limitations/implications – In the case of the heating element, the use of metallic yarns as electrodes makes the final structure a more rigid. This can be improved by choosing other conducting yarns that are more flexible, or by developing knitted structures instead of woven fabrics. Practical implications – The CPC provide a low cost solution, and the elements are usually designed so as to work with a low voltage supply. Originality/value – The CPC has been prepared with a solvent process which is especially adapted to flexible materials like textiles. This is original in comparison to the conventional melt-mixing process usually found in literature. Keywords Actuators, Electrical conductivity, Textiles, Intelligent sensors, Electrical heating elements Paper type Research paper

International Journal of Clothing Science and Technology Vol. 21 Nos 2/3, 2009 pp. 82-92 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910933808

Introduction The term “smart textile” describes a class of apparel that has active functions in addition to the traditional properties of clothing. These innovative functions or properties are obtained by utilizing special textiles or electronic devices, or a combination of the two. Thus, a sweater that changes colour under the effect of heat could be regarded as smart textile. Amongst other technical elements or building blocks for integration in smart textile structures, sensors and actuators are very important. Actuators can be used as a heating element in applications where there is a need for thermal comfort or warmth, while sensors that are truly flexible and adapted to textiles structures may be used in communicative textile materials or aeronautical applications. Clothes could be used for

example to detect different actions, in particular the recognition of posture of the driver and passengers and their gestures in order to facilitate certain commands that are intuitive. Moreover, when these sensors and actuators are associated with computing and with the control unit, they may allow the recognition of situation and context for a better interpretation of reality. Sensors integrated in textile structures could also be used as psychological sensors for various parameters. This term refers to the sensors used to record health or personal parameters in a broad sense and to alert if necessary. The applications rising from the use of these sensors are numerous. We can, for example, use sensors to provide a physical performance analysis of a driver or even a car racer pilot or to conduct elder people’s medical follow-up in real time. A conductive polymer composite (CPC) based on carbon black particles has been chosen in this work to develop electro-conductive elements. The characteristics of such a material are particularly interesting for textile applications: they are flexible, lightweight and have a good general mechanical resistance. This paper focuses on two applications: a strain sensor for textile fabrics and a textile heating element. Textile sensors A sensor based on a thermoplastic elastomer (Evoprene, Alpha Gary, Pineville, NC, USA)/carbon black nanoparticle composite has been developed. This sensor presents general mechanical properties that are strongly compatible with any flexible and soft material such as a textile structure. It offers a great potential for use in clothing or home textiles. To improve the precision and reliability of these sensors, it is important to know the influence of external parameters such as rate of strain deformation, temperature or humidity on the sensor response and on its mechanical behaviour. The design and calibration of the sensor on a Nylon fabric have been demonstrated in previous papers (Cochrane et al., 2007, 2006a, b), which also describe in full details the sensor fabrication and optimization. The CPC of which is made the sensor material is prepared via a solvent process. Carbon black particles (Printex L6, Degussa, Frankfurt, Germany) are dispersed with a thermoplastic elastomer, Evoprene 007 from Alpha Gary, which is a styrene-butadiene-styrene co-polymer, in chloroform (Aldrich, Milwaukee, WI, USA). The textile substrate is then coated with the blend and left to dry at room temperature. The final concentration of the carbon black particles is 27.6 vol. per cent in the polymer matrix, after total evaporation of the solvent. This concentration is greater than the percolation threshold to ensure a good conducting network throughout the composite. Finally, a protective latex film is deposited on the sensor. Figure 1 shows schematically the dimensions of the sensor on the fabric which is a light Nylon fabric with a mass per unit area of 42 g/m2. The linear density of the weft and warp yarns is equal to 3.3 Tex. The final sensor system (CPC þ latex film) has the following dimensions: 10 mm £ 110 mm £ 46 mm. The sensor track is parallel to the weft direction of the fabric (sample dimensions: 50 £ 300 mm). Stainless steel yarns used as electrical connections are added at the ends of the sensor track in order to measure its electrical response (Figure 2). The electromechanical calibration of the sensor on the fabric sample has been performed on a tensile testing machine (MTS 2/M). The sensor track is positioned parallel to the direction of fabric extension, and at the centre of the test specimen (fabric test length ¼ 200 mm). Different rates of extension, 16, 200 and 500 mm/min,

Conductive polymer composites 83

IJCST 21,2/3

84

were applied to different samples, and the electrical resistance of the sensor was recorded during elongation until the sample breaks. Since the different sensor specimens will necessarily present slight variations in their intrinsic resistance values, a normalized relative electrical resistivity is defined to characterize the sensor’s electrical property (with Rr ¼ DR/R): ðR 2 Ri Þ Rr ¼ ð1Þ Ri where Ri is the initial resistance of the sensor measured before extension, and R the resistance at a certain length l of the sample. The sample is left at rest for about 1 min before beginning the test, and Ri represents the mean value of resistance measured during this lapse of time. The electrical resistance was measured until the breaking up of the fabric which occurs at around 45 per cent elongation. In general, textile applications such as weaving, the elongation zone of interest is within the 0-10 per cent range. An example of the electrical behaviour of the sensor in this zone is given in Figure 3. To be able to compare our gauge sensitivity to existing strain gauges, a first assumption of a linear variation was made. In this case, the gauge factor (K) is defined as follows: K¼

Conductive track (16 µm)

ðDR=RÞ 1

ð2Þ

10 mm 2 mm

Figure 1. Schematic representation of the structure and dimensions of the sensor integrated on the fabric

Figure 2. Set-up of sensor on fabric, with electrical connections

Latex film (30 µm)

Nylon fabric (45 µm)

Conductive polymer composites

Rr 4 3.5

Model

Data

3

85

2.5 2 1.5

Figure 3. Dependence of relative electrical resistivity of sensor on elongation 1, at an extension rate of 16 mm/min

1 0.5 ε(%)

0 0

2

4

6

8

10

In the elongation range (0-10 per cent), K is greater than 30. In comparison, the K coefficient of a classic metal gauge (copper-nickel) is 2.1. Moreover, a classic metal gauge has a range of elongation between 0.1 and 0.5 per cent. The linear relationship (equation (2)) applied to our sensor is however not very satisfying. In fact, the electromechanical behaviour of the sensor can be better modelled by the following equation: DR ð3Þ ¼ a1 b R Here a represents the sensitivity of the sensor, while b relates to the deviation from the linear model, in which ideal case b ¼ 1 and a ¼ K. The good fitting of this model is illustrated in Figure 3 by the solid line. In this case, a ¼ 99.9 and b ¼ 1.51. The two parameters depend on the intrinsic characteristics of the sensor (initial resistance, length to width ratio, composite quality etc.) but also on the elongation rate. Table I gives the values of a and b obtained at other strain rates. In all cases, there is good accordance between the experimental data and the model (equation (3)). The influence of climatic conditions on sensor behaviour was next studied. The tests were carried out in a climatic chamber (Excal 2221-HA, Climatsw, St Me´dard d’Eyrans, France). The samples mounted on a metallic grid are placed at the centre of the chamber (Figure 4). The electrical resistance R of each sensor is recorded by an ohmmeter during the experiment. In the aim of studying separately the two parameters, temperature T and relative humidity RH, two types of tests have been performed in which one of the parameter is

a b

16

Elongation rate (mm/min) 200

500

99.9 1.51

200.8 1.69

264.2 1.79

Table I. Elongation rates

IJCST 21,2/3

86

Figure 4. Set-up of samples in climatic chamber

kept constant while the other one undergoes an cycle. In the iso-humidity cycle, RH is kept constant, and T is increased from 10 to 508C and decreased back to 508C, in successive 108C steps. At each step, the temperature is maintained constant for 150 min, time during which the resistance of the sensor was recorded. In the isothermal cycles, T is kept constant while RH undergoes a cycle from 20 to 90 per cent in increment steps of 10 per cent, and again with a 150 min equilibrium time at each step. At the beginning of each cycle, a reference value of electrical resistance R0 is measured at 208C/40 per cent RH. As in the electromechanical characterization (equation (1)), the resistance is expressed as a normalized relative resistance Rrc: Rrc ¼

ðR 2 R0 Þ R0

ð4Þ

For each cycle experiment, two values of Rrc are measured for each T/RH couple, one during the upward direction, and the second during the downward direction of the cycle. The main results are summarized in Figures 5 and 6, for the upward cycles only. 20

Rrc (%)

15

a b c

10 5

d

0 0 –5

20

40

60

80 HR (%)

–10 –15

Figure 5. Isothermal cycles

–20 –25

a: 20˚C b: 30˚C c: 40˚C d: 50˚C

Rrc (%) 20

HR (%)

15

a 80

10

b 70

5

c 60 a

0 12 –5 –10 –15 –20

Conductive polymer composites

b c

17

d e

22

27

32

37

42

47 T (˚C)

d 50

87

e 40 f 32

f g h

g 20 h 12

–25

The first observation in the isothermal cycles (Figure 5) is that the same behaviour is obtained at all temperatures. Rrc increases with RH, with somewhat a sharper dependence for RH lower than 35 per cent (sensitivity around 0.5 per cent/per cent RH) and higher than 55 per cent (sensitivity around 1.2 per cent/per cent RH). In between these two values, the resistance does not change much, and the sensor sensitivity amounts to only 0.2 per cent/per cent RH. There are very few studies found in literature on the influence of humidity on conductive polymer resistance. The only study is about sensors used to measure hygrometry and in this particular application, the electrical resistance varies linearly with RH (Barkauskas, 1997; Barkauskas and Vinslovaite, 2003). The dependence of electrical resistance on temperature varies for different RH values (Figure 6), but the main result is that the variations of Rrc are relatively small (0.33 per cent/8C) compared to the sensor’s electrical sensitivity to strain (increase of Rr ¼ 3 per cent/per cent strain). Despite the fact that the carbon conducts electricity following a “metallic” behaviour, in the sensor, the carbon particles are surrounded by a polymer matrix, an insulator. RH and T probably have an impact on the molecular arrangement in the polymer, leading to a modification in the arrangement of the conductive carbon particles (“de-percolation”). Textile actuators Heating textiles can find an application in numerous and varied fields such as sports and leisure, medical, or automotive (Droval et al., 2005; El-Trantawy et al., 2002; Kirkpatrick, 1973). The situation is quite complex with actuators because they need important power supplies that are rarely flexible and lightweight. The heating element developed and presented in this work is designed to be specifically adapted to flexible structures. It comprises a plain woven textile element in which are integrated electrodes composed of metallic yarns woven or stranded directly in the fabric in a comb architecture arrangement. These electrodes are connected to a power supply. In order to ensure a uniform heat distribution, a thin conductive coating is applied on the fabric surface and electrode arrangement. The coating is again based on a carbon black CPC. The heating element can thus be shaped in a desired pattern, by choosing specific

Figure 6. Isohumidity cycles

IJCST 21,2/3

88

dimensions of electrode structure and area of coating layer. Moreover, the whole assembly can be placed in selected areas depending on the application. The electrical resistance of the heating element (electrodes and conductive coating, Figure 7) can be calculated. The two stainless steel yarns parallel to the warp direction are used to supply the power that is necessary to heat the fabric and also create – together with the stainless steel yarns on the weft direction – a comb structure. This comb structure electrode design is made to create a parallel resistance (decreases the resistance).Considering that the element is composed of a number of parallel resistances, the distance between the comb teeth being constant: number of parallel resistances ¼ L/lp, with L, conductive coating width; lp, distance between two adjacent teeth. The total resistance is thus: Rtot ¼

R r £ ðlp=SÞ r £ lp 2 ¼ ¼ LS L=lp L=lp

ð5Þ

where l, comb teeth superposition width; S, coating cross-section area; e, coating thickness; r, coating resistivity. The coating resistivity depends on the amount of carbon black loading in the composite. It can be predicted by the following relation:

r ¼ r0ðV 2 CpÞ2t

ð6Þ

where r0, constant; V, vol.% of carbon black; t, critical index; Cp, concentration of carbon black (vol. per cent) at percolation threshold. The textile fabric has been made on a hand weaving machine (ARM, electronic control Selectron). A plain weave has been chosen, and the warp and weft are made of cotton yarns with warp and weft densities of, respectively, 27 and 10 yarns/cm. For the electrodes, stainless steel 2-ply yarns (Cre´afibres, France) are used, each ply consisting of 275 filaments. The overall yarn count is 500 Tex, with a resistivity of 14 ohm/m. The distance between the teeth of the comb (lp, Figure 5) is 0.7 cm, while the distance between the two parallel steel yarns in the warp direction is around 7 cm. The steel yarns are introduced manually during the weaving process according to the pattern in Figure 7. l

Conductive coating

Warp direction

L

Combstructure electrode lp

Figure 7. Pattern of heating element

Cotton fabric

The conductive carbon black coating composite was made with the following ingredients: a synthetic rubber latex solution (Kraton IR-401, Kraton Polymers), which is an anionic dispersion of polyisoprene, a dispersing agent (Disperbyk-2010, SPCI) and carbon black particles (Printex L6, Degussa). The preparation procedure is as follows: the dispersing agent is put into water and the CB particles are gradually added while mixing continuously. A paste is obtained and left to rest for some time. The polymer is finally added while mixing gently in order to avoid too strong shearing. The coating was then applied on the fabric with a magnetic coating table equipped with a magnetic bar as scraper. The thickness of the coating was however hard to adjust with this technique. Moreover, the quantity of coating deposited depends on the viscosity of the solution, which itself depends on the amount of water in the coating solution. If the viscosity is different, even if the same thickness is deposited on the fabric, the final quantity of composite will differ since the absorption of the coating will be different. Hence, the final resistance will also be different. After drying at room temperature, the resistance of the actuator was measured with an ohmmeter, and was found to be equal to 12 ohm. The fabric with the electrodes is shown in Figure 8, before and after application of conductive coating. The final coated surface is approximately 5 £ 7 cm. The thermo-electrical properties of the element were studied. The electrical resistance of 3 £ 3 cm coated samples with various carbon black contents was measured. It was found to depend of course, on the CB loading level: the higher the amount of CB, the lower the resistance. The optimum CB loading is thus around 35-45 wt per cent. A coating containing 37.5 wt per cent was used for the heating property study. The coated fabric was connected to an ammeter and a power supply according to Figure 9. It is important to note that the sample temperature may change in function of the time even when applying the same voltage. Because of this, the temperature of each sample was measured for 60-80 min for a given voltage. The temperature of the sample surface was recorded every 5 min with a temperature sensor (Testow 925). The temperature was not the same in every point of the heating element, and the highest temperature was obtained at the centre of the sample, but the variation was not very important.

Conductive polymer composites 89

Figure 8. Fabric with heating element: before and after coating

IJCST 21,2/3

Heating fabric A

90 V

Figure 9. Power circuit with fabric heating element, and temperature sensor

T (˚C)

The results are presented graphically in Figure 10. The measurement was made at room temperature (228C). The increase in temperature is important, about 308C, from an initial temperature of 228C, for the highest power supply (6 W). The heating process was found to be quite rapid; the time to reach the highest temperature was about 10 min. The increase of heating efficiency requires a quite important percentage of carbon black, more than 35 wt per cent. It was found that the temperature was not homogeneous over the whole coated surface, the sample being hotter at the centre than at the corners. So this point has to be improved, maybe by modifying the pattern design of the electrodes. The heating is moreover limited to the coated area; another possibility using intelligent design can result in the dissipation of the excess heat generated to other parts of the fabric. After coating with the CB solution, the woven samples became quite rigid. Other studies have therefore to be done to obtain a more flexible final structure, and we have

Figure 10. Temperature of heating fabric as a function of power supply

0

1

2

3 4 Power (W)

5

6

7

extended this research work to knitted structures (Lewandowski et al., 2007). The steel yarns are integrated inside the knitted fabric, and this type of textile structure being intrinsically more deformable that woven structures, the final heating element will be normally more flexible. Conclusion Different building blocks (sensors and actuators) have been presented in order to show development and integration possibilities of smart textile structures. CPCs prepared via the solvent process constitute a very interesting solution to design electro-conductive materials for flexible structures such as textile fabrics and garments. This work has shown their potential use in two applications: sensors and heating elements. For the first application, the CPC sensor is used to measure the strain deformations of a lightweight Nylon fabric. The influence of temperature and relative humidity on the electrical properties has been investigated in order to predict sensor behaviour in different environmental conditions. In the heating application, there exist non textile conventional sensors and actuators based on wire heating technologies that may be integrated in textile structures, but they are often not appropriate because of their rigidity, and may also present hot spots. The solution proposed uses a thin CPC conductive coating applied on the fabric surface and electrode arrangement in order to ensure a uniform heat distribution. The heating performance obtained is interesting but fabric feel and flexibility have to be improved. A wide range of applications exist for the two cases studied. Sensors based on conductive composites may be used for example in the design of car interior: to detect fabric deformation, passenger presence or to evaluate safety belts ageing. They also may be used to control air bags deployments (Jayaraman et al., 2001). On the other hand, heating fabrics are suitable for car seats or winter clothes. The advantage of our solution is that it is low cost, but with a long life time and a good reliability. The other advantage is that the same actuator may be used as a sensor helping the temperature control and regulation. References Barkauskas, J. (1997), “Investigation of conductometric humidity sensors”, Talanta, Vol. 44, pp. 1107-12. Barkauskas, J. and Vinslovaite, A. (2003), “Investigation of electroconductive films composed of polyvinyl alcohol and graphitized carbon black”, Materials Research Bulletin, Vol. 38, pp. 1437-47. Cochrane, C., Koncar, V. and Dufour, C. (2006a), “Cre´ation d’un capteur d’allongement souple, compatible textile”, paper presented at Mate´riaux 2006 Conference, Dijon, 13-17 November. Cochrane, C., Koncar, V. and Dufour, C. (2006b), “Nanocomposite material sensors for textiles”, paper presented at AUTEX 2006 World Textile Conference, Raleigh, NC, 11-14 June. Cochrane, C., Koncar, V., Lewandowski, M. and Dufour, C. (2007), “Design and development of a flexible strain sensor for textile structures based on a conductive polymer composite”, Sensors, Vol. 7, pp. 473-92. Droval, G., Glouannec, P., Feller, J.F. and Salagnac, P. (2005), “Simulation of electrical and thermal behaviour of conductive polymer composites heating elements”, Journal of Thermophysics and Heat Transfer, Vol. 19 No. 3, pp. 375-81.

Conductive polymer composites 91

IJCST 21,2/3

92

El-Trantawy, F., Kamada, K. and Ohnabe, H. (2002), “In situ network structure, electrical and thermal properties of conductive epoxy resin-carbon black composites for electrical heater applications”, Materials Letters, Vol. 56, pp. 112-26. Jayaraman, S., Rajamanickam, R., Gopalsam, C. and Park, S. (2001), “Method and apparatus for controlling air bag deployment”, US Patent, 6254130. Kirkpatrick, S. (1973), “Percolation and conduction”, Reviews of Modern Physics, Vol. 45 No. 4, pp. 574-88. Lewandowski, M., Koncar, V., Cochrane, C. and Giraud, S. (2007), “Electro-conductive heating textile elements based on a carbon black conductive polymer composite”, paper presented at Intelligent Textiles and Mass Customisation International Conference, Casablanca, 15-17 November. Corresponding author V. Koncar can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

New bioactive textile dressing materials from dibutyrylchitin

Textile dressing materials from DBC

Gustaaf Schoukens and Paul Kiekens Department of Textiles, Faculty of Engineering, Ghent University, Ghent, Belgium, and

93

Izabella Krucinska Department of Textile Metrology, Technical University of Lodz, Lodz, Poland Abstract Purpose – Dibutyrylchitin (DBC) is an ester derivative of a natural polysaccharide – chitin. DBC is obtained by reaction of chitin with butyric anhydride in the presence of a catalyst. The production methods of DBC have been elaborated and optimized. DBC is easily soluble in common organic solvents and has film – and fibre forming properties. Such characteristics allow obtaining classical fibres from the polymer solutions. DBC is also a raw material for manufacturing yarn and for a broad range of textile dressing materials. Fibres with good mechanical properties are obtained by an optimized spinning process from the DBC solutions. The purpose of this paper is to present a further optimization of the mechanical properties of DBC-fibres and yarns. Design/methodology/approach – The excellent biomedical properties of the DBC are confirmed by different experimental results which prove that DBC is a biocompatible and biodegradable polymer and stimulates regeneration of damaged tissues. Tests of these DBC dressing materials under clinical conditions prove the excellent results of DBC-based dressing materials for the ordered healing of tissues and wounds. The DBC dressing materials accelerate the healing of the wound and are biodegraded during the healing process. From the clinical tests, it can be clearly observed that the DBC dressing materials are absorbed into the fresh tissue formed during the healing process of the wounds. Findings – The DBC and DBC-based dressing materials are good bioactive textile materials for wound healing and for understanding the biological properties of chitin derivatives. The obtained results prove the importance of the O-substitution of the hydroxyl groups present in chitin, not only for the solubility of the derivatives and the mechanical properties of the produced fibres, but still more important for the biological properties of these ester derivatives of chitin containing butyric acid. This development creates a link between textile products, based on material properties and human health, based on the biological properties of the basic material. Originality/value – The mechanical properties of DBC are further optimized by blending it with poly(1-caprolactone). Good transparent and flexible products, such as films, with a high elongation to break are obtained by blending 10-20 wt per cent of poly(1-caprolactone) with DBC. This creates new possible bioactive applications for DBC or poly(1-caprolactone). Keywords Polymers, Mechanical properties of materials, Medical sciences, Injuries Paper type Research paper

Introduction Chitin, the natural most abundant polysaccharide containing nitrogen, undergoes degradation by enzymes, induces human cells to promote the restoration of wounds, enhance and enhances the healing process of wounds and has high permeability for substances of medium molar masses in serum (Oshimoma et al., 1987; Muzzarelli, 1993; This work has been partly supported by the European Commission as part of the Project CHITOMED, QLK5-CT-2002-01330.

International Journal of Clothing Science and Technology Vol. 21 Nos 2/3, 2009 pp. 93-101 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910933817

IJCST 21,2/3

94

Muzzarelli, 1996). Chitin might be an ideal raw material for bioactive dressing materials, but due to its low solubility can be only hardly converted into useful forms like films, fibre or non-woven matrices. An original method of synthesis of di-O-butyrylchitin (DBC), the soluble derivative of chitin, was worked out at the Technical University of Ło´dz´, Poland (Szosland, 1996). The proposed method applied to chitin of different origin (crab, shrimp & krill shells and insect chitin) gave the products of definite chemical structure with a degree of esterification very close to two. DBC is easily soluble in common organic solvents and has both film and fibre-forming properties (Szosland and Ste˛plewski, 1998). Such properties of DBC created the possibility of manufacturing a wide assortment of DBC materials suitable for medical applications in the form of films, fibres, non-woven, knitted materials and woven fabrics. Thus, O-butyrylation of chitin gives the possibility of practically unlimited manufacturing of DBC dressing materials comfortable and easy in use. The first investigations of biological properties of DBC materials, carried out in vitro and in vivo in accordance with the European standards EN ISO 10993 (“Biological evaluation of medical devices”), showed good biocompatibility of DBC (Szosland et al., 2002, 2001) and his ability to accelerate wound healing (Szosland et al., 2002; Pelka et al., 2003). The recent investigations published by Muzzarelli et al. (2005) confirms the biocompatibility of DBC. The presented results indicated that DBC is not cytotoxic for fibroblasts and keratinocytes. The first clinical investigations into medical properties of DBC have been carried out at the Polish Mother’s Health Institute in Ło´dz´, Poland. DBC samples under investigation have been used in the form of non-woven materials. Results of their clinical investigations were presented on the 6th International Conference of the European Chitin Society and published (Chilarski et al., 2004). Manufacturing of wound dressings Preparation of dibutyrylchitin (DBC) and DBC characteristics DBC as a raw material used for manufacturing of non-woven dressing materials was prepared from shrimp chitin. Shrimp shell chitin powder with particle size of 200 mesh and viscosity average molar mass Mv ¼ 454.6 £ 103 g/mol was delivered by France Chitin, Marseille, France (Figure 1). O O

O O

C C3H7

O

C3H7

O

O NH

O

Figure 1. DBC: chemical structure after complete O-butyrylation

C

O C

C3H7

CH3

C C3H7

O

O NH

O O

O O

C

C

O

C3H7

Note: Degree of substitution = 2

C

O NH

O O

CH3

C

O

C3H7

C

O

CH3

The synthesis of DBC was carried out under heterogeneous conditions using chitin, butyric anhydride and 72 per cent perchloric acid in approximate proportion equal to 10:50:6.8 (g/g). The intrinsic viscosity value of DBC determined in DMAc solutions at 258C was 1.70 dl/g, the weight average molar mass, determined by SEC method coupled with light scattering and viscometry, was Mw ¼ 132 £ 103 g/mol. Another possible reaction to produce DBC is the so-called homogeneous reaction in methanesulphonic acid. Intrinsic viscosity value of the obtained DBC by the homogeneous reaction determined in DMAc solutions with DBC was 5.70 dL/g, the corresponding weight average molar mass was 456 £ 103 g/mol. A DBC with a higher molecular weight was obtained by the homogeneous reaction in methanesulphonic acid. The starting chitin has a mean amount of 2,800 monomers in the polymer chain. The mean amount of monomers for the DBC obtained by the homogeneous reaction is 1,340 and for the DBC obtained from the heterogeneous reaction is 450. There is less degradation of the polymer chains for the homogeneous reaction than for the heterogeneous reaction.

Textile dressing materials from DBC 95

Mechanical properties The mechanical properties were measured on films cast from an acetone solution. The elasticity modulus, elongation and maximum stress were obtained from the tensile stress-strain data on the films using an Instron 3369 tensile tester. The product TUL-DBC was the DBC with an average molecular weight Mw of 153 £ 103 g/mol and UGT-DBC the DBC with an average molecular weight Mw of 456 £ 103 g/mol. The films were conditioned at room temperature and a relative humidity of 65 per cent. The measured stress-strain curves are reproduced in Figure 2. The elasticity modulus, equal to 1,025 MPa, was the same for the two samples, independent of the molecular weight, but the maximum elongation is increasing with increasing molecular weight. The maximum elongation is increasing from 4.3 per cent for the lower molecular weight to 12.5 per cent for the higher molecular weight. This resulted in better strength characteristics of the films, due to the more plastic behaviour of the DBC with the highest molecular weight.

TENSILE TESTS ON DBC FILMS: AVERAGE CURVES

30 25 Stress (MPa)

TUL-DBC cast from acetone 20 UGT-DBC cast from acetone 15 10 5 0 0

3

6 9 Strain (per cent)

12

15

Figure 2. Stress-strain data for two different DBCs at room temperature

IJCST 21,2/3

96

Preparation of dibutyrylchitin fibres Wet spinning of DBC fibres was made at the Institute of Chemical Fibres, Ło´dz´, Poland, on an apparatus commonly used for preparation of rayon fibres. Dopes containing 15 per cent of DBC from krill chitin in dimethylformamide or 12 per cent of DBC from shrimp chitin in dimethylsulphoxide were added to the reservoir of the spinning system and extruded through a spinneret (300 holes, 80 mm diameter of the hole) to a coagulation bath. The filaments were coagulated in water, drawn in hot water, collected on rollers with a rate of 40 m/min and dried (Table I). Low susceptibility to deformation during the drawing stage at about 708C is linked to the rigid structure of the DBC macromolecules. The results of this are rather poor strength properties in the fibres, especially the maximum elongation for bending. By using ethanol as the solvent for DBC, porous fibres were obtained. The presence of very large pores in the cross-section view is visible in the SEM micrograph (Figure 3). The DBC fibres from DMSO-solution were round without visible pores. These tests confirmed that the supramolecular structure of fibre-forming polymers depends upon the solvent and the character of the solvent-polymer interactions. Blends of DBC and poly(1-caprolactone) Blending natural polymers with chemosynthetic polymers is a very interesting methodology by which to modify the properties of natural polymers and to develop novel composite materials based on natural polymers. Blending may be used effectively to modify physical and mechanical properties that each individual polymer does not have. The extent and rate of biodegradation of polymer blends are determined not only by the

Fibre’s symbol

Table I. Dibutyrylchitin fibre

Figure 3. SEM photos of DBC Fibres (solvent-ethanol); cross-section

8 E/1 22/3 35/1 68

Dtex

Solvent

Tenacity (cN/tex)

Loop (cN/tex)

Elongation at break (%)

Maximum elongation (loop; %)

3.3 2.1 2.8 2.9

Ethanol DMF N-MP DMSO

6.5 15.7 13.0 11.2

2.9 3.1 2.2 3.5

4.8 6.2 9.4 12.6

2.1 2.3 3.4 2.6

intrinsic degradability of the blend compositions themselves, but also by the blend composition, phase structure (miscibility, crystallinity of the components) and surface blend composition. In particular, hydration plays an important role in determining polymer degradation via hydrolysis of ester or b-glycosidic bonds (Park et al., 1992). Thus, the blending strategy is an important technique for biodegradable polymers, such as chitin, chitosan and DBC, not only to improve the properties of the components, but also to control the biodegradation profile. In the last two decades, numerous biodegradable polyesters have been developed chemosynthetically. Among them, aliphatic polyesters including poly(1-caprolactone) (PCL), poly(glycolic acid), poly(lactic acid) and their copolyesters, have considerable importance in several medical applications. Recently, blends of PCL and chitin or chitosan were studied. DSC thermal analysis revealed that the crystallization of PCL was suppressed by blending with chitin and chitosan (Wu, 2005). All film samples were prepared by the solution-casting technique, using methylene chloride as the common solvent. PCL and DBC were dissolved separately in CH2Cl2 before blending. After the solutions had been homogenized, there were mixed. The blend films were prepared by casting the mixture on an aluminium dish, and then dried at room temperature for 2 days. Throughout this paper blend compositions are given in wt per cent of PCL. Weight ratios of PCL to DBC were fixed at 100/0, 80/20, 60/40, 50/50, 40/60, 20/80 and 0/100. The mechanical properties were measured on films. The elasticity modulus, elongation and maximum stress were obtained from the tensile stress-strain data on the films using an Instron 3369 tensile tester. The results of the elasticity modulus and elongation at break as function of the compositions of the blends are reproduced in Figures 4 and 5. The elasticity modulus of the blend containing 20 wt per cent PCL decreases from 1,025 to 770 MPa and the elongation to break increases from 4.3 to 47.6 per cent. In the DSC-measurements, no crystallinity of PCL is observed. This is an indication that the PCL is good miscible with DBC and behaves as a plasticizer in those blends.

Textile dressing materials from DBC 97

1600 1400

E-modulus (MPa)

1200 1000 800 600 400 200 0

0

20

40 60 PCL (wt%)

80

100

Figure 4. The elasticity modulus of the blends DBC-PCL as function of the composition of the blends

IJCST 21,2/3

Figure 5. The elongation at break of the DBC-PCL blends as function of the composition

Elongation at break (percent)

98

1,000

100

10

1 0

20

40 60 PCL (wt%)

80

100

The blends containing between 10 and 20 wt per cent PCL are plasticized by PCL, transparent and characterized by a high elongation to break. A good flexibility of the DBC-derived films and fibres can be obtained by adding 10-20 wt per cent of PCL to DBC. Clinical investigations of DBC non-woven dressing materials Rationale for the possible provision of butyrate Short chain fatty acids, especially butyrate, play central metabolic roles in maintaining the mucosal barrier in the gut. A lack of butyrate, leading to endogenous starvation of enterocytes, may be the cause of ulcerative colitis and other inflammatory conditions. The main source of butyrate is dietary fibre, but they can also be derived from structured biopolymers like DBC. Butyrate has been shown to increase wound healing and to reduce inflammation in the small intestine (Wa¨chtersha¨user and Stein, 2000). In the colon, butyrate is the dominant energy source for epithelial cells and affects cellular proliferation and differentiation by yet unknown mechanisms. Recent data suggest that the luminal provision of butyrate may be an appropriate means to improve wound healing in intestinal surgery and to ameliorate symptoms of inflammatory diseases. It was also suggested that butyrate may inhibit the development of colon cancer (Hinnebusch et al., 2002; Emenaker et al., 2001). Butyrate has a relatively short metabolic half life. The half-life of butyrate in plasma is extremely short, as peak plasma butyrate concentrations occurred between 0.25 and 3 h after application and disappeared from plasma by 5 h after the application. Results of clinical investigations The surgical staff of the Department of Paediatric Surgery was provided with a number of DBC petals for medical application (Figure 6). The first group of patients, composed of 10 persons, suffered from burns. The total area of thermal burns of patients changed within the range of 5-20 per cent. The depth of burns was classified in each case as 2a. All the burns healed up within 1-2 weeks

Textile dressing materials from DBC 99

Figure 6. The flake of DBC

after DBC petal application. The healing processes are documented in Figures 7(a) and (b). The result of the healing process was very good. Conclusions As far as thermal burn patients are concerned, in all cases DBC dressings have been applied to the clean wound and not removed till the end of the healing process, while DBC has been disintegrated in the area of the wound. No other medical products have been applied for the wound healing. The presented observations are preliminary and further evaluation is necessary.

Figure 7. (a) Burn of right lower limb and (b) healing nearly completed (a)

(b)

IJCST 21,2/3

100

Summarising, it is possible to conclude from the preliminary results of DBC application that DBC seems to promote wounds’ healing. Further randomised trials with referential groups should be completed to obtain evidence-based proofs of beneficial effects of DBC wound dressings. Blending DBC with PCL gave very good flexible products, with a good value of elasticity modulus, a high elongation to break and a good transparency for films containing between 10 and 20 wt per cent PCL. References Chilarski, A., Szosland, L., Krucinska, I., Błasinska, A. and Cisło, R. (2004), “The application of chitin derivatives as biological dressing in treatment of thermal and mechanical skin injuries”, The Annual of Pediatric Traumatic Surgery, The Division of Pediatric Traumatic Surgery, Vol. 8, XXXII, pp. 58-61. Emenaker, N.J., Calaf, G.M., Cox, D., Basson, M.D. and Qureshi, N. (2001), “Short-chain fatty acids inhibit invasive human colon cancer by modulating uPA, TIMP-1, TIMP-2, Bcl-2, Bax, p21 and PCNa protein expression in an in vitro cell culture model”, Journal of Nutrition, Vol. 131, pp. 3041S-6S (Supplement 11). Hinnebusch, B.F., Meng, S., Wu, J.T., Archer, S.Y. and Hodin, R.A. (2002), “The effects of short-chain fatty acids on human colon cancer phenotype are associated with histone hyperacetylation”, Journal of Nutrition, Vol. 132, pp. 1012-7. Muzzarelli, R.A.A. (1993), “In vivo biochemical significance of chitin-based medical items”, in Dimitriu, S. (Ed.), Polymeric Biomaterials, Marcel Dekker, New York, NY, pp. 179-97. Muzzarelli, R.A.A. (1996), “Chitin”, in Salamone, J.C. (Ed.), Polymeric Materials Encyclopedia, CRC, Boca Raton, FL. Muzzarelli, R.A.A., Guerrieri, M., Goteri, G., Muzzarelli, C., Armeni, T., Ghiselli, R. and Cornelissen, M. (2005), “The biocompatibilityof dibutyryl chitin in the context of wound dressings”, Biomaterials, Vol. 26, pp. 5844-54. Oshimoma, Y., Nishino, K., Yonekura, Y., Kishimoto, S. and Wakabayashi, S. (1987), “Clinical application of chitin non-woven fabric as wound dressings”, European Journal of Plastic Surgery, Vol. 10, pp. 66-9. Park, T.G., Cohen, S. and Langer, R. (1992), “Poly(L -lactic acid)/pluronic blends”, Macromolecules, Vol. 25, pp. 116-22. Pelka, S., Paluch, D., Staniszewska-Kus´, J., Zywicka, B., Solski, L., Szosland, L., Czarny, A. and Zaczyn´ska, E. (2003), “Wound healing accelerating by a textile dressing containing dibutyrylchitin and chitin”, Fibres & Textiles in Eastern Europe, Vol. 11, pp. 79-84. Szosland, L. (1996), “Synthesis of highly substituted butyrylchitin in the presence of perchloric acid”, Journal of Bioactive and Compatible Polymer, Vol. 11, pp. 61-71. Szosland, L. and Ste˛plewski, W. (1998), “Rheological characteristic of dibutyrylchitin semi-concentrated solutions and wet spinning of dibutyrylchitin fibre”, in Domard, A., Roberts, G.A.F. and Va˚rum, K.M. (Eds), Advances in Chitin Science, Vol. II, pp. 531-6. Szosland, L., Krucin´ska, I., Cisło, R., Paluch, D., Staniszewska-Kus´, J., Solski, L. and Szymonowicz, M. (2001), “Synthesis of dibutyrylchitin and preparation of new textiles made from dibutyrylchitin and chitin for medical applications”, Fibres & Textiles in Eastern Europe, Vol. 9 No. 34, pp. 54-7. Szosland, L., Cisło, R., Krucin´ska, I., Paluch, D., Staniszewska-Kus´, J., Pielka, S., Solski, L. and Z˙ywicka, B. (2002), “Dressings made from dibutyrylchitin and chitin accelerating wound healing”, Proceedings of International Conference MEDTEX’2002, Ło´dz´.

Wa¨chtersha¨user, A. and Stein, J. (2000), “Rationale for the luminal provision of butyrate in intestinal diseases”, European Journal of Nutrition, Vol. 39, pp. 164-71. Wu, C. (2005), “A comparison of the structure, thermal properties, and biodegradability of polycaprolactone/chitosan and acrylic acid grafted polycaprolactone/chitosan”, Polymer, Vol. 46 No. 1, pp. 147-55. Corresponding author Gustaaf Schoukens can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

Textile dressing materials from DBC 101

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

IJCST 21,2/3

The application of microcapsules of PCM in flame resistant non-woven materials

102

Isabel Cardoso Micropolis SA, Braga, Portugal, and

Jaime Rocha Gomes Department of Textile Engineering, University of Minho, Guimara¨es, Portugal Abstract Purpose – The use of organic phase change materials microcapsules (mPCM) has been gaining ground in technical textiles and clothing as a temperature regulating medium and hence a means of keeping the body at a comfortable temperature when wearing impermeable protective clothes. However, for such applications as fire fighter’s protective clothes, the standards require that all the material composing the material be fire resistant. The purpose of this paper is to produce a lining containing fire resistant microcapsules of PCM without using flammable binders. Design/methodology/approach – This work tests other ways of fixing mPCM to the fibres with a lot less binder present. Washfastness is evaluated in SEM photographs and by weight. The thermal effect is evaluated in a prototype plate calorimeter. Findings – This method is first tested for fixing mPCM but the non-woven still does not pass the test according to the standard EN532. Microcapsules are alternatively fixed with MF resin, non-flammable, and by applying flame retardant recipes it is possible for the samples to pass the test. Research limitations/implications – Since the amount of flame retardant necessary for the mPCM to stand the test, and the resin to thermo fix it is very high, the material becomes unacceptably stiff. Originality/value – Based on a new approach where reactive microcapsules without any binder are used, it is possible to use a lot less flame retardant and resin, and the material is resistant to the standard EN532. In this standard the material has to resist washing and still be flame retardant. Keywords Flame retardants, Thermal efficiency, Protective clothing Paper type Research paper

International Journal of Clothing Science and Technology Vol. 21 Nos 2/3, 2009 pp. 102-108 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910933826

Introduction Protection clothes for firefighters are made in a way that they protect them from external heat. The materials used are usually thick and heavy so as to provide the necessary insulation to heat. Whereas they perform as expected during the firefighting, the energy the firefighter spends on the exercise of firefighting and accessory tasks, aggravated by the fact that the firefighting clothes are heavy and non-breathable, makes him experience discomfort and extra stress. The use of organic microcapsules of phase change materials (mPCM) has been gaining ground in technical textiles as a temperature regulating medium and hence as a means of keeping the body at a comfortable temperature when wearing impermeable protective clothes. They work by absorbing the extra energy the body releases, and then giving it back when the body cools down. This exchange of energy happens around the phase change temperature (melting point) of the PCM which is this case is 288C, which is also considered the

comfort temperature for the body. However, for such applications as fire fighter’s protective clothes, the standards require that all the material composing the material be fire resistant. Since PCM are flammable paraffins, they should be microencapsulated in heat resistant material so that they do not become exposed to the flames due to deterioration of the microcapsule shell. The most common material used as shell material is melamine-formaldehyde, since it is heat resistant. It is also in itself a fairly good flame resistant material (Kim and Cho, 2002; Choi et al., 2004), In this work the mPCM had to be applied to a non-woven material made of aramid fibre, which is a flame resistant, in such a way as not to bind it with flammable binders. Most applications of microcapsules are with thermoplastic binders such as acrylic or polyurethane. In a first attempt to avoid this problem, instead of immersing the microcapsules in a binder, the microcapsules were coated with a second wall made of binder material (Su et al., 2007), acrylates, and this second wall was treated with flame retardant products. To fix the microcapsules and the fire retardant a cross-linking resin, also based on melamine-formaldehyde was used, which altered the non-woven material making it too stiff. Other solutions were tested such as the direct binding of the microcapsules with melamine-formaldehyde resins without the use of acrylate thermoplastic binders and the application of alternative fire retardant products. The different fire retardant products tested were of different types, and chosen so as to bind onto the microcapsules or/and bind onto the material. In the first case a halogenated copolymer was used on the coating of the MF mPCM for one trial and a metal oxide was bound by ion exchange to a polymer coated onto the MF mPCM on another trial. For the non coated mPCM, finishing products based on phosphorous were applied on the material containing the microcapsules, Pyrovatex (n-hydroxy-methyl-3-dimethylphosphonopropionamide) and phosphoric acid. To complement the fire retardant treatment, boric acid was first applied to the microcapsules as it could react with the MF wall of the PCM microcapsules. The flame retardancy compared to the non-treated microcapsules was tested according to the standard test EN532 which included washing (five times in washing machine). The washfastness was evaluated by weight measurement, before and after washing. The stiffness of the material was screened by different observers and only those materials which did not alter too much their drape and tactile properties were accepted. The thermal effect was tested in a prototype plate calorimeter. Experimental Equipment The microcapsules’ emulsion was applied in a Werner Mathis laboratory machine, the pressure being pneumatically controlled. Thermal fusion of double walled microcapsules and curing of melamine-formaldehyde resin was carried out in a Werner Mathis drying/curing chamber at 1508C. The calorimeter used was a prototype consisting of two plates through which a heat flux is produced. The top plate is heated to 458C and the bottom plate kept at 258C. The heat sensor in the bottom plate measures the heat flux through the non-woven placed between the plates (Pushaw, 1997). Starting materials . Non-Woven (needled) supplied by Duflot Industries (France). . Fibre: Nomex (aramid).

Application of mPCM

103

IJCST 21,2/3

. .

.

104

.

Mass: 110 g/m2. Microcapsules of octadecane PCM with a Melamine-formaldehyde shell were supplied by Micropolis (Portugal). The melamine-formaldehyde precondensate was supplied by BASF (Lyofix MLF and Kaurit TX). Pyrovatex was supplied by CIBA (Pyrovatex CP new).

Fire retardant microcapsules mPCM coated with PMMA-PPBBA co-polymer Microcapsules were coated with (polymethylmethacrylate) PMMA copolymerized with a fire retardant monomer (poly-pentabromobenzylacrylate) PPBBA by radical polymerisation around the MF microcapsules of PCM (Gomes, 2007). mPCM coated with PMMA-PMA co-polymer treated with zinc oxide A copolymer of PMMA with (polymethacrylic acid) PAM coated in the same way as above onto the MF mPCM, was treated with zinc oxide so as to attach zinc ions onto the carboxylic groups of the PAM polymer. mPCM with boric acid Boric acid was applied to the PCM microcapsules, and the mPCM were subsequentely washed and applied to the non-woven. Results for aramid non-woven containing mPCM Figure 1 shows a SEM photograph of non-woven containing mPCM treated with boric acid shows acid. Thermal effect of non-woven A thermograph for aramid non-woven was taken in the plate calorimeter, first for the sample without mPCM and then for comparison, on the same graph, for the non-woven containing double walled mPCM, fire retardant (Pyrovatex) and melamine-formaldehyde resin.

Figure 1.

Figure 2 shows the thermograph for non-woven containing double walled mPCM. The time the flux takes to reach half the maximum flux was measured. The graph is on Figure 3 and the results for the flux are on Table I and II. Washfastness results Double walled microcapsules (Figure 4). Single walled microcapsules with boric acid (Figure 5).

Application of mPCM

105

Flame-retardancy results The non-woven samples with different types of double walled mPCM were tested according to standard EN532. The results are on Table III.

Figure 2. Thermograph for non-woven containing double walled mPCM Note: Reproduced from the only available original

Tempo (s)

0

Tempo (s)

20

30

40

Fluxo Termico (°C) Note: Reproduced from the only available original

50

60

70

Fluxo Termico (°C)

Figure 3. Thermograph for non-woven containing single walled mPCM treated with boric acid

IJCST 21,2/3

106 Table I. Half-time of flux

Sample

Half-time

Non-woven Non-woven with double walled mPCM

Half-time

Non-woven Non-woven with double walled mPCM

Figure 4. Weight loss of non-woven with double walled mPCM after five washes

100 90 80 70 60 50 40 30 20 10 0

44.5

0

Weight loss (%)

Figure 5. Weight loss of non-woven containing single walled MF mPCM treated with boric acid

13.2 26.2

Sample

Weight loss (%)

Table II. Half-time of flux for non-woven containing singe walled mPCM treated boric acid

The non-woven samples containing a flame retardant mixture of Pyrovatex (n-hydroxy-methyl-3-dimethylpropionamide) þ phosphoric acid and two types of MF mPCM were tested according to standard EN532. The results are on Table IV.

100 90 80 70 60 50 40 30 20 10 0 0

10 23

47.7

50.9

51.9

51.9

1

2

3 No. of washes

4

5

27

29

30

31

32

1

2

3 No. of washes

4

5

6

6

Tactile (handle), drape (rigidity) and other properties of materials Other properties important for the commercial viability of the material, such as tactile properties (handle), adhesion of microcapsules when shaking and applying friction to the material, other noticeable characteristics such as smell, were evaluated by independent observers. The results are on Table V.

Application of mPCM

Discussion of the results Treated double walled microcapsules were not sufficiently flame-retardant to withstand the standard test EN532 as can be seen in Table III and Figure 4. Single walled MF mcirocapsules bound by MF resin and non-woven treated with flame retardancy products, resisted the flame test EN532, as seen in Table IV, but altered considerably the properties of the original non-woven material with respect to handle and rigidity. They also released a strong unpleasant fish odour, due to the large quantity of Pyrovatex needed for withstanding the flame test. When this product was replaced partially by boric acid, these properties improved a lot (Figure 5). Boric acid gave the best results and did not wash-off from the microcapsules nor from the non-woven material. The reason must be that they react with both the microcapsules MF wall and with the MF resin, according to the equation:

107

CH2

CH2

NH N CH2 HN

NH HO

N N

+ NH

CH2OH n

OH B OH

N CH2 HN

Microcapsules

Speed of flame propagation

Double walled mPCM Double walled mPCM with PMBBA Double walled mPCM with zinc

Fast Fast Medium

Microcapsules

Speed of flame propagation

MF mPCM MF mPCM in non-woven with Pyrovatex þ phosphoric acid mPCM with boric acid in non-woven with Pyrovatex þ phosphoric acid

N N

NH

CH2 n

Time of burning (s) .5 .5 3-5

Time of burning (s)

HO – OH + O B OH (1) H

Extent of burning Total Partial Partial

Table III. Flame retardancy double walled PCM

Extent of burning

Fast None

.5 0

Total None

None

0

None

Table IV. Flame retardancy of MF mPCM

IJCST 21,2/3

108 Table V.

Microcapsules

Handle

Rigidity

Non-woven with double walled mPCM treated with zinc oxide MF mPCM in non-woven with Pyrovatex þ phosphoric acid mPCM with boric acid in non-woven with Pyrovatex þ phosphoric acid

Harsh

Medium

Other

Low adhesion of mPCM to fibres Medium Hard Strong odour (fish) Soft Low (high drape) None

Note: Properties of non-woven material containing microcapsules

The washfastness of the single walled MF microcapsules with boric acid, bound by MF resin and with Pyrovatex and phosphoric acid, also gave the best results for washfastness which suggests that there was a reaction between the MF wall of the mPCM, the MF resin and the n-hydroxy-methyl-3-dimethylphosphonopropionamide (Pyrvatex). Conclusions For achieving the flame retardancy standard in a non-woven lining inside firefighters’ protective clothes, of the different technologies tested, it was found that only a conjugation of both microcapsule protection and non-woven flame retardant finishing, was appropriate. It was also found that even the use of small quantities of thermoplastic binder around the microcapsules was sufficient for the material to fail the test. Boric acid previously applied on the mPCM was found to be a good complement to the standard flame retardancy finishing based on phosphorous and applied on the non-woven. In this way tactile and other properties of the original non-woven material were preserved. References Choi, K., Cho, G., Kim, P. and Cho, C. (2004), Textile Research Journal, Vol. 74, pp. 292-6. Gomes, J. (2007), Patent GB2417495, 18 April. Kim, J. and Cho, G. (2002), Textile Research Journal, Vol. 72 No. 12, pp. 1093-8. Pushaw, J. (1997), US Patent US5677048. Su, J., Wang, L., Ren, L. and Huang (2007), Journal of Applied Polymer Science, Vol. 103, p. 1295. Corresponding author Jaime Rocha Gomes can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

Recycled wool-based nonwoven material for decolorisation of dyehouse effluents Maja Radetic, Darinka Radojevic, Vesna Ilic, Darka Mihailovic and Petar Jovancic

RWNM for decolorisation

109

Textile Engineering Department, Faculty of Technology and Metallurgy, University of Belgrade, Belgrade, Serbia Abstract Purpose – The purpose of this paper is to investigate the possible application of recycled wool-based nonwoven material (RWNM) for removal of different dyes that are used in textile dye houses. Design/methodology/approach – The sorption kinetics, the influence of initial dye concentration, pH and temperature are analyzed. Basic, reactive, direct and metal complex dyes are studied. Findings – The sorption properties are highly influenced by the type of the dye owing to differences in their chemical structure and thus, the mechanism of binding to wool. Modification of material with chitosan and hydrogen peroxide improves the sorption capacities and sorption rates but no general trend can be established. Consequently, the sorption behaviour is analyzed separately for each type of the dye. Originality/value – The results indicate that RWNM can be used as an efficient, low-cost sorbent for decolorisation of effluents. Keywords Wool, Dyes, Recycling, Trade effluents, Textile waste processing Paper type Research paper

Introduction Nowadays, textile industry must deal with increasing environmental demands for purification and control of effluents since it generates huge quantities of wastewater. In addition to limitation of BOD, COD, metal ion and some organic compounds in industrial effluents, the request for almost complete decolorisation of wastewater was introduced (Ho¨nings et al., 1995). Therefore, these regulations had strong impact on the textile dyehouses, which produce very complex effluents containing a wide range of dyes and auxiliaries. It is estimated that the BOD of mixed wastes from dyehouses is typically 200-3,000 mg/l, COD 500-5,000 mg/l, suspended solids 50-500 mg/l, with pH in the range 4-12 (Laing, 1991). The amount of dye released in effluent through the exhaust and wash baths strongly depends on the type of dye, depth of shade, dyeing method, liquor ratio, etc. (Laing, 1991). Many efforts have been made to develop treatment or combination of treatments for decolorisation and removal of dyes from wastewater that would meet the demands of The authors gratefully acknowledge the support from European Community FP6 Programme through financing the EMCO project INCO CT 2004-509188 and Ministry of Science and Environmental Protection of the Republic of Serbia for project TD-7017B. This research reflects only the authors’ views and the European Community is not liable for any use that may be made on the information contained therein.

International Journal of Clothing Science and Technology Vol. 21 Nos 2/3, 2009 pp. 109-116 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910933835

IJCST 21,2/3

110

high efficiency and economical feasibility (Tu¨nay et al., 1996; Smith et al., 1993). Dyes are usually very difficult to decolorise because of their complex structure and synthetic origin (Robinson et al., 2001). There is a wide range of dyes differing in chemical structure, which require specific and selective treatments. Dyehouse effluents cannot be aerobically treated in municipal sewerage systems (Willmot et al., 1998). Textile effluent decolorisation treatments are divided into three categories: biological, chemical and physical. Currently, physical and chemical treatments are mainly used for removal of dyes from effluents (Robinson et al., 2001). Adsorption seems to be a viable treatment as it provides removal of dyes that are too stable for conventional methods. It is highly affected by numerous physico-chemical factors such as pH, sorption time, temperature, sorbent/dye interaction, sorbent surface area, particle size, etc. (Rao et al., 1994). Many different sorbents are currently available, but activated carbon in powder and granulated form is the most commonly used (Laing, 1991; Kumar et al., 1998). Although it is extremely efficient for removal of different kinds of dyes, it is also very expensive (Robinson et al., 2001). Thus, many low-cost natural sorbents such as chitin, chitosan (CHT), peat, wood chips, fly ash and coal mixture, corn cobs, silica gel, natural clay, rice hulls, palm kernel fibre, coconut hard shell are studied (Laing, 1991; Robinson et al., 2001; McKay et al., 1984; Carlough et al., 1991; Yoshida et al., 1991; Ofomaja and Ho, 2007; Pathak et al., 2006). It is well documented that wool has excellent sorption properties for metal ions and oils (Maclaren and Milligan, 1981; Choi and Moreau, 1993) but there are only few available data on the sorption properties of wool for dyes (Weltrowski et al., 1995). This study was aimed at investigating the possibility of using the recycled wool-based nonwoven material (RWNM) developed by our group as a sorbent for removal of basic, reactive, metal complex and direct dyes. To improve the sorption properties, the material was treated with biopolymer CHT and hydrogen peroxide. The influence of pH, temperature and initial dye concentration on dye uptake as well as sorption kinetics were studied. Experimental Material The RWNM (78 per cent wool/22 per cent polyester) was produced from secondhand military knitted pullovers of the constant characteristics that were torn off, washed, decolorised with reducing agent, dried and garneted in industry. To avoid the effect of chemical binders on dye sorption, needlepunch process was chosen to produce the nonwoven material. The material was produced from recycled fibres on Dilo (Germany) needle loom (Radetic et al., 2003). CI numbers and commercial names of investigated dyes are given in Table I.

Table I. CI numbers and commercial names of investigated dyes

Commercial name

CI number

Manufacturer

Tubantin blue GLL 300 per cent Maxilon blue TRL Lanaset yellow 4GN Lanaset gray G

CI CI CI CI

Bezema Ciba Ciba Ciba

direct blue 78 basic blue 145 reactive yellow 39 acid blue 317

Treatments The treatment of RWNM with biopolymer CHT was based on immersion of samples in 0.3 per cent CHT (liquor ratio 30:1), which were shaken for 20 min, squeezed out through laboratory squeeze rolls and dried at room temperature. Subsequently, they were washed with tap water and dried at room temperature. The solution of CHT (0.3 per cent) was prepared according to the following procedure: 3.00 g of CHT was stirred in 0.4 per cent acetic acid. Volumetric flask of 1 l was filled up with 0.4 per cent acetic acid and solution left overnight ready for the application to material. Hydrogen peroxide treatment (H2O2, 20 ml/l; Na4P2O7, 1.5 g/l and NH3, 2.5 ml/l) was done in static conditions (without shaking). Samples were treated in the solution for 1 h (liquor ratio 30:1) at 708C and pH 9.40, washed with tap water and dried at room temperature. Methods The dye uptake (q, mg/g) was determined as a difference between the initial concentration of dye in the solution (C0, mg/l) and the final concentration of dye in the solution (Cf, mg/l) (equation (1)). UV/VIS spectrophotometer (Shimatzu 1700, Japan) was used for determination of the concentration of dyes in the solution: q¼

ðC 0 2 C f Þ · V m

ð1Þ

The symbol V presents the solution volume (l) and m is the mass of sorbent material (g). The following processes and parameters have been studied: . The sorption kinetics – 1.00 g of material was shaken in 50 ml of dye solution (C0 ¼ 100 mg/l) for 0.25, 0.5, 1, 3, 6, 12 and 24 h at pH 5.00 for basic dyes and pH 3.30 for metal complex, reactive and direct dyes. . The influence of initial dye concentration on sorption process – 1.00 g of material was shaken in 50 ml of dye solution of different concentrations (C0 ¼ 10, 50, 100 and 500 mg/l) for 3 h at pH 5.00 for basic dyes and pH 3.30 for metal complex, reactive and direct dyes. . The influence of temperature on sorption process – 1.00 g of material was shaken in 50 ml of dye solution (C0 ¼ 100 mg/l) for 1 h at pH 5.00 for basic dyes and pH 3.30 for metal complex, reactive and direct dyes, in water bath WB14 (Memmert, Germany) supplied with shaking device. Temperature of the solutions were maintained at 20, 40 and 608C. . The influence of pH on the sorption process – 1.00 g of material was shaken in 50 ml of dye solution (C0 ¼ 100 mg/l) for 1 h. Appropriate initial pH values of the solutions were adjusted to 3.30, 5.00 and 7.00 with CH3COOH (1.00 g/l) and KOH (0.100 M). pH values of the solutions were measured using an Inolab 730 (WTW, Germany) pH-meter. Results and discussion Sorption kinetics for direct, basic, reactive and metal complex dyes is shown in Figure 1. Sorption of direct dye (Figure 1(a)) on untreated and H2O2 treated RWNM is almost unaffected by prolongation of sorption time. After the rapid sorption on CHT treated RWNM in the first 3 h, process slowed down and equilibrium was reached after 6 h.

RWNM for decolorisation

111

IJCST 21,2/3

5.5

5.5

5.0

5.0 4.5

4.5 Untreated CHT H2O2

3.5

3.0 2.5

2.5 2.0

1.5

1.5

1.0

1.0

0.5

0.5 0

5

10 15 Time, h (a)

20

0.0

25

5.5

5.5

5.0

5.0

4.5

4.5

4.0

4.0

3.5

3.5

3.0 2.5 2.0 1.0

0

5

10 15 Time, h (b)

20

25

3.0 2.5 Untreated CHT H2O2

1.5 1.0

0.5 0.0

Untreated CHT H2O2

2.0

Untreated CHT H2O2

1.5

Figure 1. Sorption kinetics

3.0

2.0

0.0

q, mg/g

4.0

q, mg/g

112

q, mg/g

3.5

q, mg/g

4.0

0.5 0

5

10 15 Time, h

(c)

20

25

0.0

0

5

10 15 Time, h

20

25

(d)

Notes: (a) Tubantin blue GLL; (b) Maxilon blue TRL; (c) Lanaset yellow 4GN; (d) Lanaset grey G

Uptake of direct dye on untreated RWNM was low. Selected direct dye has high molecular weight with several sulphonate groups. The affinity of dye might decrease with an increase in degree of sulphonation of the dye and after the first sulphonate group, each additional group likely has a negative influence on the sorption, facilitating desorption of the dye from RWNM. However, H2O2 and particularly CHT treatment significantly improved uptake of direct dyes. Superior sorption behaviour of CHT treated RWNM can be attributed to existence of new amino groups on wool originating from CHT that contribute to the increase in the positive z-potential of the fibre surface (Radetic et al., 2003). Amino groups in acidic conditions are protonated and ionic interaction between sulphonate groups of direct dyes and protonated amino groups of wool are expected. Basic dye uptake on H2O2 treated RWNM (Figure 1(b)) was higher compared to untreated and CHT treated RWNM. In the case of H2O2 treated RWNM equilibrium was reached already after 1 h of sorption. Untreated and CHT treated samples performed similar behaviour. Higher dye uptake on H2O2 treated RWNM is suggested to be due to wool oxidation and formation of appropriate functional groups ( Jovancic et al., 2001). The binding of basic dyes to wool is carried out via carboxylic groups. Electrostatic attraction between deprotonated, negatively charged carboxylic groups of wool and positively charged cationic basic dyes could be established at operated pH.

Each modification of RWNM brought about increase in uptake of reactive dye (Figure 1(c)). In the first several hours, CHT treated sample showed higher uptakes in comparison with H2O2 treated RWNM, but the sorption decreased within last 12 h, causing almost the same final uptake (24 h) for the both samples. Similar behaviour occurred during the sorption of metal complex dye (Figure 1(d)). The influence of initial dye concentration in solution on dye uptake of direct, basic, reactive and metal complex dyes is demonstrated in Figure 2. The results on sorption kinetics pointed out that high percentage of dyes was removed within the first 3 h of sorption, this being the reason to select the sorption time of 3 h for a study on influence of initial concentration on dye uptake. The increase in initial concentration brought about rise of dye uptake. At low initial dye concentrations (50 mg/l), there was a slight difference between untreated and differently treated samples. However, at higher initial dye concentrations treated samples started to demonstrate their advantages. Although the uptake of dyes increased, the percentage of dyes adsorbed by untreated and differently treated materials decreased. Reduction of the fraction of dyes adsorbed indicates that higher amounts of dyes are left in the solution. The influence of temperature on dye uptake of direct, basic, reactive and metal complex dyes is demonstrated in Figure 3. Temperature positively affected sorption of

113

25

15

Untreated CHT H2O2

20 10 15

Untreated CHT H2O2

q, mg/g

q, mg/g

RWNM for decolorisation

5

10 5

0 0 0

100

200 300 Co, mg/L (a)

400

0

500

100

200 300 Co, mg/L (b)

400

500

15

15 Untreated CHT H2O2

10

q, mg/g

q, mg/g

10

Untreated CHT H2O2

5

5

0

0 0

100

200 300 Co, mg/L (c)

400

500

0

100

200 300 Co, mg/L (d)

400

Notes: (a) Tubantin blue GLL; (b) Maxilon blue TRL; (c) Lanaset yellow 4GN; (d) Lanaset grey G

500

Figure 2. The influence of initial concentration on uptake

IJCST 21,2/3

6 6 5

5

3

2

1

1 0 20

40 Temperature,°C (a)

20

60

40 Temperature,°C (b)

60

6

6 Untreated CHT H2O2

5

5

Untreated CHT H2O2

4 q, mg/g

4 q, mg/g

3

2

0

Figure 3. The influence of temperature on uptake

Untreated CHT H2O2

4 q, mg/g

114

q, mg/g

4

Untreated CHT H2O2

3

3

2

2

1

1

0

0 20

40 Temperature,°C (c)

60

20

40 Temperature,°C (d)

60

Notes: (a) Tubantin blue GLL; (b) Maxilon blue TRL; (c) Lanaset yellow 4GN; (d) Lanaset grey G

direct dye on untreated and H2O2 treated RWNM. The rise of temperature induced the increase in direct dye uptake at 408C on CHT treated RWNM. However, at 608C dye uptake decreased to the level of sorption at 208C. Basic dye uptake is not considerably affected by temperature independently on RWNM studied. In the case of reactive and metal complex dyes, the rise of temperature led to an increase in dye uptake for untreated and H2O2 treated RWNM. On the contrary, sorption on CHT treated sample decreased with an increase in temperature. The influence of pH on dye uptake of direct, basic, reactive and metal complex dyes is shown in Figure 4. The increase in pH negatively influenced the sorption of direct dyes. Such behaviour was expected since the rise of pH caused the inhibition of the protonation of amino groups that are the main sites on wool for establishing the ionic interaction with sulphonate groups of dyes. On the contrary, the rise of pH induced significant increase in basic dye uptake. This effect was particularly pronounced in the case of H2O2 treated RWNM. Such behaviour could be anticipated since at higher pH values, carboxilyc groups are deprotonated and thus negatively charged, providing the electrostatic attraction with cationic basic dyes. The influence of pH on sorption of reactive and metal complex dyes is similar to that obtained for direct dyes. CHT treated samples are not remarkably affected by pH.

4.0

5.0

Untreated CHT H2O2

3.5

4.5 4.0

3.0

RWNM for decolorisation

Untreated CHT H2O2

3.5 q, mg/g

q, mg/g

2.5 2.0 1.5

3.0 2.5

115

2.0 1.5

1.0

1.0 0.5 0.0

0.5 3.30

5.00 pH (a)

0.0

7.00

Untreated CHT H2O2

4.5 4.0

7.00

Untreated CHT H2O2

5.5 5.0 4.5

3.5

4.0

3.0

3.5

q, mg/g

q, mg/g

5.00 pH (b)

6.0

5.0

2.5 2.0

3.0 2.5 2.0

1.5

1.5

1.0

1.0

0.5 0.0

3.30

0.5 3.30

5.00 pH (c)

7.00

0.0

3.30

5.00 pH (d)

7.00

Notes: (a) Tubantin blue GLL; (b) Maxilon blue TRL; (c) Lanaset yellow 4GN; (d) Lanaset grey G

Conclusion The RWNM whether in fabric or garment form can efficiently remove basic dye Maxilon blue TRL, reactive dye Lanaset yellow 4GN and metal complex dye Lanaset grey G from water. However, insufficient sorption of direct dye Tubantin blue GLL occurred either on untreated sorbent or sorbent treated with biopolymer CHT. Sorption of studied dyes was significantly affected by initial dye concentration, pH and temperature. The latest results indicated that RWNM can be efficiently used for purification of real effluents from textile dyehouses. References Carlough, M., Hudson, S., Smith, B. and Spadgenske, D. (1991), “Diffusion coefficients of direct dyes in chitosan”, Journal of Applied Polymer Science, Vol. 42, pp. 3035-8. Choi, H.M. and Moreau, J.P. (1993), “Oil sorption behavior of various sorbents studied by sorption capacity measurement and environmental scanning electron microscopy”, Microscopy Research and Techniques, Vol. 25, pp. 447-55. ¨ Honings, R., Peters, R., Mu¨ller, B.M., Thomas, H. and Ho¨cker, H. (1995), “Reduction of effluent contaminants from wool dying processes by use of ion exchange”, Proceedings of 9th

Figure 4. The influence of pH on uptake

IJCST 21,2/3

116

International Wool Textile Research Conference, Vol. 1, International Wool Secretariat, Biella, pp. 300-9. Jovancic, P., Jocic, D., Molina, R., Julia´, M.R. and Erra, P. (2001), “Shrinkage properties of peroxide-enzyme-biopolymer treated wool”, Textile Research Journal, Vol. 71, pp. 948-53. Kumar, M.N.V.R., Sridhari, T.R., Bhavani, K.D. and Dutta, P.K. (1998), “Trends in color removal from textile mill effluents”, Colorage, Vol. 40, pp. 25-34. Laing, I.G. (1991), “The impact of effluent regulations on the dying industry”, Review of Progress in Coloration, Vol. 21, pp. 56-71. McKay, G., Blair, H.S. and Gardner, J.R. (1984), “The adsorption of dyes onto chitin in fixed bed columns and batch adsorbers”, Journal of Applied Polymer Science, Vol. 29, pp. 1499-514. Maclaren, J.A. and Milligan, B. (1981), Wool Science – The Chemical Reactivity of the Wool Fibre, Science Press, Marrickville. Ofomaja, O.E. and Ho, Y.S. (2007), “Equilibrium sorption of anionic dye from aqueous solution by palm kernel fibre as sorbent”, Dyes and Pigments, Vol. 74, pp. 60-6. Pathak, J., Rupainwar, D.C., Talat, M. and Hasan, S.H. (2006), “Removal of basic dyes from aqueous solutions using coconut hask shell powder as a sorbent”, Journal of Indian Chemical Society, Vol. 83, pp. 1253-5. Radetic, M., Jocic, D., Jovancic, P., Rajakovic, Lj., Thomas, H. and Petrovic, Z.Lj. (2003), “Recycled wool based non-woven material as sorbent for lead cations”, Journal of Applied Polymer Science, Vol. 90, pp. 379-86. Rao, K.L.L.N., Krishnaiah, K. and Ashutush, N. (1994), “Color removal from a dye stuff industry effluent using activated carbon”, Indian Journal of Chemical Technology, Vol. 1, pp. 13-19. Robinson, T., McMullan, G., Marchant, R. and Nigam, P. (2001), “Remediation of dyes in textile effluent: a critical review on current treatment technologies with a proposed alternative”, Bioresource Technology, Vol. 77, pp. 247-55. Smith, B., Koonce, T. and Hudson, S. (1993), “Decolorizing dye wastewater using chitisan”, American Dyestuff Reporter, Vol. 82, pp. 18-35. Tu¨nay, O., Kabdasli, I., Eremektar, G. and Orhon, D. (1996), “Color removal from textile wastewaters”, Water Science and Technology, Vol. 34, pp. 9-16. Weltrowski, M., Patry, J. and Beaudoin, B. (1995), “Wool bases filter for adsorption of heavy metals and textile dyes”, Proceedings of 9th International Wool Textile Research Conference, International Wool Secretariat, Biella, Vol. 4, pp. 343-50. Willmot, N., Guthrie, J. and Nelson, G. (1998), “The biotechnology approach to color removal from textile effluent”, Journal of the Society of Dyers and Colourists, Vol. 114, pp. 38-41. Yoshida, H., Fukuda, S., Okamoto, A. and Kataoka, T. (1991), “Recovery of direct dye and acid dye by adsorption on chitosan fiber”, Water Science and Technology, Vol. 23, pp. 1667-76.

Corresponding author Maja Radetic can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

Characterization of conducting polymer coated fabrics at microwave frequencies Akif Kaynak

Conducting polymer coated fabrics 117

School of Engineering and Information Technology, Deakin University, Geelong, Australia, and

Eva Ha˚kansson Centre for Materials and Fiber Innovation, Geelong Technology Precinct, Deakin University, Geelong, Australia Abstract Purpose – The purpose of this paper is to investigate microwave reflection, transmission, and complex permittivity of p-toluene-2-sulfonic acid doped conducting polypyrrole coated nylon-lycra textiles in the 1-18 GHz frequency with a view to potential applications in the interaction of electromagnetic radiation with such coated fabrics. Design/methodology/approach – The chemical polymerization of pyrrole is achieved by an oxidant, ferric chloride and doped with p-toluene sulfonic acid (pTSA) to enhance the conductivity and improve stability. Permittivity of the conducting textile substrates is performed using a free space transmission method accompanied by a mathematical diffraction reduction method. Findings – The real part of permittivity increases with polymerization time and dopant concentration, reaching a plateau at certain dopant concentration and polymerization time. The imaginary part of permittivity shows a frequency dependent change throughout the test range. All the samples have higher values of absorption than reflection. The total electromagnetic shielding effectiveness exceeds 80 percent for the highly pTSA doped samples coated for 3 h. Originality/value – A non-contact, non-destructive free space method thin flexible specimens to be tested with high accuracy across large frequency range. The non-destructive nature of the experiments enables investigation of the stability of the microwave transmission, reflection, absorption and complex permittivity values. Moreover, mathematical removal of the diffraction enables higher accuracy. Keywords Ageing (materials), Permittivity, Electromagnetic fields, Microwaves Paper type Research paper

Introduction Conducting polymers may be used as alternatives to some commonly used metallic shielding materials. In contrast to metallic shielding materials, conducting polymers not only reflect but also absorb electromagnetic radiation in the microwave frequency range of 30 MHz-30 GHz (Kaynak et al., 1993; Kaynak, 1996). Conducting polymer coated textiles may have potential applications in the minimization of EMI interference by fabricating frequency selective fabric absorbers. A non-contact non-destructive free space method has been refined and used for investigations on thin and flexible specimens of sample sizes greater than 150 £ 150 mm across broad microwave bandwidths (Amiet, 2003). Investigation of a range of materials with the free-space transmission method has yielded satisfactory results (Amiet and Jewsbury, 2000; Truong et al., 1998) while an extensive investigation of conducting textiles using this method was lacking. In this paper,

International Journal of Clothing Science and Technology Vol. 21 Nos 2/3, 2009 pp. 117-126 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910933844

IJCST 21,2/3

118

the reflection, transmission, and complex permittivity of p-toluene-2-sulfonic acid doped conducting polypyrrole (PPy/pTSA) coated nylon-lycra textiles in the 1-18 GHz frequency were investigated using the free space transmission methods. Experimental Synthesis of polypyrrole on textile substrates A 0.53 mm thick double-sided basket-weave Nylon-Lycraw with an average lycra content of 20 percent was used as substrate textile. After drying fabric samples in a Binder FED 115 Lab oven at 1058C and cooling to room temperature, the samples were introduced into an aqueous solution containing pyrrole monomer (98 percent, Aldrich, Milwaukee, WI, USA), the dopant p-toluene sulfonic acid (pTSA) (98 percent, Sigma-Aldrich), and the wetting agent Albegal FFA (Ciba, Australia) at a concentration of 0.01 percent (w/w). Pyrrole was distilled under vacuum and used at a fixed concentration of 0.045 mol/l in all experiments. The oxidant ferric chloride hexahydrate (minimum 98 percent, Fluka) was added to the solution and a film of conducting polypyrrole (PPy) was formed directly on the textile substrate through oxidative polymerization (Malinauskas, 2001). The optimized ratio of the oxidant ferric chloride hexahydrate (FeCl3, Aldrich) to monomer of 1:2.23 was used (Kaynak and Beltran, 2003), resulting in a fixed concentration of 0.1 mol/l FeCl3, while pTSA was used in concentrations up to 0.036 mol/l. The synthesis was performed at room temperature with a varying polymerization times from 60 to 240 min. After PPy coating and drying, the samples were cut to size (305 £ 305 mm or 500 £ 500 mm) and stored at 208C at 65 percent RH. Characterization A LEO 1530 FEG-SEM (scanning electron microscopy, SEM) was used to perform a surface morphology study of the conducting textiles. Due to the sufficient conductivity of the PPy coatings, no further conductive coating prior to imaging was necessary. Fabric thickness measurements were made on preconditioned textile samples in a standard atmosphere using a textile thickness tester (DGTW01B, Mitutoyo, Japan) in accordance with ISO 9073-2 standard (0.5 kPa). The average thickness value from 20 measurements on each sample was used to obtain good accuracy. The original thickness of the uncoated substrate material was 0.53 mm. After in situ polymerization, an approximate thickness of 0:53 , 0:55 mm was obtained. Fiber diameter measurements of the uncoated and coated fibers were carried out using an optical fiber diameter measurement analysis (OFDA) using an OFDA 2000, which uses optical image analysis to measure the fiber diameters of 2 mm snippets of fibers between two 70 mm square glass slides. A minimum of 35,000 fibers from each sample were measured and averaged. Optical transmission microscopy analysis was performed on 8 mm microtome transverse sections of PPy coated fabric set in Technovit 7100 resin using an Olympus BX51 equipped with a DP12 camera (3.34 Mpixel). The surface resistivity of the conducting fabrics was measured according to AATCC test method 76-1995 (American Association of Textile Chemists and Colorists, 1996) using a 34401A multimeter (Agilent Technologies) after conditioning in a standard atmosphere (208C, 65 percent RH). In the case of coated fabrics, rather than resistivity in V/m, surface resistivity Rs in V/sq is used to express the electrical property.

Dielectric measurements Permittivity of the conducting textile substrates was performed using a free space transmission method in conjuntion with a mathematical diffraction reduction method as described by Amiet and Jewsbury (2000). The conducting textile sample was placed horizontally flat between two broadband DRG118-A horns operating in the 1-18 GHz frequency range. A radiation output system consisting of an 8510C vector network analyzer (Agilent Technologies) and an 8517A S-parameter test set (Agilent Technologies) with an 83651B synthesized frequency source (Agilent Technologies) generated a swept signal across the pre-set frequency range and collected the data from the measurement. The system covers the frequency range of 45 MHz-50 GHz, where the vector network analyzer has a dynamic range of greater than 100 dB with resolutions of 0.01 dB in magnitude and 0.018 in phase. Software written by Amiet controls the system and to improve accuracy of the permittivity measurements, the diffraction signal was removed using a mathematical method involving two fast Fourier transforms (FFT) and one inverse FFT (Amiet, 2003). A time-gate of 1.0 ns, using a Kaiser-Bessel window, was applied to the frequency domain measurement data, which was converted to time domain to include only signals that reached the output port within the predetermined amount of time. The system was calibrated at least every 5 min during the transmission measurements in order to adjust for changes in the ambient temperature and humidity with time. The interaction of a material with electromagnetic fields can be characterized by complex permittivity 1* (related to the electrical component of field), complex permeability m* (related to the magnetic component of field), and total (ac þ dc) conductivity s tot . The relative complex permittivity (1* ¼ 10 1r where 10 ¼ 8.854 £ 10-12 F/m) of a material consists of a real part (10 ), mainly associated with amount of polarization occurring in the material, and an imaginary part (100 ), related to dissipation of energy in the material as per: 1r ¼ 10r þ i100r

ð1Þ

The shielding effectiveness (SE) in [dB] can be calculated as per Balanis (1989): SE ¼ 10 · log10 ðT½ percent=100Þ

ð2Þ

A material is considered a good conductor if stot q 2v10 1r and a good dielectric if stot p 2v10 1r (stot . 0). This inequality determines how to approximate the skin depth as shown. For good conductors: sﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ 2 ½m ð3Þ dø vm0 stot For good dielectrics: 2 dø stot

sﬃﬃﬃﬃﬃﬃﬃﬃ 10 10 ½m m0

In the case of the conducting polymers investigated stot < 2v10 1r .

ð4Þ

Conducting polymer coated fabrics 119

IJCST 21,2/3

The materials tested in this work are thin conductive films deposited on non-conductive fabrics, where the skin depth almost exclusively is larger than the total thickness of the material. The free space method used here produces information about the magnitude and percentages of all amounts of radiation being reflected, absorbed and transmitted through the fabric.

120

Results and discussion The morphology of the nylon-lycra filaments coated with PPy-pTSA was smooth and free from cracks (Figure 1). As the polymerization time was increased, the amount of nodular particles on the surface increased. These nodular particles are not strongly adherent to the coating surface, and they can be minimized via control of the process parameters. A thin layer of conducting PPy on the filament is shown in Figure 2. When the elastomeric lycra fibers are cut off for imaging, they contract slightly, thus causing wrinkling of the coating in the vicinity of the cut ends. The wrinkling, instead of cracking or delaminating of the conducting layer, indicates good adherence and elasticity of the applied coating. Upon coating by PPy doped with pTSA a color change occurs in the nylon textile as can be seen in Figure 3. As the coating progressed the fabric colour changed from light grey (uncoated), to yellowish towards and finally towards brown. The handle and flexibility of the fabric after coating were very close to that of the untreated fabrics. The PPy coatings had little influence on the mechanical properties of the substrate on which it was formed. The color change of the fabric was accompanied by an increase in the fiber diameter as the polymerization time was increased. The fiber diameter analysis showed that the mean diameter of the pristine nylon-lycra fibers was 23.84 ^ 0.1 mm. During polymerization,

Figure 1. SEM images of PPy-pTSA (0.027 mol/l) coated nylon-lycra

Note: Polymerization time = 180 min

Conducting polymer coated fabrics 121

Note: Polymerization time 180min, concentration 0.018 mol/1 pTSA; magnification 2500 × ; scale bar = 10µm

Note: From left to right uncoated, 5, 15, 30, 60, 120, 300 min

the mean fiber diameter increased with time as the polymer deposited on the fabric. The rate of mean fiber diameter increase was higher in the initial stages of polymerization and slower at longer polymerization times due to the higher polymer formation rate in the early stages of coating. The variation of coating thickness with polymerization time was pﬃﬃ approximated by d ¼ ð1=20Þ t where, d is the coating thickness in mm and t is the time in minutes. Optical transmission microscopy analysis showed that the coating of PPy surrounded each fiber (Figure 4). The coating was adherent to the fiber even after setting in resin, microtome sectioning and application onto glass slides. The thickness of the coating layer increased as the polymerization time increased and the appearance of bulk polymer (nodular) particles became more pronounced. At long polymerization times of 180 min or more, the coating started to crack due to its larger thickness and incipient brittleness, as indicated by arrow in Figure 4(b). The surface resistivity as a function of concentration of pTSA at a set polymerization time showed that the surface resistivity dropped quickly as even small

Figure 2. Micrograph of coated Lycraw fiber (B) ends

Figure 3. Color change in the PPy-pTSA coated nylon-lycra at different polymerization times

IJCST 21,2/3

122

Figure 4. Optical transmission microscope images of transverse sectioned Ppy-pTSA coated nylon-lycra with different polymerization times

amounts of pTSA was added (Figure 5). The initial decrease in surface resistivity for FeCl3 when compared with FeCl3 þ 0.004 mol/l pTSA was 65 percent (from 1,010 to 350 V/sq). The surface resistivity continued to drop with an increase in dopant concentration, however at a much lower rate than at the initial drop. Microwave frequency dielectric results Permittivity of conducting textiles. The imaginary part of permittivity was almost 200 percent larger for the long polymerization time than that of the short time (Figure 6). The difference in surface resistivity between the two samples was approximately one order of magnitude. The real part of permittivity was between 60 percent (0.027 mol/l) and 80 percent (0.015 mol/l) higher for a long polymerization time than the short time. The increase in the real part of permittivity upon addition of larger amounts of dopant

(b)

(a) Note: (a) 60 min; (b)180 min; scale bar = 50 µm

1,000

Surface resisitivtiy [Ω/sq]

No dopant, only FeCl3 used

Figure 5. Surface resistivity versus dopant concentration for PPy-pTSA coated nylon-lycra

100 0.00

0.01

0.02 Concentration pTSA [mol/l]

Note: Polymerization time 180 min

0.03

0.04

Conducting polymer coated fabrics

ε′ (240 min) ε′ (60 min)

Permittivity [relative units]

0

ε′′ (60 min) ε′′ (240 min)

–50

123

0.027 mol/l Real 0.015 mol/l Real 0.027 mol/l Real 0.015 mol/l Real 0.015 mol/l Imaginary 0.027 mol/l Imaginary 0.015 mol/l Imaginary 0.027 mol/l Imaginary

–100

–150

–200 2

4

6

8

10

12

14

16

18

Frequency [GHz]

may be indicative of the added dopant actively taking part in the charge storage in the material. The same situation was present at the longer polymerization time, however the differences in magnitude between the different dopant concentrations were less pronounced. The smaller effects of higher dopant concentrations on the real part of permittivity at long polymerization times indicates that the increase in energy storage capacity may be possible only up to a certain level of added dopant. Above this concentration, the addition of further dopant ions does not contribute to any further capacitive behaviour of the material. Reflection, transmission, and absorption of conducting textiles. The results of reflection and absorption of PPy coated nylon is in accordance with those of Kim et al. up to 1.5 GHz. PPy-coated plain-weave PET fabrics had a reported contribution to shielding from absorption of up to 7 percent at frequencies up to 1.5 GHz. The absorption contributions to shielding obtained for PPy coated fabrics investigated here were up to six times higher. Another study of similar materials showed a maximum absorption of 33 percent at a resistivity of 2.85 S/cm at 1.5 GHz (Kim et al., 2002; Lee et al., 2002). A common feature of the results is that the longer the polymerization time and the higher the dopant concentration, the lower the transmission and the higher the reflection in the conducting textile samples. The reflection values increase with an increase in frequency. This is expected because the higher frequency will have a smaller penetration depth of radiation and hence less interaction with the polymer. Results indicated that the combination of high values of both real and imaginary parts of permittivity were responsible for the resulting total shielding of radiation. When pTSA was excluded, significantly lower value of reflection, and almost half the level of absorption was obtained. This is indicative of the dopant ions in the material actively contributing to the conductivity, which has been confirmed by published data (Rodriguez et al., 1997; Nalwa, 1997). The absorption increases by over 40 percent (average value over whole frequency range) with the addition of only small amounts of dopant ions. This is an effective way of improving the microwave absorption in PPy-coated fabrics. The absorption level for a sample doped with

Figure 6. Permittivity response for PPy-pTSA coated nylon-lycra with 0.015 mol/l or 0.027 mol/l pTSA and 60 or 240 min polymerization time

IJCST 21,2/3

124

0.018 mol/l pTSA was twice that of a sample with no additional dopant. The amount of reflection steadily increased with the increase in concentration of the dopant in the polymerization bath, reaching maximum levels of 31.36 percent, whereas the sample without dopant displayed an average reflection value of less than 5 percent. The transmission decreased with the increase in the dopant concentration and frequency. Conversely, the reflection levels increased with frequency. Figure 7 shows the average values of reflection, transmission and absorption for samples with different dopant concentrations, but a constant polymerization time of 180 min. As discussed above, the conductivity increased with an increase in the dopant concentration and/or polymerization time. There sas an upper limit for the conductivity of PPy-coated textiles and some samples with long polymerization times would have approached their threshold conductivity. This is why the total transmission loss reached a plateau at polymerization times above 180 min and dopant concentrations beyond 0.027 mol/l. The variations in total transmission loss with frequency decreased with an increase in dopant concentration, especially at long polymerization times. This is in accordance with previous reported results for PPy-loaded paper in the 2-18 GHz range. The transmission loss for the samples with a 60 min polymerization time was the lowest, and an increase in total transmission loss was obtained when the polymerization time was extended. However, the increase in the total transmission loss at a very long polymerization time of 240 or 300 min was only a few percent higher compared to that obtained at 180 min.

Figure 7. Reflection, transmission, and absorption values for PPy-pTSA coated nylon-lycra in 1-18 GHz frequency range for different dopant concentrations

Reflection/Transmission/Absorption [percent]

Conclusion The permittivity of PPy-coated textiles was measured using a free space transmission measurement technique over the frequency range 1-18 GHz. The measurements are relatively free from diffraction aberration, which indicates that the free space Reflection Transmission Absorption

60 50 40 30 20 10

0.000

0.009

0.018 Concentration pTSA [mol/l]

Note: Polmerization time 180 min

0.027

0.036

transmission measurement method is suitable for use with the flexible and thin conducting textiles. The real part of complex relative permittivity increased with polymerization time but stabilized after 120 min. The change in the real part of permittivity was not significant beyond 12 GHz irrespective of polymerization time. The imaginary part of permittivity changed with an extension of the polymerization time and varied throughout the frequency range. The influence of an increase in dopant concentration on the permittivity response was also confirmed. Both the real and imaginary parts of permittivity remained stable above dopant concentrations of 0.018 mol/l pTSA. The polymerization time, dopant concentration and choice of dopant influenced the permittivity and hence reflection, transmission, absorption as well as total SE of the conducting textiles. However, it was difficult to distinguish either of these two factors as being exclusively deterministic of shielding behaviour. All conducting nylon-lycra textiles had higher values of absorption than reflection. The highest absorption levels of around 48.27 to 48.78 percent are obtained for samples with a polymerization times of 120 or 180 min in combination with a dopant concentration of 0.018 or 0.027 mol/l pTSA. The shielding analysis showed that chemical structure influences the SE. The absorption dominated, considerably high shielding proves that conducting polymer coated textiles are good light-weight candidates as shielding materials. References AATCC (1996), “Electrical resistivity of fabrics”, AATCC Technical Manual, American Association of Textile Chemists and Colorists, Research Triangle Park, NC, pp. 100-1. Amiet, A. (2003), “Free space permittivity and permeability measurements at microwave frequencies”, PhD thesis, Monash University, Melbourne. Amiet, A. and Jewsbury, P. (2000), “Free space permittivity measurements at microwave frequencies”, paper presented at 2000 Asia Pacific Microwave Conference, Sydney. Balanis, C.A. (1989), Advanced Engineering Electromagnetics, Wiley, New York, NY. Kaynak, A. (1996), “Electromagnetic shielding effectiveness of conducting polypyrrole films in the 300-2,000 MHz frequency region”, Materials Research Bulletin, Vol. 31 No. 7, pp. 845-60. Kaynak, A. and Beltran, R. (2003), “Effect of synthesis parameters on the electrical conductivity of polypyrrole-coated poly(ethylene terephthalate) fabrics”, Polymer International, Vol. 52, pp. 1021-6. Kaynak, A., Unsworth, J., Beard, G.E. and Clout, R. (1993), “A study of conducting polypyrrole films in the microwave region”, Materials Research Bulletin, Vol. 28 No. 11, pp. 1109-25. Kim, M.S., Kim, H.K., Byun, S.W., Jeong, S.H., Hong, Y.K., Joo, J.S., Song, K.T., Kim, J.K., Lee, C.J. and Lee, J.Y. (2002), “PET fabric/polpyrrole composite with high electrical conductivity for EMI shielding”, Synthetic Metals, Vol. 126, pp. 233-9. Lee, J.Y., Kim, H.K., Kim, M.S., Joo, J.S., Jeong, S.H., Kim, S.H. and Byun, S.W. (2002), paper presented at American Chemical Society Division of Polymeric Materials, Science and Engineering Meeting, Washington, DC, August 20-24, 2000. Malinauskas, A. (2001), “Chemical deposition of conducting polymers”, Polymer, Vol. 42 No. 9, pp. 3957-72. Nalwa, H.S. (Ed.) (1997), “Charge transport in conducting polymers”, Handbook of Organic Conductive Molecules and Polymers, Vol. 4, Wiley, New York, NY.

Conducting polymer coated fabrics 125

IJCST 21,2/3

Rodriguez, J., Grande, H.-J. and Otero, T.F. (1997), “Polypyrroles: from basic research to technological applications”, in Nalwa, H.S. (Ed.), Handbook of Organic Conductive Molecules and Polymers, Wiley, New York, NY, pp. 415-42. Truong, V.-T., Riddell, S.Z. and Muscat, R.F. (1998), “Polypyrrole based microwave absorbers”, Journal of Materials Science, Vol. 33 No. 20, pp. 4971-6.

126

Further reading Ha˚kansson, E., Lin, T., Wang, H. and Kaynak, A. (2006), “Effect of dye dopants on the conductivity and optical absorption properties of polypyrrole”, Synthetic Metals, Vol. 156 Nos 18/20, pp. 1194-202. Corresponding author Akif Kaynak can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

Heating behaviors of metallic textile structures

Heating behaviors

Ozan Kayacan and Ender Yazgan Bulgun Textile Engineering Department, Dokuz Eylul University, Izmir, Turkey

127

Abstract Purpose – The purpose of this paper is to investigate the concept of “electrically conductive fabrics”. The primer applications that import electrical conductivity properties to textiles and clothing are summarized. Also the heated fabric panels produced by steel yarns are evaluated. Single and multi-ply steel fabrics are applied to electrical current and their heating behaviors are observed and compared. Design/methodology/approach – The integration of electronic components with textiles to create very smart structures is getting more and more attention in recent years. Most of the textile materials are electrical insulators. Hence, various types of fibers and fabrics having reasonably good electrical conductivity are required especially for electronically functional apparel products. The textile-based materials being flexible and easily workable are the most preferred one in such cases. In this study, the steel yarns are placed in the fabric construction owing to their flexible characteristics. The heating panels used in this study are produced by conventional textile processes and applied to electrical current. For this purpose, an electronic circuit that contains textile-based warming panels connected to a power supply, has been developed. Findings – The heated steel fabric panels with different number of plies provide different heating degree intervals owing to the different resistance levels, Therefore, in the applications of textile-based heating elements it is suggested that the electrical characterization of conductive materials should be examined and the materials that have the most appropriate electrical resistance characteristic must be applied. Furthermore, in the circuits used for heating function, the current amount depends on the electrical features of heating structures. Consequently, the pads with different plies have various efficient heating in point of time. It is recommended that the appropriate heating pad dimensions, ply or conductive yarn amounts and sufficient power supply conditions should be evaluated and chosen according to the desired heating level. Originality/value – Electrically conductive stainless steel yarns are processed to form a heating panel that can be used within an electronic circuit as a warming mechanism. Keywords Electrical conductivity, Yarn, Textiles, Stainless steel Paper type Research paper

1. Introduction Since the nineteenth century, revolutionary changes have been occurring at an unprecedented rate in science and technology with a profound impact on our lives. Inventions in science and technology have transformed the entire world. As technology has progressed, it has been applied rapidly. Technological improvements in the field of smart and interactive materials have attracted more and more attention in recent years. There will be a boom in the use of these new ideas and components in the next few years. These new achievements of the textile industry enable electronic devices to be directly integrated into the structure of textile, therefore modifying the functionality of the apparel. The worldwide smart materials market is estimated at $8.1 billion in 2005 and is expected to rise at an average annual growth rate (AAGR) of 8.6 percent to $12.3 billion in 2010. On the other hand, the global market for smart/interactive textiles reached

International Journal of Clothing Science and Technology Vol. 21 Nos 2/3, 2009 pp. 127-136 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910933853

IJCST 21,2/3

128

$304.0 million in 2005 and is expected to grow to $642,1 million by 2008 a compound annual growth rate of 28.3 percent. The materials of our surroundings are being “intellectualized”. Whereas, in the past, we needed several components to satisfy a certain function. Today, technology has allowed us to satisfy the same function with less components. The concept of “miniaturization” not only means the production of smaller components, but the elimination of components. Intelligent textile systems, integrated to electronics, have the capacity of improving the user’s performance by sensing, adopting itself and responding to a situational need. The smart/interactive textile structures that integrate electronics and textile materials and the materials that react to the external stimuli physically and chemically have been developed. These products, which are called “the garments of the future”, involve different functions such as protection, actuation, communication, etc. The garments, which can heat the body, will possibly be one of the most widely used products for future use in daily life. Beside the medical usage, these products are developed especially for use by people who work outside during their day, such as military applications, security services, and country duties. The heating functions of the garments occur in the thermal panels. The thermal panels can be produced by using electrically conductive textile-based materials by weaving or knitting. Stitching is another method for placing the conductive yarns into the fabric surface. There are different textile-based materials that can be used as a heating material such as silver coated, copper-plated yarns, carbon-based fibers, tiny inox cables, etc. In this study, conductive stainless steel yarns are chosen as a sample for metallic textile structure. The heated fabric panels produced by steel yarns are evaluated. Single and multi-ply steel fabrics are applied to electrical current and their heating behaviors are observed and compared. For this purpose, an electronic circuit that contains textile-based warming panels connected to a power supply, has been developed. 2. Conductive yarns for smart textile applications Natural fibers have insulative characters. The integration of electronic functions in textiles can be realized in two extreme ways. One way is to produce the apparel or technical textile and then integrate electronic components. The other way is to process conductive yarns when producing textiles and create textile structures with electronic functions. There is a lot of research and development activity regarding the needs of conductive textile materials. The studies about transformation of the natural materials into conductive structures by different chemical processes such as coating, lamination and modification are progressing. One of the major electrically conductive textile-based materials is carbon fiber. In a similar way, metallic fibers and steel-based yarns are also used for the same purpose. If examined by production types, there are different types of materials: knitted fabric, woven fabric, and non-woven structure. Electrically conductive yarns can be used for: (1) Transfer of power. (2) Transmission of signals between: . Sensors.

Transmitters. Microcontrollers, etc. (3) Traditional use in: . Heating. . Electromagnetic protection. . .

High conductive yarns are made from different materials and exist in a lot of different forms. Materials are metals like copper or stainless steel or alloys such as Nitinol, coated metals, carbon fibers, and polymers. Insulating polymers can be coated or doped. There are several criteria for selecting a conductive yarn for a certain application. The criteria can be based on the product and the processing. Costs, conductivity, chemical resistance (e.g. against moisture, wash ability), mechanical properties and contact ability (e.g. thermal resistance) of the product is the main factors regarding to selecting a conductive yarn. The electrical resistance of the yarns is described in V/m, a value which in one case is a material parameter, the resistance depends on length and diameter, and in other case a yarn parameter, the resistance depends only on the length. The metallic wires and tapes are the most typical materials that can be applied to textiles for their conductivity properties. They can be interlaced into the fabric structure. However, there are some disadvantages about the material characteristics such as limited flexibility, increased weight and cutting problems. Owing to the isolating characteristics of textile polymers, electrical conductivity is reached by modification of the polymer structure and/or addition of conductive materials. The modification of the fabric structure can be performed in two ways: Impregnation using antistatic agents and coating using conductive substances. Moreover, electrically conductive fibers were produced by wet spinning, melt spinning from conductive polymers; or coating fibers with electrically conductive materials such as metal powders and carbon black or intrinsically conductive polymers. Polyaniline, polyamide-11, polyvinilalcohol are the most common conductive polymers using such processes. However, the major drawbacks are “limited flexibility” and “use only in blends with conventional fibers”. Modification of polymers can be carry out by filling with electro-conductive powder, vacuuming spread metal, galvanic- and chemical-coating. Among the manufacturing processes, various coating techniques have been attractive due to simple in process and easy to handle. The textiles produced not only gain controllable electrical properties, but also maintain their excellent physical properties of the textiles such as mechanical strength and flexibility. 3. Material and method 3.1 Heated fabric In previous studies, its been indicated that, the resistance of heating panels produced by weaving techniques is lower than the knitted structures with the same dimensions because of the structural characteristics. When compared to knitted fabric, woven fabrics have less surface characters and quality properties. So they are more appropriate for implications, as a heating panel.

Heating behaviors

129

IJCST 21,2/3

130

The heating panels used in this study were produced by using industrial narrow weft knitting machine. In the panel construction, the conductive yarns were placed in coarse direction while polyester yarns were placed in wale direction. As seen on Figure 1., the conductive yarns were placed in a parallel form like the weft yarns in the woven fabrics. After laying process, the yarns leave the fabric without any cut, brake or any interruption. In a regular heating element the electrical wires have a similar structure. So, this is the most suitable fabric formation for heating panel production. Ten centimeter-long fabric bands were chosen as a unit ply for heating panels. To compare the different heating levels; single-, two-, three- and four-ply heating pads have been produced. For each ply amount, four heating pads have been connected to the circuit and 12 V. electric current have been applied. The heating behaviors were observed in point of temperature and electrical characteristics. The system can be illustrated as Figure 2. 3.2 Stainless stain yarns Metal fiber combines flexibility of traditional fiber with high temperatures resistance and thermal-electric conductivity of steel. Metal fibers are obtained by successive wire bundle-drawings. Because of its mechanical properties, AISI 316L alloy is the most common alloy. In this study, a special range of continuous stainless steel filament yarns was used to obtain heating function. These yarns have a very precise electrical resistance. Some characteristics of special yarns can be seen on Table I.

(a)

Figure 1.

(b)

(c)

Note: (a) General production procedure of heating panel; (b) the parallel structure of conducive yarns; (c) general view of narrow width conductive fabric band

single-ply

Heating Pad

Heating Pad

two-ply CIRCUIT

Figure 2. Heating system of the stainless steel fabrics

three-ply four-ply

+

− 12 V 3300 mAh

Heating Pad

Heating Pad

3.3 Electronic circuit design for activating the heating panel In this study, to observe the heating behavior of metallic textile structure, an electronic circuit is designed to meet the electrical requirements. The circuit activates the heating panel, produced by stainless steel yarns. The small dimension, lightness and low-cost parameters of the circuit are considered owing to the needs of smart-interactive garment design in the future works. Owing to the location in the garment, the dimensional parameters are very important. The circuit should be placed in as small an area as possible. The use of a high linear temperature sensor for observing the heating behavior with digital output is accurate. The temperature measurement system is implemented by using DS1820 (a one-wire digital temperature sensor). The property of one-wire bus has the capability to connect up to several different one-wire devices into the same bus. This is possible due to a unique address that every one-wire device has. Implementing the one-wire interface with one I/O pin is possible, but because that would have significantly increased the complexity of the software, the one-wire sensors should be connected to the serial multiplexer through a serial one-wire line driver. The sensors have a range of ^ 0.58C. Temperature sensor should be used with a microcontroller. Both devices should be low-cost and small in dimensions. The temperature control values are loaded into the microcontroller by software and the heating functions runs according to these pre-set values. The general scheme of the circuit is shown on Figure 3.

Heating behaviors

131

4. Results Figures 4-7 indicate the heating behavior of panels containing stainless steel yarns and shows the mean temperature values of the system. When we consider the first 10 min of the application, single-ply heating pad system provides approximately 158C heating as seen on Figure 4. On the other hand, two-ply system supplies nearly 208C within the same period as shown on Figure 5. As seen on Figure 6, approximately 258C heating was obtained by using three-ply heating pad system within the first 10-min period. Four-ply heating pad system provides approximately 358C within about the same period as shown on Figure 7. According to the comparison between different-ply heating pads, four-ply ones reach the maximum value (about 608C) (Figure 8). As expected, the heating levels of the different plies can be arranged as “four-, three-, two-, and one-ply”, respectively. Electrical current for the circuit have been supplied from a battery-pack having a capacity of 12 V. Consequently, all pad system starts their performance from this level. However, they have been served for different time period with same power source because of the electrical characteristics of the panels such as resistance, as seen on Figure 9.

Average linear resistivity Variation linear resistivity Yarn count Average Breaking load

71 (V/m) ^ percent 14 110 tex 23 N

Table I. Some physical characteristics of stainless steel fiber used in heating panels

IJCST 21,2/3

132

5. Conclusion Nowadays, conductive fibers are of special interests due to their application possibilities especially for interactive electro-textile products for technical applications such as wearable electronic clothes. Various metallic-based yarns have been applied for different function. In this study, the heating behaviors of conductive steel fibers were observed and evaluated. A special electronic circuit was designed to assist the system. For each ply, four heating pads have been connected to the circuit and 12 V. electric current have been applied. Mean heating value of the pads were observed. Each plies of the heating pad

LCD VDD = 5V

7 6 5 4 3 2 1 0 E R/w RS VoVDD GND

5V

RESET

100nF

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

+5V 10K 20pF 5V 20pF

4Hz

DS1820 1 2 3 470

RB7 40 39 38 37 36 35 34 R50 33 32 31 30 29 28 27 26 25 24 23 22 21

16F877

270 270 270 270

1

2

3

4

5

6

7

8

9

*

0

#

5V

DS1820 1 2 3 470

R1

DS1820

470

R2 470

R3 470

R4 470

R5 470

1 2 3 470 DS1820 1 2 3 470

+5V

Figure 3. General scheme of electronic circuit board

TRANSISTORLER : BD×50

1 : BASE 2 : COLLECTOR 3 : EMITER

Heating behaviors

40

Temperature (°C)

35 30 25 20

133

15 10 5 0 0

5

10

15

20

25

30

35

40

45

Figure 4. Heating behavior of single-ply heating pads

Time (min)

50

Temperature (°C)

40 30 20 10 0 0

2

4

6

8

10

12

14

16

18

20

22

Time (min)

Figure 5. Heating behavior of two-ply heating pads

50

Temperature (°C)

40 30 20 10 0 0

2

4

6

8

10

Time (min)

12

14

16

18

Figure 6. Heating behavior of three-ply heating pads

IJCST 21,2/3

50 Temperature (°C)

134

60

40 30 20 10

Figure 7. Heating behavior of four-ply heating pads

0 0

2

1

3

4

5

6

7

8

9

10

11

12

13

14

Time (min)

Temperature (°C)

60 50 40 30 20 10 0

Figure 8. Comparisons of heating pads in terms of warming period

0

5

10

15

20

25

30

35

40

45

Time (min) Single-Ply Pad

Two-Ply Pad

Three-Ply Pad

Four-Ply Pad

14 12

V

10 8 6 4 2 0 0

Figure 9. Voltage levels of heating pads

5

10

15

20

25

30

35

40

45

Time (min) Single-Ply Pad

Two-Ply Pad

Three-Ply Pad

Four-Ply Pad

contains the same amount of steel yarns. Its observed that owing to the different electrical resistance levels, the pads with different number of plies provide different heating degree intervals. Therefore, in the applications of textile-based heating elements its suggested that, the electrical characterization of conductive materials should be examined and the materials that have the most appropriate electrical resistance characteristic must be applied. Furthermore, in the circuits that used for heating function, the current amount depends on the electrical features of heating structures. Consequently, the pads with different plies have various efficient heating in point of time. Its recommend that the appropriate heating pad dimensions, ply or conductive yarn amounts and sufficient power supply conditions should be evaluated and chosen according to the desired heating level. On the other hand, in the early stages of the investigations, the combination of electronics and textiles were seemed not to be practicable in view of their opposite properties. With the successful results of scientific studies, the integration of electronic components into textiles offers great advantages. The first samples that combined these systems together were observed in the military and other outdoor applications. As a sample of such an application, the temperature controlled heating garment design should have two main divisions. The first step should meet the functional needs while the second should cover the electronic functions. Being a mobile device, the stability, the time required to operate, and the choice of an adequate mobile power supply unit are the major factors regarding to the design. Research into the electronic control circuit using pre-set temperature values is on going. In this study, an electronic circuit was designed with the help of previous research. This device will be used in designing a smart garment with a heating capability in the near future. Further reading BS EN ISO 9886 (2004), Ergonomics. Evaluation of Thermal Strain by Physiological Measurements, British Standard/European Standard/International Organization for Standardization, International Organisation for Standardisation, Geneva. Cottet, D., Grzyb, J., Kirstein, T. and Tro¨ster, G. (2003), “Electrical characterization of textile transmission lines”, IEEE Transactions on Advanced Packaging, Vol. 26 No. 2, pp. 182-90. DS18B20 (N.D.), Datasheet for DS18B20 Programmable Resolution 1-wire Thermometer, Dallas-Maxim Semiconductor Inc., available at: http://datasheets.maxim-ic.com/en/ds/ DS18B20.pdf Kayacan, O. (2008), “An investigation about smart garment design”, PhD thesis, Dokuz Eylul University, Izmir. ¨ . (2006), “Designing an electronic body temperature control Kayacan, O., Bulgun, E. and Sahin, O unit for smart garments”, paper presented at International Conference Futurotextiles, 23-24/11/2006 Lille. Kukkonen, K., Vuorela, T. and Rantanen, J. (2001), “The design and implementation of electrically heated clothing”, Proceedings of IEEE International Symposium on Wearable Computers, pp. 180-1. Mauch, H.P. and Nusko, R. (2005), “New possibilities with special conductive yarns”, Melliand International, No. 3, pp. 224-5. Vassiliadis, S., Provatidis, C., Prekas, C. and Rangussi, M. (2005), “Novel fabrics with conductive fibres”, Intelligent Textile Structures – Application, Production & Testing International Workshop, Thessaloniki.

Heating behaviors

135

IJCST 21,2/3

136

VDC (2006), Smart Fabrics and Interactive Textiles: OEM and End-User Requirements, Preferences and Solution Analysis, 2nd ed, Venture Development Corporation Publication, Research Group, Natick, MA. Xue, P., Park, K.H., Tao, X.M., Chen, W. and Cheng, X.Y. (2007), “Electrically conductive yarns based on PVA/carbon nanotubes”, Composite Structuresm, Vol. 78 No. 2, pp. 271-7. Yazici, T. (2004), “Temperature control unit of an electrically heated clothing”, DEU Electrical and Electronics Department, BSc Project, Advisor: O. Sahin. Corresponding author Ender Yazgan Bulgun can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

Surface modification and characterization of cotton and polyamide fabrics by plasma polymerization of hexamethyldisilane and hexamethyldisiloxane

Surface modification and characterization 137

Bengi Kilic and Aysun Cireli Aksit Textile Engineering Department, Engineering Faculty, Dokuz Eylul University, Izmir, Turkey, and

Mehmet Mutlu Plasma Aided Biotechnology and Bioengineering Research Group (PABB), Engineering Faculty, Hacettepe University, Ankara, Turkey Abstract Purpose – Plasma polymerization is a very promising technique to produce functional textile materials for any textile end uses as well as for high performance clothing. It can be possible to obtain highly cross-linked, pinhole free and very thin polymer films up to 1 mm thickness with unique physical and chemical properties. These films can be used as very effective barriers. The purpose of this paper is to investigate the influences of plasma polymerization of hexamethyldisilane (HMDS) and hexamethyldisiloxane (HMDSO) on the surface properties of cotton and polyamide fabrics. Design/methodology/approach – The methodology is based on the surface modification of the cotton and polyamide fabrics by plasma polymerization of HMDS and HMDSO. The fabrics are modified by low pressure low temperature RF (radio frequency 213.56 MHz) plasma polymerization system under different power and time conditions. The changes in surface structure and morphology of the fabrics are investigated by Fourier transform infrared spectroscopy-attenuated total reflectance (FTIR-ATR) analysis and atomic force microscopy (AFM). Findings – Water repellency of polyamide fabrics is strongly enhanced after plasma polymerization of both HMDS and HMDSO monomers. In addition to this, the treatments are found to slow down the vertical flame spread in cotton fabrics. Originality/value – Increased water repellency and decreased vertical flame spread are achieved using plasma polymerization technique in a very short time with very little amount of chemical and without water and auxiliary agent. Keywords Surface treatment, Polymerization, Cotton, Polyamides Paper type Research paper

1. Introduction Plasma, as a very reactive material, can be used to modify the surface of a certain substrate (typically known as plasma activation or plasma modification), depositing This research has been partly supported by Turkish Scientific and Technical Research Council Project, TUBITAK 105M099 and Dokuz Eylul University, Scientific Research Center, Project No. 05.KB.FEN.050.

International Journal of Clothing Science and Technology Vol. 21 Nos 2/3, 2009 pp. 137-145 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910933862

IJCST 21,2/3

138

chemical materials (plasma polymerization or plasma grafting) to impart some desired properties, removing substances (plasma cleaning or plasma etching) which were previously deposited on the substrate. Plasma technologies offer a wide spectrum of possible treatments of materials. Plasma-chemical conversion of the feed gas produces chemically active particles that are able to modify textile surface molecules via chemical reactions after impinging on the surface. The radicals generated inside the plasma region must be given the opportunity to move to the reaction place at the textile fiber surface (Arefi et al., 1992; Luo, 2002; Yasuda, 1981). In the last few years, a great deal of work has been produced in the field of plasma deposition of organo-silicon compounds (SiOx). The first target of these studies was the production of SiO2-like films devoted to micro-electronics applications. However, plasma-deposited SiO2-like materials are also characterised by other very important properties such as good biocompatibility, resistance to chemicals and abrasion. In addition to these silicone and silica based compounds are well known fire retardants and give a protective surface layer. Plasma polymerization of siloxane monomers are used to give fire retardant property to the polymer surfaces (Hegemann et al., 2003; Jama and Delobel, 2007; Quede et al., 2002, 2004; Schartel et al., 2002; Zanini et al., 2005). The aim of this study was the investigation of the influence of hexamethyldisiloxane (HMDSO) and hexamethyldisilane (HMDS) plasma conditions on the surface characteristics of cotton and polyamide fabrics being treated with plasma. For this purpose, 100 percent cotton and 100 percent polyamide fabrics were treated in the hexamethyldisiloxane and hexamethyldisilane plasmas. The modification of surfaces was carried out at low pressure (, 100 Pa) and low temperature (, 508C) plasma conditions. Plasma treatments were performed in PICO RF (radio frequency, 13.56 MHz) low pressure plasma system (Diener electronic GmbH þ Co. KG, Germany). Variables of the processes were discharge power and exposure time. The changes in the characteristics of the cotton and polyamide fabrics were evaluated by FTIR-ATR analysis, contact angle measurements and vertical flammability tests. 2. Experimental 2.1 Materials One hundred percent cotton and 100 percent polyamide fabrics were used. The sample sizes of fabrics were 36 cm in warp and 22 cm in weft direction. Hexamethyldisiloxane and hexamethyldisilane (Aldrich Chemical) were used as monomers in radio frequency glow-discharge plasma system. 2.2 Plasma treatment Plasma polymerization treatments were carried out in PICO RF (radio frequency 2 13.56 MHz) Plasma Polymerization System (Diener electronic GmbH þ Co. KG, Germany). Power loss was minimized by means of a matching network. At first, the reactor was evacuated to 26-30 Pa and then monomer inlet was opened and monomer (HMDSO or HMDS) vapor was allowed to flow through the reactor. The details of the steps of the plasma treatment of fabrics were given in our previous paper (Cireli (Aksit) et al., 2007). Cotton and polyamide fabrics were modified in various plasma polymerization conditions (discharge power: 20, 40, and 60 W and exposure time: 5, 10, and 15 min). The effects of power and exposure time parameters on the contact angles

and vertical flammability properties were evaluated by statistical software, MINITABw for Windows. After plasma treatments, one piece of each cotton fabric was subjected to a heat treatment at 1058C for 1 h to see if the increasing chain mobility of deposited surface groups changed the wettability properties of the fabrics (Hochar et al., 2003). FTIR-ATR analyses and AFM (atomic force microscopy) investigations were conducted to obtain the changes in surface structures after plasma treatments.

Surface modification and characterization 139

2.3 Wettability properties The water contact angles of polyamide and cotton fabrics were measured by the sessile drop method using a goniometer (KSV-CAM 100) before and after the treatments. 2.4 Vertical flame spread properties Flame spread (in warp direction) times of the cotton and polyamide fabrics were evaluated according to TS 5569 EN ISO 6941 with vertical flammability tester. In this method, a vertical fabric with the sizes of 560 £ 170 mm2 is exposed to the flame of a horizontal burner at a distance of 2 cm from the short bottom side. The flame spread times to the marking threads at 24, 39, and 54 cm distances from the bottom are recorded for the evaluation. However, our sample length was small than the length in the standard. Therefore, we could obtain only the flame spread time of the first thread at 24 cm. 2.5 FTIR-ATR analysis The ATR-IR spectroscopic measurements of treated and untreated cotton fabrics were performed using a Perkin-Elmer Precisely Spectrum One FTIR spectrometer equipped with ZnSe ATR crystal. 2.6 Atomic force microscopy The topography of the cotton fabrics was investigated by means of AFM using a Digital Instruments MMSPM Nanoscope IV. Measurements were carried out in tapping mode with silicon tip. For 5 £ 5 mm2 scans the images were measured in air. 3. Results and discussion 3.1 Wettability properties In order to investigate the effects of HMDS and HMDSO plasma polymerization on wet ability properties of polyamide and cotton fabrics, we measured the contact angles of treated and untreated fabrics. The effects of power and time parameters on the polyamide fabrics were evaluated by MINITAB software and presented in three-dimensional graphics in Figure 1(a) and (b). The average of the contact angle of the untreated polyamide fabrics was 388. This value rose to a mean value of 1138 after HMDSO plasma treatments and 958 after HMDS plasma treatments. In Figure 2, the contact angles of the untreated fabric (Figure 2(a)) and the fabric treated under the conditions of 20 W and 5 min HMDSO plasma (Figure 2(b)) can be seen. The water repellency of polyamide fabrics may be due to the terminating methyl groups and orientation of Si-groups on the surface of the treated fabrics (Carpentier and Grundmeier, 2005; Grundmeier et al., 2004; Hochart et al., 2004).

IJCST 21,2/3

Figure 2. Contact angles of polyamide fabrics before (a) and after (b) 20 W-5 min HMDSO plasma treatment

125 100 75 50 0

20 40 Power (W)

60

15 (a)

(a)

0 5 10 Time (min)

Surface plot of contact angle-PA-HMDS vs power (W); time (min) Contact angle-PA-HMDS

Figure 1. Contact angles of polyamide fabrics before and after plasma polymerization of (a) HMDSO and (b) HMDS

Contact angle-PA-HMDSO

140

Surface plot of contact angle-PA-HMDSO vs power (W); time (min)

125 100 75 50 0

20

Power (W)

40

60

15

10

0 5 Time (min)

(b)

(b)

Cotton fabrics were hydrophilic, so the contact angle measurements of the untreated fabrics cannot be possible. The possible orientation of Si-groups for HMDSO treated fabrics and terminating methyl groups in HMDS treated fabrics could not give water repellency properties to cotton fabrics. Therefore, after the plasma polymerization the fabrics remained hydrophilic. After the plasma polymerization of HMDSO at 20 W-5 min and the heat treatment, the contact angle of the cotton fabric was 1408 in average (Figure 3). This result cannot be obtained for other plasma conditions and for HMDS plasma treatments. With the effect of the heat treatment, Si-groups might oriented to make brush effect against water and so the surface tension might be decreased to give a contact angle of 1408 (Hochart et al., 2004).

3.2 Vertical flame spread In Figure 4, three-dimensional graphics of vertical flame spreads of HMDSO (a) and HMDS (b) plasma treated polyamide fabrics versus discharge power and time can be seen. After plasma treatments, vertical flame spreads of polyamide fabrics were decreased. The reductions were about 23 percent with HMDSO plasma polymerization

Surface modification and characterization 141 Figure 3. The contact angle of the cotton fabric after plasma þ heat treatment

Surface plot of PA-flammability time-HMDS(s) vs power (W); time(min)

10 9 8 7 0

20

Power (W)

40

60 (a)

15

10

5

0

Time (min)

mmability time-HMDS(s)

mmability time-HMDSO (s)

Surface plot of PA-flammability vs power (W); time (min)

10 9 8 7 0

20

Power (W)

40

15

60

10

0 5 Time (min)

(b)

and 29 percent with HMDS plasma polymerization. The deposition thickness may be insufficient to give flame retardancy property to polyamide fabrics. Vertical flame spread values of cotton fabrics before and after plasma polymerizations can be seen in Figure 5(a) (HMDSO) and (b) (HMDS). The best results were obtained at 20 W-15 min plasma condition for both monomers. The vertical flame spread times were increased between 1.6-2.5 times and 1.5-2.3 times the value of the untreated fabric after HMDSO and HMDS plasma treatments, respectively. The reason of this increase might be the addition of Si atoms in the surface structure by plasma treatments of Si-containing monomers. The peaks in FTIR-ATR spectra strengthened this idea. 3.3 FTIR-ATR analysis FTIR-ATR spectra of untreated and treated cotton fabrics at 20 W-15 min plasma condition with HMDSO and HMDS plasma can be seen in Figure 6. The all spectra show characteristic cellulose peak around 1,000-1,200 cm2 1 while this peak of spectra of HMDS and HMDSO plasma treated fabrics may integrated with Si-O-Si band (Carpentier and Grundmeier, 2005; Grundmeier et al., 2004; Quede et al., 2002, 2004; Wang et al., 2006). Broad absorption band exits at 3,000-3,500 cm2 1

Figure 4. Vertical flame spreads of polyamide fabrics before and after plasma polymerization of (a) HMDSO and (b) HMDS

IJCST 21,2/3

Mability time-HMDS (s)

Figure 5. Vertical flame spreads of cotton fabrics before and after plasma polymerization of (a) HMDSO and (b) HMDS

Mability time-HMDSO (s)

142

Surface plot of co-flammability time-HMDS (s) vs power (W); time (min)

Surface plot of co-flammability vs power (W); time (min)

125 100 75 50 0

20 40 Power (W)

5

10 Time (min)

15

60

0

125 100 75 50 0

20

40

Power (W)

15

60

0 5 10 Time (min)

(b)

(a)

1026.33

A

0.00450 0.0040 0.0035 0.0030 0.0025 0.0020 0.0015 0.0010 0.0005 3275.33 0.0000 3433.03 –0.0005 3654.65 1313.19 2860.50 –0.0010 3797.033540.65 2627.85 1653.91 1433.63 1522.84 3575.16 –0.0015 2305.89 2060.78 –0.0020 –0.00250 3,000 2,000 4000.0 1,500

766.88

1,000

700.0

cm–1

Untreated – 0.00120 – 0.0015

1040.43

– 0.0020 – 0.0025

A

– 0.0030 3335.59

–0.0035 – 0.0040

Figure 6. FTIR-ATR spectra of untreated treated fabrics

3815.09 3686.93 – 0.0045 3744.06

2768.95 2508.26 2128.94 2390.90

1559.17

1327.87

728.30

– 0.0050 – 0.00540 4000.0

3,000

2,000

1,500

1,000

700.0

0.00300 1013.06 0.0028 0.0026 0.0024 0.0022 0.0020 0.0018 0.0016 0.0014 0.0012 3257.01 0.0010 0.0008 2857.40 0.0006 3644.91 1427.28 817.04715.41 2609.36 2251.07 1343.16 0.0004 3815.91 2326.49 2084.07 1654.37 0.0002 3849.38 2468.08 757.06 0.0000 3986.98 – 0.0002 – 0.0004 – 0.00050 4000.0 3,000 2,000 1,500 1,000 700.0

cm–1

cm–1

After HMDSO plasma polymerization (20W-15min)

After HMDS plasma polymerization (20W-15min)

because it contains many hydroxyl groups. There are CH stretching band at 2,800-3,000 cm2 1, CH wagging at 1,312-1,316 cm2 1, CH bending at 1,420-1,430 cm2 1. These are also characteristic bands related to cellulose. A peak around 1,650 cm2 1 may be due to the adsorbed water (Adebajo and Frost, 2004; Chung et al., 2004; El-Shafei et al., 2006; Wang et al., 2006). The peak approximately at 817 cm2 1 in spectra of treated fabrics may be indicated Si-C bonds in the surface structure, because this peak is not seen at the spectrum of untreated fabric (Chang et al., 2002; Quede et al., 2002, 2004). Height of the peak at the spectrum of HMDS plasma treated fabric is bigger than the peak at the spectrum of HMDSO treated fabric. This was expected because HMDS might be broken from the Si-Si bond, which was the weakest bond in the structure with bond energy of 193 kJ mol2 1, while HMDSO might be broken from the C-Si bond, which was the weakest bond with the bond energy of 318 kJ mol2 1 (De Buyl, 2001).

This also helps the explanation of the orientation after heat treatment for HMDSO plasma treated fabrics. If HMDSO had been broken from the Si-C bond, the Si-O-Si groups, which were bonded to the fabric surface, might have had free Si-end to orient. Besides, if HMDS had been broken from SI˙-Si bond, the Si-(CH3)3 groups might have been bonded to the surface. Therefore, HMDS plasma treated fabrics might have no free Si-end to orient and might have more Si-C bond than HMDSO plasma treated fabrics. A peak at approximately 1,430 cm2 1, can be seen in the spectra of HMDS treated and untreated fabrics. This peak may correspond to the absorption of the CH2 group present in cellulose. For HMDS plasma treatment, the intensity of the peak was increased. This may be due to the -Si-(CH3)3 groups from HMDS, which may be bonded to the surface. For HMDSO plasma treatment, the peak was disappeared. This may be attributed to the ablation process during plasma polymerization of HMDSO (Carpentier and Grundmeier, 2005; Grundmeier et al., 2004).

Surface modification and characterization 143

3.4 Atomic force microscopy investigation AFM of untreated and treated cotton fabrics at 20 W-5 min plasma condition with HMDSO and HMDS plasma can be seen in Figure 7. The mean roughness value after HMDSO plasma polymerization was 56 percent higher than the mean roughness of the untreated fabric. This may indicate the ablation effect mentioned in the previous section. In addition to this, mean roughness value after HMDS plasma polymerization was 51 percent lower than that of untreated fabric. This may be attributed to the deposition of -Si-(CH3)3. If -CH3-groups cover the surface, this

Untreated

After HMDSO plasma polymerization

After HMDS plasma polymerization

Figure 7. AFM of untreated and treated fabrics

IJCST 21,2/3

144

may reduced the roughness of the surface. This result is also parallel with the results of the FTIR-ATR study. 4. Conclusion Hexamethyldisiloxane and hexamethyldisilane monomers were used to plasma surface modification of polyamide and cotton fabrics. Water repellency and vertical flame spread were selected as the main criteria for obtaining glow-discharge plasma modification of cotton and polyamide fabric surfaces. FTIR-ATR analysis and atomic force microscopy were applied to characterize the changes on the fabric surfaces. Water repellency of polyamide fabrics were improved by the plasma polymerization of the hydrophobic monomers, HMDSO and HMDS. For the cotton fabric, water repellency property was enhanced with a heat treatment after HMDSO plasma polymerization. This may result from the orientation of Si-groups. Flame retardant properties of cotton fabrics were improved by plasma treatments. These results were affirmed by FTIR-ATR and AFM studies. These findings have potential of application in many areas of technical textiles and clothing. References Adebajo, M.O. and Frost, R.L. (2004), “Infrared and 13C MAS nuclear magnetic resonance spectroscopic study of acetylation of cotton”, Spectrochimica Acta Part A, Vol. 60, pp. 449-53. Arefi, F., Andre, V., Montazer-Rahmati, P. and Amoroux, J. (1992), “Plasma polymerization and surface treatment of polymers”, Pure and Applied Chemistry, Vol. 64 No. 5, pp. 715-23. Carpentier, J. and Grundmeier, G. (2005), “Chemical structure and morphology of thin bilayer and composite organosilicon and fluorocarbon microwave plasma polymer films”, Surface & Coatings Technology, Vol. 192, pp. 189-98. Chang, T.C., Mor, Y.S., Liu, P.T., Tsai, T.M., Chen, C.W., Chu, C.J., Pan, F.M., Lur, W. and Sze, S.M. (2002), “Trimethylchlorosilane treatment of ultra low dielectric constant after photo resist removal processing”, Journal of The Electrochemical Society, Vol. 149 No. 10, pp. F145-8. Chung, C., Lee, M. and Choe, E.K. (2004), “Cheracterization of cotton fabric scouring by FTIR-ATR spectroscopy”, Carbohydrate Polymers, Vol. 58, pp. 417-20. Cireli (Aksit), A., Kutlu (Kilic), B. and Mutlu, M. (2007), “Surface modification of polyester and polyamide fabrics by low frequency plasma polymerization of acrylic acid”, Journal of Applied Polymer Science, Vol. 104, pp. 2318-22. De Buyl, F. (2001), “Silicone sealants and structural adhesives”, International Journal of Adhesion & Adhesives, Vol. 21, pp. 411-22. El-Shafei, A., Knittel, D. and Schollmeyer, E. (2006), “Anionically-modified cotton and surface layer formation with polyelectrolytes”, AUTEX Research Journal, Vol. 6 No. 3, pp. 175-81. Grundmeier, G., Thiemann, P., Carpentier, J., Shirtcliffe, N. and Stratmann, M. (2004), “Tailoring of the morphology and chemical composition of thin organosilane microwave plasma polymer layers on metal substrates”, Thin Solid Films, Vol. 446, pp. 61-71. Hegemann, D., Brunner, H. and Oehr, C. (2003), “Evaluation of deposition conditions to design plasma coatings like SiOx and a-C:H on polymers”, Surface & Coatings Technology, Vol. 174-175, pp. 253-60.

Hochar, F., De Jaeger, R. and Levalois-Gru¨tzmacher, J. (2003), “Graft-polymerization of a hydrophobic monomer onto PAN textile by low-pressure plasma treatments”, Surface & Coatings Technology, Vol. 165, pp. 201-10. Jama, C. and Delobel, R. (2007), “Cold plasma technologies for surface modification and thin film deposition”, in Duquesne, S., Magniez, C. and Camino, G. (Eds), Multifunctional Barriers for Flexible Structure Textile, Leather and Paper, Vol. 97, pp. 109-24, Springer Series in Materials Science. Luo, S. (2002), “Surface modification of textile fibers and cords by plasma polymerization for improvement of adhesion to polymeric matrices”, doctoral thesis, University of Cincinati, Cincinnati, OH. Quede, A., Mutel, B., Supiot, P., Jama, C., Dessaux, O. and Delobel, R. (2004), “Characterization of organosilicon films synthesized by N2-PACVD. Application to fire retardant properties of coated polymers”, Surface & Coatings Technology, Vol. 180-181, pp. 265-70. Quede, A., Jama, C., Supiot, P., Le Bras, M., Delobel, R., Dessaux, O. and Goudmand, P. (2002), “Elaboration of fire retardant coatings on polyamide-6 using a cold plasma polymerization process”, Surface & Coatings Technology, Vol. 151-152, pp. 424-8. Schartel, B., Ku¨hn, G., Mix, R. and Friedrich, J. (2002), “Surface controlled fire retardancy of polymers using plasma polymerisation”, Macromolecular Materials and Engineering, Vol. 287, pp. 579-82. Wang, Q., Fan, X., Gao, W. and Chen, J. (2006), “Characterization of bioscoured cotton fabrics using FT-IR ATR spectroscopy and microscopy techniques”, Carbohydrate Research, Vol. 341, pp. 2170-5. Yasuda, H.J. (1981), “Glow discharge polymerization”, Journal of Polymer Science: Macromolecular Reviews, Vol. 16 No. 1, pp. 199-293. Zanini, S., Ricardi, C., Orlandi, M., Esena, P., Tontini, M., Milani, M. and Cassio, V. (2005), “Surface properties of HMDSO plasma treated polyethylene terephthalate”, Surface & Coatings Technology, Vol. 200, pp. 953-7. Corresponding author Bengi Kilic can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

Surface modification and characterization 145

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

IJCST 21,2/3

A methodology of selecting a suitable garment for sports use N. Martı´nez, J.C. Gonza´lez, D. Rosa and E. Alca´ntara

146

Institute of Biomechanics of Valencia, Universidad Politecnica De Valencia, Valencia, Spain Abstract Purpose – Choosing the adequate garment for sports practice in adverse weather conditions, either cold or hot, is an aspect of great influence on activity performance. The purpose of this paper is to describe how the Institute of Biomechanics of Valencia has developed a methodology which allows assessing the fit of the garment to the real situation of use by evaluating its influence in the thermoregulatory response of the human body. Design/methodology/approach – Under controlled environmental conditions and at fixed activity levels, two shirts are tested in the laboratory. Eight subjects performed a test which consisted of six phases of different activity level in two conditions (258C/50 percent RH and 108C/60 percent RH). Throughout the test, physiological parameters of the thermal response as well as work load indicators are registered. Skin temperature at three different locations (chest, arm, and thigh), microclimate variables in some areas of subject-garment interface (in armpit and upper back) and heart rate are measured continuously. Six samples of sweat are also collected regularly from dorsal region during the test to estimate the sweating rate and the loss of salts. Weight loss is also checked before and after performing the test to estimate the dehydration level. Subjects will be asked during the test about humidity and temperature perception on the body as a whole or by different zones. The results allowed measuring a significant influence of the shirt in skin temperature. Therefore, the methodology developed for studying of the user-product interaction through the assessment of the thermophysiological response and the subjective perception allows recommending the comfort ranges for each piece of garment as well as indicating those work load and environmental conditions for which the influence of garment on user’s performance is optimal. Findings – The user-product interaction through the assessment of the thermophysiological response and the subjective perception allows recommending the comfort ranges for each piece of garment as well as indicating those work load and environmental conditions for which the influence of garment on user’s performance is optimal. Originality/value – Choosing a suitable garment for sports practice in adverse weather conditions, either cold or hot, is an aspect of great influence on activity performance and this paper presents new results. Keywords Thermal testing, Human physiology, Clothing, Sports Paper type Research paper

International Journal of Clothing Science and Technology Vol. 21 Nos 2/3, 2009 pp. 146-154 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910933871

1. Introduction High performance is a desirable objective for all professionals, especially at sport competitions. At the same time, comfort is well-known as an important item to improve the quality and efficiency of an activity. Many people daily perform their tasks in adverse thermal conditions (for example at work like fire-fighters or people working in foundry workshops or cold-storage rooms) which increase demands on the Collaborations of the Physiology Department of the University of Leo´n (Spain) and the Research Unit for Physical and Sports Performance of the University of Valencia (Spain) have been much appreciated in this work.

thermoregulatory system. This in the end can reduce human abilities and performance with a negative effect upon comfort and even putting health and safety at risk. In this sense, thermal requirements are determined for each person by a great number of factors such as environmental conditions or the physical activity level and also intrinsic human factors but clothing can have a determinant role and therefore, exposition to cold or heat environments should be done with proper thermal protective equipment. With this respect, the textile industry’s current development has flooded various sectors (such as fashion, sports clothing, protection equipment, etc.) with a wide variety of technical textiles offering sufficiently diversified properties to be capable of satisfying the user’s needs in many situations. This has not, however, been accompanied by a development in accordance both with assignation criteria and methods for evaluating their performance. For these reasons, an appropriate clothing selection is not usually easy to be carried out as far as a correct information management does not exit at present. To select the clothing according to end-users requirements, information about thermal properties of the products and the thermal interaction between users and products in the different environments and situations must be characterised and related to users’ necessities. This paper focuses on the development of a clothing selection methodology according to thermal requirements. 2. Material and methods Two commercial shirts were tested at different situations of use in the laboratory. Both commercial shirts were made of the same textile (100 percent polyester) but with different thickness and consequently different thermal properties. Both were thermally characterised (thermal resistance Rct and evaporation resistance Ret) according to Skin Model test described in ISO 31092 standard. Thermal properties are shown in Table I. Eight trained male subjects (Average age: 25.25 years (SD: 1.38); average height: 1.76 m (SD: 0.068); average weight: 72.45 Kg (SD: 0.67)) performed a test on a tread-mill which consisted of six phases at different activity level in two environmental conditions (258C/50 percent RH and 108C/60 percent RH). The activity test designed lasted for 60 min and included different phases of activity and rest in order to cover the wide range of activity possibilities for the use of the shirts. The test is described below: (1) Acclimatization (10 min). This step includes a first phase of walking at a speed of 2-3 Mph during 5 min followed by asecond phase of jogging at 5 Mph. (2) Rest (5 min). The subject rested but he was not allowed to seat at any moment. (3) Low intensity activity (15 min). This step includes three phases of running at different speeds (5 min of duration each): 6-8 Mph. (4) Rest (5 min). The subject rested but he was not allowed to seat at any moment. (5) High intensity activity (15 min). This step includes threephases of running at different speeds (5 min of duration each): 8-10 Mph. (6) Rest (10 min). The subject rested but he was not allowed to seat at any moment. The methodology proposed consists of measuring different kinds of variables regarded to the thermal state throughout the test described above. Objective measurements as thermal comfort variables (skin temperature and microclimate variables) or

Selecting a suitable garment for sports use 147

IJCST 21,2/3

148

physiological (heart rate, fluid loss, and sweat composition) were done together with subjective variables (thermal perception and comfort degree). Combined analysis of both allowed to assess the garment fitness to user thermal needs. 2.1 Thermal comfort variables and heart rate Skin temperature at three different locations (chest, arm and thigh) and microclimate variables in subject-garment interface (at the armpit and upperbacks) were measured each 2 s during the whole test by means of digital sensors (developed by the IBV). Heart rate was measured as well using a POLARe heart rate monitor. Figure 1 shows the exact locations of the sensors. After checking the existence of a pattern in chest, arm and thigh temperature registers; average skin emperature has been calculated according to equation (Daanen, 1997): TAverage Skin Temprature ¼ 0:36TArm þ 0:25TChest þ 0:34TThigh þ 1:19

ð1Þ

From the curve of average skin temperature, 13 variables were obtained by parameterization of the calculated registered. The selected values corresponded with the phase changes of the activity test. 2.2 Sweat samples Six samples of sweat were also collected regularly from dorsal region during the test in order to estimate the sweating rate and the loss of salts. The location of the patches can be observed in the Figure 2. The patches were removed at the beginnig and at the end Table I. Thermal properties of the shirts participating in the study

Shirt Low resistance High resistance

Thermal resistance Rct (m2 K/W)

Evaporation resistance Ret (m2 Pa/W)

0.0342 ^ 0.0002 0.512 ^ 0.0005

3.56 ^ 0.03 4.89 ^ 0.01

Heart rate

Figure 1. Instrumentation

1

Selecting a suitable garment for sports use

2

3

4

5

6

149

Figure 2. Sweat patches distribution

of the resting steps for monitorizing the increment in sweating due to activity bouts and rest periods. After having been retired, each patch was inmediately encapsulated in a tube and placed in a freezer at 2 208C to avoid evaporation. The sweat samples were removed from the patches by centrifugation in order to be analysed for sodium and potassium by means of a mass spectrometer. The methodology for collecting the sweat samples has been developed according to literature (Alvear et al., 2003). 2.3 Fluid loss Fluid loss was registered by means of weighting the subject in nude conditions before and after performing the test to estimate the dehydration level. To estimate the absorption capacity of each piece of garment, all of them were also weigthed separately before and after the test. The parameters obtained from the collected information correspond to the loss of weight of the subject and the increment of weight of each piece of clothing and measurement equipment worn by the subject. 2.4 Thermal perception The aim of the subjective test was to gather the thermal perception of users and about the sensations of general comfort or the incidence of discomfort associated with a certain thermal state. In this way, the subject’s opinion was recorded at the same time as the temperature and humidity data during the different stages of the study. The survey was divided into a preliminary survey and another while the activity test was being carried out. Different questions were asked before and after the subject did the test, to assess their tendencies in relation to the thermal perception of a situation. Subjects were asked during the test about humidity and temperature perception on the body as a whole or by different zones. Global comfort degree was also registered in each case. The survey was based in a five-point scale as follows: very warm (Dry); warm (Dry); neither warm (Dry) nor cold (Wet); cold (Wet); very cold (Wet) for thermal perception and extremely uncomfortable; very uncomfortable; uncomfortableuncomfortable; slightly uncomfortable; comfortable for comfort degree. 2.4.1 Data analysis. For each parameter of the study obtained from the collected information an statistical analysis was done using SPSS 14.0 software.

IJCST 21,2/3

The analysis consists firstly of a descriptive analysis, followed by an analysis of the variance (ANOVA for parametric variables and Kruskall-Wallis analysis for non-parametric ones) to check the influence the shirt for each environmental condition on each thermophysiological variable of the body response.

150

3. Results 3.1 Thermal comfort variables and heart rate 3.1.1 Descriptive analysis. Figure 3 shows the evolution of the parameters obtained from the average temperature of the skin for all the subjects at each condition for both shirts in cold and heat conditions. Analysis of the variance (ANOVA) results showed that the shirt with higher thermal resistance produces higher average skin temperature significantly ( p ¼ 0.05) during the test with the exception for high level of activity (end of the test) independently from the environmental conditions. Regarding the heart rate and after the parameterization of the curves, a descriptive analysis was done for estimating the average heart rate values for each test configuration. After the descriptive and having checked that the ANOVA did not show significant differences for any heart rate parameter due to the shirt or the environmental conditions (in this case, heart rate for 258C/50 percent RH was systematically higher but not significantly), individual differences were considered. The data shows a great variability in heart rate between subjects. It is discussed to be a possible reason for not detecting significant differences due to study factors. 3.2 Sweat samples Having collected a sample at each phase activity change in case of both shirts has shown in a first place how for shirt with higher thermal resistance sweat amount has been systematically higher at every instant collection. Besides, sodium and potassium concentration were found to be also higher in this case. Regarding to the sodium and potassium contents it can be observed that the concentration modifies during the Mean skin temperature 36 35

25°C

Temperature (°C)

34 33 32 31 30

10°C

29

Figure 3. Evolution of the average skin temperature in each case of the study

28

running

rest

running

rest

running

Low Resistance High Resistance rest

27 T0

T660

T1260

T1860 Time (s)

T2460

T3060

T3660

test: after having sweat for some time during the test, the salts concentration found in sweat slightly decreases (Figure 4). 3.3 Fluid loss Descriptive analysis gave an estimation of the weight loss. Differences in weight loss have been found due to the environmental conditions by means of ANOVA ( p , 0.05) but not due to the shirts. Exercising in warm conditions (258C/50 percent RH) results in a bigger weight loss than exercising in cool conditions (108C/60 percent) regardless of the garment (Figure 5).

Selecting a suitable garment for sports use 151

3.4 Thermal perception Data representation for “Perception of temperature” and “Thermal comfort degree” was carried out by means of histograms done separately for each test condition (environmental condition and shirt) and for the survey at the end of each activity step. “Perception of temperature” was systematically found “Warmer” at 258C than at 108C regardless of the shirt and the instant of the test. In the next figures a histogram for the “Perception of temperature” and another one for the associated “Comfort degree” are presented.

High thermal resistance shirt

2,000 1,800 1,600

Low thermal resistance shirt

1,400 35

40

55

Potasium concentration (ppm)

2,200

300 High thermal resistance shirt

270 240 Low thermal resistance shirt

210 180

65

35

Time

40

55

65

Figure 4. Evolution of sodium and potassium concentration in sweat during the test for each shirt

Time

Environmental temperature

1200.0

10˚C 25˚C

1000.0

APSU

Sodium concentration (ppm)

2,400

800.0 600.0 400.0 200.0 1

2 Shirt

Figure 5. Box and whiskers diagram for weight loss for each condition and shirt

IJCST 21,2/3

The relationships between both allowed to know which sensation was pleasant for the user. Furthermore, adding the objective thermal comfort variables is possible to decide which level of thermophysiological variables as skin temperature are found satisfactory and give comfort sensation to the user (Figure 6). Analysis of the variance (Kruskal-Wallis) showed that, only the environmental conditions caused significant differences ( p , 0.05) at any activity level.

152 4. Discussion and conclusion After this study, it has been concluded that it is possible to estimate the adaptation of the garment for a particular situation of use defined by the user, the environmental conditions and the activity performed with different degrees of accuracy. The proposed methodology consists of different steps with an increasing degree of complexity (Figure 7). It starts by gathering subjective opinion of users. This level of the methodology in its own is capable of giving enough information about the product in order to know the adequation or not of the product by means of a simple survey to the user. The results of the thermal perception study have allowed us to find differences between tested garments and between the performances of a garment under different thermal requirements. However, and in spite of that this type of information is determinant in the purchase decision, information presented in this form do not provide designers and manufactures with designs criteria for products development. Second step consist of combining this with objective measurement of thermoregulatory response. In the test, we measured skin temperature and microclimate what provides more useful information to include in the product development process. Thermal comfort variables analysis has shown that indeed skin temperatures are different due to the garment and the environments and has also shown that the influence of the garment on them is presented as a function of the activity level. By means of this values and their relationship with the user perception enough information is available to know what objetive values of thermal comfort variables are perceived as comfortable by the users. In fact, the point of measuring both simultaneously is having a translation of the user perception into quantified variables. Third level of complexity consider measuring variables related to user health that CODCAM

CODCAM 2

1

Frequency

Figure 6. Histograms of the thermal perception (left) and the degree of comfort (right) at the end of the last phase of activity for each condition and garment

2 0 8 6 10

4 2 0 0

1

2

3

4 0 T3060

1

2

3

4

Frequency

25

4

ENVIRONMENTAL CONDITION

6

2

6 5 4 3 2 1 0 6 5 4 3 2 1 0

25

10

0

1

2

3

4 0 ST3060

1

2

3

4

ENVIRONMENTAL CONDITION

1 8

Disadvantages

Advandages

Subjective response

Traditional methods. Minimal instrumentation requriments.

Not much solid. No provision of design criteria for textile and garment industry.

Objective response

More solid. More information about the product. Criteria about the effect of the garment on the body thermal response.

More complexity. Instrumentation and measurement aids requirements.

Sport performance and health

More information. Criteria about the effect of the garment on the physiological and work response (sweet analysis, lactase, heart rate)

More complexity. Increase in the instrumentation needs.

includes heart rate and sweat composition. Our results did not show differences in heart rate due to the garment. In this case, a high individual variability was detected and it was discussed it may conceal any possible difference due to the shirt. However, the effect of the environmental condition had enough size to be independent of the individual variability (anyway it was not in a significant way). Regarding sweat composition (sodium and potassium contents), with the methodology proposed we could measure both parameters on the sweat at different moments throughout the test, getting information about the dynamical behavior of the sweating mechanism. In our study, it was found the same trends for the evolution of sodium and potassium concentrations. Besides, levels of both parameters differ between garments possibly due to the sweat rate. In this sense, it has been reported in many occasions in the literature that lower sweat rate let the re-absorption of sodium in the duct of the gland (Sato and Dobson, 1970). The decrease seen in Figure 4 is supported by numerous studies in the literature, and reveals the adaptive nature of the perspiration mechanism. During prolonged physical exercise, the reduction in volemia and the loss of salts stimulates the secretion of aldosterone, which is responsible for keeping sodium at adequate levels (Willmore and Costill, 1994). The sodium retention induced by the rise in aldosterone levels has been extensively documented for prolonged periods of exercise (over 6 h) (Patterson et al., 2000), though the effect does not appear so well documented for intermediate periods of activity. The research team postulates the existence of a lesser effect of this hormone (aldosterone), which in combination with the effect of training upon the sweat glands in subjects adapted to situations of thermal stress (Ahlman and Karvonen, 1961; Patterson et al., 2000), could explain why beyond a given moment during perspiration the secreted sweat becomes more hypotonic as a defence measure against salt depletion – though the evidence in this sense is still inconclusive. By improving and implementing this methodology, people involved in sports world and other professionals as safety managers of the enterprises, would be able to choose the most suitable gear for each person, according to the kind and duration of the activity which

Selecting a suitable garment for sports use 153

Figure 7. Levels of complexity in thermal comfort studies

IJCST 21,2/3

154

is performing and the environmental conditions in which it is developed. Choosing the right equipment enhance the efficiency and performance by means of a higher user’s comfort and satisfaction. References Ahlman, K.H. and Karvonen, M. (1961), “Stimulation of sweating by exercise after heat induced ‘fatigue’ of sweating mechanism”, Acta Physiologica Scandinavica, Vol. 53, pp. 381-6. Alvear-ordenes, I. (2003), “Efectos del ejercicio fı´sico sobre los niveles de amonı´aco y urea en el sudor y en la sangre: relacio´n con el dan˜o muscular”, Tesis de Doctorado, Departamento de Fisiologı´a, Universidad de Leo´n. Daanen, H. (1997), “Central and peripherial control of finger blood flow in the cold”, Thesis, Vrije Universiteit. Patterson, M.J., Galloway, S.D. and Immo, M. (2000), “Variations in regional sweat composition in normal human males”, Experimental Physiology, Vol. 85 No. 6, pp. 869-75. Sato, K. and Dobson, R.L. (1970), “The effect of intracutaneous daldosterone and hydrocortisone on human eccrine sweat gland function”, The Journal of Investigative Dermatology, Vol. 54, pp. 450-62. Willmore, J. and Costill, D. (1994), Physiology of Sport And Exercise, Human Kinetics, Windsor. Further reading Alvear-ordenes, I. (1998), “I´ndice de excrec¸a˜o de amonı´aco no suor em atletas de fundo durante exercı´cio ate´ exausta˜o”, Tesis de Ma´ster em Cieˆncias, Universidade federal do Rio de Janeiro. Alvear-ordenes, I. (1998), “I´ndice de excrecio´n de amonı´aco en el sudor de atletas de fondo durante ejercicio hasta la ratiga”, Tesis de Maestria, Universidad federal de Rı´o de Janeiro. Alvear-ordenes, I., Flegner, A.J. and Gonza´lez-Gallego, J. (2001), “I´ndice de excrecio´n de amonı´aco en el sudor de atletas de fondo durante ejercicio hasta la fatiga”, Arch. Med. Dep., Vol. XVIII No. 86, pp. 593-9. Alvear-ordenes, I., Garcı´a-lo´pez, D., De paz, J.A. and Gonza´lezgallego, J. (2005), “Sweat lactate, ammonia, and urea in rugby players”, International Journal of Sports Medicine, Vol. 26 No. 1, pp. 632-7, 1-6. Morgan, R.M., Pattterson, M.J. and Nimmo, M.A. (2004), “Acute effects of dehydration on sweat composition in men during prolonged exercise in the heat”, Acta Physiologica Scandinavica, Vol. 182, pp. 37-43. Rav-acha, M., Hadad, E., Epstein, Y., Heled, Y. and Moran, DS (2004), “Fatal exertional heat stoke: a case series”, The American Journal of Medical Sciences, Vol. 328, pp. 84-7. Corresponding author N. Martı´nez can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

IJCST 21,4

C2CAD: a sustainable apparel design and production model

166

Department of Family and Consumer Sciences, Illinois State University, Normal, Illinois, USA

Hae Jin Gam

Received 19 April 2008 Revised 20 October 2008 Accepted 20 October 2008

Huantian Cao Department of Fashion and Apparel Studies, University of Delaware, Newark, Delaware, USA

Cheryl Farr Department of Design, Housing and Merchandising, Oklahoma State University, Stillwater, Oklahoma, USA, and

Lauren Heine Lauren Heine Group LLC, Bellingham, Washington, USA Abstract Purpose – The purpose of this paper is to develop and implement a new sustainable apparel design and production model, cradle to cradle apparel design (C2CAD), that provides guidelines for apparel designers and manufacturers to solve some of the sustainability problems related to apparel production. Design/methodology/approach – The C2CAD model was developed by integrating McDonough and Braungart’s “cradle to cradle” model into existing apparel design and production models. Knitwear design and production was used to implement the C2CAD model as a proof of concept. The performance and cost of the C2CAD knitwear were evaluated. Findings – The C2CAD model has four main steps: problem definition and research; sample making; solution development and collaboration; and production. Following the four steps and with an international collaboration similar to current apparel industry practices, “Four-season sustainability” children’s knitwear prototypes were developed. Produced with an acceptable manufacturing cost, the products have good mechanical and color fastness performance. Practical implications – The C2CAD model provides practical guidelines for apparel designers and manufacturers and allows them to address all three pillars in sustainable development: economic development, social development, and environmental protection. Originality/value – The C2CAD is the first apparel design and production model that emphasizes sustainability in addition to functional, expressive, and aesthetic considerations. The production process of “Four-season sustainability” children’s knitwear demonstrated the implementation of C2CAD model in sustainable apparel design and production. Keywords Sustainable design, Design and development, Modelling, Knitwear, Clothing Paper type Research paper

International Journal of Clothing Science and Technology Vol. 21 No. 4, 2009 pp. 166-179 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910959954

The work was funded by The US Environmental Protection Agency under a STAR Research Assistance Agreement No. SU832483 (P3 Award: A National Student Design Competition for Sustainability Focusing on People, Prosperity, and the Planet). The authors thank Mr Albin Ka¨lin and Dr Alain Riviere of Environmental Protection Encouragement Agency (EPEA), Germany for the help in the assessment of the synthetic dyes.

Introduction The goal for sustainable development is human health and well-being in economic, environment, and social systems (Crofton, 2000). However, the apparel industry is a major contributor to environmental problems from textile material manufacturing through apparel production to landfills replete with synthetic fabrics. The production of cotton can cause major environment damage since a large quantity of pesticides, fertilizers, and defoliants are used in cotton fields. In 1999, cotton was the second most heavily pesticide sprayed crop (behind only corn) with approximately 81 million pounds of pesticide applied to upland cotton in the USA (Marquardt, 2001). Some of these chemicals are carcinogens and have severely contaminated our water supply. In manufacturing, the textile industry consumes a large quantity of water and generates large volumes of waste. Textiles are also a chemical-intensive industry and the wastewater from textile processing contains processing bath residues from preparation, dyeing, finishing, slashing, and other operations (US Environmental Protection Agency, 1996). There are several process models very useful in apparel designing and manufacturing (Watkins, 1988; Lamb and Kallal, 1992; May-Plumlee and Little, 1998; LaBat and Sokolowski, 1999). Lamb and Kallal (1992) proposed the functional, expressive, and aesthetic (FEA) consumer needs model that set FEA considerations as the design criteria for different users/markets. This model has been widely used in apparel design and production. However, as far as this project reviewed, no apparel design and production model puts the designer’s role in environmental sustainability into consideration. The purpose of this paper is to document the development and implementation of a new apparel design and production model, which integrates the sustainable design into existing apparel design and production models. The intention is that the new sustainable apparel design and production model will provide guidelines for apparel designers and manufacturers in their work and solve some of the environmental problems related to apparel production. In this paper, knitwear design and production is used in the implementation of the new model as a proof of concept. Development of the sustainable apparel design and production model Apparel design and production models LaBat and Sokolowski (1999) reviewed a variety of design processes, including architecture and environment design, engineering design, industrial product design and clothing design, and found common factors among these processes. As a result, they developed a three-stage textile product design process that provides guidelines for how creative thinking evolves in the textile product design process (LaBat and Sokoloskwi, 1999). According to this model, the apparel design process is divided into three phases: (1) Problem definition and research. (2) Creative exploration. (3) Implementation. LaBat and Sokolowski (1999) demonstrated the application of the three-stage design process in a cooperative industry-university project to redesign an athletic ankle brace. May-Plumlee and Little (1998) recognized the importance of the product development process for apparel manufacturing firms. They reviewed models of the product development process used in apparel industry and developed a comprehensive model, no-interval coherently phased product development (NICPPD) model for

Cradle to cradle apparel design

167

IJCST 21,4

168

apparel, based on the current practices found in the US apparel industry (May-Plumlee and Little, 1998). This model has six phases: (1) Line planning and research. (2) Design/concept development. (3) Design development and style selection. (4) Marketing the line. (5) Pre-production. (6) Line optimization. The NICPPD model is a non-sequential model that includes approved and not approved stages and emphasizes communication. For instance, when the design team or production team discovers problems, the production process can move backward, allowing them to find solutions. Cradle to cradle model Contrary to the traditional one-way “cradle to grave” manufacturing model that does not consider the materials after the use and creates a large amount of waste and pollution, McDonough and Braungart (2002) introduced a model of “cradle to cradle” design in which products can be designed from the onset so that they will provide “nourishment” for something new after useful lives. Materials can be conceived as “biological nutrients” that will easily reenter the water or soil without depositing synthetic materials and toxins, or as “technical nutrients” that will continuously circulate as pure and valuable materials within “closed-loop” industrial cycles, rather than being “downcycled” into lower-grade materials (McDonough and Braungart, 2002). Key to “cradle to cradle” design is the chemical and material assessment protocol that allows designers to assess chemical ingredients against multiple human and environmental health and safety criteria. Once assessed, the ingredients are flagged using color-coding to facilitate decision making. Red indicates an ingredient of potentially high hazard. Yellow is moderate to low inherent hazard and green indicates that the ingredient is inherently benign for the application. Orange designates ingredients for which necessary data are missing (McDonough et al., 2003). “Cradle to cradle” design has been successfully applied by some textile product manufacturers such as Nike, DesignTex, and Shaw Industries (McDonough and Braungart, 2002; Cao et al., 2006). Development of the sustainable apparel design and production model: C2CAD In this paper, “cradle to cradle” model (McDonough and Braungart, 2002) was integrated into existing apparel design and production models (LaBat and Sokoloskwi, 1999; May-Plumlee and Little, 1998) to develop a sustainable apparel design and production model, cradle to cradle apparel design (C2CAD). C2CAD, as illustrated in Figure 1, has four main steps: (1) Problem definition and research. (2) Sample making. (3) Solution development and collaboration. (4) Production.

Step 2: Sample making Step 1: Idea generation

Research User needs -Function -Aesthetic -Economic

Ingredient analysis -Use chemical assessment protocol Green Yellow Orange Red Approved Style selected by design team

Biological nutrients

Safe disposal by consumer

-Find and select product design ideas based on the research

Step 3

Design for disassembly

Technical nutrients

-Cost estimation -Design evaluation • Fit and style -Quality evaluation • Function and performance

Collaboration With supply chain or other companies

Step 4

169

Need modification

Reuse recycle

Accept

Not approved

Develop solution

Cost and design evaluation

Material selection and testing

Problem definition and research -Define problem -Analyses the market and company situation

Cradle to cradle apparel design

Reject

Approved

Production -Sustainability • Energy use • Consider waste: air emission, waste water, solid waste • Transportation

Not approved: Production cannot eliminate or reduce harmful impact Approved: Final production

In step (1) of C2CAD, problem definition and research, designers define problems and analyze market and company situations. Designers need to understand the users’ functional, aesthetical, and economical needs. Conducting research to satisfy these needs and generating design ideas are necessary in this step. At the end of this step, an apparel style should be decided. Step (2), sample making, includes “material selection and testing” and “cost and design evaluation.” According to Pitimaneeyakul et al. (2004), the sample making process is essential to determining whether products can be marketable and producible. A “sample making” step can help companies evaluate the design before they invest significant money and time on real production. In material selection and testing, designers assess chemical ingredients based on the “cradle to cradle” chemical assessment protocol and materials based on their feasibility as biological or technical nutrients. Designers will phase out “red” materials and use more “green” materials. Materials are defined as either biological or technical nutrients. For biological nutrients, disposal without negative environmental impact is necessary. For technical nutrients, designers must decide upfront on the pathways for reuse or recycling of materials after the apparel’s use. If a product is made from a mixture of biological and technical nutrients, or a mixture of different technical nutrients, then separation processes (design for disassembly) are considered so that after separation, different nutrients can follow different pathways for disposing, reuse, or recycle (McDonough and Braungart, 2002). In cost and design evaluation, apparel producers will evaluate function, performance, fit, style, and estimate cost. If the sample does not meet the requirements for these criteria, the design will be modified and re-evaluated.

Figure 1. C2CAD model for sustainable apparel design and production

IJCST 21,4

170

Step (3) of C2CAD is solution development and collaboration. In current industrial divisions, most apparel manufacturers do not produce the textile fabrics, dyes, and other apparel materials. Co-development of fabric with vendors is a component in May-Plumlee and Little’s NICPPD model (May-Plumlee and Little, 1998). In the “cradle to cradle” model, Braungart (2002) proposed “intelligent materials pooling,” which emphasizes collaborative approaches, such as sharing knowledge and resources, as important strategies in sustainable development. In the C2CAD model, apparel designers and manufacturers will collaborate with other companies in the supply chain to solve materials problems, such as phasing out a “red” material. The last step, step (4) in C2CAD, is production. Unlike other apparel production models, C2CAD considers sustainability in production. In addition to safety of material inputs and sustainable material flows, considerations regarding sustainability in production include energy use, air emissions, water, and solid waste. Collaborations with other industries, such as companies that produce renewable energy or use solid waste from apparel production as their biological nutrients or raw materials, are needed to reduce or eliminate harmful impacts during production. The fundamental concept of C2CAD model is that sustainable aspects are considered whenever designers and manufacturers make decisions. Currently many apparel companies use software packages to facilitate design, sourcing and production (Corcoran, 2005). Key features for some industry standard software packages such as WebPDM (Gerber Technology), Visual 2000 ERP (Visual 2000) and ABS (Apparel Business System) include creating design, selecting and ordering materials, allocating inventory, sourcing, calculating cost, and/or producing products. The C2CAD model can be incorporated into the industry software packages. Step (1) in C2CAD model is related to market research and creating designs part in industry software and step (2) is connected with material selecting and ordering. Current software packages provide designers with materials’ features, performance and cost. With the incorporation of the color coded ingredients’ environmental and human health characteristics, designers’ material selection decision is driven by not only aesthetic and economic aspects but also sustainable aspects. Step 3 is connected with outsourcing or collaborating part in software. When designers or manufacturers select other manufacturers for outsourcing or collaboration, the information about these manufacturers regarding sustainability issues will be easily obtained and considered. Step 4 can be connected to the production part of the software package. Implementation of C2CAD model in knitwear design and production Knitwear design and production is used to implement C2CAD model as a proof of concept. The reason of selecting knitwear is that knitwear is made by intertwining yarns in a series of connected loops. Therefore, it is possible to observe almost all processes from the yarn to the final product in knitwear production. Also, the knitwear industry is one of most important sections in the apparel industry. Knitwear, like other apparel products, generates environmental problems throughout its life cycle from raw materials through production to using and disposing. This study followed the steps in C2CAD model in prototype knitwear design and production. Step 1: problem definition and research Since children are vulnerable to the potential toxins such as unsafe dyes used in apparel and one of key points of sustainable development is to preserve the environment for our

children, the consumers of children’s knitwear were selected as our target market. In the past, parents who had small children were very value conscious and did not want to spend a lot of money on clothing for children because children quickly grow out of their clothing (Frings, 2005). However, recent trends for small children’s clothing have changed. Sales of infant and toddler apparel have increased over the last five years while the birth rate has decreased (Verdon, 2003). One of the reasons for this trend could be that there are smaller numbers of children in families; thus, children’s parents and grandparents spend more money on clothing gifts (Verdon, 2003). Another possible reason is that many of today’s parents have their first child when they are in their mid-30s after they have achieved stable careers earning more disposable income (Prendergast and Wong, 2003). Annual birthrate in the USA is 4 million. The steady birthrate attracts companies to pursue children’s market. The design requirements for children’s wear include ease of dressing, washability, durability, and versatility (Frings, 2005). The design theme for this project was “Four-season sustainability” – spring, summer, fall, and winter knitwear apparel for children using the C2CAD model. Step 2: sample making Materials selection and testing along with cost and design evaluation were conducted for the development of C2CAD samples. This project used 100 percent organic cotton fibers, which were grown without harmful chemicals. This made the whole knitwear product a biological nutrient. For biological nutrients, all chemicals used should be able to easily re-enter the water or soil without depositing toxins (McDonough and Braungart, 2002). Eight different colors were incorporated in the design and five natural dyes and three synthetic dyes were used. The five natural dyes were indigo (light blue), brazilwood (pink), logwood (brown), weld (light yellow), and fustic (dark green). The three synthetic dyes were obtained from Ciba Specialty Chemicals. All eight dyes were batch dyed on cotton yarns. The only mordant used to help fix natural dyes on the yarns was salt (NaCl). By collaborating with project partners, the researchers evaluated the three synthetic dyes to make sure they are categorized into “green” (see step 3, next paragraph, for details). In the evaluation process, this study compared the organic and traditional cotton yarns and fabrics and evaluated the dyeing performance as discussed in, Evaluation of the “Four-season sustainability” children’s knitwear section. Step 3: solution development and collaboration While evaluating the three synthetic dyes, the researchers collaborated with the partners Green Blue Institute (GreenBlue) in Virginia, USA, and Environmental Protection Encouragement Agency (EPEA) in Hamburg, Germany. The EPEA further partnered with dye manufacturer Ciba Specialty Chemicals. Ciba provided three synthetic direct dyes, solophenyl blue FGLE 220 percent, solophenyl yellow ARLE 154 percent, solophenyl scarlet BNLE 200 percent, and their material safety data sheets (MSDS), based on European regulations, which contain more documentation on eco-toxicological properties than those used in the US Ciba also provided EPEA with proprietary information on their dye structures and syntheses. Though EPEA did not release Ciba’s proprietary information, EPEA assured that they support the use of these three dyes for biological cycles based on knowledge on structure and synthesis pathways. With the advising from GreenBlue and based on the toxicological information on MSDS sheets and EPEA’s biological nutrient assurance, it was

Cradle to cradle apparel design

171

IJCST 21,4

172

concluded that these three synthetic dyes have no known adverse effect on human and environmental health (“green” category). This process demonstrated how apparel designers and producers can collaborate with chemical manufacturers and third party institutions on chemical evaluation to assure inherent safety. Step 4: production The knitwear production process is divided into four parts: dyeing yarns, knitting fabrics, assembly, and setting. It is possible to dye yarns and knit fabrics by the designers. However, for other processes, special equipment and facilities are needed. Thus, “Four-season sustainability” knitwear working sketches and dyed organic cotton yarns were sent to the production partner, Maeil, a knitwear manufacturer in Kyunggi, Korea. Using an industrial full-fashion intarsia knitting machine, Maeil produced the “Four-season sustainability” knitwear prototypes as shown in Figure 2. Apparel production is a global industry, with manufacturers in both developing and developed worlds (Glock and Kunz, 1995). This project demonstrated the importance of international partnership, as illustrated in Figure 3, in the implementation of C2CAD model for sustainable apparel production. This project had apparel designers and performance evaluators in the USA, third party chemical evaluation consultants in the USA (GreenBlue) and Germany (EPEA), dye manufacturer (Ciba) headquartered in Switzerland with branches in the USA, and apparel producer (Maeil) in Korea. This global sourcing, in which clothing is designed in the USA and produced in another country, is similar to current industry practices. A close collaboration among all the partners, directly and indirectly, is critical in the successful implementation in C2CAD to accomplish sustainable apparel production. Evaluation of the “Four-season sustainability” children’s knitwear Performance evaluation In order to evaluate the performance of the “Four-season sustainability” children’s knitwear, the yarns to be used for the prototypes were dyed, knitted, and tested. Four tests, tensile strength and elongation, pilling, abrasion resistance, and color fastness of

Figure 2. “Four-season sustainability” children’s knitwear

Designer (USA)

Design work forms and yarns

“Cradle to cradle” approach and assessment information

3rd Party chemical consultant (GreenBlue, USA)

Cradle to cradle apparel design

Dye suggestions Products

Dyes and MSDS

3rd Party chemical consultant (EPEA, Germany)

173 MSDS

MSDS Apparel manufacturer (Maeil, Korea)

Dye manufacturer (Ciba, Switzerland & USA)

“biological nutrients” yarns and knit fabrics were conducted. To compare organic and traditional cotton yarns, 100 percent organic and traditional cotton carded yarns with the same thickness (10/2: cotton number 10 and 2 plies) were purchased. Both types of yarns were open end, unwaxed, undyed, and unmercerized. The testing methods were in accordance with American Society for Testing and Materials (ASTM, 2004) and American Association of Textile Chemists and Colorists (AATCC, 2004) standards as summarized in Table I. Yarns were used in tensile strength and elongation tests. For pilling, abrasion resistance and color fastness tests, yarns were knitted on a flatbed-knitting machine in the same gauge to produce knit fabrics and then fabrics were tested. The test results are summarized in Table II. The t-test was used to statistically compare the tensile strength, elongation, pilling, abrasion resistance, and color fastness of the organic and non-organic cottons. For yarn strength, non-organic cotton yarn is significantly stronger and has significantly higher elongation than organic cotton yarn. Both organic and traditional cotton knit fabrics have no pilling after the tumble pilling test. There is no significant difference between organic and traditional cotton knit fabrics in abrasion resistance. Because children grow rapidly, their clothing is used for a relatively short time; therefore, researchers for this project believe the mechanical properties of organic cotton yarns and fabrics are acceptable for or exceed the expectations for the target market. For color fastness, there is one significant difference between organic cotton and traditional cotton: color change after laundry. The evaluation of color fastness is a subjective process with a rating of 5 for no color changing or staining and 1 for significant color changing or staining. The results showed that after laundry, organic cotton knit fabric has significantly less color change (better color fastness) than traditional cotton knit fabric. No significant differences exist between organic and traditional cotton knit fabrics with regards to colorfastness to light and to crocking. Except laundry color fastness for traditional cotton fabric, all color fastness ratings, organic or traditional cotton fabrics, are 3 or higher. This indicates that the natural dyes and “green” synthetic dyes used in this project can deliver good dyeing quality that is acceptable for apparel products.

Figure 3. International collaboration in the “Four-season sustainability” knitwear production

Same as above

Crocking fastness

Taber Model 503 Abraser Atlas Random Tumble Pilling Tester Atlas Launderometer

Atlas Suntest XLS Atlas Crockmeter CM-5

ASTM D3884b ASTM D3512c

AATCC 16e AATCC 8f

d

Thawing-Albert EJA Universal Materials Testing Instrument

ASTM D2256a

AATCC 61

Equipment

Test method

9 (three evaluators, three ratings) 6 (three evaluators, two ratings)

9 (three evaluators, three ratings)

3

3

5

Replicates

Notes: ASTM D2256: standard test method for tensile properties of yarns by the single-strand method (ASTM, 2004); bASTM D3884: standard guide for abrasion resistance of textile fabrics (rotary platform, double-head method) (ASTM, 2004); cASTM D3512: standard test method for pilling resistance and other related surface changes of textile fabrics: random tumble pilling tester (ASTM, 2004); dAATCC 61: colorfastness to laundering, home and commercial: accelerated (AATCC, 2004); eAATCC 16: colorfastness to light (AATCC, 2004); fAATCC 8: colorfastness to crocking: AATCC crockmeter method (AATCC, 2004)

a

Light fastness

Eight fabric samples for each type of yarn (dyed with three synthetic dyes and five natural dyes) Same as above

Laundry fastness

Resistance to pilling

Abrasion resistance

Seven yarn samples for each type of yarn (undyed, dyed with three synthetic dyes and three natural dyes) One fabric sample (undyed) for each type of yarn Same as above

Strength

Table I. Cotton yarns and fabrics performance test procedures Samples

174

Performance

IJCST 21,4

Yarn type Tensile strength and elongation (n ¼ 35)

Strength (pounds) Elongation (%)

Abrasion resistance (n ¼ 12)

Loss in breaking Load (%)

Color fastness after laundry (n ¼ 72) Color transference To multifiber test fabric (n ¼ 72)

Color fading Acetate fibers Cotton fibers Nylon fibers Polyester fibers Acrylic fibers Wool fibers

Light fastness (n ¼ 72) Crocking (n ¼ 48)

Color fading Dry Wet

Mean

t

Organic Non organic Organic Non organic Organic Non organic

2.61371 3.26057 10.96266 11.81909 1.25 1.21

2 11.080 *

Organc Non organic Organic Non organic Organic Non organic Organic Non organic Organic Non organic Organic Non organic Organic Non organic Organic Non organic Organic Non organic Organic Non organic

3.174 2.715 4.403 4.271 3.257 3.201 3.701 3.701 4.403 4.354 4.458 4.556 4.604 4.576 4.326 4.139 4.458 4.417 3.615 3.375

Cradle to cradle apparel design

2 3.558 * 0.266

175

3.524 * 1.372 0.266 0.000 0.484 2 1.064 0.425 1.304 0.292 1.366

Note: *p , 0.05

Cost analysis Owing to the globalization of the apparel industry, it is difficult to analyze the production cost for a specific garment. The cost of a garment can vary significantly depending on manufacturers and suppliers of raw materials and chemicals. Generally, volume manufacturers supply the lowest priced goods (Frings, 2005). In the apparel industry, materials, production pattern making, assembly, finishing, freight, and duty are all cost considerations (Frings, 2005). In order to compare the cost of organic and traditional cotton yarns, the researchers ordered the same amount, 100 pounds, of the same thickness (10/2) of organic and traditional cotton yarns. The cost of the organic cotton yarn was $5.95/pound, and the traditional cotton yarn was $8.00/pound. Both types of yarns were produced in the USA. At the beginning of the project, project team members expected that organic cotton yarns would cost more than traditional cotton yarns. However, the organic cotton yarns were purchased from a company with a shorter supply chain, which resulted in a lower cost. This cost difference between organic and traditional cotton yarns used in this project demonstrates that environmental friendly materials do not necessarily be more expensive for apparel producers if materials go through a proper or short supply chain. Except for yarn cost, all other cost considerations were the same for both organic and traditional cotton knitwear.

Table II. Summary of performance test results

IJCST 21,4

176

The cost analysis for “Four-season sustainability” children’s knitwear is summarized in Table III. In this analysis, the cost of dyeing was not considered because it is very difficult to evaluate the cost of a laboratory-scale dyeing. If dyed yarns were ordered for apparel production, a typical process in the apparel industry, the prices for undyed and dyed cotton yarns were about the same. This knitwear was produced in Korea, and the total cost was about $7.08 for a short sleeve knitwear and $7.38 for a long sleeve one. Compared with mass production in the US apparel industry, a smaller order is typical in Korea, which resulted in a relatively higher price for each garment. Economic, social and environmental benifits According to the 2005 World Summit Outcome Document (United Nations, 2005), the interdependent and mutually reinforcing pillars of sustainable development are economic development, social development, and environmental protection. This paper proposed a sustainable model C2CAD by integrating “cradle to cradle” model into existing apparel design and production models. As summarized in Table IV, C2CAD model allows apparel designers and manufacturers address all three pillars in sustainable development. Using the C2CAD model, apparel designers and manufacturers select chemicals and materials based on their inherent human and environmental health and safety. Therefore, employee occupational safety and the living quality of the people living in the local communities will be improved. Apparel products made from inherently benign materials and chemicals, the health of consumers, especially these people vulnerable to toxins such as children, can be improved. With materials designed to cycle safely at the end of the products’ life, the C2CAD model also helps diminish resource consumption by the apparel industry. Without harmful air, water, and solid

Table III. Cost analysis of children’s knitwear produced in this project

Process

Cost

Place

Organic cotton yarn

$5.95 per pound Short sleeve (0.35 pound: 0.35 £ 5.95 ¼ $2.08) Long sleeve (0.40 pound: 0.40 £ 5.95 ¼ $2.38) $2.5-$3 per piece (300 pieces) $2.5 per piece (300 pieces) $7.08 short sleeve $7.38 long sleeve

USA

Production pattern making Assembly and finishing Total

Elements Economic development

Table IV. Addressing three sustainability pillars in the C2CAD model

Korea Korea

Features and issues

Save apparel manufacturers lots of money in pollution prevention and treatment; improved company image in society and manufacturer competitive edge in the apparel market Social development Better occupational safety and health for employees; better environment and living quality for local communities; better health for users Environmental protection Reduced environmental impact of the apparel industry by reducing the using of toxic chemicals; cyclic material management in the apparel industry, thus diminishing resource consumption

waste release from apparel manufacturers, both the manufacturer and the local community will save a lot of money in pollution prevention and treatment for the short and long-term. With current knowledge, implementing material assessment protocol in “cradle to cradle” model (McDonough et al., 2003) costs much for chemical toxicity research and new material development. Like “cradle to cradle” model and “intelligent materials pooling” (Braungart, 2002), the C2CAD model emphasizes the importance of industrial collaboration and knowledge sharing (step 3 of C2CAD). So, this short-term cost in material research will eventually turn out to be a long-term saving in many aspects such as pollution treatment and material cost. Future study The material used in the “Four-season sustainability” children’s knitwear is 100 percent cotton, which is a biological nutrient. The children’s knitwear production implemented C2CAD in one section of apparel industry, the knitwear production, due to the relative simplicity of knitwear production and the possibility of using biological nutrients only for the complete knitwear product line. There are differences in design and production between knitwear production and other sections, such as woven fabric products, of apparel production. In many cases, garments are made from more than one material using more complicated processes. For example, men’s suit needs approximately 200 different processing and sewing operations (Frings, 2005). Men’s suit consists of main fabrics, interfacing, lining, and buttons. These apparel components use different natural and synthetic materials and the garments have to consist of a mixture of biological (natural) and technical (synthetic) nutrients. Annually, 4.5 million tons of clothing and footwear are produced in the USA and only 1.25 million tons of postconsumer textiles are recovered for next use (US Environmental Protection Agency, 1997). One of obstacles for reusing and recycling materials from post-consumer clothing is that most of apparels are made from more than one material and constructed with many permanent junctions using stitches and adhesives. According to the “cradle to cradle” model (McDonough and Braungart, 2002) and green engineering principles (Anastas and Zimmerman, 2003), the best strategy for effective material management regarding a mixture of biological and technical nutrients in a product is to “design for disassembly,” or design a product to be dismantled for easier maintenance, repair, recovery, and reuse of components and materials. The concept of “design for disassembly” is incorporated into the step 2 of C2CAD model. However, there is no research on the implementation of “design for disassembly” in apparel design and production. In future study, C2CAD model, especially “design for disassembly,” will be implemented in design and production of men’s jacket, a woven apparel product composed of natural and synthetic materials. According to the green engineering principles (Anastas and Zimmerman, 2003), material diversity in multi-component products should be minimized to promote disassembly and value retention. In the jacket design, material diversity can be minimized by having two main components, natural out shell and synthetic lining, and the design for disassembly will focus on the easy separation of these two main components. For instance, 100 percent wool fabric, a natural material or biological nutrient, will be used as the out shell of the jacket. The method to recover biological nutrients is returning the natural materials to nature through composting. Since a mixture of natural materials can be biodegraded in the composting process, different types of natural materials, such as 100 percent cotton

Cradle to cradle apparel design

177

IJCST 21,4

178

threads and natural material buttons, will be used on the out shell of 100 percent wool fabric. For the lining of the jacket, 100 percent polyester fabric, a synthetic material or technical nutrient, will be used. The method to recover technical nutrients is recycling. To recover the same quality materials after recycling rather than “downcycle” to lower quality materials, it is important to maintain the purity of the material and avoid mixing different types of synthetic materials. Therefore, all the components that are affixed on the lining, such as threads and inner pockets, will be 100 percent polyester. Without sacrificing performance, function and aesthetics of the jacket, ease to disassemble different components will be investigated and identified. After disassembly, there will be two main components: natural out shell will enter the process of composting and polyester lining will be recycled. References AATCC (2004), AATCC Technical Manual, American Association of Textile Chemists and Colorists, Durham, NC. Anastas, P.T. and Zimmerman, J.B. (2003), “Design through the 12 principles of green engineering”, Environmental Science & Technology, Vol. 37, pp. 95A-101A. ASTM (2004), Annual Book of ASTM Standards, Vol. 7, American Society for Testing and Materials, West Conshohocken, PA. Braungart, M. (2002), “Intelligent materials pooling: evolving a profitable technical metabolism”, available at: www.mbdc.com/features/feature_sep2002.htm Cao, H., Frey, L.V., Farr, C.A. and Gam, H. (2006), “An environmental sustainability course for design and merchandising students”, Journal of Family and Consumer Sciences, Vol. 98 No. 2, pp. 75-80. Corcoran, C.T. (2005), “Rocket science: not quite yet”, Women’s Wear Daily, June 15, available at: www.freeborders.com/news/art20050615.html Crofton, F.S. (2000), “Educating for sustainability: opportunities in undergraduate engineering”, Journal of Cleaner Production, Vol. 8 No. 5, pp. 379-405. Frings, G.S. (2005), Fashion: From Concept to Consumer, 8th ed., Prentice-Hall, Upper Saddle River, NJ. Glock, R.E. and Kunz, G.I. (1995), Apparel Manufacturing, 2nd ed., Prentice-Hall, New York, NY. LaBat, K.L. and Sokolowski, S.L. (1999), “A three-stage design process applied to an industry-university textile product design project”, Clothing and Textiles Research Journal, Vol. 17 No. 1, pp. 11-20. Lamb, J.M. and Kallal, M.J. (1992), “A conceptual framework for apparel design”, Clothing and Textiles Research Journal, Vol. 10 No. 2, pp. 42-7. McDonough, W. and Braungart, M. (2002), Remarking the Way We Make Things: Cradle to Cradle, North Point Press, New York, NY. McDonough, W., Braungart, M., Anastas, P.T. and Zimmerman, J.B. (2003), “Applying the principles of green engineering to cradle-to-cradle design”, Environmental Science & Technology, Vol. 37, pp. 435A-41A. Marquardt, S. (2001), “Organic cotton: production and marketing trends”, Proceedings of Beltwide Cotton Conference, Anaheim, CA, January 9-13. May-Plumlee, T. and Little, T.J. (1998), “No-interval coherently phased product development model for apparel”, International Journal of Clothing Science and Technology, Vol. 10 No. 5, pp. 342-64.

Pitimaneeyakul, U., LaBat, K.L. and DeLong, M.R. (2004), “Knitwear product development process: a case study”, Clothing and Textiles Research Journal, Vol. 22 No. 3, pp. 113-21. Prendergast, G. and Wong, C. (2003), “Parental influence on the purchase of luxury brands of infant apparel: an exploratory study in Hong Kong”, Journal of Consumer Marketing, Vol. 20 No. 2, pp. 157-69. US Environmental Protection Agency (1996), “Best management practices for pollution prevention in the textile industry”, EPA Manual 625-R-96-004, available at: www.p2pays. org/ref%5C02/01099/0109900.pdf US Environmental Protection Agency (1997), “Source reduction program potential manual: a planning tool”, EPA Manual 530-R-97-002, available at: www.epa.gov/epaoswer/non-hw/ reduce/source.pdf United Nations (2005), “World summit outcome”, available at: http://ec.europa.eu/comm/external_ relations/un/docs/050915_un_summit.pdf Verdon, J. (2003), “Sales of baby clothing increases over 20 percent in 2002”, The Record (Newspaper), Hackensack, NJ, October 4. Watkins, S.M. (1988), “Using the design process to teach functional apparel design”, Clothing and Textiles Research Journal, Vol. 7 No. 1, pp. 10-14. Corresponding author Huantian Cao can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

Cradle to cradle apparel design

179

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

IJCST 21,4

Improvement of dentist gowns – new constraints and new risks Marie Schacher and Youssef Haı¨kel

180

Faculte´ de Chirurgie Dentaire, Universite´ de Strasbourg, Strasbourg, France, and

Ste´phane Berger, Laurence Schacher and Dominique C. Adolphe ENSITM – LPMT, Mulhouse, France

Received 7 July 2008 Accepted 18 November 2008

Abstract Purpose – For years, the main reason for using textiles in the health care sector was to protect the patient from the medical staff. Nowadays, the garment has to play another role and protect the wearer. For dentists, risks can come from saliva which is considered potentially infectious because it frequently contains blood. This paper aims to define dentist gown specifications according to the new situation, and to propose new garments providing safety protective function as well as comfort. Design/methodology/approach – Enquiries, direct interviews as well as internet forums have been used to extract dentists’ requirements taking into account their need of barrier and their comfort concerns. Studies of the spraying area on the gowns have been performed to define the location of the required protection. A study of the warmer zone of the garment via IR camera has been done. Two prototypes have then been constructed and tested. Findings – Images of impacts of drops that could cause cross-infection allow defining the zones which are to be specifically protected. Thermographic images provide maps of hot zones of the garment when worn in working conditions, and information is obtained of desired open space zones which have been designed to create preferential ventilation required for comfort improvement. A second prototype was designed to improve results of the first one. Practical implications – Replacement of current dentists garment in routine situation. Originality/value – Dentists’ gowns used in dental care have not been studied and not been redesigned yet, whereas new dentists are facing new risks and eagerly looking for personal protective equipment providing safety protective function as well as comfort. Keywords Dentists, Design, Health services, Textiles, Protective clothing Paper type Research paper

International Journal of Clothing Science and Technology Vol. 21 No. 4, 2009 pp. 180-192 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910959963

Introduction For years, the main reason for using textiles in the health care sector was to protect the patient from the medical staff. Nowadays, the garment used in the medical area has to play another role because job classifications in which employees have potential for occupational exposure include staff with any instrument handling, or with patient care activities. An occupational exposure can be defined as a specific eye, mouth, nose, or skin contact with blood or saliva, other body fluids, human tissue, or contaminated instruments, resulting from the performance of an employee’s job duties (Fayolle et al., 1981). For dentists, even if there is no epidemiologic evidence to support HIV infection by saliva alone, in dental settings, saliva is considered potentially infectious because it frequently contains blood. Those risks can increase when using new devices (high speed drills) which can increase non-percutaneaous route susceptible to transmit the virus. To fulfil these new requirements, new textile products, with more effective protective functions have to be developed by the scientific community and the textile

researchers in collaboration with the medical staff in order to respond to their need of barrier and their comfort concerns. The current situation For dentist personnel in care situations, two types of garment can be encountered: surgical gowns used in the operating theatre and gowns for routine practice. The products are either used once or laundered and used several times. A rapid increase in penetration of single-use gowns has been observed in the last decades and more and more often, this type of gowns is selected by the surgical teams in the hospital or by the dentist for, e.g. teeth extraction or implant surgery. Single-use products are composed of various materials, among them non-woven fluid resistant fabrics which exhibit good barrier properties and minimise infection transmission. According to the proposed mandatory standard EN 13795 (CEN, 2001), they are now considered as medical devices. Therefore, many studies have been carried out in order to define new gown specifications and propose new garments (Abreu et al., 2003; Rigby et al., 1993; Barker et al., 2000). On the other hand, gowns used for routine application in dentist care have not been studied yet and have not been redesigned. However, they have always been considered as a relatively important workwear able to fulfil many functions: easy disinfection, identified or common appearance, protection against dirt let lone reasonable cost. For a majority of the French dentists, two pieces of cloth are used: a pair of pants and a jacket. The jacket generally has short sleeves, classical style, a front opening and pockets. Depending on the model, different kinds of collar can be found. Hooks, snaps or buttons are mostly used to secure the jacket. The garment is usually made of routine apparel fabrics, cotton or conventional mixed cotton polyester fibers. But nowadays, the performance and protective levels of the gown must go up and there is a need for new personal protective equipment for dentists providing vital or safety protective function as well as comfort. Unfortunately, the cotton barrier, when dry, exhibits some ability to prevent transmission of infection by direct contact, but its effectiveness is compromised once it becomes wet. We are faced to a dual demand: on the one hand a new demand for an efficient bacteriological barrier, which necessarily requires a moisture-repellent (hydrophobic) fibrous material, and on the other hand an optimum comfort for the wearer, which is provided only by a material whereby heat and moisture can easily pass though it: in other words a hydrophilic material. Specific solution There is a number of factor that need to be considered to determine what type of solution to propose in a specific use context. General requirements include technical parameters as well as aesthetic, subjective and financial aspects. Environmental consideration can also be taken into account when deciding between recycling potentially contaminated products and disposing of the products in dumps (Abreu et al., 2006). To evaluate the specific needs of dentists in routine work, enquiries have been carried out among 112 French dentists from September 2006 to December 2006. We also used direct interview strategies and test surveys in some local dentists’ chairs, so as to test them in terms of efficiency and accuracy. Internet forum (Dentalespace, 2007; Eugenol, 2007) has also been required. Several options were proposed.

Improvement of dentist gowns

181

IJCST 21,4

182

Participants were asked questions related to how significant the proposed items were and how much some proposed improvements were accepted. It appears, according to the obtained answers (Figure 1), that the garments have to fulfil the following specifications: comfort, protection, easy care, low price (e50 for traditional materials is considered as a maximum price), and last but not least, the dentist’ image appears to be of utmost importance. Surprisingly, the average life time of reusable products is long: three years and washing temperatures are low (308C for 5 per cent, and 608C for 35 per cent of dentists), which does not involve that the living organisms should be destroyed. Our paper will concentrate on the two most important demands: protection and comfort. Consequently, the new garment has to offer two important functions: be an efficient bacteriological barrier which will be necessarily obtained by using a moisture-repellent layer and propose an optimum comfort to the wearer, which could be provided only by a material allowing heat and moisture transfer. We can observe that these two goals are somehow contradictory. Then a compromise has to be found and a combination of the two functionalities has to be managed. In a garment, there can be vastly different insulation properties depending on how its construction is carried out. The solution was to find a compromise by tailoring the gown so that the cut and make up present protective (hydrophobic) zones where risks of contamination appear as critical; but with fit and openings able to create a micro-climate around the body for the dentist to feel comfortable. Relative to the protection against infection, each pathway that would be a susceptible route to infection should be recognised. For this purpose, a study of the spraying area has been performed to define accurately the location of the need of protection. At the same time, a study of the warmer zone of the garment during the dentist work has been done to assess the comfort of the wearer. Experiments First experiments have been performed to locate the impacts of drops susceptible to cause non-percutaneous cross-infection due to drill or other dental instruments and then define the zones of the garment which are to be specifically protected. 100

Yes, it is important

No, it is not important

80 60 40 20

Pr ot

n Sh or ts le ev es St ai n re pe lle nt Bi oba rri er

io iz at St er il

er al ) D isp os ab le

en (g n

ec t

io

pr ot ec tio n X -R ay

Figure 1. Results of dentists’ inquiries

Co m fo rt

0

The tested garments, white in colour, were commercially available routine cotton/polyester gowns currently used in French dental hospitals. The tests were firstly conducted with the help of a right-hand dentist using classical dental chair where water supply path has been modified to dispense dilute dark dyes (successively green dye food and black reactive dyes). A dentist high-speed turbine has been used – this model is the same as the one normally used to remove tooth decay. In practice, such drills rotate at speeds between about 320,000–rpm and about 500,000–rpm. The turbine is cooled thanks to water. The water, which is then mixed with blood and saliva, forms a septic aerosol. This aerosol, composed of invisible particles less than 10 microns across, is sprayed in the atmosphere due to the velocity of the turbine and can spoil the gowns or part of the body of the dentist. Ultrasonic scalers can show the same phenomenon. In our experiments, the dental drill was firstly used on artificial denture (lab model presented in Figure 2). Additional white sleeves have been added to the initial short sleeves jacket in order to locate any possible spraying or splashing areas on the forearms. Tests were replicated and each test was carried out for 15 min. An image analysis of dye points was performed on white dentist garment to precisely extract arms and torso zones with drop impacts lay-out (Figure 3). The chest zone and the upper part of both arms were coloured to some extent, but it has been observed that the left hand part of the chest zone of the jacket and the front part of right arm from elbow to cuff were more intensively marked by the coloured water. Experimental tests have then been conducted under more realistic conditions, i.e. with a real patient in hospital (Figure 4). The operation performed for this purpose was the scaling of patient mouth. For obvious reasons, no dyestuffs can be used in the mouth of a patient for spraying zone analysis. We have then used an IR camera, with a sensitivity of 0.028C, and an instantaneous field view of 3.3 mm to detect droplets impacts of water, assuming that these droplets will have a colder temperature than dentist jacket temperature (Berger, 2005). The obtained images were analysed and clearly confirm the zones previously defined (Figure 5).

Improvement of dentist gowns

183

Figure 2. Dental drill test on lab model

IJCST 21,4

184

Figure 3. Drops and splash impacts

Figure 4. Tests on human patient

36.7°C 35

30

25

Figure 5. Drops impacts seen through IR camera

21.3°C

A second set of experiments were done aiming at detecting the evolution of the thermal phenomena of the clothing systems which help the thermophysiological balance to be regulated and maintained. The first experiments were the measurements of the temperature of the outer surface of the gowns (front and back). The measurements were obtained via an Infrared-Digital-Video-Camera under standard textile atmosphere, i.e. 65 per cent Relative Humidity and 208C (NF EN 20139, 1992). This method is a non-contact one and includes simple measurement setup and speed. It is based on the thermal response to changes in the heat flux at the surface of the fabric. Lab model and classical dental chair were used to reproduce dentist work position. Thermographical images obtained (Figures 6 and 7) provide maps of hot zones and valuable information of desired open space zones which will be designed for creation of preferential ventilation required for comfort improvement. Films of temperature variation of some specific zones were also recorded to analyse accurately the transient temperature phenomena. Tests were also conducted using wear trial protocols deliberately chosen to reproduce usual conditions of activity and environment: dentist chair located in a room of 5 m2 in hospital (this parameter has been determined as an average surface for dentists’ booth (Suprun et al., 2003), scyalitic lamp, etc. in order to provide

Improvement of dentist gowns

185

35.3°C 35

30

25

20 19.9°C

Figure 6. Thermographical image – back

36.7°C 35

30

25

20 19.5°C

Figure 7. Thermographical image – front

IJCST 21,4

186

comprehensive and informative data for assessing thermophysiological comfort. Unfortunately, no conditioning of the room was available, but room temperature remained stable around 208C throughout the experiments. In both cases, we have observed that the heating parts of the body are mainly located around the shoulders where the fabric is in direct contact with the skin, and on the back part of the body. A lateral analysis has not highlighted difference between left and right side. Similar tests conducted on men and women have given similar results. Proposed solution To address the end-use requirements and balance bio-barrier with comfort function, new pattern cutting of garment used for dentists has been developed. The new garment will be composed of two pieces as already used in everyday practice and trousers have not been modified. The jacket, will be constructed with mandarin collar and longitudinal back openings closable via snaps or buttons, because it will provide more protection in primary exposure areas. No pocket will be provided, and short sleeves will be adopted, taking the large majority of dentists’ demands (97.5 per cent) into account. Dentists claim that they need to frequently wash their arms and hands with water and specific disinfecting soap. The protection of the front part of the jacket has to be improved. Several solutions can be proposed for that purpose. This can be achieved for instance by re-enforcing the fabric, using a waterproof layer. In our case, a polyethylene sheet, like in the case of the single-use surgical gowns, with a sewn assembly on classical Polyester/cotton jacket has been chosen. To improve the heat transfer, the garment has to be constructed to have more room to allow better airflow. The jacket will only be modified because the torso area of a person’s body undoubtedly must be able to breathe more than the area below the waist. The jacket will have features that facilitate heat evacuation during the dentists’ motion, and enable a “mechanical breathe”. This type of feature was obtained in our case thanks to two mesh pieces of fabrics, located on the upper part of the back of the jacket whereby hot air can escape, and fresh air enter the jacket as the user moves about. This creates some the “chimney effect”, pumping out hot air and pulling in cooler outside air. The zones will be chosen thanks to the IR images previously obtained. A first prototype (Figure 8) has been constructed with two vertical stripes of mesh knit fabrics of 10 cm £ 20 cm made up of polyester. These stripes were located on the heating zone that is to say on the shoulder areas. IR pictures obtained from jacket of the prototype worn by dentists showed that, as expected, the stripes show relatively high temperature; that is to say the heat transfer was mainly initiated there. Figure 9 shows the image after one minute wear. However, an analysis of the films acquired through IR camera along the first 5 min (Figures 10 and 11) showed that a heating pipe has been created between the two shoulders and that this pipe, closed by the collar of the gown, traps and accumulates the hot air without any chance to release it. A second prototype has, then, been constructed, with larger breathable areas made of the same mesh knit, starting from underarms and covering shoulders. An additional breathable opening made of the same textile product has been added on the middle waist line zone (Figure 12). It was hypothesized that all air movements between garment and body may reduce body temperatures from rising due to increased

Improvement of dentist gowns

187

Figure 8. First prototype construction

35.2°C 34 32 30 28 26 24 23.6°C

chimney effect: fresh air from the outside could be used to “flush” heated air from the clothing. To compare the two prototypes, the variations of surface temperature of the jacket have been, as previously, recorded by an IR camera. It appears that, as expected, no trapped hot air can be noticed between the shoulders (Figure 13). Two areas have been selected on the two prototypes (Figures 14 and 15): one area is located on the shoulder zone; the second one is in the middle of the back, between the shoulders, where hot air has been previously detected. On the shoulder area, the maximum of temperature has been plotted, and between shoulders, the minimum of temperature has been plotted too (Figure 16).

Figure 9. IR image of the first prototype after 1 min. wear

IJCST 21,4

34.3°C 34 32

188

30 28 26

Figure 10. IR image of jacket temperature after 0 min.

24 22.1°C 34.4°C 34 32 30 28 26

Figure 11. IR image of jacket temperature after 5 min.

Figure 12. Second prototype construction

24

22.1°C

34.2° C 34

Improvement of dentist gowns

32 30

189

28 26 24

22.2° C

Figure 13. IR image of the second prototype after 1 min. wear

34.4° C 34 ARO1

32 ARO2

30 28 26 24 22.1° C

Figure 14. Jacket temperature – prototype 1

34.4° C 34 ARO1

32 ARO2

30 28 26 24 22.1° C

Figure 15. Jacket temperature – prototype 2

IJCST 21,4

Proto1 maxi Proto1 mini Proto2 maxi Proto2 mini

36

190

Temperature (°C)

34

32

30

28

Figure 16. Plotted temperature of the two prototypes

26 0

50

100

150 Time (s)

200

250

300

It has been observed that very quickly, the temperature of the breathable zone seems to be stable and constant for the two prototypes. Maximum temperatures of the two prototypes reflect average man skin temperature (around 348C), which can be explained by the knit mesh structure, presenting an important open mesh area of 37 per cent (Figure 17). However, some differences appear between the two prototypes: the maximum temperature of the second prototype is significantly higher than the temperature of the

Figure 17. Knit mesh structure

first one. The increase in temperature is theoretically beneficial in terms of the heat balance, because it shows a good heat transfer and an improved skin sensory comfort in comparison to the first prototype. Conversely, the minimum temperature in the zone located in the middle of the back starts immediately to rise for the first prototype. For this zone, when the steady state is reached, a plateau appears where the minimum of temperature of the first prototype is always higher than the temperature measured for the second prototype. That is to say that the warm feeling could be considered as minimum for the second prototype in comparison with the first. A clear cold area (less than 248C) can been noticed on the lower part of the jacket, under the large opening located on the middle waist line of the second prototype, suggesting that, as expected, thermal evacuation by ventilation was achieved. Conclusion The new generation of routine dentist gowns can both meet new protection demands and acceptable comfort. To achieve optimal functional product, the garment design and fabrication can be modified to provide optimum comfort and good microbial protection in everyday practice. Some other solutions may be found in textile material and treatment such as adding metals, metal oxides, metallic salts, and quaternary ammonium salts which can be initially applied on the fabric. Their durability, however, must be long enough, the average life time of the product taking into account. The importance which will be attached to these aspects of comfort/protection in the next future is clearly dependent on the sales arguments of the companies which are marketing these products and moreover on the attitudes of the wearers themselves. References Abreu, M.J., Silva, M.E., Schacher, L. and Adolphe, D. (2003), “Designing surgical clothing and drapes according to the new technical standards”, International Journal of Clothing Science and Technology, Vol. 15 No. 1, pp. 69-74. Abreu, M.J., Silva, M.E., Schacher, L. and Adolphe, D. (2006), “Recycling of textiles used in the operation theatre”, Recycling Textiles, Woodhead, Cambridge. Barker, R.L., Scrugg, B.J. and Shalev, I. (2000), “Evaluating operating room gowns: comparing confort of woven and nonwoven material”, International Nonwoven, Journal, Spring, pp. 23-9. Berger, S. (2005), “Contribution a` l’e´tude des proprie´te´s me´caniques des structures textiles par thermographie”, PhD thesis, Mulhouse. CEN/TC 205/WG 14, prEN 13795 (2001), “Surgical drapes, gowns and clean air suits, used as medical devices, for patients, clinical staff and equipment – Part 1: general requirements for manufacturers, processors and products”. Dentalespace (2007), “Dentalespace”, available at: www.dentalespace.com/dentiste/forum (accessed 3 March 2007). Eugenol (2007), “Eugenol”, available at: www.eugenol.com/eugenol (accessed 3 March 2007). Fayolle, P., Herr, P., Holz, J. and Baume, L.J. (1981), “La protection oculaire du me´decin dentiste a` l’e´gard des projections de particules”, SWISS DENT, Vol. 4, pp. 29-40. NF EN 20139 (1992), “Textiles – Atmosphe`res normales de conditionnement et d’essais”.

Improvement of dentist gowns

191

IJCST 21,4

Rigby, A.J., Anand, S.C. and Miraftab, M. (1993), “Medical textiles medical textiles in medicine and surgery”, Textile Horizons, December, pp. 42-6. Suprun, N., lasenko, V. and Ostrovetchkhaya, Y. (2003), “Some aspects of medical clothing manufacturing”, International Journal of Clothing Science and Technology, Vol. 15 Nos 3/4, pp. 220-4.

192

Corresponding author Marie Schacher can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

Effect of high-speed sewing on the tensile properties of sewing threads at different stages of sewing Vinay Kumar Midha Department of Textile Technology, National Institute of Technology, Jalandhar, India

The tensile properties of sewing threads 217 Received 12 July 2008 Revised 12 December 2008 Accepted 12 December 2008

V.K. Kothari and R. Chatopadhyay Department of Textile Technology, Indian Institute of Technology, New Delhi, India, and

A. Mukhopadhyay Department of Textile Technology, National Institute of Technology, Jalandhar, India Abstract Purpose – In this paper, the contribution of dynamic loading, needle and fabric, and the bobbin thread interaction on the changes in the tensile properties of the needle thread are to be investigated. Design/methodology/approach – Tensile properties of the needle thread have been studied at four sewing stages, namely before being subjected to any loading, after dynamic loading, before bobbin thread interaction and after sewing. Findings – It is observed that bobbin thread interaction plays a dominant role in the reduction of tensile properties except breaking elongation in cotton threads. Dynamic loading is mainly responsible for reduction in the breaking elongation of cotton threads. During sewing, there is an increase in initial modulus due to the dynamic loading, which is more in the case of cotton threads than polyester threads. However, the impact of dynamic loading on tenacity, breaking elongation and breaking energy is greater for coarser cotton thread. The contribution of bobbin thread interaction is more for fine threads as compared to coarse threads. Practical implications – Since seam strength is dependent on the thread strength, a reduction in thread strength during sewing will lead to lower seam strength than expected. Therefore, in order to minimize the thread strength reduction, it is important to understand the contribution of different machine elements or processes during sewing. During high-speed sewing, the dynamic and thermal loading are found to be the major causes of strength reduction of needle thread, which can go up to 30-40 per cent. However, the extent of strength loss at different sewing stages is unknown. Originality/value – The study will help in engineering sewing threads, designing of sewing machines and selection of process parameters for controlling loss of useful properties of sewing threads. Keywords Thread, Dynamic loading, Tensile loading, Textile making-up processes Paper type Research paper

Introduction During sewing at high-speed, the needle thread is subjected to repeated tensile stresses at very high rates. The thread also comes under the influence of heat, bending, pressure,

International Journal of Clothing Science and Technology Vol. 21 No. 4, 2009 pp. 217-238 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910959981

IJCST 21,4

218

torsion and wearing. These stresses act on the thread repeatedly and for a fairly prolonged period of time. Depending upon the stitch length, a length of thread before being incorporated in a stitch may pass 50-80 times through the fabric, the needle eye and the bobbin case mechanism (Ukponmwan et al., 2000). Further, local abrasion and cutting of the needle thread can occur, due to impact and rubbing at the top of the needle eye on the short length of thread, and cutting of the thread by the upper edge of the needle at the side of needle eye. The level of these stresses is influenced by sewing speed, machine settings, tensile, bending and surface properties of the needle thread and its interaction with machine elements, the fabric and the bobbin thread. Such severe working conditions in combination with the heat generated in the needle may cause some critical flaws and damages in the fibres or yarn structure, which would ultimately affect the subsequent seam performance of the thread. Further this may also produce potential weak spot in the thread to result in its breakage at high-speed sewing. The stresses created within the thread have a negative effect on the processing and functional characteristics of the thread. In an early research work, Crow and Chamberlain (1969) reported that there is up to 60 per cent reduction in thread strength after sewing. A number of researchers have observed that there is about 30 per cent to 40 per cent reduction in the thread strength after sewing and various reasons including the structural damage, dynamic and thermal loading have been assigned (Sundaresan et al., 1997; Sundaresan et al., 1998; Gersak and Knez, 1991; Rudolf and Gersak, 2006; Lojen and Gersak, 2005). The breaking elongation, initial modulus and toughness also reduce considerably. Scanning electron microscopic studies of sewn thread confirm structural damage in the thread, with displacement of the plies, twist alterations at specific locations and the surface fibres get pulled out of the structure. The damages are mostly concentrated at the interlocking portion of the needle thread in the stitch, where maximum tension, bending and thread-thread abrasion takes place (Sundaresan et al., 1997; Rudolf and Gersak, 2006). All the researchers studied the strength reduction of the needle threads, but very little has been reported about the other tensile properties and their variability during the sewing process. Breaking elongation, work of rupture and initial modulus of the needle thread have significant influence on the performance of the thread during sewing. Moreover, the contribution of the dynamic loading due to the action of take up lever, needle and fabric assembly, and the bobbin thread interaction in the total loss in tensile properties is unexplored. An understanding of these aspects will be useful for machine design, selection of the right sewing thread and machine parameters so as to minimize the strength loss. It may be added that Stylios et al. (1994, 1995) has contributed significantly as regards developing control systems for intelligent sewing machine (Stylios and Sotomi, 1994, 1996). Based on the proposed research and its outcome, the sewing parameters can be monitored and re-adjusted for intelligent garment manufacturing process. In this paper, tensile properties of the needle thread like tenacity, breaking elongation, initial modulus and breaking energy have been studied at four stages during thread’s passage through the machine and the contribution of dynamic loading, needle and fabric assembly, and the bobbin thread interaction on the changes in these tensile properties of the needle thread have been investigated. The effect of thread linear density has also been studied on the contribution of dynamic loading, passage through needle and fabric and bobbin thread interaction.

Background The total change in the tensile property of the needle thread can be conceived due to the dynamic loading due to the action of take up lever, passage through the needle and interaction with the fabric assembly and the bobbin thread. To adjudge the contribution of these factors, tensile properties of the needle thread are measured at four different stages (S1, S2, S3 and S4) during the sewing process as shown in Figure 1: S1.

Zone extending from the cone to guide G1. The thread in this zone is similar to the parent thread. S1 – Parent thread (Cone to guide G1) S2 – Thread zone after dynamic loading (between T2 G1 and point A) S3 – Thread zone after dynamic loading and passage through needle and fabric (between point A and B) S4 – Thread after seam formation (between B and C) M – Take up lever N – Hole in needle bar O – Needle eye T1 – Pre-tensioner T2 – Tension regulator T3 – Tension spring G1-G6 – Guides A – Mark on needle thread at 90 mm from B B – Mark on needle thread after last stitch C – A mark in the seam

The tensile properties of sewing threads 219

S1

Cone T1

S2

M G2

G4

G5

T3 T2

G3

A G6 N S3

O

Seam S4

B

C

Figure 1. Passage of the needle thread through the sewing machine

IJCST 21,4

S2.

Zone extending from tension regulator T2 to a point A (90 mm from the last stitch). The thread in this zone has been subjected to dynamic loading at the tension regulator but has not passed through the needle. The segment closer to the point A undergoes more number of dynamic loadings than the segment closer to T2.

220

S3.

Zone extending from the last stitch at B to A. The thread in this zone has undergone dynamic loading at the tension regulator and repeated passage through needle eye and fabric assembly. The segment closer to point B experiences more number of passages through the needle eye and fabric assembly than the part closer to the point A.

S4.

Zone extending from the last stitch at point B to point C in the seam. The thread in this zone has undergone all the previous processes, interacted with the bobbin thread and got incorporated into the seam.

It was observed that a specific length of the needle thread (in our case 9 cm) goes inside the machine (over the bobbin case holder) during each stitch formation. When sewing at 3 mm stitch length, the 90 mm of thread gets consumed in 22 stitches and after each stitch into the fabric, 4.1 mm of needle thread is pulled from the spool through the tension regulator. Therefore, each segment of the needle thread (4.1 mm) passes 22 times through guides, tension regulator, take up lever, needle and the fabric before getting incorporated into the seam. At any time during sewing, a segment that is just close to the seam (S3), but has not gone into the seam, has passed 21 times through all the machine parts and the fabric and in the next cycle, it would interact with the bobbin thread and get incorporated into the seam. Each previous segment of the thread has passed through the machine elements for decreased number of cycles, and a segment (S2) which is 9 cm away from the segment S3 towards the spool, has passed through the needle and the fabric for zero number of cycles. It is known from the literature (Lojen and Gersak, 2005; Ferriera et al., 1994; Lojen and Gersak, 2001; Kamata et al., 1984) that there are four tension peaks in the needle thread during each stitch cycle, with highest tension being during the tightening of the stitch, when the take up lever pulls the necessary thread length over the tension regulator for the next stitch. Lojen and Gersak (2005), in a study on the thread loading at different measuring positions on the sewing machine observed that the tension in the length between the tension regulator and the take up lever is about 1.5 to 2 times higher than in the length between the take up lever and the needle, because of the pull out of the 4.1 mm through the tension regulator during each stitch formation. The thread length between the take up lever and the tension regulator (90 mm in the said machine), experiences 22 dynamic loading cycles at the highest level of tensile forces during the sewing cycle. The thread segment S2 mentioned above has passed through all the dynamic loading cycles but has not passed through the needle and the fabric. A difference in tensile property between any two stages is due to various types of stresses acting during that stage, i.e. between stages: S1 and S2.

Dynamic loading due to the action of take up lever.

S2 and S3.

Abrasion due to passage through needle and fabric assembly and thermal damage at the needle.

S3 and S4.

Bending, abrasion and tensile deformation during needle and bobbin thread interaction.

The change (%) at different stages and the contribution of the stresses in total change in tensile property is calculated from the following expressions: Change ð%Þ ¼

Tn 2 T1 £ 100 T1

Contribution ð%Þ ¼

ð1Þ

T n 2 T n21 £ 100 T1 2 T4

The tensile properties of sewing threads 221

ð2Þ

where, Tn ¼ Tensile property at nth sewing stage, n ¼ 2, 3, 4 corresponding to sewing stage S2, S3 and S4, respectively, T1 ¼ Tensile property of parent thread, at sewing stage S1. Negative (2 ) change (per cent) indicates the loss in tensile property, whereas a positive change (per cent) indicates the gain in tensile property. Experimental Materials In this study, 12 commercially available threads of different type and linear density are used. The threads were characterized for number of plies, twist level, twist direction, actual linear density and tensile properties. Tensile testing of the threads was performed at a gauge length of 250 mm on Zwick tensile testing machine as per ASTM standard D2256. 30 tests were carried out and the error at 95 per cent confidence level was found to be less than 4 per cent. The physical characteristics of the threads are shown in Table I. Standard denim fabric of 355 g/m2 weight, 27 ends/cm and 17 picks/cm is used for sewing. Seam preparation Juki industrial lockstitch sewing machine was run at a speed of 4,000 stitches/min for producing a balanced seam with 3 mm stitch length, on three layers of denim fabric. Characteristic Sample

Thread type

A B C D E F G H I J K L

Cotton Cotton Cotton Polyester Polyester Polyester Poly-cot Poly-cot Poly-cot Poly-poly Poly-poly Poly-poly

Linear density (tex)

Plies

Twist/ length (tpm)

30 40 60 27 40 60 24 40 60 24 40 60

3 3 3 3 2 2 2 2 2 2 2 3

840 704 580 689 567 560 1123 668 591 1072 575 504

Twist direction Z Z Z Z Z Z Z Z Z Z Z Z

Tenacity (cN/tex)

Breaking elongation (%)

Initial modulus (N/tex)

27.21 32.85 30.08 34.22 35.87 35.16 44.67 44.73 29.18 44.94 52.66 50.83

6.45 6.31 6.98 14.22 18.59 19.04 22.09 25.78 22.78 21.54 24.65 25.82

4.24 4.46 4.03 2.05 2.92 2.01 2.27 2.64 2.30 2.2 3.26 3.32

Table I. Physical characteristics of threads

IJCST 21,4

222

The stitching speed was controlled by using a tachometer. Most appropriate needle size was used for each thread, i.e. 75 Nm for 24, 27 and 30 tex; 100 Nm for 40 tex and 110 Nm for 60 tex thread. The static tension for the bobbin threads was kept constant at 50 cN and the needle thread tension was adjusted to obtain a balanced stitch. The seam balance ratio for all the threads, calculated as a ratio of needle and bobbin thread consumption, is shown in Table II. Same type of the thread was used as needle thread and the bobbin thread for preparing the seams. Methodology After sewing over a meter of seam, the machine was stopped and a mark was placed on the thread near the last stitch (B). The needle thread was cut after the tension regulator at guide G2 as shown in Figure 1. The thread was carefully removed from the guides and tensioner to avoid any abrasion or loss of twist. This length of thread is 445 mm and a number of such pieces were collected. The placement of the jaws of tensile tester for testing the tensile properties of the thread at different sewing stages at a gauge length of 250 mm is shown in Figure 2. Tensile testing of the parent thread corresponds to that of zone S1. For determining tensile strength of thread in zone S2, 250 mm length of thread from mark A towards G2 is mounted in the jaws. Several tests were carried out. For tensile testing of thread in zone S3, a length of 250 mm from mark B towards G2 is mounted in the jaws. The length in between the jaws therefore contained some length, which has not undergone any passage through the needle and the fabric. A sufficient length of thread was removed from the seam for gripping in the lower jaw. In order to establish the fact that the tensile properties obtained in this thread zone belong to the zone S3, the distance of break from the mark B was measured after the tensile test. The thread in zone S3 is relatively weak; consequently during tensile testing greater number of breaks (60 per cent) occur in this 90 mm length of thread (placed near to lower jaw) in zone S3, whereas relatively less breaks occur in the additional 160 mm of the specimen length. To assess the strength of thread in zone S4, the thread was extracted from the sewn fabrics after cutting the bobbin threads, removed carefully to avoid any extension and Thread consumption/10 stitches Sample Thread type Linear density (tex) Needle thread Bobbin thread Seam balance ratio (%)

Table II. Needle and bobbin thread consumption for different threads

A B C D E F G H I J K L

Cotton Cotton Cotton Polyester Polyester Polyester Poly-cot Poly-cot Poly-cot Poly-poly Poly-poly Poly-poly

30 40 60 27 40 60 24 40 60 24 40 60

4.44 2.50 3.89 3.89 2.56 3.83 4.17 2.28 4.94 4.17 2.39 4.72

4.17 2.39 3.50 3.78 2.67 4.28 4.50 2.33 4.83 4.44 2.28 4.67

106.7 104.7 111.1 102.9 95.8 89.6 92.6 97.6 102.3 93.8 104.9 101.2

The tensile properties of sewing threads

Upper jaw Cone S1 – Parent thread (Cone to guide G1)

S1

S2 –Thread zone after dynamic loading (between T2 and point A) S3 – Thread zone after dynamic loading and passage through needle and fabric (between point A and B)

223

S4 – Thread after seam formation (between B and C)

G1 Lower jaw G1-G2 – Guides T2 – Tension regulator A – Mark on needle thread at 90 mm from B B – Mark on needle thread after last stitch C – A mark in the seam G2 Upper jaw T2

Upper jaw S2

S3

A Lower jaw

90 mm Upper jaw

B Lower jaw

S4

C Needle thread

Lower jaw

Figure 2. Schematic views of the jaw positions for tensile testing of the threads at four sewing stages

IJCST 21,4

loss of twist. 30 tests were performed at each stage and the error at 95 per cent confidence level was found to be less than 4 per cent. Statistical analysis was carried out, based on t-test. Using this test we investigated whether the mean values of tensile properties in successive sewing stages were statistically significant with respect to the tensile properties in the previous stage. All measurements were carried out under standard testing conditions.

224 Results and discussion The tenacity, breaking elongation, initial modulus and breaking energy measured at different sewing stages, for staple spun threads and core spun threads are shown in Tables III and IV, respectively. The change per cent at each stage and the contribution of dynamic loading due to the action of take up lever, passage through needle and fabric assembly and bobbin thread interaction in total change in tensile properties as calculated from Equations 1 and 2 is also shown in the Tables III and IV. As the thread moves from cone to the seam, it undergoes various stresses and strains as mentioned earlier under background. In general, a significant loss in tensile properties has been observed for all the threads. Tenacity A substantial loss in tenacity is observed for all the threads (Tables III and IV). However, cotton threads show highest loss in tenacity (23 per cent to 32 per cent) as compared to polyester staple spun (8 per cent to 12 per cent) and core spun threads (5 per cent to 9 per cent). The strength of core spun threads depends on the filament core, which does not get affected during repeated passage of the thread through the various machine elements, needle eye and fabric assembly. Therefore, the core spun thread shows lower loss in tenacity after sewing. Weaker cotton fibres get more damage due to abrasion and fatigue during the sewing process leading to higher loss in tenacity for these threads. Since cotton threads show a substantial loss in tenacity as compared to other threads, it is important to study the changes in cotton thread’s properties in detail. Figure 3 shows the tenacity of cotton threads at different sewing stages and their distribution. A progressive decrease in tenacity is observed as the thread passes through these stages. Polyester staple spun and core spun threads show a slight increase in tenacity after dynamic loading at stage S2, followed by decrease at stage S3 and S4 (Tables III and IV). The effect of thread type on changes in tensile properties of threads at different sewing stages has been discussed elsewhere (Midha et al., 2008). Figure 4 shows the contribution of dynamic loading due to the action of take up lever, passage through needle and fabric assembly and bobbin thread interaction in total loss in tenacity of all threads. It is observed that about 15 per cent tenacity loss (which is 50 per cent of the total tenacity loss) in cotton threads, takes place at the time of interaction between the needle thread and the bobbin thread. Similar results are obtained for polyester staple spun and core spun threads. In these threads the contribution of bobbin thread interaction in total loss is even more than 50 per cent. Displacement of plies and loosening of the structure at the time of bobbin thread interaction may be responsible for this loss. Dynamic loading of the needle thread at the time of pulling of the thread from the disc tensioner is the second largest contributor to the tenacity loss in cotton threads, whereas polyester staple spun and

%

%

%

%

%

%

%

%

Linear ! density (tex)

Mean Mean Change % Contribution S3 Mean Change % Contribution S4 Mean Change % Contribution Breaking elongation (%) S1 Mean S2 Mean Change % Contribution S3 Mean Change % Contribution S4 Mean Change % Contribution Initial modulus (N/tex) S1 Mean S2 Mean Change % Contribution S3 Mean Change % Contribution

Tenacity (cN/tex) S1 S2

Sewing stage 32.85 30.35 * 2 7.61 2 26.68 29.04 * 2 11.60 2 13.99 23.48 *a 2 28.52 2 59.33 6.31 5.25 * 2 16.80 2 79.10 5.03 * 2 20.29 2 16.43 4.97 a 2 21.24 2 4.47 4.46 5.69 * 27.65 352.23 5.92 * 32.72 64.59

6.45 5.92 * 2 8.22 251.99 5.63 * 212.71 228.40 5.51 a 215.81 219.61 4.24 4.38 * 3.20 18.17 4.50 * 7.32 23.40

Cotton 40

27.21 26.46 * 2 2.76 212.09 25.15 * 2 7.57 221.08 21.17 *a 222.82 266.83

30

4.03 5.46 * 35.27 184.0 5.21 * 29.22 231.58

6.98 5.40 * 222.64 298.78 5.38 222.92 21.22 5.38a 222.92 0.00

30.08 26.03 * 213.46 242.57 25.58 * 214.96 24.74 20.57 *a 231.62 252.69

60

2.05 2.58 * 25.18 499.6 2.61 * 26.62 28.57

14.22 14.31 0.63 10.18 14.16 2 0.42 2 16.96 13.34 *a 2 6.19 2 93.21

34.22 35.80 4.63 60.44 35.29 * 3.12 2 19.71 31.61 *a 2 7.66 2140.73

27

2.92 2.86 2 1.98 2 10.39 2.82 2 3.57 2 8.35

18.59 18.56 2 0.16 2 1.10 17.64 * 2 5.11 2 34.04 15.89 *a 2 14.54 2 64.86

35.87 36.33 1.28 12.81 35.09 2 2.18 2 34.63 32.29 *a 2 9.99 2 78.18

Polyester 40

2.01 2.37 * 17.91 116.15 2.31 * 14.93 219.33 (continued)

19.04 18.76 21.47 214.97 18.65 22.05 25.91 17.17 *a 29.82 279.12

35.16 35.71 1.56 13.09 35.38 0.63 27.80 30.97 *a 211.92 2105.29

60

The tensile properties of sewing threads 225

Table III. Tensile properties of staple spun threads at different sewing stages

Mean Mean Change % Contribution % Mean Change % Contribution % Mean Change % Contribution %

*a

0.065 0.056 * 213.85 239.14 0.052 * 220.00 217.38 0.043 *a 235.39 243.49

3.50 217.61 2141.5

30 *a

0.098 0.078 * 2 20.41 2 47.60 0.073 * 2 25.51 2 11.89 0.056 *a 2 42.88 2 40.51

4.11 2 7.85 2 516.82

Cotton 40 *a

0.153 0.112 * 226.80 257.75 0.109 228.76 24.22 0.082 *a 246.41 238.03

3.26 219.16 2252.51

60

*a

0.137 0.151 * 10.15 78.08 0.147 * 7.45 2 20.77 0.119 *a 2 13.0 2157.31

1.96 2 5.04 2628.17

27

0.293 0.292 2 0.11 2 0.51 0.267 * 2 8.88 2 40.94 0.230 *a 2 21.42 2 58.54

2.36 2 19.05 2 81.26

*a

Polyester 40

0.417 0.433 3.79 20.40 0.429 2.76 25.54 0.340 *a 218.58 2114.85

1.70 *a 215.42 2196.82

60

Notes: Change % is w.r.t mean at stage S1; minus sign (2 ) indicates the loss in property; * indicates the significant difference of means at 95 per cent confidence level, from previous stage; a indicates the significant difference of means at 95 per cent confidence level from stage S1

S4

S3

Breaking energy (J) S1 S2

Mean Change % Contribution %

S4

Table III.

Linear ! density (tex)

226

Sewing stage

IJCST 21,4

%

%

%

%

%

%

%

%

Linear ! density (tex)

Mean Mean Change % Contribution S3 Mean Change % Contribution S4 Mean Change % Contribution Breaking elongation (%) S1 Mean S2 Mean Change % Contribution S3 Mean Change % Contribution S4 Mean Change % Contribution Initial modulus (N/tex) S1 Mean S2 Mean Change % Contribution S3 Mean Change % Contribution

Tenacity (cN/tex) S1 S2

Sewing stage 44.73 46.17 * 3.22 68.66 45.54 * 1.82 2 29.85 42.63 *a 2 4.69 2 138.81 25.78 25.69 2 0.36 2 3.24 24.99 * 2 3.05 2 24.19 22.91 *a 2 11.12 2 72.57 2.64 2.73 * 3.38 20.53 2.62 * 2 0.84 2 25.64

22.09 20.79 * 25.89 230.39 20.51 27.15 26.50 17.81 *a 219.38 263.11 2.27 1.87 * 217.62 307.5 1.90 216.3 223.04

Polyester-cotton 40

44.67 42.77 * 24.25 251.64 42.89 23.99 3.16 40.99 *a 28.23 251.52

24

2.3 2.31 0.44 2.66 1.98 * 2 13.91 2 86.86

22.78 22.19 * 2 2.59 2 15.36 21.55 * 2 5.4 2 16.67 18.94 *a 2 16.86 2 67.97

29.18 29.28 0.34 2 4.61 28.13 * 2 3.6 2 53.46 27.03 *a 2 7.37 2 51.15

60

2.2 1.98 * 210.00 110.0 2.14 * 22.73 279.98

21.54 21.22 21.49 214.80 20.73 23.76 222.54 19.37 *a 210.07 262.66

44.94 44.81 20.29 24.30 44.91 20.07 3.26 41.91 *a 26.74 298.96

24

60 50.83 50.95 0.24 3.86 49.42 * 2 2.77 2 48.39 47.67 *a 2 6.22 2 55.47 25.82 25.11 * 2 2.75 2 60.71 25.21 2 2.36 8.61 24.65a 2 4.53 2 47.90 3.32 3.39 2.11 6.94 3.01 * 2 9.34 2 37.64 (continued)

Polyester-polyester 40 52.66 53.53 1.65 26.92 52.61 20.10 228.55 49.43 *a 26.13 298.37 24.65 24.59 20.24 22.61 24.18 21.90 218.06 22.38 *a 29.19 279.33 3.26 3.42 * 4.88 16.10 3.34 2.44 28.05

The tensile properties of sewing threads 227

Table IV. Tensile properties of core spun threads at different sewing stages

*a

0.240 0.210 * 212.50 260.01 0.210 212.50 0.00 0.190 *a 220.83 239.99

2.40 5.73 2384.4

24

0.540 0.562 * 4.05 30.25 0.534 * 2 1.25 2 39.58 0.468 *a 2 13.39 2 90.66

2.21 2 16.46 2 94.90

*a

Polyester-cotton 40 *a

0.45 0.42 * 2 6.67 2 33.35 0.4 2 11.11 2 22.20 0.36 *a 2 20 2 44.45

1.92 2 16.52 2 15.80

60

*a

0.240 0.230 24.17 233.36 0.230 24.17 0.00 0.210 *a 212.50 266.64

2.40 9.09 2130.03

24

0.559 0.571 2.16 14.51 0.550 21.67 225.72 0.476 *a 214.89 288.78

2.27 230.31 2108.05

*a

Polyester-polyester 40

0.83 0.76 * 2 8.43 2 58.30 0.78 2 6.02 16.67 0.71 *a 2 14.46 2 58.37

2.31 *a 2 30.42 2 69.30

60

Notes: Change % is w.r.t mean at stage S1; minus sign (2) indicates the loss in property; * indicates the significant difference of means at 95 per cent confidence level, from previous stage; aindicates the significant difference of means at 95 per cent confidence level from stage S1

Breaking energy (J) S1 Mean S2 Mean Change % Contribution % S3 Mean Change % Contribution % S4 Mean Change % Contribution %

Mean Change % Contribution %

S4

Table IV.

Linear ! density (tex)

228

Sewing stage

IJCST 21,4

The tensile properties of sewing threads

35 S1 S2 S3 S4

31

31

29

29

Tenacity (cN/tex)

Tenacity (cN/tex)

33

27 25 23 21

27 25 21

19

19

17

17

15

S1

S2 S3 Sewing stages

229

23

0

S4

2

4

6 8 Frequency

10

12

38

36

36

34

34

32

32

Tenacity (cN/tex)

Tenacity (cN/tex)

(a) 38

30 28 26 24

30 28 24

22

22

20

20

18

S1

S2 S3 Sewing stage

S1 S2 S3 S4

26

S4

0

2

4

6

8 10 12 14 16 18 20 22 Frequency

(b)

34

34

32

32

30

30

Tenacity (cN/tex)

Tenacity (cN/tex)

36

28 26 24 22

26 24

S1 S2 S3 S4

22 20

20

18

18 16

28

16 S1

S2 S3 Sewing stages

0

S4

(c)

5

10

15

20

25

30

35

Frequency

core spun threads do not show any loss in tenacity after dynamic loading (stage S2). Thread and fibre fatigue during repeated loading causes significant reduction in cotton fibres’ mechanical properties and therefore cotton thread tenacity loss. Passage through needle eye and the fabric assembly has small contribution to the loss in tenacity (Figure 4). It may be noted that the most appropriate size of the needles was used in accordance to the size of the sewing thread. Further, it is observed that coarser cotton and polyester staple spun threads show higher tenacity loss as compared to finer threads (Table III). In core spun threads,

Figure 3. Mean tenacity and its distribution at different sewing stages for cotton threads: (a) 30 tex; (b) 40 tex; (c) 60 tex

IJCST 21,4

Dynamic loading

Passage through needle and fabric assembly

Bobbin thread interaction

100%

230

Figure 4. Contribution of dynamic loading, passage through needle and fabric assembly and bobbin thread interaction in tenacity loss of threads

Tenacity loss (%)

80% 60% 40% 20% 0% –20% –40%

A

B

C

D

E

F G Sample

H

I

J

K

L

the fineness of threads does not have any significant effect on tenacity loss. It may be noted here that the needle size also changes as the thread linear density changes. If we compare the contributions of different stages towards the loss in tenacity, it can be inferred that bobbin thread interaction has greater contribution in tenacity loss of finer threads (Figure 4). The finer thread in spite of its lower contact area exhibits greater tenacity loss during bobbin thread interaction. This may be attributed to sharper bending of finer thread (fine needle size is used for finer thread) at the time of interaction with bobbin thread. Further, the ratio of damaged surface to the total exposed surface is likely to be higher for finer thread, affecting the loss of tensile strength. In cotton threads, the contribution of dynamic loading is more for coarser threads as compared to fine threads. Usually, shorter fibres are used in coarser threads, which make the thread intrinsically weak due to fibre-fibre slippage under dynamic loading. Therefore, coarser threads exhibit greater loss in tenacity due to dynamic loading. Passage through needle and fabric assembly causes bending deformation, repeated abrasion and heat loading of threads. In polyester threads, passage through needle and fabric assembly has higher contribution in tenacity loss for coarser threads, whereas in cotton threads the trend is reverse, i.e. passage through needle eye and fabric assembly has lower contribution in tenacity loss. With increase in thread linear density, the needle size also increases; the needle penetration force and hence needle temperature is known to increase (Hurt and Tyler, 1971, 1972; Dorkin and Chamberlain, 1963), leading to increased thermal loading of the threads. Therefore, the contribution of passage through needle and fabric assembly increases as the thread size increases, for polyester threads. However, cotton is insensitive to thermal damages, the higher contribution of passage through needle and fabric assembly in finer threads may be due to the reason that sharp bending of the finer thread during its passage through needle eye and fabric assembly leads to greater bending deformation which in turn results in greater strength loss. However, it is noteworthy that the tension level in the needle thread during its passage through needle and fabric

assembly is much lower than the tension during bobbin thread interaction (Ferriera et al., 1994), influencing the extent of damage in the respective cases. Breaking elongation It is observed from Table III that there is considerable reduction in breaking elongation after sewing for all threads. Cotton threads show highest loss in breaking elongation (16 per cent to 23 per cent) as compared to polyester staple spun (6 per cent to 15 per cent) and core spun threads (8 per cent to 19 per cent). Figure 5 shows the 8.5 7.5

Breaking elongation (%)

Breaking elongation (%)

8.0 7.0 6.5 6.0 5.5 5.0 4.5

The tensile properties of sewing threads 231

S1 S2 S3 S4

7.8 7.4 7.0 6.6 6.2 5.8 5.4 5.0 4.6

4.0 3.5

S1

S2

S3

S4

0

2

3

8.0

7.5

7.5

7.0

7.0

6.5 6.0 5.5 5.0 4.5 4.0 3.5

4

5

6

7

8

9

10

Frequency

8.0

3.0

1

(a)

Breaking elongation (%)

Breaking elongation (%)

Sewing stages

S1 S2 S3 S4

6.5 6.0 5.5 5.0 4.5 4.0 3.5

S1

S2

S3

S4

Sewing stages

0 2 4 6 8 10 12 14 16 18 20 22 24 26 Frequency

(b)

8.5

8.5

8.0

8.0 Breaking elongation (%)

Breaking elongation (%)

9.0

7.5 7.0 6.5 6.0 5.5 5.0

7.5 7.0 6.5 6.0 5.5 5.0 4.5

4.5 4.0

S1 S2 S3 S4

4.0 S1

S2

S3

Sewing stages

S4

0

(c)

2

4

6

8 10 12 14 16 18 20 22 Frequency

Figure 5. Mean breaking elongation and its distribution at different sewing stages for cotton threads: (a) 30 tex; (b) 40 tex; (c) 60 tex

IJCST 21,4

232

breaking elongation of cotton threads at different sewing stages. It is observed that there is progressive fall in breaking elongation as the thread passes from stage S1 to S4. Similar results are obtained for polyester staple spun and core spun threads. Figure 6 shows the contribution of dynamic loading, passage through needle and fabric, and bobbin thread interaction in total breaking elongation loss of threads. In cotton threads more than 50 per cent loss in breaking elongation takes place at stage S2 due to dynamic loading of the threads (Table III), whereas in polyester staple and core spun threads, dynamic loading has relatively small contribution in the total breaking elongation loss. This is due to the possible realignment of fibres in the yarn during dynamic loading (Reumann and Offermann, 1993). During yarn spinning, higher bending rigidity and convoluted cross-section of cotton fibres leaves scope for fibre readjustment and during dynamic loading realignment of fibres takes place, which causes a substantial fall in breaking elongation. Bobbin thread interaction is the major contributor in breaking elongation loss (about 60 per cent) of polyester staple spun and core spun threads. Passage through needle and fabric assembly has relatively small contribution in breaking elongation loss. With increase in thread linear density, the breaking elongation loss increases for cotton threads and decreases for polyester staple spun and core spun threads. In cotton threads, the contribution of dynamic loading, passage through needle and fabric assembly and bobbin thread interaction also show a significant trend with increase in thread linear density. Dynamic loading has more contribution in the breaking elongation loss for coarse threads, which is again due to readjustment of relatively shorter and more number of fibres in the yarn cross section. Similar to tenacity results, passage through the needle and fabric assembly has minimum effect for the coarser thread. Bobbin thread interaction has more effect on the finer threads than coarser threads, which is again due to the sharper bending deformation of threads during interaction with bobbin threads. Dynamic loading

Passage through needle and fabric assembly Bobbin thread interaction

100%

Figure 6. Contribution of dynamic loading, passage through needle and fabric assembly and bobbin thread interaction in breaking elongation loss of threads

Breaking elongation loss (%)

80% 60% 40% 20% 0% –20%

A

B

C

D

E

F G Sample

H

I

J

K

L

Initial modulus It is observed from Tables III and IV that there is a significant loss in initial modulus of threads after sewing, for all threads. Figure 7 shows the mean initial modulus and its distribution for cotton threads at different sewing stages. A significant increase in initial modulus is observed at stage S2 after dynamic loading for cotton threads;

233

7.0 6.5 6.0

Initial modulus (N/tex)

Initial modulus (N/tex)

The tensile properties of sewing threads

5.5 5.0 4.5 4.0 3.5 3.0

S1 S2 S3 S4

5.4 5.0 4.6 4.2 3.8 3.4 3.0

2.5 S1

S2 S3 Sewing stages

S4

0

7.5

7.5

7.0

7.0

6.5

6.5

6.0 5.5 5.0 4.5 4.0

8 10 12 14 16 18 20 22 Frequency

5.0 4.5 S1 S2 S3 S4

4.0

3.0

3.0 S2 S3 Sewing stage

6

5.5

3.5

S1

4

6.0

3.5 2.5

2

(a)

Initial modulus (N/tex)

Initial modulus (N/tex)

2.0

S4

0

(b)

2

4

6

8 10 12 14 16 18 20 Frequency

6.5

6.5

6.0

6.0

Initial modulus (N/tex)

Initial modulus (N/tex)

7.0

5.5 5.0 4.5 4.0 3.5

5.5 5.0 4.0 3.5

3.0

3.0

2.5

2.5

2.0

S1

S2

S3

Sewing stages

2.0

S4

(c)

S1 S2 S3 S4

4.5

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 Frequency

Figure 7. Mean initial modulus and its distribution at different sewing stages for cotton threads: (a) 30 tex; (b) 40 tex; (c) 60 tex

IJCST 21,4

234

polyester staple spun and core spun threads show only a slight increase or no change in initial modulus at stage S2 after dynamic loading. This is due to the possible realignment of fibres in the yarn during dynamic loading (Reumann and Offermann, 1993), which is relatively more in case of threads made from cotton fibres than polyester fibres. The increase in initial modulus before loop formation is helpful for clear loop formation and avoiding the skipped stitches during high-speed sewing. A marginal change is observed in initial modulus, due to the passage through the needle and fabric assembly (stage S3), whereas a significant decrease in the values is observed at stage S4 due to the interaction with bobbin thread. Loosening of structure and displacement of plies occurs at the time of bobbin thread interaction and the non contribution of surface fibres to thread tension causes significant loss in the initial modulus. The contribution of dynamic loading, passage through needle and fabric and bobbin thread interaction in total loss in initial modulus is shown in Figure 8. It is observed that loss in initial modulus mainly takes place due to bobbin thread interaction. Dynamic loading and passage through needle and fabric assembly has only small contribution in initial modulus loss. The contribution of bobbin thread interaction decreases for coarser threads as compared to finer threads, the trend is similar to the tenacity and breaking elongation loss. Breaking energy Tables III and IV show a substantial loss in breaking energy after sewing for all threads. Cotton threads show highest loss in breaking energy (35 per cent to 46 per cent) as compared to polyester staple and core spun threads (12 per cent to 21 per cent). Figure 9 shows breaking energy of cotton threads and their distribution at different sewing stages. A progressive fall in breaking energy is observed at successive sewing stages. In polyester staple spun and core spun threads, the breaking energy increases slightly after dynamic loading (stage S2) and then decreases after passage through needle and fabric assembly (stage S3) and bobbin thread interaction (stage S4). Dynamic loading Passage through needle and fabric assembly Bobbin thread interaction 100%

Figure 8. Contribution of dynamic loading, passage through needle and fabric and bobbin thread interaction on initial modulus loss

Initial modulus loss (%)

80% 60% 40% 20% 0% –20% – 40% – 60% –80% A

B

C

D

E

F G Sample

H

I

J

K

L

The tensile properties of sewing threads

0.16

0.12

Breaking energy (J)

Breaking energy (J)

0.14

0.10 0.08 0.06 0.04 0.02 0.00

S1 S2 S3 S4 0.080

235

0.065 0.050 0.035

S1

S2 S3 Sewing stages

S4

0

2

4

(a)

6 8 10 Frequency

12

14

0.16 S1 S2 S3 S4

0.12

Breaking energy (J)

Breaking energy (J)

0.14

0.10 0.08 0.06 0.04

0.12 0.10 0.08 0.06 0.04

0.02 0.00

S1

S2 S3 Sewing stage

S4

0

2

4

6

(b)

8 10 12 14 16 18 20 22 Frequency

0.19

0.19

0.17

0.17

Breaking energy (J)

Breaking energy (J)

0.21

0.15 0.13 0.11 0.09 0.07 0.05

S1 S2 S3 S4

0.15 0.13 0.11 0.09 0.07

S1

S2 S3 Sewing stage

0.05

S4

0

2

4

6 8 10 Frequency

12

14

(c)

The contribution of dynamic loading, passage through needle and fabric assembly and bobbin thread interaction in total loss in breaking energy is shown in Figure 10. Dynamic loading has greatest contribution in the loss in breaking energy of cotton threads; 39 per cent for 30 tex, 48 per cent for 40 tex and 58 per cent for 60 tex, whereas it does not cause significant loss in breaking energy for polyester staple spun and core spun threads. Dynamic loading of thread causes significant thread and fibre fatigue,

Figure 9. Breaking energy and its distribution at different sewing stages for sewing thread of different linear densities: (a) 30 tex; (b) 40 tex; (c) 60 tex

IJCST 21,4

Dynamic loading

Passage through needle and fabric assembly Bobbin thread interaction

100%

236

Figure 10. Contribution of dynamic loading, passage through needle and fabric assembly and bobbin thread interaction in breaking energy loss

Breaking energy loss (%)

80% 60% 40% 20% 0% –20% –40%

A

B

C

D

E

F G Sample

H

I

J

K

L

leading to reduction in both strength and elongation. The loss in both strength and elongation usually leads to decrease in breaking energy. Bobbin thread interaction contributes 40 per cent in the total loss in breaking energy for all the threads. In polyester staple spun and core spun threads, bobbin thread interaction is a mainly responsible for the loss in breaking energy, which is again due to significant loss in strength and elongation of these threads during rubbing action between bobbin and needle thread. However, increase in initial modulus during dynamic loading is likely to compensate the extent of decrease to some extent. Further, it is observed that the passage through the needle and the fabric assembly has only small influence on the loss in breaking energy. In cotton threads, the loss in breaking energy increases for the coarser threads as in case of tenacity loss and the contribution of dynamic loading also increases for coarser threads. The contributions of bobbin thread interaction and passage through needle and fabric assembly decrease with increase in thread linear density. Conclusions High-speed sewing operation leads to overall reduction in tensile properties of the needle thread. However, the change in tensile properties is different at three distinct stages viz. dynamic loading (S2), passage through needle and fabric assembly (S3) and bobbin thread interaction (S4). In cotton threads, tenacity, breaking elongation and breaking energy reduce progressively at all sewing stages, whereas for polyester staple spun and core spun threads, tenacity and breaking energy increase slightly after dynamic loading at stage S2. Initial modulus increases at stage S2, remains unchanged at stage S3 and then decreases substantially at stage S4 for all threads. Bobbin thread interaction is the major cause of loss in tensile properties of all the threads, except breaking elongation in cotton threads. Dynamic loading also causes significant loss to tenacity, breaking elongation and breaking energy of cotton threads. Passage through needle and fabric assembly has relatively small contribution in reducing the tensile properties.

Among all the threads, cotton threads show highest loss in tenacity, breaking elongation and breaking energy. In cotton threads, the loss in tenacity, breaking elongation and breaking energy is more for the coarser threads as compared to finer threads, whereas change in thread fineness does not have any impact on loss in tensile properties for polyester staple spun and core spun threads. Dynamic loading has more influence on the coarser threads than finer threads. Contribution of bobbin thread interaction is lower on the loss in tensile properties for coarse thread as compared to finer threads. In case of polyester staple spun and core spun threads, the contribution of passage through needle and fabric assembly is higher for coarser threads, whereas the trend is reverse for cotton threads. The contribution of bobbin thread interaction is also lower for coarser polyester threads. References Crow, R.H. and Chamberlain, N.H. (1969), “The performance of sewing threads in industrial sewing machines”, Technological Report No. 21, The Clothing Institute, London. Dorkin, C.M.C. and Chamberlain, N.H. (1963), “The facts about needle heating”, Technological Report No. 13, The Clothing Institute, London. Ferriera, F.B.N., Harlock, S.C. and Grosberg, P. (1994), “A study of thread tensions on lockstitch sewing machine, Part-I”, International Journal of Clothing Science and Technology, Vol. 6, pp. 14-19. Gersak, J. and Knez, B. (1991), “Reduction in thread strength as a cause of loading in the sewing process”, International Journal of Clothing Science and Technology, Vol. 3, pp. 6-12. Hurt, F.N. and Tyler, D.J. (1971), “An investigation of needle heating and associated problems in machine sewing-I”, HATRA Research and Report No. 19. Hurt, F.N. and Tyler, D.J. (1972), “An investigation of needle heating and associated problems in machine sewing-III”, HATRA Research and Report No. 21. Kamata, Y., Kinoshita, R., Ishikawa, S. and Fujisaki, K. (1984), “Disengagement of needle thread from rotating hook, effects of its timing on tightening tension, industrial single needle lockstitch machine”, Journal of Textile Machinery Society of Japan, Vol. 30, pp. 40-9. Lojen, D.Z. and Gersak, J. (2001), “Study of the tensile force of thread in relation to its pre tension”, International Journal of Clothing Science and Technology, Vol. 13, pp. 240-50. Lojen, D.Z. and Gersak, J. (2005), “Thread loading in different positions on the sewing machine”, Textile Research Journal, Vol. 75, pp. 498-506. Midha, V.K., Chatopadhyay, R., Kothari, V.K. and Mukhopadhayay, A. (2008), “Studies on the changes in tensile properties of sewing thread at different sewing stages”, Textile Research Journal. Reumann, D. and Offermann, P. (1993), “Changes in yarn properties following preloading”, Melliand-Textilberichte, No. 6, pp. E200-1. Rudolf, A. and Gersak, J. (2006), “Influence of twist on alterations in fibers’ mechanical properties”, Textile Research Journal, Vol. 76, pp. 134-44. Stylios, G. and Sotomi, J.O. (1994), “A neuro-fuzzy control system for intelligent sewing machines”, Intelligent Systems Engineering Technology, IEEE Publication No. 395, pp. 241-6. Stylios, G. and Sotomi, J.O. (1996), “Thinking sewing machines for intelligent garment manufacture”, International Journal of Clothing Science and Technology, Vol. 8 Nos 1/2, pp. 44-55.

The tensile properties of sewing threads 237

IJCST 21,4

238

Stylios, G., Sotomi, J.O., Zhu, R., Xu, Y.M. and Deacon, R. (1995), “The mechatronic principles for intelligent sewing environments”, Mechatronics, Vol. 5 Nos 2/3, pp. 309-19. Stylios, G., Fan, J., Sotomi, J.O., Zhu, R., Fan, J., Xu, Y.M. and Deacon, R. (1994), “A sewability integrated environment for intelligent garment manufacture”, Factory 2000; Advanced Factory Automation, IEE Proceedings No. 398, pp. 543-51. Sundaresan, G., Hari, P.K. and Salhotra, K.R. (1997), “Strength reduction in sewing threads during high speed sewing in an industrial lockstitch machine: Part I-mechanism of thread strength reduction”, International Journal of Clothing Science and Technology, Vol. 9, pp. 334-45. Sundaresan, G., Salhotra, K.R. and Hari, P.K. (1998), “Strength reduction in sewing threads during high speed sewing in industrial lockstitch machine, Part II: effect of thread and fabric properties”, International Journal of Clothing Science and Technology, Vol. 10, pp. 64-79. Ukponmwan, J.O., Mukhopadhyay, A. and Chaterjee, K.N. (2000), Sewing Threads, Textile Progress, Vol. 30 3/4, Published by The Textile Institute, pp. 1-94. Corresponding author Vinay Kumar Midha can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

Selection of ERP suppliers using AHP tools in the clothing industry

Selection of ERP suppliers using AHP tools

Can U¨nal and Mu¨cella G. Gu¨ner Department of Textile Engineering, Faculty of Engineering, University of Ege, Izmir, Turkey Abstract

239 Received 9 April 2008 Revised 19 November 2008 Accepted 19 November 2008

Purpose – The purpose of this paper is to explore selection of the best ERP suppliers in the clothing industry by using analytic hierarchy process (AHP). Design/methodology/approach – AHP is used in order to achieve the paper’s purpose; selection criteria are determined by managers and experts. Findings – Three different enterprise resource planning (ERP) suppliers are investigated and best alternative is selected by using AHP. After the best alternative is selected, cost benefit analysis is calculated in order to define decisive result. All calculations are verified by performing the consistency test. Research limitations/implications – Selection criteria and their evaluations can be changed depending on size of the clothing manufacturer and product type. Originality/value – The results of the study will be helpful to clothing manufacturers which plan to implement an ERP system in their organizations. Furthermore, they can use AHP in other decision problems as well. Keywords Manufacturing resource planning, Decision making, Analytic hierarchy process, Garment industry Paper type Research paper

Introduction Over the last decade, our world has changed dramatically due to the growing phenomenon of globalization and revolution in information technology (IT). These changes have obliged the clothing industry to make costs lower, enlarge product assortment, improve product quality, and provide reliable delivery dates through effective and efficient coordination of production. To achieve these conflicting changes, companies must constantly re-engineer or change their business practices and employ information systems (ISs) such as enterprise resource planning (ERP) (Mahesh and Amarpreet, 2006). An ERP system integrates all necessary business functions, such as product planning, purchasing, inventory control, sales, financial and human resources, into a single system with a shared database. According to Waters (1996), the following are some of the benefits of using an ERP package: . reduced stock and inventories, lower stock levels, with saving in capital, space, and warehousing; . higher stock turnover, reduced cycle-time, and increased productivity; . better communication with customers and suppliers; . more reliable and faster delivery time; . higher utilization of facilities, as materials are always available when needed; and . better control over business.

International Journal of Clothing Science and Technology Vol. 21 No. 4, 2009 pp. 239-251 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910959990

IJCST 21,4

240

In the past few years, thousands of companies around the world have implemented ERP systems. The number of companies that plan to implement ERP is growing rapidly. Since the early to mid-1990s, the ERP software market had been and now is and will be one of the fastest growing segments of the IT industry (Verville and Halingten, 2003). AMR Research, an authoritative market forecast institution in America, indicated that the ERP market would grow at annual rate of 37 percent in recent five years. The sales of the ERP packaged software were around $20 billion by the year 2000 and the eventual market size is predicted to be around $1 trillion by the year 2010 (Rajagopal, 2002). Surprisingly, given the significant investment in resources and time, many companies did not achieve success in ERP implementation. It is estimated that the failure rate of ERP implementation ranges from 40 to 60 percent or higher (Umble et al., 2003). Because, most of the ERP systems have modular construction. At first glance, this suggests that ERP systems can be customized to fit a firm’s specialized business processes. Unfortunately, the constraints on customization are massive; few ERP suppliers undertake customization efforts that would have significant strategic consequences. According to Davenport (1998), most firms find that they need to adapt or even completely reconfigure their business processes to conform to the requirements of the ERP system. In another point of view, ERP systems are designed to reflect “best practices,” but it is suppliers who determine a single “best practice” for an industry (Davenport, 2000). Owing to limitations in available resources, the complexity of ERP systems, and the diversity of alternatives, it is often difficult for an organization to select a suitable ERP system. The complexity of ERP system makes it difficult for a single decision maker to consider all aspects of problem (Liao et al., 2007). In this paper, this complexity is tried to be eliminated by using analytic hierarchy process (AHP) in clothing industry. Research methodology The AHP, developed by Saaty is designed to solve complex multi-criteria decision problems. It is a flexible and powerful tool for handling both qualitative and quantitative multi-criteria problems (Ngai and Chan, 2005). The strength of this approach is that it organizes tangible and intangible factors in a systematic way, and provides a structured yet relatively simple solution to the decision-making problems (Skibniewski and Chao, 1992). In addition, by breaking a problem down in a logical fashion from the large, descending in gradual steps, to the smaller and smaller, one is able to connect, through simple paired comparison judgments, the small to the large (Al-Harbi, 2001). In the application of AHP, the relative importance or weights of the criteria are determined after the consultation session. To determine local weights of the components in the AHP hierarchy, each set of components is compared using a pairwise comparison method with respect to their immediate higher-level (parent) component in the hierarchy (Korpela and Tuominen, 1996a, b). Quantitative or qualitative assessments can be used in the comparisons. In general, a nine-point numerical scale, shown in Table I, is recommended for the comparisons (Forman, 1996; Saaty, 1980; Saaty and Alexander, 1981). In the pairwise comparison matrix, rows and columns of the pairwise comparison matrix are allocated to the components of the set belonging to the same parent in the hierarchy. Thus, if the number of components belonging to the same parent is m; the pairwise comparison matrix is an “m £ m” square matrix. The relative importance of component i compared to component j with

Intensity of weight

Verbal judgment of preference

1 3

Equal importance Moderate importance

Explanation

Two activities contribute equally to the objective Experience and judgement slightly favor one over another 5 Strong importance Experience and judgement strongly favor one over another 7 Very strong importance An activity is strongly favored and its dominance is demonstrated in practice 9 Absolute importance The importance of one over another affirmed on the highest possibble order 2,4,6,8 Intermediate values Used to present compromise between the priorities listed above Reciprocals of above If activity i has one of the above non-zero numbers assigned to it when non-zero numbers compared with activity j, then j has the reciprocal value when compared with I

Selection of ERP suppliers using AHP tools 241

Table I. Saaty’s 1-9 scale for pairwise comprasion

regard to their parent component in the AHP hierarchy is determined using Saaty’s scale and assigned to the (i, j)th position of the pairwise comparison matrix. Automatically, the reciprocal of the assigned number is assigned to the ( j,i )th position according to the following rule (Chang et al., 2007): aij . 0;

aji ¼

1 ; aij

aii ¼ 1 for all i

Once the pairwise comparison matrix is formed, local priorities are calculated by solving for the eigenvector of the pairwise comparison matrix. In this paper, the computational procedure described in Wabalickis (1987) and Cheng and Li (2001) is used to get the eigenvector. In this procedure, first, every element of the pairwise comparison matrix is divided by its column sum. Then, the row sums are calculated and normalized by dividing each of them by their sum to make the sum equal to 1. Normalized row sums give the eigenvector. Once the local weights are calculated, global weights can be calculated by combining local weights with respect to all successive hierarchical levels (Korpela and Tuominen, 1996a, b). The synthesized global weights express the contribution of each component to the overall goal of the AHP hierarchy. In the application of the AHP, it is also possible to check the consistency of judgments by calculating consistency ratio (CR), defined as: CR ¼

CI RI

where CI is the consistency index and RI is the random index (Hajeeh and AI-Othman, 2005) which is taken by Table II related to size of the matrix. CI defined as: Size of matrix

1

2

3

4

5

6

7

8

9

10

RI

0

0

0.58

0.9

1.12

1.24

1.32

1.41

1.45

1.49

Table II. Random index values for matrices

IJCST 21,4

242

CI ¼

ðlmax 2 nÞ ðn 2 1Þ

The CR value should be less than 10 percent if judgments are to be considered consistent (Yurdakul and Tansel, 2004). In contrast, if the CR value is larger than the acceptable value, the matrix results are inconsistent and were exempted for the further analysis (Wong and Li, 2007). Literature review AHP has been applied to a wide variety of decisions such as car purchasing (Byun, 2001), supplier selection (Tam and Tummala, 2001), IS project selection (Muralidar and Santhanam, 1990; Schniedejans and Wilson, 1991) and software selection (Kim and Yoon, 1992; Mamaghani, 2002). Yang et al. (2000) applied AHP to determine systematic layout planning on semiconductor wafer fabrication facilities. Ho applied the AHP model to strategically evaluate emerging technologies in the semiconductor foundry industry (Ho, 2004). Furthermore, Yurdakul (2004) adopted AHP for selecting machine tool alternatives so as to help the manufacturing strategy of a manufacturing organization. And also it has been widely utilized in various fields: software selection problems (Min, 1992), economic and management problem solving (Yang and Lee, 1997), plant location selection (Chan et al., 2004), supplier selection (Kahraman et al., 2003), evaluation of project termination or continuation, based on the benchmarking method (Liang, 2003), selection of the best alternative between different outsourcing contracts in terms of maintenance services (Bertolini et al., 2004) and so on. A number of methods have been applied to ERP or other IS selection including scoring, ranking, mathematical optimization, and multi-criteria decision analysis. The scoring method is intuitive, but too simple to truly reflect opinions of the decision makers (Lucas and Moore, 1976). Buss (1983) proposed a ranking approach to compare computer projects. Mathematical optimization such as goal programming, 0-1 programming, and nonlinear programming have been applied to resource optimization for IS selection. Santhanam and Kyparisis (1995, 1996) proposed a nonlinear programming model to optimize resource allocation allowing for the interaction of factors; their model considered interdependencies between projects in the IS selection process. Lee and Kim (2000) claimed that Santhanam and Kyparisis’ model dealt with IS selection problems with limited criteria. They combined the analytic network process and a 0-1 goal-programming model to select an IS project. Badri et al. (2001) presented a 0-1 goal programming model to select an IS project considering multiple criteria including benefits, hardware, software and other costs, risk factors, preferences of decision makers and users, completiontime, and training time constraints. However, the applicability of these methods is often weakened by sophisticated mathematic models or limited attributes to carry out in a real-world ERP system selection decision, especially when some attributes are not readily quantifiable, as well as not too easy for managers to understand (Wei et al., 2005). Research findings In order to select ERP suppliers for clothing industry following steps are used, which are developed by Saaty (1980):

Step 1. Define the evaluative criteria used to select the suitable ERP supplier, and establish a hierarchical framework At the beginning of the paper, all modules of ERP systems, which are participated in current market are investigated in details and it is tried to determine which modules are used in which departments of clothing firms. The criteria of choosing the right supplier are discussed with the firms’ managers and experts. According to these discussions and literature reviews nine criteria are determined. These are: (1) functionality; (2) implementation approach; (3) support; (4) costs; (5) organizational credibility; (6) experience; (7) flexibility; (8) customer focused; and (9) future strategy.

Selection of ERP suppliers using AHP tools 243

In light of the criteria above, we investigate three different ERP suppliers and their softwares, which are developed for clothing industry are called as A, B, C during the study. In order to prevent emotional evaluations, “costs” criterion is kept separate from calculations. It is preferred to use this criterion in cost benefit analysis at final stage. Hierarchical scheme of problem is composed according the criteria below (Figure 1). Level one represents the goal, which is to select the best ERP supplier. The second level represents selection criteria, followed by the alternatives in the lower level. Step 2. Establish each factor of the pairwise comparison matrix A set of pairwise comparison matrices is constructed (size 8 £ 8) by using the relative scale measurement shown in Table I. The pairwise comparisons are done in terms of which element dominates the other. Evaluations of criteria, which are done by experts and clothing firm managers take place in Table III. There are n(n 2 1) 56 judgments required to develop the set of matrices in Step 2. Reciprocals are automatically assigned in each pairwise comparison. This matrix is called as “A” matrix: Selection of ERP supplier

Functionality

Implementation approach

Firm A

Support

Organizational credibility

Firm B

Experience

Flexibility

Firm C

Customer focused

Future strategy

Figure 1. Decision hierarchy for the selection ERP supplier

Table III. Pairwise comparison matrix for criteria

Functionality Implementation approach Support Organizational credibility Experience Flexibility Customer focused Future strategy Sum of colums 1/2 1/3 3 2 1/3 13.1667

1 3

1/3 1/2 1/5 1/6 1/2 1/6 1/4 2.8667

3

Implementation approach

1

Functionality

1/5 1/3 3 1/2 1/4 9.6167

1/3 1

4

Support

1 2 5 3 2 25.0000

2 5

5

Organizational credibility

1/2 1 4 2 3 22.5000

3 3

6

1/5 1/4 1 1/3 1/4 4.7000

1/3 1/3

2

Experience Flexibility

1/3 1/2 3 1 1/3 13.6667

1/2 2

6

Customer focused

244

Criteria

1/2 1/3 4 3 1 19.8333

3 4

4

Future strategy

IJCST 21,4

Step 3. Calculate the eigenvalue and eigenvector Pairwise comparison matrix (A) is normalized by the computational procedure described in Wabalickis and Cheng and Li (Table IV). In order to get the eigenvector (w), average of each line is calculated. To get the local priority vector (Aw), matrix A is multiplied with eigenvector of criteria. After dividing Aw to eigenvector value for every criterion, average of these results gives us largest eigenvalue of matrix (lmax, Table V). Step 4. Perform the consistency test After having all the pairwise comparisons, the consistency is determined by using the eigenvalue, lmax, to calculate the consistency index, CI as follows: CI ¼ ðlmax 2 nÞ=ðn 2 1Þ ¼ ð8:6600 2 8Þ=7 ¼ 0:0943 where n is the matrix size. Judgment consistency can be checked by taking the CR of CI with the appropriate value in Table II: CR ¼ CI=RI ¼ 0:0943=1:41 ¼ 0:0669 ¼ 6:69 percent The CR is acceptable, since CR # 0.10. Step 5. Steps 2-4 are performed for all levels in the hierarchy (Al-Harbi, 2001) After the consistency test for criteria, all steps are repeated in order to calculate eigenvector of criteria for each alternative. Also consistency test is performed for all calculations. The results are shown in Table VI. Best ERP supplier is gained by multiplying eigenvector of criteria and eigenvector of alternavites. As it is seen in Table VII, firm A is the best of selection. The weights of each criterion for three alternatives are shown in Figure 2. As it is seen in Figure 2 functionality, flexibility, and support criteria have the biggest slice in all alternatives. As it is mentioned before, in order to have fair evaluation “costs” criterion is kept separate from calculations. Because “costs” criterion is the most important criterion naturally, thus it has to affect the results after the best alternative is selected. In order to add this criterion to the whole calculations, first the prices have to be normalized by dividing each price to sum of all prices. Then cost benefit analysis can easily be performed by dividing synthesis value to normalized costs (Table VIII). The result will help us in our definitive decision. Thus, firm A is selected as the best supplier according to the calculations above. Discussions and conclusions Recently, companies face the challenge of increasing competition, expanding markets, and rising customer expectations. Especially, clothing manufacturers must be able to address rapidly changing consumer needs. The consumer focus for clothing manufacturers requires a shortened product life cycle and increased diversification of fashion. The clothing industry in the changing market must obtain the capability to produce many different types of products in small quantities in shorter lead times. To remain competitive, clothing companies are increasingly turning to ERP systems.

Selection of ERP suppliers using AHP tools 245

Table IV. Normalized pairwise comparison matrix

Functionality Implementation approach Support Organizational credibility Experience Flexibility Customer focused Future strategy

0.3488 0.1163 0.0872 0.0698 0.0581 0.1744 0.0581 0.0872

Functionality 0.2278 0.0759 0.2278 0.0380 0.0253 0.2278 0.1519 0.0253

Implementation approach 0.4159 0.0347 0.1040 0.0208 0.0347 0.3120 0.0520 0.0260

Support 0.2000 0.0800 0.2000 0.0400 0.0800 0.2000 0.1200 0.0800

Organizational credibility

0.2667 0.1333 0.1333 0.0222 0.0444 0.1778 0.0889 0.1333

Experience

0.4255 0.0709 0.0709 0.0426 0.0532 0.2128 0.0709 0.0532

Flexibility

0.4390 0.0366 0.1463 0.0244 0.0366 0.2195 0.0732 0.0244

Customer focused

246

Criteria

0.2017 0.1513 0.2017 0.0252 0.0168 0.2017 0.1513 0.0504

Future strategy

IJCST 21,4

Criteria Functionality Implementation approach Support Organizational credibility Experience Flexibility Customer focused Future strategy lmax

A B C A B C

Firms A B C

Functionality 0.428571429 0.285714286 0.285714286 Experience 0.1667 0.5000 0.3333

Eigenvector (w)

(Aw)

Aw/w

0.3157 0.0874 0.1464 0.0354 0.0436 0.2157 0.0958 0.0600

2.8483 0.7428 1.2986 0.2984 0.3667 1.9536 0.8416 0.4922

9.0223 8.5016 8.8695 8.4368 8.4031 9.0553 8.7864 8.2050 8.6600

Implementation approach 0.3750 0.3750 0.2500 Flexibility 0.6000 0.2000 0.2000

Support 0.3333 0.5000 0.1667 Customer focused 0.5000 0.1667 0.3333

Organizational credibility 0.2857 0.2857 0.4286 Future strategy 0.5 0.25 0.25

Selection of ERP suppliers using AHP tools 247 Table V. Calculation of lmax

Table VI. Eigenvectors of criteria for all alternatives

Priority values 0.441575014 0.302204781 0.256220204

Selecting a suitable ERP system is the basis of implementing ERP project successfully. This paper presents AHP for ERP supplier selection, which is based on criteria, which are determined by firm managers and experts. Best alternative of three ERP suppliers is selected easily with the help of AHP. The main advantages of using the AHP methodology are: . the hierarchical structure definition permits to understand all the variables involved and their relationship; . the decisional problem is represented in a structured way; . the method does not replace the personnel involved in the resolution process but integrates all the judgments with structured links; and . from simple choice, the decision becomes process (Bertolini et al., 2006). The results of the paper are hoped to be helpful to clothing manufacturers in both developed and developing economies with reference to selecting the suitable ERP systems in their organizations.

Table VII. Selection of best ERP supplier

IJCST 21,4

0.5 Future strategy Customer focused Flexibility

0.45 0.4

248

Experience Organizational credibility Support

Implementation approach Functionality

0.35 0.3 0.25 0.2 0.15 0.1

Figure 2. The weights of each criterion for three alternatives

0.05 0 A

Alternatives

Table VIII. Cost benefit analysis

A B C Total

B

C

Synthesis value

Costs ($)

Normalized costs

Cost benefit rate

0.441575014 0.302204781 0.256220204

48.000 45.000 40.000 133.000

0.360902256 0.338345865 0.30075188

1.223531 0.893183 0.851932

References Al-Harbi, K.M.A.-S. (2001), “Application of the AHP in project management”, International Journal of Project Management, Vol. 19, pp. 19-27. Badri, M.A., Davis, D. and Davis, D. (2001), “A comprehensive 0-1 goal programming model for project selection”, International Journal of Project Management, Vol. 19, pp. 243-52. Bertolini, M., Braglia, M. and Carmignani, G. (2006), “Application of the AHP methodology in making a proposal for a public work contract”, International Journal of Project Management, Vol. 24, pp. 422-30. Bertolini, M., Bevilacqua, M., Braglia, M. and Frosolini, M. (2004), “An analytical method for maintenance outsourcing service selection”, International Journal of Quality & Reliability Management, Vol. 21 No. 7, pp. 772-88. Buss, M.D.J. (1983), “How to rank computer projects”, Harvard Business Review, Vol. 61 No. 1, pp. 118-25.

Byun, D.H. (2001), “The AHP approach of selecting an automobile purchase model”, Information & Management, Vol. 38, pp. 289-97. Chan, A.H.S., Kwok, W.Y. and Duffy, V.G. (2004), “Using AHP for determining priority in a safety management system”, Industrial Management & Data Systems, Vol. 104 No. 5, pp. 430-45. Chang, C.W., Wu, C.R., Lin, C.T. and Chen, H.C. (2007), “An application of AHP and sensitivity analysis for selecting the best slicing machine”, Computers & Industrial Engineering, Vol. 52, pp. 296-307. Cheng, E.W.L. and Li, H. (2001), “Analytic hierarchy process: an approach to determine measures for business performance”, Measuring Business Excellence, Vol. 5, pp. 30-6. Davenport, T.H. (1998), “Putting the enterprise into the enterprise system”, Harvard Business Review, Vol. 76 No. 4, pp. 121-31. Davenport, T.H. (2000), Mission Critical: Realizing the Promise of Enterprise Systems, Harvard Business School Press, Boston, MA. Forman, E.H. (1996), “Decision by objectives”, manuscript, Expert Choice, Pittsburg, PA. Hajeeh, M. and AI-Othman, A. (2005), “Application of the analytical hierarchy process in the selection of desalination plants”, Desalination, Vol. 174, pp. 97-108. Ho, C.T. (2004), “Strategic evaluation of emerging technologies in the semiconductor foundry industry”, PhD thesis, Portland State University, Portland, OR. Kahraman, C., Cebeci, U. and Ulukan, Z. (2003), “Multi-criteria supplier selection using fuzzy AHP”, Logistics Information Management, Vol. 16 No. 6, pp. 382-94. Kim, C.S. and Yoon, Y. (1992), “Selection of a good expert system shell for instructional purposes in business”, Information & Management, Vol. 23 No. 5, pp. 249-62. Korpela, J. and Tuominen, M. (1996a), “A decision aid in warehouse site selection”, International Journal of Production Economics, Vol. 45, pp. 169-80. Korpela, J. and Tuominen, M. (1996b), “Inventory forecasting with a multiple criteria decision tool”, International Journal of Production Economics, Vol. 45, pp. 159-68. Lee, J.W. and Kim, S.H. (2000), “Using analytic network process and goal programming for interdependent information system project selection”, Computers & Operations Research, Vol. 27, pp. 367-82. Liang, W.Y. (2003), “The analytic hierarchy process in project evaluation. An R&D case study in Taiwan”, Benchmarking: An International Journal, Vol. 10 No. 5, pp. 45-56. Liao, X., Li, Y. and Lu, B. (2007), “A model for selecting an ERP system based on linguistic information processing”, Information Systems, Vol. 32 No. 7, pp. 1005-17. Lucas, H.C. and Moore, J.R. Jr (1976), “A multiple-criterion scoring approach to information system project selection”, Systems and Operational Research, Vol. 14 No. 1, pp. 1-12. Mahesh, G. and Amarpreet, K. (2006), “Enterprise resource planning systems and its implications for operations function”, Technovation, Vol. 26, pp. 687-96. Mamaghani, F. (2002), “Evaluation and selection of an antivirus and content filtering software”, Information Management & Computer Security, Vol. 10 No. 1, pp. 28-32. Min, H. (1992), “Selection of software: the analytic hierarchy process”, International Journal of Physical Distribution & Logistics Management, Vol. 22 No. 1, pp. 42-53. Muralidar, K. and Santhanam, R. (1990), “Using the analytic hierarchy process for information system project selection”, Information & Management, Vol. 18 No. 1, pp. 87-95.

Selection of ERP suppliers using AHP tools 249

IJCST 21,4

250

Ngai, E.W.T. and Chan, E.W.C. (2005), “Evaluation of knowledge management tools using AHP”, Expert Systems with Applications, Vol. 29, pp. 889-99. Rajagopal, P. (2002), “An innovation-diffusion view of implementation of enterprise resource planning (ERP) systems and development of research model”, Information & Management, Vol. 40, pp. 87-114. Saaty, T.L. (1980), The Analytic Hierarchy Process, McGraw-Hill, New York, NY. Saaty, T.L. and Alexander, T.M. (1981), Thinking with Models: Mathematical Models in the Physical, Biological and Social Sciences, Pergamon, London. Santhanam, R. and Kyparisis, G.J. (1995), “A multiple criteria decision model for information system project selection”, Computers & Operations Research, Vol. 22 No. 8, pp. 807-18. Santhanam, R. and Kyparisis, G.J. (1996), “A decision model for interdependent information system project selection”, European Journal of Operational Research, Vol. 89, pp. 380-99. Schniedejans, M.J. and Wilson, R.L. (1991), “Using the analytic hierarchy process and goal programming for information system project selection”, Information & Management, Vol. 20 No. 5, pp. 333-42. Skibniewski, M.J. and Chao, L. (1992), “Evaluation of advanced construction technology with AHP method”, Journal of Construction Engineering and Management, ASCE, Vol. 118 No. 3, pp. 577-93. Tam, M.C.Y. and Tummala, V.M.R. (2001), “An application of the AHP in supplier selection of a telecommunications system”, Omega, Vol. 29 No. 2, pp. 171-82. Umble, E., Haft, R. and Umble, M. (2003), “Enterprise resource planning: implementation procedures and critical success factors”, European Journal of Operational Research, Vol. 146, pp. 241-57. Verville, J. and Halingten, A. (2003), “A six-stage model of the buying process for ERP software”, Industrial Marketing Management, Vol. 32, pp. 585-94. Wabalickis, R.N. (1987), “Justification of FMS with the analytic hierarchy process”, Journal of Manufacturing Systems, Vol. 7 No. 3, pp. 175-82. Waters, D. (1996), Operations Management, Addison-Wesley, Wokingham. Wei, C.C., Chien, C.F. and Wang, M.J. (2005), “An AHP-based approach to ERP system selection”, International Journal of Production Economics, Vol. 96, pp. 47-62. Wong, J.K. and Li, H. (2007), “Application of the analytic hierarchy process (AHP) in multi-criteria analysis of the selection of intelligent building systems”, Building and Environment, Vol. 43 No. 1, pp. 108-25. Yang, J. and Lee, H. (1997), “An AHP decision model for facility location selection”, Facilities, Vol. 15 No. 9, pp. 241-54. Yang, T., Su, C.T. and Hsu, Y.R. (2000), “Systematic layout planning: a study on semiconductor wafer fabrication facilities”, International Journal of Operations & Production Management, Vol. 20 No. 11, pp. 1360-72. Yurdakul, M. (2004), “AHP as a strategic decision-making tool to justify machine tool selection”, Journal of Materials Processing Technology, Vol. 146 No. 3, pp. 365-76. Yurdakul, M. and Tansel, Y. (2004), “AHP approach in the credit evaluation of the manufacturing firms in Turkey”, International Journal of Production Economics, Vol. 88, pp. 269-89.

Further reading King, S.F. and Burgess, T.F. (2006), “Beyond critical success factors: a dynamic model of enterprise system innovation”, International Journal of Information Management, Vol. 26, pp. 59-69.

Corresponding author Can U¨nal can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

Selection of ERP suppliers using AHP tools 251

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

IJCST 21,4

Parallel garment drape simulation of triangular mesh using GPU programming

252

In Hwan Sul

Received 22 August 2008 Accepted 16 November 2008

i-Fashion Technology Center, Konkuk University, Seoul, South Korea Abstract Purpose – The purpose of this paper is to determine the possibility of implementing parallel processing feature of graphic processor unit (GPU) in garment drape simulation. Design/methodology/approach – Velocity-Verlet method based on explicit integration is used to drape triangular table cloth meshes. Both drape simulation and collision detection engines are converted to GPU version. Simulation speeds of simple linear algebra and actual free fall table cloth simulation are compared with those of the central processing unit (CPU) version. Findings – There is apparent calculation speed increase when the parallel computation of GPU is implemented. But the current GPUs have several limits for general purpose computation, so the original CPU version algorithm should be split and modified to be used in GPU. Originality/value – This paper implemented GPU parallel processing technique in both drape simulation and collision detection. Voxel-based method is used to find possible collision pairs. Triangular meshes, which are more difficult to implement than quadrilateral ones in GPU programming, are successfully implemented. Keywords Garment industry, Meshes, Cloth, Simulation, Programming and algorithm theory Paper type Research paper

International Journal of Clothing Science and Technology Vol. 21 No. 4, 2009 pp. 252-269 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910960006

1. Introduction Fast and exact prediction of garment behavior is both theoretically and practically important topic for textile researchers. Not to mention, there have been numerous researches for the exact prediction. First, finite element method (Kang et al., 1995) was implemented for garment simulation but it was replaced by particle-based method (Baraff and Witkin, 1998; Volino and Magnenat-Thalmann, 2000; Choi and Ko, 2003) which has less computational cost and is easier to deal with boundary conditions. Volino and Magnenat-Thalmann (2000) showed virtual fashion show using particle-based method. Several commercial engines are available now and they are used for special effects in movies. Various fabric material mechanical properties measured by KES-FB or FAST system or even anisotropy could be expressed successfully. For 3D apparel computer-aided design (CAD) developers, simulation quality is not negligible but calculation speed is more important factor because the pattern designing step needs repetitive works. Garment patterns are not determined by a single drawing trial but should be modified until it fits the manikin or human body well. Therefore, fast drape simulation calculation is an essential factor for three dimensional apparel CAD system. Drape simulation is composed of two major processes. The one is prediction of the deformation and the other is the collision detection among clothes or between clothes and bodies. For the former, Baraff and Witkin (1998) introduced semi-implicit drape simulation formula to use more large time step. Their work was brilliant, but still the

practically suitable time step is limited to 0.01-0.02 s/frame so 100 of iteration is needed for 1 s simulation. For the latter, many researchers tried to find alternative way for expensive cloth triangle-to-body triangle pair collision tests. Volino and Magnenat-Thalmann (2000) used hierarchical fabric structure so that one fabric was decomposed to several sub-groups. They also used octree-based hierarchical modeling of body data for collisions. Such modified approaches show more improved results than raw triangle pair test method, but it is far from real time simulation when the number of cloth layer increases. If the speed of the original algorithm cannot be improved, then it would be wiser to seek another way to speed up the performance. The most suitable and practical alternative would be using the parallel processing technique. Cloth simulation is based on triangular or quadrilateral meshes and their calculations can be easily vectorized. In mechanics or aerodynamics area, very large systems are solved by supercomputer or multi-node computer network Kim et al. (1994). But such method is not suitable for apparel CAD area because of not only economical reason but also communication problem. Garment simulation is a dynamic simulation so the communication between each noode computers would be more burdensome. The more economical way to implement parallel computation is to use graphic processor unit (GPU) of graphics cards instead of central processing unit (CPU). Current GPU’s of 3D graphic cards are inherently designed for parallel works such as vertex transformation and pixel shading. Their original purpose is to shade the pixels in a short time for real time 3D graphics but researchers found out they can be utilized to other general purpose programming Luebke (2005). The use of GPU programming covers wide areas from matrix solving using conjugate gradient method to medical image processing. Boltz et al. (2003) used GPU for sparse matrix solvers using conjugate gradient and multigrid method. Krueger and Westermann (2003) devised linear algebra operators using GPU. Skjermo and Eidheim (2005) used GPU for image analysis of medical data. The first application areas were limited to linear algebra and image analysis because vectors and image data were easy to deal with in GPU programming. Later the area was extended to various fields such as particle-based modeling. Tejada and Ertl (2005) implemented implicit integration of quadrilateral mesh. Wong and Baciu (2005) used GPU for collision detection test on one piece dress and multi-layer clothes on a sphere. Georgii and Westermann (2005) compared point based and edge based GPU mass spring system. The previous researches tried to adopt GPU in either deformation prediction or collision detection and showed remarkable speedup factor compared with CPU. But there was no integrated approach, which contains both deformation prediction and collision detection yet. Moreover, only quadrilateral meshes were used because it was easier to deal with than triangular ones. The paper deals with drape simulation algorithm modified for GPU programming. Deformation prediction was done using velocity-Verlet integration scheme (Trobec and Janezic, 1995) and collision test was done between cloth and body. Triangular elements were used for meshes so that more detailed description of patterns shapes is feasible. 2. Overview of GPU programming 2.1 GPU internal structure Figure 1 shows internal structure of general graphic cards. To display 3D objects, 3D mesh data should be projected on to pixels of image plane which lies on the monitor and then each pixel should be given suitable colors and shades according to the normal

Garment drape simulation of triangular mesh 253

IJCST 21,4

CPU

GPU

254 VS

VS

VS

VS

VS

FS FS FS FS FS FS FS FS FS FS FS FS FS FS FS FS

VS

Frame buffer

Monitor

Figure 1. Internal structure of GPU

VS

Vertex shader

FS

Fragment shader

vector arithmetic (OpenGL ARB, 2003). It is the vertex shaders that take care of projection and rasterization of 3D vertices data. Meanwhile fragment shaders calculate shading and texturing of each pixel. To accelerate displaying speed, vertex and fragment shaders have multiple numbers of units so that data are processed in parallel. Each shader unit can do only simple linear algebra operations and is a little slower than CPU, but the parallel work of the overall multiple shaders can be many times faster than CPU. Generally, fragment shader has more loads and they have more number of units than fragment shaders do. Thus, it is the fragment shaders that we are going to use as means of parallel processing unit. Figure 2 represents procedure of displaying 3D mesh, whose information includes node coordinates, element indices, and normal vectors, with texture data if exists. Mesh geometry is transformed and rasterized by model view, projection, perspective and viewport matrices which comes from default functionality of OpenGLw/DirectXw or user defined shading language if the graphic card supports it. Rasterized pixels are given color and shade from normal vector and texture maps by suitable linear algebraic arithmetic whose exact equation comes be from fixed functionality or user

CPU Vertex info glBegin(GL_TRIANGLE); glVertex3f(..); glEnd(); Texture info glBindTexture(..); glMultiTexCoord(..);

Mesh coordinates & normal vector

Texture coordiates & texture data

Garment drape simulation of triangular mesh 255

GPU Vertex shader

Image plane

Fragment shader

Add shade & texture

Figure 2. Flowchart of 3D object display

defined one. Example of default shading equation of OpenGLw in RGBA mode is like the following (OpenGL ARB, 2003): Vertexcolor ¼ emissionmaterial þ ambientlightmodel *ambient material N 21 X 1 *ðspotlighteffectÞi *½ambientlight *ambientmaterial þ kc þ kl d þ kq d 2 i¼0 þ ðmax{L·n;0}Þ*diffuselight *diffusematerial þ ðmax{s·n;0}Þshininess* specularlight *specularmaterial I ð1Þ which is apparently a combination of linear algebra. The idea of general purpose GPU programming is to replace equation (1) with other scientific governing equation and to let GPU calculate linear algebra of scientific data such as matrices or cloth internal forces.

IJCST 21,4

256

2.2 Shading language description Control of GPU behavior is possible by using several shading language such as GLSL, Cg, Sh, Brook, and so on (Luebke, 2005). They differ in detailed language grammar but the actual functions of graphic card are not limited by the choice of language. But even the newest GPU’s have several common limitations for general purpose application: L1.

Input data should be in texture form which has (R,G,B,A) float quadruples.

L2.

Output data also should be in texture form.

L3.

Lengths of vertex and fragment shading languages are limited (to about 1,024 instructions).

L4.

Number of input data (textures) are limited (about 16 maximum).

L5.

Texture data can be referenced only via texture coordinates.

L6.

Calculations are done for each texel units.

These limitations come from the fact that graphic card were originally designed for rendering, not for general purpose computation. A texture means 2D or 3D matrix whose each element has (R,G,B,A) floating point numbers which follow IEEE 754 standard. As each texture element has only four components, every input data should be split and reorganized to fit the texture. Because of L1 and L4, cloth and body mesh data should have GPU optimized form. The more severe problem is the limit of shading language length (L3), which results into splitting of big governing equation into small sub-programs. Furthermore, each shading loop can generate only up to four kinds of outputs at a time, therefore any program that requires several different sets of output should be also split to several sub-programs. L2 and L3 is the current bottleneck for free GPU programming. Nevertheless, it has a strong advantage in speed over CPU. 3. Data structure 3.1 Particle-based modeling Particle-based method (Sul and Kang, 2004) with velocity-Verlet integration (Trobec and Janezic, 1995) was used for fabric drape simulation. Velocity-Verlet is the most stable algorithm known among explicit integration (Trobec and Janezic, 1995). Implicit integration can give more time step, but detailed coding of implicit integration is very complex, so the whole procedure should be split too many subprograms (L3). Therefore, we chose explicit method for simplicity. The vectors needed for velocity-Verlet method are node coordinate position vector (R), velocity (V), acceleration (A), total force acting on each node (F) and tension vector (T). Mechanical forces implemented were tension, bending and air resistance force but they shared common vector T during computation to reduce number of input data texture. Triangular mesh element was used and the tension calculation was identical to Breen et al.’s (1994). Bending force came from product of curvature and bending stiffness where curvature was approximated from angle of two adjacent triangles (Sul and Kang, 2004). 3.1.1 Mechanical forces. Implemented mechanical forces were tension, bending, and air resistance forces. Shear force was omitted because its effect was negligible. The triangular mesh configuration is shown in Figure 3.

Garment drape simulation of triangular mesh

Bending neighbor node, q Bending neighbor element Neighbor node

Particle Dashpot

Neighbor edge

257 Neighbor node

Neighbor element

Figure 3. Nomenclature of triangular mesh for bending calculation

Current node, t

~ i of a node i from N neighbor edges: Suppose we want to get tension vector T ~i ¼ T

N 21 X

E · 1ij · n~ ij

ð2Þ

j¼0

where E, elastic constant; 1ij , strain; and n~ ij , directional vector of edge ij. Bending force is expressed as: ~i ¼ B

N 21 X

· ~ E B · kij þ hB · kij · N i0

ð3Þ

j¼0

where EB, bending stiffness; hB, damping coefficient for bending; kij, curvature from · ~ i0 , normal vector of neighbor triangle; kij , backward difference of curvature; and N neighbor element. Air resistance force is: ~ i ¼ E Air · Area · V ~iÞ ~ i · ð2V Air ð4Þ where EAir, arbitrary aerodynamic constant; Area, cross-sectional area of element with ~ i , wind vector at node i. respect to wind vector; V 3.1.2 Collision detection. Collision detection and collision response are the most time determining step in explicit integration based cloth simulation. The complexity is determined by the number of possible collision pairs. Cloth-to-cloth self collision is the more severe problem than cloth-to-body collision. Self collision can be detected by the same method for the cloth-to-body collision, but it was not feasible to implement multi-layer cloth information in the current 16 textures limit of the current commercial GPU’s. So we checked only cloth-to-body collision because the main purpose of the paper is to study the possibility of GPU programming in cloth simulation. Once cloth and body triangle pair to detect collision is chosen, two kinds of triangular collision tests were done. Figure 4(a) shows NODE_INFILTRATION case which means a cloth node is under body mesh element and Figure 4(b) shows EDGE_PENETRATION case which means a cloth edge goes under or through body mesh element. Another third case (FACE_PENETRATION) which may be collision of two triangle faces may be used, but it could not be implemented because of current

IJCST 21,4

p (p'x, p'y, p'z)

p (px , py, pz) n (nx , ny, nz) = (r = 1,θ,φ)

p' (p'x , p'y, 0)

z

258 x

n' (0,0,1) = (r' = 1,0, π ) 2

Rotation & Translation

y (a) Node infiltration p (px, py, pz) Rotation & Translation

n

Figure 4. Types of collision checking

z x

p' (p'x, p'y, p'z)

q (qx, qy, qz)

n' (0,0,1)

q' (q'x, q'y, q'z)

y (b) Edge infiltration

instruction length limit (L3). Rotation and translation of both triangles are needed to make the problem two dimensional and this was done by MapToXYPlane() function which transforms triangles so that one vertex lies on origin and normal vector faces þ z direction. And then PointInTriangle2D() test was done to check whether a node or an edge penetrates the other triangle or not. 3.2 Voxel-based encoding As mentioned earlier, time bottleneck of drape simulation is the detection of cloth-to-cloth or cloth-to-body collisions. In GPU version of collision detection, nearest body elements of a cloth node which can be possible candidate for collision cannot be searched because scanning all the texture data is not possible (due to L5 and L6). Therefore, an alternative data format should be devised for finding possible collision set of body elements. We voxelized 3D space in which body and cloth mesh lies and recorded element indices that each voxel possesses (Figure 5). Number of voxel size is limited by maximum texture size (4,096 £ 4,096, at the time of writing). This voxel information was contained in BodyVoxel texture. Table I shows the type and components of input data textures. As the Cartesian coordinate system is used, voxel ID Vi,j,k can be easily known from node coordinate as in Figure 6. Once voxel ID is known in the shading language, its nearby body element list can be known from BodyVoxel texture. AABB0 and AABB1 are global constants that represent the size of axis aligned bounding box of total voxels: AABB0 ¼ ðxmin ; ymin ; zmin ; 0Þ

ð5Þ

AABB1 ¼ ðxmax ; ymax ; zmax ; 0Þ

ð6Þ

Garment drape simulation of triangular mesh 259

Figure 5. Texture data example in GPU version drape calculation

ID

Texture map name R

0 1 2 3 4 5 6 7 8 9 10-13 14-16 17-20

R F A V R0 HalfEdge PreviousStepInfo Tension ClothNormal Collision BodyElement1-4 BodyVoxel1-3 Temporary1-4

G

Component data type B A

Position x Position y Position z Initial edge length Vector x Vector y Vector z – Vector x Vector y Vector z – Vector x Vector y Vector z – Position x Position y Position z Initial bending curvature Cloth halfedge (start node, end node, left element, right element) Last edge length Last curvature – – Vector x Vector y Vector z – Vector x Vector y Vector z – Vector x Vector y Vector z – 4 £ 4 collision matrix Number of elements in voxel, element ID no. 0-no. 11 – – – –

Table I. Types of texture data

IJCST 21,4

Figure 6. Finding 2D voxel ID from node coordinate

Voxel index i

260

Voxel space

Voxel index k

Vi–1,j,k

Vi,j,k

Vi–1,j–1,k

Vi,j–1,k

→ Clothnode R = (x,y,z)

Voxel index i index_i = int |x – x_min) / VoxelSixe| index_ j = int |y – y_min) / VoxelSixe| index_k = int |z – z_min) / VoxelSixe|

Vi,j,k = index_i*nVoxelY*nVoxelZ + index_ j*nVoxelZ + index_k

3.3 Vector representation Using recent FramebufferObject (FBO) or pBuffer extension (Luebke, 2005), the result data can be sent not to the screen but to other texture memory without rounding off error. Because those extensions are supported only in recent cards, some older cards sends data only to screen which results into clamping of values to [0,1]. In case those extensions are not supported, we encoded position vector as: P~ texture ¼ P~ real * ðAABB1 2 AABB0 Þ þ AABB1 ð7Þ ~ ~ where Preal and Ptexture are position vector in real coordinates and in texture each. Force vectors were also encoded to [0,1] interval using approximate large value MaxVectorSize: F~ texture ¼ F~ real * ðMaxVectorSizeÞ* 2 þ 1 ð8Þ But such redundant encoding caused rounding off error and the simulation showed much error. So we chose nVidia 7900GT card which supports FBO extension for our simulation. 3.4 Texture size Width and height of textures should be identical to rendering context size in general purpose GPU calculation so that each pixel corresponds to one texture element exactly. The maximum size of textures is generally 4,096 £ 4,096 in 2D textures which corresponds to about 1.6 £ 107 texels (texture elements). But due to L2 condition, complex data formats such as body element indices should have multiple numbers (designated as MAXNeighbor) of texels for each data. Suppose number of cloth nodes, elements, body node and elements are CN, CE, BN, and BE, respectively. Generally CN . CE and BN . BE, then the size of texture TW (width) or TH (height) should be: qﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ TW ¼ TH $ MAXneighbor * maxðCN; BNÞ ð9Þ

All the textures have the same size except for the body voxel texture which should have size dependent on the total number of voxels. Body voxel has the following width (TBW) and height (TBH): pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ TBW ¼ TBH $ N voxel * EMAXvoxel ð10Þ where Nvoxel is number of total voxels and EMAXvoxel is maximum element number of voxel.

Garment drape simulation of triangular mesh 261

3.5 Shading language descrition Figure 7 represents GPU version drape simulation procedure. Each operation contains one pair of vertex and fragment shading operation. Vertex shaders were only used to deliver texture coordinate to fragment shaders. Texture coordinates are essential data varying vec4 TexCoord0; void main() { TexCoord0 = gl_MultiTexCoord0; gl_Position = ftransform(); }

(a) Vertex shader uniform sampler2D uRTexMap; uniform sampler2D uATexMap; uniform sampler2D uVTexMap; varying vec4 TexCoord0; void main() { vec4 pCloth_R; vec4 pCloth_V; vec4 pCloth_A; vec4 vecdT; vec4 newR; pCloth_R pCloth_V pCloth_A

= texture2D( uRTexMap, TexCoord0.xy ); = texture2D( uVTexMap, TexCoord0.xy ); = texture2D( uATexMap, TexCoord0.xy );

newR = PositionVectorToRealCoord( pCloth_R); pCloth_V = ForceVectorToReal( pCloth_V); pCloth_A = ForceVectorToReal( pCloth_A); newR = newR + pCloth_V * vec4(dT) + pCloth_A * vec4(0.5 * dT * dT); newR = PositionVectorToTexture(newR); gl_FragColor = newR; }

(b) Fragment shader

Figure 7. Example of shading language

IJCST 21,4

262

for referencing specific data from textures. Thus, the vertex shader has the simpler form than the fragment shader as shown in Figure 7(a). Most of the actual calculations are done in fragment shading as shown in Figure 7(b) where ‘Input initial R’ operation of Table II is shown for example. Some sub-functions were omitted for conciseness. After fragment shader operation is done, the result is updated to proper output textures for next step. 3.6 Gathering texture data GPU operations occur per pixel and summative operation such as tension calculation of a node should be split to the several texels as many as number of neighbor edges because input data type is restricted to four float values of RGBA. Therefore, tension calculation is composed of getting stress vector per each edge and summing those vectors into single tension vector. The latter operation is generally called as “gathering”. Figure 8 shows two consecutive steps of tension calculation of Section 3.1.1. Bending and air resistance forces are gathered in the same way. 4. Results and discussion 4.1 Simple linear algebra speed comparison between GPU and CPU To see the actual speedup factor of GPU, test results of simple multiplication and addition (AXPY, y ¼ a *x þ y) operations is shown in Figure 9. As conventional C language uses main() function in.cpp file, GPU programming needs two shader text files with each main() functions. The times taken for the CPU operations are from the following loop operation: float x½SIZE; ð11Þ float y ¼ new float *[SIZE]; for(i ¼ 0; i , SIZE; iþ þ ) { y[i] þ ¼ a *x[i]; }

Table II. Hardware specifications of CPU’s

ID

CPU type

CPU1 CPU2 CPU3

Intel Pentium 3.0 GHz Intel Pentium D 3.0 GHz AMD Opteron 265 £ 2

Number of cores

Clock speed (GHz)

1 2 4

3.0 3.0 1.8

Taxel group for node i

Figure 8. Example of gathering data

Step 7 : Get stress vectors

Eεi0 → . ni0

Eεi1 .→ ni1

Eεi2 → . ni2

EεiN–1 → . niN–1

Step 8 : Gather tension T

→ Ti

→ Ti

→ Ti

→ Ti

Garment drape simulation of triangular mesh

14,000

Calculation time (ms/frame)

12,000 CPU1 GPU1

10,000 8,000

263

6,000 4,000 2,000 0 0

1

2

3 4 5 6 Ln (no. of nodes) (a) Calculation time versus number of mesh nodes

7

Operation time (msec)

700

600

Figure 9. Calculation speed comparison of GPU programming

0 0 1,000 2,000 3,000 4,000 5,000 Number of trigonometric function (sine) calls (b) Calculation time versus different number of functions calls

which is a familiar C language description. Meanwhile the time taken for the GPU operations is for the only one pair of vertex and fragment shader operation, where vertex shader language has: Void main() { gl_TexCoord½0 ¼ gl_MultiTexCoord0; ð12Þ gl_Position ¼ ftransformðÞ; } And the fragment shader language has: uniform sampler2D uXTexMap; uniform sampler2D uYTexMap; uniform float A;

ð13Þ

IJCST 21,4

void main() { vec4 X ¼ texture2DðuXTexMap; gl_TexCoord½0:xyÞ;

ð14Þ

vec4Y ¼ texture2DðuYTexMap; gl_TexCoord½0:xyÞ; gl_FragData½0 ¼ vec4ðAÞ* X þ Y;

264

ð15Þ

} The above two text files are typical vertex/fragment shader sets for GPU calculation. As the GPU rendering is generally pair of two processes, which are rasterization of each vertex coordinates and giving colors and shaders to the pixels. The operations are called vertex and fragment shaders, respectively. The major difference between conventional CPU programming (C language or else) and GPU programming is the data storage style. Unlike conventional CPU memory usage using static or dynamic allocation (as shown in equation (1)), GPU programming stores large vector data only in textures. Texture is a rectangular matrix with four sub components (x,y,z,w). So it is necessary to store data in textures in advance. Once the data are ready in textures, they can be accessed in fragment shader (as in equation (4)), and the calculation results are stored in another texture (as in equation (5)). Figure 9(a) shows the calculation time of CPU increases with the number of data, while that of GPU is almost constant. Once the GPU calculation is activated, the calculation time is dependent on the number of textures the shader language accesses and on the texture sizes. In the simple case of AXPY operation, the size of texture is the dominant factor for GPU calculation speed and it shows very slow increment compared with that of CPU. This is due to the parallel computing aspect of GPU. Figure 9(b) shows the effect of number of operations in shader text. Different numbers of trigonometric function calls (sine and cosine) were used instead of equation (15) and the result shows almost constant calculation time. This means that the internal operations inside the shader do not affect the overall performance. Figure 10 shows the

Execution time (ms/frame)

10

Figure 10. Time taken for each shaders

8

6

4

2

0 0

2

4

6 8 Shader ID

10

12

14

16

calculation speed of each 13 shaders. It shows generally “gather” operation takes much time, because each gathering operation accesses eight pixels simultaneously while other simple operation such as ‘InputR’ accesses only one (R) texture. This again shows the GPU calculation speed is not dependent on the internal function complexity, but on the number of texture accesses. But current conventional GPU’s allow only limited number of function calls in main() (about 4,096 in the case of sine function) and to not allow loop unrolling, so very lengthy algorithm should be split into smaller ones. Otherwise there is a side effect that the GPU operation becomes very slow. So our drape simulation algorithm for GPU was composed of 13 short sub programs (shaders, Table III-VI).

ID

GPU type

GPU1 GPU2 GPU3

nVidia Geforce 6800 nVidia Geforce 7600GT nVidia Geforce 7900

Number of mesh vertices 10 106 1,087 10,395 116,277

Number of fragment shaders

Core speed (MHz)

12 12 24

350 560 525

AGP 8 £ PCI-Express 16 £ PCI-Express 16 £

GPU time

0 0 15 94 1,060

330 470 516 630 636

Note: Unit, ms/frame

330 330 330 630 9,601

Note: Unit, ms/frame

3

5 103 153 203 403 803 Note: Unit, ms/frame

Table III. Hardware specifications of GPU’s

Table IV. Calculation speed comparison of CPU and GPU for simple AXPY(y ¼ y þ a *x) operation

GPU time

1 10 100 1,000 4,000

Total number of voxels

265

Interface

CPU time

Number of sine function calls

Garment drape simulation of triangular mesh

Size of texture map for voxel data

GPU time

12 32 59 90 253 716

44.1 42.3 44.1 44.2 46.4 46.9

Table V. Execution speed of a single fragment shader with different number of internal trigonometric function calls

Table VI. Effect of voxel size on drape simulation speed of 1,087 vertices table cloth

IJCST 21,4

266

4.2 Drape simulation speed comparison To implement explicit drape simulation with collision detection in GPU programming, both the algorithm and the data structure were reorganized. The algorithm is shown in Table VII, which is actually composed of 13 shader text files. The data were classified and stored in textures as shown in Table I. Each pixel of textures contains information of each mesh vertex coordinate (R), force vectors (F), acceleration vector (A) and so on. Vertex based representation of the mesh structure was used. It is suitable to represent node coordinates but it cannot represent edge information, which is needed for tension and bending force calculation. To implement edge information for each node, eight pixels were grouped into same group and each group were associated to the same mesh vertex. An ith group contains the ith vertex information and/or neighboring edge information of vertex i. For example, first eight pixels of texture R contains the same coordinates(x,y,z) of cloth vertex 0. But they have different initial edge lengths in the w component. This data structure seems redundant and complex but the current GPU data structure makes it inevitable. To compare the GPU drape simulation speed with conventional CPU programming, several CPUs and GPUs were tested. The hardware specifications are shown in Table II. Figure 11 shows the result of each drape simulation speed for free falling table cloth. The number of cloth mesh vertices was varied from 10 to 106. The time step of the simulation was fixed at 0.02 s and the material properties were fixed at E ¼ 10, EB ¼ 10, and hB ¼ 0.1 (in arbitrary dimensions). A totla of 1,000 iterations were done and the average frame rate (frame/seconds) was measured as calculation time. Rendering time were excluded because switching to and from rendering can affect GPU operation. Figure 11 shows that the calculation time of GPU is almost constant while that of CPU increase linearly. So the speedup factor of GPU versus CPU is almost 30 when the number of mesh vertices are more than 105. Even if the CPU has multiple ID

Table VII. List of split sub programs of explicit drape simulation for GPU

Fragment shader name

Operation

1

InputR

2 3 4 5 6 7

InputV InputFA Normal GatherNormal Tension GatherTension

8 9

Bending GatherBending

10

MapToXYPlane

11 12 13

Infiltration UpdateAV UpdateFR

Input initial force to F map Push R to R0 texture Get new R Get new V Get new F and A Update ClothNormal Gather ClothNormal Get new tension T Gather tension T Add T to F Get new bending force B Gather bending force B Add B to F Get voxel ID Get nearest body element ID Transform cloth vertex coordinate onto xy plane Check cloth vertex collieds with the body element triangle Update A and V map Update F and R map

Garment drape simulation of triangular mesh

Calculation time (ms/frame)

2,500 CPU1 CPU2 CPU3 GPU1 GPU2 GPU3

2,000

1,500

267

1,000

500

0 1e+1

1e+2

1e+3 1e+4 Log (number of vertices)

1e+5

Figure 11. Drape simulation speed of CPU’s and GPU’s

core (CPU2 versus CPU1) or high clock speed, it cannot follow the speed of GPU as shown in the result. Figure 12 is the final drape simulation result using GPU. There were some noises around the texture boundary so the cloth edges had a little fluctuation.

Figure 12. Snapshot of the table drape simulation using GPU

IJCST 21,4

268

5. Conclusions A possibility of the GPU programming for fast drape simulation was verified. The result showed tens of speedup factors with respect to conventional CPU calculation. The speed advantage was more valid when the cloth mesh has more vertices. But there were also disadvantages. The algorithm and data structure should be reorganized to fit GPU environment and texture memory. This paper voxelized the space so that scanning collision candidate vertices become easier, therefore evaded the use of difficult “gather” process of GPU. To represent both vertex based data and edge based data, eight pixels of a texture were grouped and contained same vertex data and different edge value. The original algorithm also should be split into many sub algorithms and it is not easy to update texture data frequently. Therefore, further advancement should be made for the GPU hardware to implement complex general purpose calculation including implicit integration, collision detection with self collision and dynamic drape simulation using motion data. Nonetheless, the speed advantage was huge and the GPU can be adapted for general purpose scientific programming with large mesh data, especially for cloth drape simulation. References Baraff, D. and Witkin, A. (1998), “A large steps in cloth simulation”, Proceedings of Computer Graphics, Annual Conference Series, pp. 43-54. Boltz, J., Farmer, I., Grinspun, E. and Scho¨der, P. (2003), “Spare matrix solvers on the GPU: conjugate gradients and multigrid”, ACM Transactions on Graphics (Proceedings of SIGGRAPH 2003), Vol. 22 No. 3, pp. 102-11. Breen, D.E., House, D.H. and Wozny, M.J. (1994), “Predicting the drape of woven cloth using interacting particles”, SIGGRAPH ’94 Conference Proceedings, July, Orlando, FL,USA, pp. 365-72. Choi, K.J. and Ko, H.-S. (2003), “Stable but responsive cloth”, ACM Transactions on Graphics, Vol. 21 No. 3, pp. 604-11. Georgii, J. and Westermann, R. (2005), “Mass-spring systems on the GPU”, Simulation Modelling Practice and Theory, Vol. 13 No. 8, pp. 693-702. Kang, T.J., Yu, W.R. and Chung, K.S. (1995), “Drape simulation of woven fabric by using the finite-element method”, Journal of Textile Institute, Vol. 86 No. 4, pp. 635-48. Kim, J.R., Kim, W.D. and Kim, S.J. (1994), “Parallel computing using semi-analytical finite element method”, AIAA Journal, Vol. 33 No. 5, pp. 1066-71. Krueger, J. and Westermann, R. (2003), “Linear algebra operators for GPU implementation of numerical algorithms”, ACM Transactions on Graphics, Vol. 22 No. 3, pp. 908-16. Luebke, D. (2005), “General-purpose computation on graphics hardware”, Proceedings of SIGGRAPH 2005, August, Dublin, Ireland. OpenGL ARB (2003), OpenGL Programming Guide, Addison-Wesley, Reading, MA. Skjermo, J. and Eidheim, O.C. (2005), “Real-time analysis of ultrasound images using GPU”, paper presented at Computer Assisted Radiology 19th International Congress and Exhibition. Sul, I.H. and Kang, T.J. (2004), “Improvement of drape simulation speed using constrained fabric collision”, International Journal of Clothing Science and Technology, Vol. 16 Nos 1/2, pp. 43-50.

Tejada, E. and Ertl, T. (2005), “Large steps in GPU-based deformable bodies simulation”, Simulation Modelling Practice and Theory, Vol. 13 No. 8, pp. 703-15. Trobec, R. and Janezic, D. (1995), “Comparison of parallel Verlet and implicit Runge-Kutta methods for molecular dynamics integration”, Journal of Chemical Information and Computer Sciences, Vol. 35 No. 1, pp. 100-5. Volino, P. and Magnenat-Thalmann, N. (2000), Virtual Clothing: Theory and Practice, Springer-Verlag, Heidelburg. Wong, W.S.K. and Baciu, G. (2005), “GPU-based intrinsic collision detection for deformable surfaces”, Computer Animation and Virtual Worlds, Vol. 16, pp. 135-61. Corresponding author In Hwan Sul can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

Garment drape simulation of triangular mesh 269

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

IJCST 21,5

Key variables in the control of color in the textile supply chain

256

Lina Maria Ca´rdenas, Renzo Shamey and David Hinks

Received 2 December 2008 Revised 18 February 2009 Accepted 18 February 2009

Department of Textile Engineering, Chemistry, and Science, North Carolina State University, Raleigh, North Carolina, USA Abstract Purpose – The purpose of this paper is to address the key variables that determine the level of control of color in a typical textile supply chain, including lighting variability, color perception, and color measurement. Design/methodology/approach – A fishbone diagram is used to demonstrate the wide range of variables that affect the control and communication of color within the textile supply chain. Findings – It is important to identify the important parameters and variables that influence the control of color within various stages of the textile supply chain. In regard to visual assessment variability, the results obtained in an ongoing study at North Carolina State University based on the psychophysical testing of 50 observers demonstrate a statistical difference for visual judgments of small color differences between naı¨ve and expert observers. Results of a paired t-test between the second and the third trial conducted by naı¨ve observers indicate that the repetition of the visual observations significantly affects the assessment of small color differences. Research limitations/implications – Assessment of lighting measurements of several stores in the USA demonstrate variability in lighting, with many stores having at least two different light sources. This variability, in combination with uncontrolled lighting from external windows and entrance/exit areas, can lead to significant variability in the color perception of textile garments displayed in such areas, and may lead to consumer experience being significantly different from that intended by the designer. Practical implications – The optimization of variables that influence the assessment and communication of color is vital to achieving effective communication between all parties involved. This can significantly reduce costs and lead times resulting in improved competitiveness and cost efficiency associated with increased consumer satisfaction and confidence in the industry. Originality/value – The repetition of visual observations significantly affects the assessment of small color differences. Keywords Colours technology, Textiles, Supply chain management Paper type Research paper

1. Introduction In the textile industry, efficient color control and communication between designer, dyer, and retailer are critical to obtaining high quality and cost efficiency. For instance,

International Journal of Clothing Science and Technology Vol. 21 No. 5, 2009 pp. 256-269 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910983759

The authors acknowledge the National Textile Center (Grant No. 02-07400, C04-NS11) for financial support of this work. The authors also thank Mr John Darsey and colleagues at DyStar for generation of the dyed fabric samples, Kashif Noor for his spectroradiometric analysis of area lighting in a retail store, Professors Nancy Cassill and Richard Aspland for their valuable comments, and Professors Roger Woodard, Rolf Kuehni, and Warren Jasper for helpful discussions. The authors would also like to thank those US retail stores that facilitated store lighting measurements and to all observers who participated in this paper.

variability in lighting may mean that the color of a product in a store is perceived significantly differently from that originally intended by the designer, despite a high level of color and lighting control during the design, and production of the product. Color communication throughout the supply chain is a dynamic process, and attempts at optimizing color control must focus on the variability that arises due to the complex interaction between supplier and consumer. Digital color communication is emerging as an effective path to process optimization, especially in reducing time and costs. However, this approach requires optimization of numerical models and a clear understanding of the scope and limitations of the assessment methodology employed. The primary objective of this communication is to illustrate to all participants in the textile supply chain the most important considerations when attempting to improve efficiency of color control in textile product development. 2. Key variables in color control In color communication, a significant array of variables exist that need to be clearly identified understood and ultimately minimized. A practical approach to understanding the complexity of the number of variables in color control is the use of a fishbone diagram or cause-and-effect diagram. The fishbone diagram is a tool to identify the potential or real causes that contribute to a single outcome (Graystone, 2000; Eckes, 2003). The causes are organized in order of significance creating a hierarchical structure in relation to the outcome. The variability in the control of color through the supply chain can be broken down into the following broad categories: . concept; . human factor; . manufacturing; . color quality control; and . point-of-sale. Figure 1, developed as a part of a large color control study at North Carolina State University, shows some of the most important causes of variability in the control and communication of color within the textile supply chain. The figure incorporates various components (Parrot, 2001; Butts, 2007; Sanger, 2007) that affect the communication and control of color. The illustration of all or even the majority of variables would make such diagrams very complex, as exemplified by a comprehensive study of all the variables of one batch dyeing process (which may form one small part of the supply chain fishbone diagram shown in Figure 1) (Koksal et al., 1992). However, a less complex design can be used as a framework to evaluate the significance of procedures and to indicate stages within the supply chain where critical problems may have been overlooked. Where necessary, a complete fishbone diagram can be developed for each sub-factor. 2.1 Concept Arguably, a successful design that considers not only consumer needs, but also production requirements and constraints, can lead to significant gains in the production, handling, and retail merchandising of a product. Color plays a vital role in almost all industries, but especially the retail textile industry. Thus, the choice of color

Key variables in the control of color 257

IJCST 21,5

Concept

Color quality control

Manufacturing Accuracy & Environmental repeatability conditions Substrate Material type Visual assessment

Calibration & specification Sample preparation

258

Type of instrument

Software Machinery Control parameters Lab trials Recipes Dyes and pigments Digital Atlases data

Geometry Human factor Booth & Tolerances specifications Dyeing Finishing

Geometry

Accuracy & Tolerances repeatability Software & Color definition equations & color difference

Hardware Equipment Color production system Trends

Human Factor

See lab trials

Market

Standards Substrate Materials Specifications Technical See lab specifications trials

Cost

(batch & continuous) See lab trials Printing Preparation

Key variables in the control of color in the supply chain

Customer Television Resolution Non Store

Color production system Monitor Design Paper

Figure 1. Variables in the control of color in the textile supply chain

Viewing conditions

Browser Print quality

Designer Calibration & settings Color gamut

Point of sale

Q.C personnel

Observers Surround

Adaptation mechanisms

Operator & manufacturer Viewing conditions

Design

Display & Floor plan surrounds

Internet

Catalog

Original design

Market place & trends

Bulk production

Instrumental Assessment

Color management system CAD

Store lighting Handling & Rotation of storage samples

Acquired color deficiency

Fatigue Appearance phenomena

Physiological Age Gender Inherited color deficiency

Emotions

Lighting

Training Psychological factors

Past experience Culture Human factor

at the point of concept must correspond with production capabilities, as well as consumer needs and trends (Sanger, 2007). The selection of colors in a design initially involves careful planning and assessment of the components of the collection by the design team. Every piece in the collection must complement the others, and most importantly, the components need to represent the brand. Designers gain inspiration for their ideas, including color, from different sources such as forecast reports, seasonal trends, cultural background and of course, personal creativity and experience. A theme for inspiration can be an exotic place, a range of culturally based colors such as “colors of Africa,” a period of time, art, etc. The approach taken by each designer for the generation of the collection is often unique, but the main goal is often to ensure the timely generation of a new perspective and experience for the customer (Sanger, 2007). Another important factor in the selection and generation of a design is the end-use application and the overall cost of the product. In many cases, designers need to develop products that require matching colors on several substrates thus adding further variability, complexity and cost to the communication and reproduction of color (Koksal, 1992; Sanger, 2007). The use of technology for the development of patterns and garments has had a profound impact on the reproduction of color in recent years and statements such as “what you see is what you get” have often been used (Mahale and Townsend, 2007). However, communication of color ideas between designers and manufacturers can be frustrating for both parties and matching the desired color attributes is often a challenge. The representation of the “original” design color in a digital format is bounded within the limits of production color gamut for the medium of choice. Textile and fashion design software, whether off the shelf or proprietary, come with a variety of color options. Computer-aided design (CAD) tools, monitors, scanners, and desktop printers also have a range of colors each with their own device dependant color gamut,

thus limiting the range of overall production colors. In addition, despite significant research to overcome variations there are inherent differences in the technology used to reproduce a color on-screen (based on red, green, and blue emission signals) compared to that used for the generation of a physical sample (based on cyan, yellow, magenta, and black primaries). Furthermore, the on-screen color produced with CAD systems can vary from monitor to monitor and printed samples can also vary according to the type of device/printer or substrate used (e.g. paper, cotton, nylon, or polyester) (Mahale and Townsend, 2007). Recent developments in color communication include the introduction of calibration devices (Tippet, 2005) that allow the use of more accurately reproduced digital palettes potentially resulting in better communication of color between two or more parties in the supply chain. However, to date, there is no standard methodology to ensure the communication is consistent and optimized. This troublesome problem in the control of color in a textile supply chain stems from differences in perspective, understanding, and communicating the “language” of color between a designer, dyer, retailer, and consumer. Therefore, all those involved in the decision-making process within the supply chain should receive appropriate training on a continuous basis and should develop and utilize an agreed upon (internally standardized) communication protocol. One example of communication breakdown is the use of color difference terms used by the designer and dyer when a modification to a dyed sample is required. Recently, Wardman et al. (2006) reported a promising approach to define dyers’ terms of depth, brightness, and hue. This approach is currently a new work item within an International Organization for Standardization (ISO) (ISO TC 38/SCI, 2007) color measurement committee, and may lead to a useful standard communication protocol that relates descriptive color terms to measured attributes of color. 2.2 Manufacturing Reproduction of a color according to specific criteria that matches a target (standard) color is one of the most challenging aspects in the control of colored products within the textile supply chain. Despite technological advances, interpreting a color which may be based on the conceptual inspirations of a designer, who may or may not be familiar with technical limitations in the production of textiles, is not an easy task (Sanger, 2007; Strickland, 2007). The communication of digital color data between a designer and a dyer and the reproduction of a target color based on such communications is dependant on both parties using the same “color language,” as well as the same set of standard methods, including calibrated instruments for measurement and standardized viewing and lighting conditions. The optimized control of the myriad variables often requires complex control models. In addition, matching the color of two or more substrates according to the specifications of a textile designer can be difficult – or in some cases impossible, depending on the limitations of the available colorants. Often, the primary goal of the dyer or printer is to obtain a match to a specified color (the standard) that is as close and as inexpensive as possible. The match must be within a specified tolerance, be cost effective, and the finished product must meet technical requirements such as wash, rub, and light fastness (Koksal, 1992; McDonald, 1997). A further requirement that is now becoming of greater significance to many in the supply chain is a reduction in color inconstancy in which the color of an object appears to change due to a change in the lighting used (Luo et al., 2003). Of particular

Key variables in the control of color 259

IJCST 21,5

260

importance is, of course, the color inconstancy of the object under the lighting conditions intended for use and the lighting conditions under which the product is to be displayed to the consumer. Color inconstancy can be predicted using a color inconstancy index that was recently standardized by ISO 105-J05 (2007). In addition, the production of color standards in the laboratory under specified conditions is also usually significantly different from the bulk color production of substrates, even under identical temperature, liquor flow, pH, and other conditions. This is one of the main sources of variation in the reproducibility of colors in a production setting, and a satisfactory comprehensive solution to resolving these issues in a consistent manner has yet to be put forward in many cases (Koksal, 1992). In regard to color standards, further complication often arises as the dyer often selects the first production batch that is accepted by the customer as the standard used to match future dyeings for that particular style. Hence, in a supply chain several “pseudo” standards may be used for a given target color, unbeknownst to the retailer. 2.3 Color quality control Accurate color quality control is key to reducing lead times in delivering the final colored product. As previously stated, control in the textile industry can be carried out both visually and instrumentally. Visual color assessment requires the preparation and standardized presentation and viewing of physical samples, which inevitably involves subjective judgments by trained observers (American Association of Textile Chemists and Colorists (AATCC), 1986). Moreover, physical standard samples or production samples often have to be sent between supplier (manufacturer) and customer (e.g. retailer) for approval and feedback, which introduces significant time delay and increases production costs. Also, the storage and handling of standard samples over prolonged periods of time can result in a significant change of color, which in turn can lead to potentially erroneous judgments. Hence, considerable reproduction issues can result, particularly considering that many retailers today involve multiple suppliers for the production of the same components of a particular product. Therefore, digital color communication has attracted a considerable amount of attention by all parties involved in the textile supply chain, with a view to minimizing or eliminating some of these variables. It should be noted, however, that instrumental assessment of samples also requires specification of all variables including measurement conditions using calibrated instruments such as spectrophotometers. Moreover, the models used to predict color differences in use today are not optimized. Again, standardization of methodologies used by various sectors is necessary to ensure that all parties use the same “color language” when communicating within the supply chain (Connelly, 1997; Laidlaw, 1997; Butts, 2007). It is evident that in order to create an optimized digital color communication system, all variables that contribute to the change in color of substrates should be identified and minimized. 2.3.1 Lighting variability. In a retail store when consumers inspect the color of a commercial product, the spectral power distribution (SPD) of the light source where the product is located may differ significantly from the standard illuminant(s) that was used in the original color specification. Variability in lighting may mean that the color of a product in a store is perceived significantly differently from that originally intended by the designer, despite a high level of color control during the design and production of the product. Also, the background viewing environment may have

a significant effect on the perceived color. Possible ways in which variability in lighting may occur include: . incorrect lamp installed; . mixing different lamps together in the same region of space (e.g. fluorescent and incandescent); . variability in lamp emission; . light pollution, e.g. from entrance/exit areas, as well as exterior windows; and . use of strongly colored surfaces surrounding the product. No standardized method exists for the spectroradiometric measurement of lighting in a retail store such that the spectral data can be used as custom illuminant data for the calculation of color differences of products displayed in the store. The SPD of key areas in more than 12 stores that sell the products of leading US retail companies were measured in a previous study to ascertain the level of variability within typical stores in the USA, to obtain inter-store variability, and also to compare the area lighting to standard illuminant data used in colorimetric calculations (Hinks et al., 2000, 2001; Noor, 2003). In addition, the lighting variability in standard viewing booths currently in use in various sectors of the textile supply chain, including design, color standard development, dyeing and finishing quality control, and retail color management has also been measured to evaluate the potential need and feasibility of the following developments (Hinks et al., 2000): . new standard illuminants that more accurately reflect real lighting conditions; . new standard light sources used in standard viewing booths that more accurately reflect real lighting conditions; and . a set of recommendations to retailers on improving lighting to optimize the color appearance of textile materials. A standard method of area lighting measurement has been proposed (Hinks et al., 2001). The control of lighting in retail stores must be an integral part of the color quality control within the textile supply chain. In fact, without an integrated and well-controlled color management process, the variability in communication and production of color would likely be high, leading to long, lead times, and higher than necessary production costs. 2.4 Point-of-sale Figure 1 shows some of the variables that influence color control at the point-of-sale. The colors of commercial products are never seen in isolation by the consumer. The viewing conditions of colored samples have a tremendous impact on the perception of colored products and phenomena such as simultaneous contrast affect each individual’s color experience. The effect of simultaneous contrast on perceived color of textiles is well documented (Fairchild, 2005; Kuehni, 2005). The lighting used to display products in a retail store also critically affects the perceived color of multi-component products, which typically involve varying levels of color inconstancy and metamerism. Importantly, as already pointed out research indicates that many US retail chain stores employ varied and uncontrolled lighting conditions for the display of

Key variables in the control of color 261

IJCST 21,5

262

their merchandise. Also, store designers may change the lighting design of the floor due to seasonal or other design/managerial specifications (Noor, 2003). Such variations, which may result in a significant change of perceived product color compared to the original design, are likely to be seen in many retail stores. However, several large retail stores have been found to control their lighting precisely, both within different areas of a store and between stores in a retail chain. By way of example of poor lighting control, the variability in SPD of light sources used and measured at various locations of store “A” that contained merchandize of the retailer along with many other brands is shown in Figure 2. From the SPD data for this store, it is seen that two types of lighting are used, namely fluorescent sources in combination with incandescent lamps. Figure 3 shows the variability in illuminance (lx) at the measured locations in the store. Clearly, since the brightness in the store can vary between approximately 250 lx and over 1,000 lx, considerable variability exists that could impact significantly the viewing experience of a consumer. For instance, the details of very dark samples viewed under low-illuminance conditions (e.g. 300 lx) will not be easily discerned. In addition, an analysis of color inconstancy for a pair of blue metameric textile samples in different parts of the store based on the CIE (2004) color inconstancy index, D65 as the reference illuminant, and the SPD of the lighting in a particular location of the store as the test illuminant is shown in Figure 4. The blue standard used was considerably color inconstant under standard illuminants, and although Figure 4 shows a high-color inconstancy index value for all locations in the store, the value is relatively consistent, ranging from 3.2 to 5.7. It would be valuable to further determine the significance of the color inconstancy index values 2,000 1,800 1,600 1,400

SPD

1,200 1,000 800 600 400 200 0 36 0 38 0 40 0 42 0 44 0 46 0 48 0 50 0 52 0 54 0 56 0 58 0 60 0 62 0 64 0 66 0 68 0 70 0 72 0 74 0

Figure 2. Normalized SPD data for measurements taken at store A

Wavelength

Key variables in the control of color

Illumination levels Indoor mall lighting

Changing rooms

529

513

532 925

386

490

263

662 696

306

1,010

460

Check out counter 312 428

249

678

367

387

827 798 817

Figure 3. Illuminance values for the various locations of store A

364

675 Indoor store lighting

Color inconstancy Indoor mall lighting

Changing rooms

3.4

4.0

3.6 4.6

4.4

3.4

3.2

3.4

4.1

4.1

3.4

Check out counter 4.2 4.2

4.2

4.0

4.0

4.2 4.2 3.3

4.1

3.4

3.5

Indoor store lighting

on variability in visual color difference assessment; and establish a color inconstancy acceptability tolerance for a typical commercial product. In addition to the above factors, increasing use of non-store shopping including catalogue as well as the internet and television shopping has introduced different color control and communication issues. Among some of the important issues include accurate color reproduction via paper printing, resolution and size of color images,

Figure 4. Color inconstancy index for standard blue metamer for store A

IJCST 21,5

264

monitor type, monitor color gamut and calibration, potential to control background viewing conditions, and metamerism (Kuehni, 2005; Butts, 2007). While the variables in the control of color for these media may be entirely different compared to those in a store, they have to be identified and controlled to ensure that the consumer experiences closely what the product designer intended. 2.5 Human factor The human factor is arguably the most fundamental aspect in the fishbone diagram shown in Figure 1. Within the textile supply chain, there are many stages where decisions and processes rely on individual subjective skills and performances. Even with standardized procedures, humans often perform and make subjective assessments differently. Such high levels of variability lead to poor or inconsistent decisions and interpretations within the supply chain. For instance, although the light reflecting properties of objects are physically quantifiable, color is a human experience that can never be defined in absolute terms (Berns, 2000; Kuehni, 2005). Moreover, practically all factors shown in Figure 1 are affected, to certain extent, by the human factor, which in turn leads to variations in the assessment of colored materials. Some of the factors that influence color decisions made by observers include psychological and environmental factors, age, gender, past experience in handling color-related decisions, as well as viewing conditions. A good example of the impact of the human factor in color control and communication is the visual assessment of color. Evidently, all individuals responsible for color assessment must first be tested for normal color vision (Ishihara, 1917; Farnsworth, 1943; Neitz, 2001). Yet, not all color-related companies test employees for color vision, and there have been cases of color assessors in industry that had defective color vision. The evaluation of color is also subject to individual’s perception of color as well as various cognitive aspects. Perceptual aspects of the evaluation of color include the immediate recording in the brain of the stimuli in terms of lightness, hue, and chroma, while the cognitive aspects include later processing of the information which include color memory, color meaning, etc. (Gao and Xin, 2006). In that sense, a color normal observer’s experience is a response of the sensory system that is impossible to define. Moreover, certain aspects of color psychology, such as color preference, can influence the assessment of color and bias individual as well as group differences. Personality, gender, race, and age have been widely investigated as a group response in relation to the evaluation of color. Whitfield and Whiltshire (1990) evaluated studies of color preference as a function of individual and group differences and state that color preference and aesthetic values are strongly influenced by cultural differences. Diagnostic tests such as Lu¨sher and Rorschach tests have also been developed to assess the relationship between color responses and personality variables. In terms of age, studies carried out by Gale (1933), Granger (1955) and Staples and Walton (1933) suggested that young children have different color preferences than adults, but the influence of age on color is somewhat ambiguous (Whitfield and Whiltshire, 1990). Moreover, the optimization of digital color communication involves high levels of technical capability and expert technical knowledge at each stage of the supply chain. At the heart of color control are mathematical models that correlate visual assessment of color to measured values. In order to achieve efficacy in digital communication,

the scope and limitations of these models, as well as the myriad variables that influence this relationship, should be elucidated. 2.5.1 Issues surrounding visual assessments. Arguably, the most important factor within the textile supply chain is the accurate assessment of color differences between two textile materials. Optimization of the correlation between the visual assessment of color and mathematical models that predict color differences is fundamental to any digital system of color management. A number of visual methods of assessing color differences between two textile samples have been developed (CIE, 2001; Luo et al., 2001; Aspland and Shanbhag, 2004; Gay and Hirschler, 2003). In an ongoing study at North Carolina State University, the sources and extent of variability in visual assessment of color differences of textile samples is being studied (Hinks et al., 2006). The goal is to optimize the experimental methodology and establish the minimum repeatable variability possible among a statistically significant set of visual observers under highly controlled conditions of observation. In this study, color difference assessments are being carried out using a Macbeth Spectralight III viewing booth equipped with a filtered tungsten artificial daylight simulator. About 50 color normal observers were used in the first stage of the study and a total of 3,100 assessments were obtained using the AATCC (1986) gray scale for change in shade based on 31 textile samples. Sample pairs were produced on 100 percent polyester knitted fabric and contained small color differences in lightness, chroma, and hue. The gray scale consists of nine pairs ranging from 1 to 5 in half steps. A ranking of five represents no perceived difference between the trial and the standard (AATCC, 1986). Of the observers selected for the study 25 were naı¨ve observers tested for normal color vision using the Neitz (2001) test that had no prior knowledge of commercial pass/fail color difference assessments, while 25 were expert observers whose employment involved, or has involved, commercial shade matching in the textile industry. Naı¨ve observers were mostly students of North Carolina State University and included 11 females and 14 males ranging from 18 to 25 years of age. The 25 expert observers who were mostly industrial colorists from the US textile industry included 10 females and 15 males and ranged from 25 to 70 years of age. The observers were adapted to the illumination conditions by observing the illuminated viewing booth for two minutes and were then presented with textile samples with small color differences compared to a standard and asked to determine the perceived color difference of samples using the gray scale. Naı¨ve observers repeated the assessment three times. However, due to constraints in availability and geographic location of the test expert observers assessed samples one time. Figure 5 shows the conditions employed during visual assessments. The average grey scale rating for each pair was compared for naı¨ve and expert assessors. Tables I and II show the preliminary analysis of results using t-tests (Figure 6). Results of a paired t-test between the second and the third trials conducted by naı¨ve observers indicate that the repetition of the visual observations significantly affects the assessment of small color differences. In addition, the comparison of average naı¨ve vs experts’ assessments shows a statistical difference at the 95 percent confidence interval, with expert observers generally perceiving a larger color difference than naı¨ve observers. Since experts did not repeat the assessment the inter-observer variability between naı¨ve and experts cannot be established at this point and requires further study (Cardenas et al., 2006).

Key variables in the control of color 265

IJCST 21,5

266

Figure 5. Visual assessment experimental setup

Group Table I. Summary statistics for repeat assessments carried out by naı¨ve observers

Table II. Summary statistics for assessments carried out by naı¨ve and experts observers

t

p

Trial 1 vs trial 2

1.950

0.060

Trial 2 vs trial 3

23.350

0.002

Trial 1 vs trial 3

0.010

0.920

Group

t

p

Trial 1 vs experts

10.88

, 0.0001

Trial 2 vs experts

9.14

, 0.0001

Trial 3 vs experts

11.46

, 0.0001

Significance No significant difference at 95 percent confidence interval Significant difference at 95 percent confidence interval No significant difference at 95 percent confidence interval

Significance Significant difference at 95 percent confidence interval Significant difference at 95 percent confidence interval Significant difference at 95 percent confidence interval

3. Conclusions It is important to identify the important parameters and variables that influence the control of color within various stages of the textile supply chain. For effective color control, the goal must be to build confidence in the data that quantifies the sources of

Key variables in the control of color

5

Gray scale grade

4

267 3

2

Experts Average naive

1 1

3

5

7

9

11

13

15

17

19

21

23

25

27

29

31

# Pair

variation within the process and to minimize variations. Initially, variations in visual assessment of small color differences in textile materials as well as those due to lighting employed during design, production and display of textiles should be examined to establish the optimum conditions and the level of observer variability. A fishbone diagram was used to demonstrate the wide range of variables that affect the control and communication of color within the textile supply chain. The optimization of variables that influence the assessment and communication of color is vital to achieving effective communication between all parties involved. This can significantly reduce costs and lead times resulting in improved competitiveness and cost efficiency associated with increased consumer satisfaction and confidence in the industry. Assessment of lighting measurements of several stores in the USA demonstrated variability in lighting, with many stores having at least two different light sources. This variability, in combination with uncontrolled lighting from external windows and entrance/exit areas, can lead to significant variability in the color perception of textile garments displayed in such areas, and may lead to consumer experience being significantly different from that intended by the designer. In regard to visual assessment variability, the results obtained in an ongoing study at North Carolina State University based on the psychophysical testing of 50 observers demonstrate a statistical difference for visual judgments of small color differences between naı¨ve and expert observers (Cardenas et al., 2006). Results of a paired t-test between the second and the third trial conducted by naı¨ve observers indicate that the repetition of the visual observations significantly affects the assessment of small color differences.

Figure 6. Average results in grade units for the visual assessments

IJCST 21,5

268

References AATCC (1986), “Evaluation procedure 9. Visual assessment of the change in color”, AATCC Technical Manual, Vol. 16, American Association of Textile Chemists and Colorists, Research Triangle, NC. Aspland, J.R. and Shanbhag, P. (2004), “Comparison of color difference equations for textiles: CMC(2:1) and CIEDE2000”, AATCC Review, Vol. 4 No. 6, pp. 26-30. Berns, R.S. (2000), Billmeyer and Saltzman’s Principles of Color Technology, 3rd ed., Wiley, New York, NY. Butts, K. (2007), “Why don’t my numbers match yours?”, paper presented at the Color Management Workshop, Raleigh, NC. Cardenas, L., Hinks, D., Shamey, R., Kuehni, R., Jasper, W. and Gunay, M. (2006), “Comparison of naı¨ve and expert observers in the assessment of small color differences between textile samples”, paper presented at the CGIV Conference, Leeds. CIE (2001), CIE Technical Report: Improvement to Industrial Colour-difference Evaluation, CIE, Vienna. CIE (2004), Colorimetry, 3rd ed., CIE Publication 15:2004, CIE, Vienna. Connelly, R. (1997), “Good sample presentation: how to get color measurement results that make sense”, Color Technology in the Textile Industry, 2nd ed., American Association of Textile Chemists and Colorists, Research Triangle, NC. Eckes, G. (2003), Six Sigma for Everyone, Wiley, Hoboken, NJ. Fairchild, M.D. (2005), Color Appearance Models, 2nd ed., Wiley, Hoboken, NJ. Farnsworth, D. (1943), “The Farnsworth-Munsell 100 hue dichotomous tests for color vision”, Journal of the Optical Society of America, Vol. 33, pp. 568-74. Gale, A.V. (1933), Children’s Preferences for Colors, Color Combinations, and Color Arrangements, University of Chicago Press, Chicago, IL. Gao, X.P. and Xin, J.H. (2006), “Investigation of human’s emotional responses on colors”, Color Research and Application, Vol. 31 No. 5, pp. 411-7. Gay, J. and Hirschler, R. (2003), “Field trials for CIEDE2000 – correlation of visual and instrumental pass/fail decisions in industry”, Proceedings of the 25th Session of the CIE, San Diego, CA, p. 26. Granger, G.W. (1955), “An experimental study of color preferences”, Journal of General Psychology, Vol. 52, pp. 3-20. Graystone, J. (2000), “Integrating colour delivery skills”, paper presented at PRA Conference on The Colour Delivery Challenge, Leeds University, Leeds. Hinks, D., Draper, S., Che, Q., Nakpathom, M., El-Shafei, A. and Conelly, R. (2000), “Effect of lighting variability on color difference”, AATCC Review, Vol. 1 No. 11, pp. 16-20. Hinks, D., El-Shafei, A., Draper, S., Che, Q., Nakpathom, M. and Conelly, R. (2001), “Radiometric measurement of area lighting critical to color assessment in the textile industry”, AATCC Review, Vol. 11 No. 1, pp. 35-9. Hinks, D., Shamey, R., Aspland, J.R., Cassill, N., Jasper, W. and Kuehni, R.G. (2006), “Optimizing color control throughout the textile supply chain NTC”, available at: www.ntcresearch. org/projectapp/index.cfm?project¼C04-NS11 (accessed June 2007). Ishihara, S. (1917), Series of Plates Designed as Tests for Colour-blindness, Handaya Company, Tokyo.

ISO 105-J05 (2007), “Textiles-tests for colour fastness – Part J05: method for the instrumental assessment of the colour inconstancy of a specimen with a change in illuminant”, CMCCON02, pp. 1-5. ISO TC 38/SCI (2007) Report of Working Group 7, Color Measurement, Las Vegas, NA. Koksal, G. (1992), “Robust design of batch dyeing process”, PhD thesis, North Carolina State University, Raleigh, NC. Koksal, G., Smith, W. and Smith, B. (1992), “A system analysis of textile operations”, Textile Chemist & Colorist, Vol. 24 No. 10, pp. 30-5. Kuehni, R.G. (2005), Color: An Introduction to Practice and Principles, 2nd ed., Wiley, Hoboken, NJ. Laidlaw, A.C. (1997), “Care and feeding of color measuring instrumentation: how to implement a system for maintaining its integrity”, Color Technology in the Textile Industry, 2nd ed., American Association of Textile Chemists and Colorists, Research Triangle, NC. Luo, M.R., Cui, G. and Rigg, B. (2001), “The development of the CIE 2000 colour-difference formula: CIEDE2000”, Color Research & Application, Vol. 26 No. 5, pp. 340-50. Luo, M.R., Li, C.J., Hunt, R.W.G., Rigg, B. and Smith, K.J. (2003), “CMC 2002 colour inconstancy index: CMCCON02”, Coloration Technology, Vol. 119 No. 5, pp. 280-5. McDonald, R. (Ed.) (1997), Colour Physics for Industry, 2nd ed., Society of Dyers and Colourists, Bradford. Mahale, G. and Townsend, K. (2007), “Digitally printed textiles & image quality”, The Indian Textile Journal, Vol. 117 No. 4, pp. 17-24. Neitz, J. (2001), The Neitz Test of Color Vision, Western Psychological Services, Los Angeles, CA. Noor, K. (2003), “Effect of lighting variability on the color difference assessment”, MS thesis, North Carolina State University, Raleigh, NC. Parrot, K. (2001), “Instrumental colour quality control: getting the best from your system”, in Gilchrist, A. and Nobbs, J.H. (Eds), Colour Science Volume 3: Colour Physics, Department of Colour Chemistry, University of Leeds, Leeds. Sanger, A. (2007), “Creativity: getting color right”, paper presented at the Color Management Workshop, Raleigh, NC. Staples, R. and Walton, W.E. (1933), “A study of pleasurable experience as a factor in color preference”, Journal of Genetic Psychology, Vol. 43, pp. 217-23. Strickland, M. (2007), “The advantages and limitations of engineered color standards”, paper presented at the Color Management Workshop, Raleigh, NC. Tippet, B. (2005), “The color challenge”, Canadian Apparel, Vol. 29 No. 2, pp. 10-12. Wardman, R.H., Islam, S. and Smith, K.J. (2006), “Proposal for a numerical definition of standards depth”, Coloration Technology, Vol. 122 No. 6, pp. 350-5. Whitfield, T.W. and Whiltshire, T.J. (1990), “Color psychology: a critical review”, Genetic, Social, and General Psychology Monographs, Vol. 116 No. 4, pp. 384-411. Corresponding author Renzo Shamey can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

Key variables in the control of color 269

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

IJCST 21,5

Weaving thickness parameters of “8” shape 3D woven enhancing fabric

270

Liu Jihong

Received 12 November 2008 Revised 18 February 2009 Accepted 18 February 2009

Creative Digital Media Research Center, Jiangnan University, Jiangsu, China

Jiang Hongxia Faculty of Textile Science and Technology, Shinshu University, Ueda, Japan, and

Lu Yuzheng Creative Digital Media Research Center, Jiangnan University, Jiangsu, China Abstract Purpose – The purpose of this paper is to deduce the thickness property of three-dimensional (3D) composite produced by 3D woven enhancing fabric based on an academic model. Design/methodology/approach – Thickness of 3D composite is determined by the important weaving parameter – the length of binder yarn. According to the shape of pile formed by binder yarn, curve function of pile is supposed. A rapier loom is modified for the 3D woven enhancing fabrics, and the composite is produced based on the fabric. The thickness of composite is produced and the theories results are validated. Findings – The result of the analysis shows that the curve of pile formed by binder yarn can be expressed as sin function approximately, and there is linear relation between the thickness of composite and the length of pile of binder yarn. Research limitations/implications – The results cannot be provided to study the relationship of thickness based on different technology of composite. Originality/value – The paper provides an academic method of calculating the thickness of composite and the relationship between the thickness of composite and the length of binder yarn. The method can reduce the testing time. Keywords Fabric production processes, Textile products, Yarn Paper type Research paper

International Journal of Clothing Science and Technology Vol. 21 No. 5, 2009 pp. 270-278 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910983768

1. Introduction Composites, using three-dimensional (3D) woven enhancing fabrics, are created and studied recently. In these composites, “8” shape 3D woven enhancing fabric is used extensively in composites recently because of its especial structure and low cost. Luo et al. (2007) focus transverse impact behavior of 3D woven composite, the energy absorption and damage modes at different impact velocities. Sun et al. (2005) conducted the compressive properties of 3D angle-interlock woven composites at quasi-static and high-strain rates, as the result, they obtained the rate dependent properties. Tagarielli et al. (2004) have investigated the effect of simply support and clamped boundary

conditions on the load versus deflection response of beam loaded in three-point bend, obtained the peak load corresponding to the initial collapse mechanism. Vuure et al. (2000) studied the properties of composite panels, pile, pillar shape and resin distribution model of woven sandwich-basic based on finite element analysis, and a special adhesive foil stretching process is proposed to control accurately the thickness of the panel. However, there is no reference focused on how to restructure loom to produce “8” shape 3D enhancing fabric and how to control the thickness of the composite so far. The restructuring method and the relationship between the thickness and the weaving parameters will lead to quickly designing of the enhancing fabric for composite and shortening the production cycle. In this paper, geometry model of “8” shape 3D woven enhancing fabric is supposed, and then the relationship between thickness of composite and pile length is analyzed. At last, restructuring method was introduced and validation samples of fabric and composite were produced on modified rapier loom using different parameters.

Weaving thickness parameters 271

2. Theories/hypotheses 2.1 Textile geometry pattern Figure 1 shows a prototype of “8” shape 3D woven enhancing fabric. The figure expresses how the yarns interweave to the fabric. The woven fabric is composed of two skins and a core. Warp yarns as x-direction and weft yarns as y-direction interweave two skins, respectively. The binder yarns as z-direction in the core connect the skins of fabric to each other. The warp and weft yarns give the basic mechanical properties; the binder yarns constitute 3D structure and stabilize the structure of composite. Simultaneously, the binder yarns make the thickness of the panels. Figure 2 is referred to the cross-section diagram of weave construction of 3D woven fabric. Each circle indicates one weft yarn. The curves along the vertical diagram (and the page) should be visualized as the binder direction of the fabric, and the horizontal direction curves represent the warp yarn. A group of binder yarns repeat to pass over one weft and under the adjacent weft of top weft one time or more: first, then begin to interlace with bottom weft using the same pattern. After that the binder yarns interlace with bottom weft firstly, then interlace with top weft. One group of binder yarns in warp direction, connecting two skins, forms an S-shape loop or V-shape. While two binder yarns combine to form an “8” shape loop. According to interval in warp of

Skin Core Skin z x y

Figure 1. Prototype of “8” shape 3D woven enhancing fabric

IJCST 21,5

272 z

Figure 2. Cross-section diagram of weave construction x

inter-point, “8” shape 3D woven enhancing fabric is divided into two kinds: regular and oblique “8” shape loops. Regular “8” shape loop means the inter-points of upper and lower skins are aligned in the adjacent rows of weft yarn. While oblique “8” shape loop means that there is one or more weft-distance between the inter-points of upper and lower skins. Suppose the distance is c. 2.2 Calculation of fabric thickness From Figure 2, the length of binder yarn is only one S-shape longer than the other of warp yarn on the condition of omitting the diameter of yarn. Named this longer length as pile length and expressed by s. Pile length is equal to the length of one S-shape. Pile length can be determined by weaving parameters, the off-letting length of binder yarn. Assume that S-shape of the piles in the warp direction can be modeled a sine function. Therefore, one “8” shape is made from two sine function curves. At first considering of oblique “8” shape, suppose w as the axes coordinate of “8” shape, the shape functions for the piles in height direction u can be given by: pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ 2p h2 þ c2 u¼A sin pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ w 2p h2 þ c2

ð1Þ

where A is the amplitude of sine function, h is the thickness of composite panel, c is the distance interwoven points of upper and low layer. Then considering of regular “8” shape, the function can be simplified. The piles in warp direction x can be given by: h 2p x¼A z sin h 2p

ð2Þ

where z is the height coordinate. The pile length of oblique shape can be calculated by equation (1): ﬃ ﬃ sﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ 2 Z pﬃﬃﬃﬃﬃﬃﬃﬃﬃ Z pﬃﬃﬃﬃﬃﬃﬃﬃﬃ h 2 þc 2 qﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ h 2 þc 2 2p 2 0 s¼ 1 þ ðu Þ dw ¼ 1 þ A cos pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ w dw h2 þ c2 0 0

ð3Þ

According to equation (3), result of s and h is listed in Table I and plot in Figure 3.

h

pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ h2 þ c2

s

2 3 4 5 6 7 8 9 10 11 12

2.83 3.61 4.47 5.39 6.32 7.28 8.25 9.22 10.2 11.2 12.2

3.44 4.38 5.44 6.55 7.69 8.85 10.0 11.2 12.4 13.6 14.8

Weaving thickness parameters 273

Table I. Relationship of s and h

15 S Linear fit

S

10

5

Figure 3. Relationship of s and h of oblique shape

0 0

2

4

6

8

10

12

sqrt (h 2+c 2) (mm)

From Figure 3, the linear fit function will be: pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ s ¼ 1:213 h 2 þ c 2

ð4Þ

The correlation is 0.996 and available from 2 to 12 millimeters. The pile length of regular shape can be calculated by equation (2): Z h qﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ Z h sﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ 2p s¼ z dz ð5Þ 1 þ A 2 cos2 1 þ ðx0 Þ2 ¼ h 0 0 According to equation (5), result of s and h is listed in Table II and plot in Figure 4. From Figure 4, the linear fit function will be: s ¼ 1:218h

ð6Þ

The correlation is 0.999 and available from 2 to 12 millimeters. The amplitude A in equations (1) and (2) is determined by the process parameter for composite. Under the same process parameter, A can be considered as a constant. Therefore, the thickness of

IJCST 21,5

274

Table II. Relationship of s and h

h

s

2 3 4 5 6 7 8 9 10 11 12

2.43 3.65 4.86 6.08 7.30 8.51 9.73 10.9 12.2 13.4 14.6

15 S Linear fit

S

10

5

Figure 4. Relationship of s and h of regular shape

0

0

2

4

6 h (mm)

8

10

12

composite is determined by the pile length of binder yarn that is over feed by off-letting of binder warp beam. Conversely, pile length of binder yarn can be deduced by producing few composite samples. 3. Experimental 3.1 Restructuring loom and fabric manufacturing All samples of woven fabric were produced successfully in modified rapier loom. Schematic diagram of the restructured loom is shown in Figure 5, where the principal parts for five basic motions are shown while reconstructed parts of the letting off and taking up part of the loom are demonstrated for this study. Binder warps 1 from the binder warp beam 2 go round the tension regulator roller 3 and oriented roller 4. Then, the binder warps and warp yarns 5 from the weaver’s beam 6 pass round the back rest 7 and go through the drop wires 8 of the warp stop-motion to the healds 9, which are intended for separating the warp yarns for the purpose of shed formation. They then pass through the reed 10 that holds the thread at uniform spacing and is designed for beating-up the weft yarn that is inserted into the triangular warp shed. Continuing uniform 3D fabric passes over the breast beam 11, and then exports from output rollers 12 and 13. After that the fabric moves under length control roller 14, round the take-up

1 3 2

11 14 16 9 8

275

7

4

Weaving thickness parameters

5

10

12 13

6

15

Figure 5. Schematic diagram of the restructured loom

roller 15 and onto the cloth roller 16. One binder warp beam or more for binder warps can be attached to the rear. Electric control system, as shown in Figure 6, is composed of two parts: master part and slaver part of warp roll. Master part was build by producer of the loom. The slaver part of binder warp beam was equipped to provide binder yarns. A group of proximity sensors is fixed around the main beam that monitors the rotation status and angle of the main bean. A human-machine interface (HMI) is used to design weaving parameters of binder yarns. According to signals of the proximity sensors and the parameters from HMI, the programmable logic control, controls rotate speed of binder warp beam. Another proximity sensor, installed under length control roller, monitors the length of fabric and drive the take-up roller. The intrinsic warp beam holds the lengthwise yarns; simultaneously the binder warp beam releases the longer lengthwise yarns to the weaving area for binder warp according to the weaving parameter. The parameter is very important for the thickness of the composite. It is very important to control two parts working synchronization. The reconstruction method can be used comprehensively because of its lower cost. 3.2 Composite specimen The “8” shape 3D woven enhancing fabrics were manufactured with glass filament tows. The prototype of the 3D fabric has been shown in Figure 1. The epoxy resin was injected into the fabric preformed by resin transfer molding (RTM), and then consolidated at room temperature.

Proximity sensor around main beam

HMI Warp beam PLC Binder warp beam

Proximity sensor under length control roller

Figure 6. System architecture of electric control system

IJCST 21,5

276

4. Results The ten designed samples of regular “8” shape were woven successfully according to the designed parameters by controlling the letting off length of binder yarn. Twist-less E-glass filament tows were used as yarns (three-direction). Specifications of the 3D fabric are listed in Table III. Figure 7 shows the surface of a fabric sample. The size of composite specimen is 2-12 (thickness) £ 100 £ 100 millimeters in the z-x-y and z-y-x directions. The specimens were cut along the x- and z-directions as shown in Figure 8(a)-(c), respectively. The ten designed samples were woven successfully according to the designed parameters by controlling the letting-off length of binder yarn. From Figure 8, it is found that the upper portion “8” is larger than hypothetic shape. In reverse, the under portion of “8” is smaller than hypothetic shape. This unsymmetrical shape is formed by gravity of epoxy resin when the composite was processed from fabric. The thickness of composite from 2-12 millimeters was produced exactly according to the theories. 5. Conclusions A mechanical weaving system, producing “8” shape 3D woven enhancing fabric, has been developed in this research. In order to clarify the thickness characteristic of fabric, the effects of different weaving parameters have been analyzed theoretically and experimentally. It was found that any change in binder yarn length affects the thickness of a panel. In addition, there is a linear relationship, as equations (4) and (6), between the thickness of woven fabric and a binder yarn length on conditions of the same RTM parameters. Some samples of 3D fabric were produced on the modified

Table III. Specifications of “8” shape 3D woven enhancing fabric

Figure 7. Surface of a fabric sample

Parameters Type Warp density (ends/10 centimeters) Weft density (ends/10 centimeters)

Basic yarn material E-glass 100 100

Weaving thickness parameters 277

(a)

(b)

(c)

rapier loom. The thickness property is proved by the samples. Therefore, the thickness of panels produced from newly conceived fabrics could be estimated before actual production or testing. References Luo, Y., Lv, L., Sun, B., Qiu, Y. and Gu, B. (2007), “Transverse impact behavior and energy absorption of three-dimensional orthogonal hybrid woven composites”, Composite Structures, Vol. 81 No. 2, pp. 202-9. Sun, B., Gu, B. and Ding, X. (2005), “Compressive behavior of 3D angle interlock woven fabric composites at various strain rates”, Polymer Testing, Vol. 24 No. 4, pp. 447-54. Tagarielli, V.L., Fleck, N.A. and Deshpande, V.S. (2004), “Collapse of clamped and simply supported composite sandwich beams in three-point bending”, Composites Part B: Engineering, Vol. 35 Nos 6/8, pp. 523-34.

Figure 8. Specimens of composite

IJCST 21,5

278

Vuure, A.W., Pflug, J., Ivens, J.A. and Verpoest, I. (2000), “Modelling the core properties of composite panels based on woven sandwich-fabric performs”, Composites Science and Technology, Vol. 60 No. 8, pp. 1263-76. Further reading Guilleminot, J., Comas-Cardona, S., Kondo, D., Binetruy, C. and Krawczak, P. (2008), “Multiscale modelling of the composite reinforced foam core of a 3D sandwich structure”, Composites Science and Technology, Vol. 68 Nos 7/8, pp. 1777-86. Miravete, A., Bielsa, J.M., Chiminelli, A., Cuartero, J., Serrano, S., Tolosana, N. and Guzman de Villoria, R. (2006), “3D mesomechanical analysis of three-axial braided composite materials”, Composites Science and Technology, Vol. 66 No. 15, pp. 2954-64. Corresponding author Liu Jihong can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

Dynamic analysis on the thermal and electrical properties of fabrics in the process of moisture absorption and liberation Weigang Cui, Xin Wang, Wenbin Li and Weilin Xu Textile Research Center, Wuhan University of Science and Engineering, Wuhan, People’s Republic of China

Thermal and electrical properties 279 Received 9 November 2008 Revised 18 March 2009 Accepted 18 March 2009

Abstract Purpose – The purpose of this paper is to present a dynamic analysis on the thermal and electrical properties of fabrics under wet conditions. Design/methodology/approach – A purpose-built apparatus is applied to test the thermal and electrical properties of textiles in moisture absorption and liberation process. Relation between temperature and resistance of a cotton/polyester double-layer fabric is also analysed. Findings – The surface temperature of textiles shows three different stages in the process. The electrical resistance is linearly related to the reciprocal of the moisture regain of fabrics. In the moisture absorption and liberation process, surface temperature of cotton layer is higher than that of polyester layer. And the electrical resistance of cotton layer decreases more quickly than that of polyester layer. The electrical resistance changes earlier than surface temperature in the moisture-liberation process. Practical implications – The paper is helpful in not only the designing of sportswear, but also the devising of moisture-testing apparatus. Originality/value – A dynamic testing method is applied to characterize the thermal and electrical properties of textiles. Keywords Thermal properties of materials, Electrical properties, Fabric production processes, Moisture measurement Paper type Research paper

1. Introduction The interaction of moisture and textiles has many important technical consequences for which affects the thermo-physiological wear comfort. In order to keep wearer dry and hence comfortable, clothing has to be able to deal with the perspiration during many activities such as sports, working and moving, etc. Thus, the clothing must be able to absorb perspiration from the skin surface and then the moisture in the clothing layer next to the skin must dry out quickly. Many researches studied the relations between moisture and textiles (Adler and Walsh, 1984; Awano et al., 1987; Crow and Osczevski, 1998; Li, 2001; Li et al., 1995; Rousselle et al., 2005) and some standards and test methods can be employed to test a fabric’s simple absorbing and wicking properties (AATCC, 2000; BSI, 1996). Clothing in service deals with perspiration and keeps body warm, this a complicated process which affects the heat and moisture balance of apparel. The electrical properties will be greatly affected during different moisture and thermal conditions. Relation between electrical resistance and moisture of fabrics is meaningful in exploring moisture-testing apparatus. A sweating, guarded

International Journal of Clothing Science and Technology Vol. 21 No. 5, 2009 pp. 279-285 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910983777

IJCST 21,5

280

hot plate was used (Fan et al., 2003) to investigate coupled heat and moisture transfer through fibrous insulation sandwiched by two covering fabrics under low-temperature condition. Diffusion of moisture and heat in wool fabrics were tested (Li and Holcombe, 1992; Li et al., 1995) under wet conditions, and results shown the surface temperature of wool fabric varied differently with that of polyester. Electric properties were also tested under wet conditions, previously, we designed a kind of new moisture management tester (Hu et al., 2005; Li et al., 2000) based on the differences of electric resistance of fabrics with different moisture contents, this is helpful to develop quick-drying clothing. Dynamic heat-moisture transfer properties of summer fabrics were studied (Jiang and Yan, 2004) using microclimatic method, however, surface electric resistance was not tested during the same process. During the wearing of fabric, especially sportswear, sweet was absorbed and transferred to the outside layer of the fabric, and then it evaporated to the environment. It is meaningful to perform dynamic analysis of the thermal and electrical properties of fabric during the moisture absorption and liberation process. In this study, we investigated the surface temperature and electrical resistance of fabric in the process of moisture absorption and liberation, and the relation between thermal and electrical properties of a cotton/polyester double-layer fabric were analyzed on a purpose-built apparatus. This is helpful in the designing of sportswear and moisture-testing apparatus. 2. Experimental 2.1 Material Different kinds of fabrics were used in this study with details listed in Table I. 2.2 Testing methods Figure 1 shows the apparatus used in the testing. In the process of moisture absorption testing, fabrics were placed in an oven for sufficient time to attain their dry weights. While in the moisture-liberation testing process, fabrics were immersed in distilled water for 30 min to be fully wetted, after the visible water in the surface, the fabrics were placed on a bracket fixed on a electronic balance. The data of the weight of fabrics were transmitted to the interface RS232 of a computer though the data transmissions interface of the balance, the computer got the data of weight every 2 s. Data of moisture regains of all the fabrics were calculated after the testing. A temperature sensor (thermal resistor, 0.6 mm germanium) was setup above and under the fabric with a distance of 5 mm; the data of temperature were obtained after AD conversion (analog-digital conversion) and transferred to the same computer. The computer also recorded the data of temperature every 2 s. In the testing process, data of weight and temperature were recorded until the weight of fabrics did not change at the end of the test. During the process, the resistances of the fabric were tested using with Count, quantity/cm Warp Weft

Structure Table I. Characteristics of experimental fabrics

Knitted cotton Knitted polyester Cotton polyester

Plain stitch Plain stitch Interlock

210 152 118

190 138 84

Dry weight (g/m2)

Thickness (mm)

211.77 141.07 188.74

0.85 0.54 0.95

Thermal and electrical properties 281

Figure 1. Scheme of testing apparatus

varied testing distances according to Li and Holcombe (1992). The apparatus were placed in an air-conditioned room with temperature and humidity conditions of 258C and 65 percent relative humidity (RH), and another temperature sensor was fixed to test the environmental temperature to make sure the temperature was 25 ^ 0.58C. 3. Results and discussion 3.1 Thermal properties of fabrics in the process of moisture absorption and liberation Figure 2 shows the surface temperature of knitted cotton and polyester fabrics in the process of moisture absorption and liberation. It is evident that the variation of surface temperature of fabrics in the process of moisture absorption and liberation could be divided into three stages as follows. Stage one. From the beginning of the curved to t1. At the beginning point, fabric was dry and its surface temperature kept the same with surrounding temperature. When water was added slowly after that, the water content on the surface of fabric rose to a saturated level, which made the water vapor concentration of its surface much higher than that of the surrounding environment. The water vapor then diffused to the surrounding environment by evaporation to achieve a new equilibrium. It was a thermal absorption process for the water evaporated into the air, thus most thermal energy of the 28

28 27

26 25

Temperature (°C)

Temperature (°C)

27

T

24 23 T0

22 21

t1

t2

t3

26 25

T

24 23 T0

22 21 t1

20

20 0

3,000 6,000 9,000 12,000 15,000 18,000 Time (s)

(a) Notes: (a) Knitted cotton fabric; (b) polyester fabric

0

t3

t2 2,000

4,000

6,000

Time (s)

(b)

8,000

10,000

Figure 2. Surface temperatures

IJCST 21,5

282

fabric was transferred to the water vapor, which made the surface temperature of the fabric decreased swiftly to a lowest point t0. During this stage, if the surface area and structure of fabric changes, the temperature decreasing rate and speed will change accordingly. For example, if the tested surface area of fabric increases by increasing the area of fabric and changing the crimp height of yarns, more water will be evaporated at a specific period, thus the temperature decreases more quickly. Stage two. The surface temperature keeps stable from t1 to t2. A dynamic equilibrium had been achieved between the fabric and environment in this stage, and the surface temperature fluctuated slightly around t0. The water content in the fabric was also saturated which provided enough dissociative water to be evaporated. However, the thermal energy absorbed by the evaporation of water equalled to the heat transferred from air to the fabric by liquefaction of water to the fabric, heat radiation and heat convection. Also, the length of this stage was largely determined by the surface area and the components of the fabrics, as shown in Figure 2. Stage three. The surface temperature rises from t2 to t3. At the second stage, with the evaporation of water from the surface of fabrics, the moisture regains of fabrics decreased gradually. The water between and in the fibers then were involved in the evaporation process, however, the evaporation of these water was much difficult than that of the dissociative water, which made the evaporation speed slow down. At last, when the heat absorbed by the evaporation of water was lower than the heat transferred from air to the fabric, the dynamic thermal balance was destroyed and the surface temperature of fabrics increased with the decrease of moisture regains of fabrics. At the end of this stage, the moisture regains of fabrics declined to the equilibrium moisture regains, and the surface temperature of fabrics was equal with the environmental temperature, a new thermal and moisture balance was setup and the process of moisture liberation finished. 3.2 Electrical properties of fabrics in the process of moisture absorption and liberation For most sportswear, it absorbs sweat and perspired to the out layer and then evaporated to the environment. During the process of moisture absorption and liberation, moisture regain of the fabric changed regularly. As moisture in the fabrics determined the electrical conductivity of the fabric mainly, the electrical resistance of fabrics changes greatly in the same process. Normally, the resistance of fabric decreases as the moisture regain of that increases. Figure 3 shows the relation between resistance and moisture regain of knitted cotton and polyester fabrics. Regression analysis of the data shows that the electrical resistance of fabric is linear related to the reciprocal of moisture regain with the regression coefficients R 2 higher than 0.99. As the moisture regain of fabric increases, more water in the fabric could be applied as electrical conductor, thus the electrical resistance decreases linearly. Besides, the resistance increases with the increase of the testing distance from Figure 3. And moisture regain shows greater impact on the electrical resistance when the testing distance increases, as shown in Figure 3, the slope of the curves increases with the testing distance increases, as the coefficients in the regression equations increases. Compared with polyester, cotton fabric shows lower electrical resistance and slope in the curves. This is due to the water absorption differences between these two different materials. Cotton fiber is hydrophilic while polyester is hydrophobic. Electrical

conducting in polyester is mainly determined by the moisture of the fabric, thus its resistance is greatly affected by the moisture regain. On the other hand, water in cotton fabric exists not only in the fabric, but also in the fiber itself, which make the electrical resistance in a lower lever and change much slowly. 3.3 Electrical and thermal properties of fabric in the process of moisture absorption and liberation In order to investigate the relation between the surface temperature and the electrical resistance of fabric in the moisture absorption and liberation process, a kind of polyester/cotton double-layer knitted fabric was tested in experimental. In this way, the electrical and thermal properties of different materials were analysed in the specific fabric at the same time. Figure 4(a) shows the variation of fabric temperature and resistance during the moisture absorption and liberation. The trends of temperature are similar to that in Figure 2. The electrical resistance decreases quickly at the beginning when water is added into the fabric, and then increases gradually in the moisture-liberation process in which the moisture regains decreases gradually. At the beginning of the process, the fabric was under extremely dry condition, its surface temperature was the same as the surrounding temperature, and the electrical

100 80 60

Distance = 20 mm Y = 0.34859+26.91171 X 2 R = 0.9993

40 20

Resistance, 10k ohm

Resistance, 10k ohm

250

Distance = 60 mm Y = 6.03346+37.55542 X R2 = 0.9886 Distance = 40 mm Y = 2.13842+35.26885 X R2 = 0.9935

120

200 150 100

Distance = 20 mm Y = 7.42613 + 29.05813 X 2 R = 0.9991

50

1.5

2.0 1/M

2.5

3.0

3.5

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 1/M

(b)

(a)

40

30

Cotton Polyester

28

30

T

20 T1

24

1 10

T2

22

2 0

3

20 0

1,000 2,000 3,000 4,000 5,000 6,000 Time (s)

Fabric temperature (°C)

32

26

30.0

50 Fabric resistance (M ohm)

Fabric temperature (°C)

34

20

28.5 15 27.0 10 25.5 24.0 22.5

Figure 3. Linear regression between resistance and 1/M of: (a) knitted cotton; (b) knitted polyester

Cotton Polyester

5

1 2

0

Fabric resistance (M ohm)

1.0

283

Distance = 60 mm Y = 10.23581 + 47.81695 X 2 R = 0.9940 Distance = 40 mm Y = 4.77135 + 42.41028 X 2 R = 0.9924

0 0.5

Thermal and electrical properties

–100 –50 0 50 100 150 200 250 300 350 400 Time (s)

(a) (b) Notes: (a) Surface temperature and electrical resistance of polyester/cotton-knitted fabrics in the process of moisture absorption and liberation; (b) magnification of (a) at the beginning period of moisture absorption

Figure 4.

IJCST 21,5

284

resistance was so high that it could be regarded as insulator. With the adding of water into the fabric, cotton fiber absorbed water quickly and its electrical resistance decreased swiftly. However, as a kind of synthetic fiber, polyester cannot absorbed so quickly as cotton, and absorbed water stayed mainly in the interface of fibers but not the fiber itself. Thus, the electrical resistance of cotton decreased more quickly than polyester. As shown in Figure 4(b), the two layers of the fabric show different trends in the resistance variation, the resistance of the cotton layer decreased 50 s earlier than that of polyester layer. Besides, the resistance of polyester layer decreased gradually from about 5 MV to zero in about 75 s, while that of cotton layer decreased sharply. At about 300 s, the surface temperature of the cotton layer was higher than that of polyester layer. As water in the polyester could be quickly evaporated due to its hydrophobic properties, and polyester filaments have much higher surface area than that of cotton fiber, more thermal energy would be transferred through the water evaporation, which made the layer lower surface temperature, as shown in Figure 4(b). At the second stage from 350 s to about 3,500 s, the fabric held saturated water for evaporation and the electrical resistance of both layers of the fabric kept stable. And the surface temperature of the two layers also kept stable with certain difference. After about 3,500 s, the surface temperature of both layer of the fabric started to increase gradually, which showed the beginning of the third stage, as shown in Figure 4(a) the points 1 and 2. As cotton fiber absorbed water itself while polyester was hydrophobic, the point 2 was much closer to the X-axis, which shows an earlier start point of stage three. However, the electrical resistances of both the layers started to increase at point 3, which was much closer to X-axis than point 2, which meant the electrical resistance changed earlier than that of surface temperature. The electrical resistance was mainly affected by the surface water content, while the surface temperature was affect by surface water content, surface area, fabric constitution and materials, so electrical resistance changed earlier than surface temperature. As surface temperature determined the garment comfort and the surface resistance could be applied in fabric moisture and temperature test apparatus, this relation was important in the design of sportswear and moisture testers.

4. Conclusions The thermal and electrical properties of textiles were analysed in the process of moisture absorption and liberation. The surface temperature of fabrics shows three different stages in the process, it decreased swiftly and then kept stable, after that, it increased gradually and attained the surrounding temperature at last. The electrical resistance of fabric decreased with the increase of moisture regain. Regression analysis shows the electrical resistance is linearly related to the reciprocal of the moisture regain of fabrics. For cotton/polyester double-layer fabric, surface temperature of cotton layer is higher than that of polyester layer in the moisture absorption and liberation process. And the electrical resistance of cotton layer decreases more quickly than that of polyester layer. The electrical resistance changes earlier than surface temperature in the moisture-liberation process. The electrical resistance was mainly determined by the surface water content while the thermal properties was determined by the surface water content, surface area, fabric components, and the materials.

References AATCC (2000), “Absorbency of bleached textiles”, AATCC Test Method 79. Adler, M.M. and Walsh, W.K. (1984), “Mechanisms of transient moisture transport between fabrics”, Textile Research Journal, Vol. 54, pp. 334-43. Awano, M., Shirai, M. and Ishikawa, K. (1987), “The attempt to evaluate the water absorption-evaporation behavior of textile fabrics”, Sen-i Gakkaishi, Vol. 43, pp. 75-80. BSI (1996), “Methods 21A and 21B, methods for determination of resistance to wicking and lateral leakage”, BS 3424: Part 18: 1986. Crow, R.M. and Osczevski, R.J. (1998), “The interaction of water with fabrics”, Textile Research Journal, Vol. 68, pp. 280-8. Fan, J., Cheng, X. and Chen, Y.S. (2003), “An experimental investigation of moisture absorption and condensation in fibrous insulations at low temperatures”, Experimental Thermal and Fluid Science, Vol. 27, pp. 723-9. Hu, J., Li, Y., Yeung, K., Wong, A.S.W. and Xu, W. (2005), “Moisture management tester: a method to characterize fabric liquid moisture management properties”, Textile Research Journal, Vol. 75, pp. 57-62. Jiang, P. and Yan, H. (2004), “Study on dynamic heat-moisture transfer properties of summer fabrics using microclimatic method”, Journal of Donghua University, Vol. 21 No. 5, pp. 69-72. Li, Y. (2001), “The science of clothing comfort”, Textile Progress, Vol. 31, pp. 64-90. Li, Y. and Holcombe, B.V. (1992), “A two-stage sorption model of the coupled diffusion of moisture and heat in wool fabrics”, Textile Research Journal, Vol. 62, pp. 211-7. Li, Y., Planet, A.M. and Holcombe, B.V. (1995), “Fiber hygroscopicity and perceptions of dampness, Part II: physical mechanisms”, Textile Research Journal, Vol. 65, pp. 316-24. Li, Y., Xu, W. and Yeung, K.W.W. (2000), “Moisture management of textiles”, US Patent No. 6,499,338 B2. Rousselle, M.A., Thibodeaux, D.P. and French, A.D. (2005), “Cotton fiber properties and moisture: water of imbibition”, Textile Research Journal, Vol. 75, pp. 177-80. Corresponding author Weilin Xu can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

Thermal and electrical properties 285

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

IJCST 21,5

Illumination invariant seam line detection in real garments Ioannis G. Mariolis and Evangelos S. Dermatas

286 Received 12 October 2008 Revised 13 January 2009 Accepted 13 January 2009

Department of Electrical Engineering and Computer Technology, University of Patras, Patras, Greece Abstract Purpose – The purpose of this paper is to provide a robust method for automatic detection of seam lines based only on digital images of the garments. Design/methodology/approach – A local standard deviation pre-processing filter is applied to enhance the contrast between the seam line and the texture and the Prewitt operator extracts the edges of the enhanced image. The seam line is detected by a maximum at the Radon transform. The proposed method is invariant to the illumination intensity and it has been also tested with moving average and fast Fourier transform low-pass filters used in the pre-processing module. Extensive experiments are carried out in the presence of additive Gaussian and uniform noise. Findings – The proposed method detects 109 out of 118 seams when the local standard deviation is used at the pre-processing stage, giving a mean distance error between the real and the estimated line of 2 mm when the image is digitised at 97 dpi. However, in case the images are distorted by additive Gaussian noise at 20 dB signal-to-noise ratio, the moving average low-pass filtering method gives the best results, detecting 104 noisy images. Research limitations/implications – The proposed method detects seam lines that can be approximated by a continuation of straight lines. The current work can be extended in the detection of the curved parts of seam lines. Practical implications – Since the method addresses garments instead of seam specimens, the proposed approach can be imported in automatic systems for online quality control of seams. Originality/value – Local standard deviation belongs to first-order statistics, which makes it suitable for texture analysis and that is why it is mostly used in web defect detection. The novelty in the approach, however, is that by considering the seam as an abnormality of the texture, the authors applied that method at the pre-processing stage to enhance the seam before the detection. Moreover, the presented method is illumination invariant, a property that has not been addressed in similar methods. Keywords Textile industry, Quality control, Image processing Paper type Research paper

International Journal of Clothing Science and Technology Vol. 21 No. 5, 2009 pp. 286-299 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910983786

Introduction Among the most important requirements in industrial textile manufacturing is the seam quality control. The developing of automatic seam inspection methods reduces significantly the production cost, but the efficiency of these methods is reduced due to the presence of the textile texture, the illuminant type and the light variation in space and time, the image acquisition system and the noise of the electronic and optical components. In this direction, robust pre-processing methods have already been used This work is supported by the general secretariat of Research and Technology project PENED2001 No. 3049.

to eliminate the seam line detection error in the presence of uniform and Gaussian noise in low and medium signal-to-noise ratios (SNRs) (Mariolis and Dermatas, 2004). In Rankov (1999) and Crispin et al. (2000), different approaches for extracting the edge positions in the image coordinate system have been investigated based on the Hough transform, the spatial histogram, polynomial regression and the discrete first derivative. These edge detection algorithms are compared in terms of speed and precision performance using a laser-line three-dimensional scan and triangulation techniques on shoe seams. Consequently, automatic seam classification methods can be used to adjust online sewing machines producing better seams (Stylios et al., 1995; Dorrity, 1995; Clapp et al., 1995). More recently, Bahlmann et al. (1999) has presented a method for automatic quality control of textile seams using gray-scale images, based on a self-organized feature map neural network for seam classification. Special attention has been paid on the selection of the optimum set of features that is based on the amplitude of the one-dimensional Fourier coefficients. Although Bahlmann in his work refers to the need for “The continuous control during manufacturing and the end control of seams in garments”, his method aimed only to the classification of seam specimens used for setting the optimal sewing machine parameters. In this paper, an illumination invariant method for accurate detection of straight seam line segments is presented and evaluated. The method is applied on gray-scale images of real garments, aiming to both continuous and end control. Two consequent non-linear transformations enhance the seam track and simultaneously reduce the textile texture: based on a local estimation of the first- and second-order moments an image transformation is followed by the Prewitt edge detector and the discrete Radon transform. At the local maxima of the produced matrix the position of the seam lines are detected. The experiments were carried out in a set of 118 images from various types of seams, textiles and colours. The comparison shows that local standard deviation is the best pre-processing method detecting 109 out of 118 “clean” images, giving a mean distance error between the real and the estimated line of 2 mm at 97 dpi. However, the most robust pre-processing method is obtained by the moving averaging filter, detecting 104 out of 118 noisy images at 20 dB SNR using Gaussian distributed additive noise, which gives in general worst detection rate than the rate estimated in the case of distorted images by uniformly distributed noise. The structure of the paper is as follows: in the first section, the proposed pre-processing and seam detection method is described. In the next section, the image database used to evaluate the proposed seam detection method is presented, while the last section contains the experimental results and a short discussion. Seam detection The proposed seam detection method consists of three sequentially connected modules: in the image pre-processing module various types of noise and the irrelevant information (i.e. texture) is reduced, the edge detection algorithm enhances the presence of strong derivatives in the image intensity and the line detector derives the lines in the image at the local maxima of the discrete-Radon-transform. Pre-processing module The texture and the other types of noise can be reduced significantly using a vast number of alternative methodologies most of them based on statistical descriptors.

Illumination invariant seam line detection 287

IJCST 21,5

288

Among them a statistical transformation of the original image, based on a local estimation of the standard deviation, has been selected. The performance of the proposed method is evaluated and compared with the well-known low-pass filtering in the frequency domain and moving-average filtering methods. Both methods reduce the high-frequency components in the frequency and spatial domain correspondingly, since this is the bandwidth where the texture appears. Local standard deviation. Simple statistical measures of gray-level images include mean value, variance, skewness and kurtosis. These measures can be computed as moments of the gray-level histogram. A simple study of garment histograms in various scales shows that they follow Gaussian distribution. If the image contains a large area of the garment, the presence of the seam is not capable of altering significantly the distribution’s moments. However, if the area of the image is such that includes both texture and seam in a more balanced proportion, the local variance or equivalently the local standard deviation is affected, which makes perfect sense if the seam is considered as a defect of the texture. A local estimation of the standard deviation is obtained by masking the digital image G(.,.) using a window W. The standard deviation within window W is assigned to the intensity value of the center pixel for the transformed image H(.,.) ( Ja¨hne, 1997): sﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ X 1 H ðm; nÞ ¼ ðGðm 2 n0 ; n 2 n0 Þ 2 kGlmn Þ2 ; P 2 1 m0 ;n0 [W

ð1Þ

where kGlmn is the mean intensity value of the sub-image W containing P pixels: kGlmn ¼

1 P

X 0

Gðm 2 m0 ; n 2 n0 Þ:

ð2Þ

0

m ;n [W

Conceptually, applying the local standard estimation operator using the appropriate window size the area of the seam is enhanced increasing the contrast between the seam and the texture. The result of applying the local standard deviation filter to an image of a fabric containing a seam is illustrated in Figure 1(b).

(a)

Figure 1.

(b)

(c) (d) Notes: (a) Original image containing a seam; (b) filtering based on the local standard deviation; (c) binary image containing the location of strong edges; (d) superposition of the estimated seam line (white line) onto the original image

Edge detection Gray-level edge detection is a popular image transformation based on discrete gradient operations, enhancing the high frequencies and reducing the low-frequency components of an image. Conceptually, locating either the local extrema of the first derivative of the image or the zero crossings of the second derivative can reveal edges. Among the various widely used edge detectors, the Prewitt approach gives the best seam detection rates as shown in Table I, where a great number of edge detection methods are evaluated in the proposed seam detection method. Prewitt edge detector belongs to the first derivative’s extrema category. This becomes obvious by the convolution masks that follow: Row mask 2 3 21 21 21 6 7 0 0 7 Dr ¼ 16 6 4 0 5; 1 1 1

Column mask 2 3 21 0 1 6 7: 7 Dr ¼ 16 6 4 21 0 1 5 21 1 1

Illumination invariant seam line detection 289

ð3Þ

The edge magnitudes matrix Ie(.,.), is estimated using the convolution masks: qﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ Ieðm; nÞ ¼ ðDr * H ðm; nÞÞ2 þ ðDc * H ðm; nÞÞ2 ;

ð4Þ

where * denotes the two-dimensional convolution operation. The binary image L(.,.), is produced by thresholding the edge magnitudes, calculated for every pixel of the image H(.,.): ( 1; Ieðm; nÞ $ T Lðm; nÞ ¼ ð5Þ 0; Ieðm; nÞ , T: The selection of the threshold’s value T is based on the root-mean-square estimation of noise. If the size of the edge magnitudes matrix Ie(m,n), is M £ N, the threshold’s value is given by the following equation: vﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ uM N uX X u Ie 2 ðm; nÞ u t i¼1 j¼1 : ð6Þ T¼2 MN The result of applying the Prewitt edge detector on a gray-scale image in contradiction to the original image is illustrated in Figure 1(c). Seam detection meted Canny Roberts Prewitt Sobel Laplacian Zerocross

Detection rate (%) 43.33 65.00 74.17 71.67 23.33 41.67

Table I. Percentage seam detection rate using various edge detection methods

IJCST 21,5

290

Line detector In case of binary images, the Radon transform is equivalent to the Hough transform (Duda and Hart, 1972) sharing the same properties. Thus, the proposed method inherits the advantages of the Hough transform (Gonzales and Woods, 1992), since it is insensitive to imperfect data and noise and does not require any prior knowledge of the lines’ position within the image. Moreover, the fast digital Radon transform, due to its recursive nature, has sequential complexity of OðN 2 · log N Þ additions for an N £ N image (Gotz and Druckmuller, 1996). In general, the Radon transform of a two variable function f (x, y) is the line integral of f(.,.) parallel to the Y-axis: Z 1 Ru ðXÞ ¼ f ðX cos Q 2 Y sin Q; X sin Q þ Y cos QÞdY : ð7Þ 21

In digital images, the quantization process transform the two-dimensional continuous space (x,y) into the discrete image space of (n,m) and the corresponding discrete Radon transform can be interpreted as the projection of the image intensity along a radial line oriented at a specific angle (Bracewell, 1995). Therefore, the Radon transform is used to detect straight lines from binary images. If the size of the binary image L(m,n) is M £ N, by considering a polar coordinate system centered at Lð b M =2c ; b N =2c Þ, the discrete Radon transform R( p,q) corresponds to the straight line with polar distance p and polar angle array u(q), p [ {1; 2; . . . ; D}; q [ {1; 2; . . . ; TH} and is given by: D X D D Rð p; qÞ ¼ Q p2 · cosðuðqÞÞ 2 2 m · sinðuðqÞÞ ; 2 2 m¼1 D D · sinðuðqÞ þ 2 m · cosðuðqÞÞ p2 2 2

ð8Þ

where bxc denotes the integer part of x, D is given by equation (9) and is sufficient to compute the projection at unit intervals, even along the diagonal. The matrix Q(m,n) can be considered as a binary image with size D £ D containing the image L(m,n) at its center and zeros anywhere else. u(q) is a vector whose elements range from 0 to 180(TH 2 1)/TH degrees and TH is the size of the vector u(q): 7 06sﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ 1 6 7 2 2 6 7 M 2 3 N 2 3 ð9Þ D ¼ 2 · @4 M2 þ N2 þ 35 þ 1A; 2 2 ( Qðm;nÞ ¼

Lðm2aþ1;n2 g þ1Þ; m [ A ; {a;aþ1; ...; b};n [ G ; {g; g þ1; ...; d} 0;

m [ {1; ...;D}2A;n [ {1; ...;D}2G

; ð10Þ

uðqÞ ¼ ðq 2 1Þ ·

180 ; TH

q [ {1; 2; . . . ; TH};

ð11Þ

D M D M 2 þ 1; b ¼ þ M 2 2 1; 2 2 2 2 D N D N þ 1; d ¼ þ N 2 2 1: g¼ 2 2 2 2 2

a¼

ð12Þ

In the local maxima of R( p,q) the seam lines are detected. As proved in the Appendix, the proposed method, including pre-processing and edge detection is illumination invariant (Figure 2).

Illumination invariant seam line detection 291

Image database Two illumination sources have been used: the Philips C2 60 W and the Osram professional (long life). The digital images have been acquired using a low-cost digital camera Sony DSC P50 (2.1 megapixel) in 30 cm from the textile. The total size of images is 1,600 £ 1,200 pixels, which gives an equivalent space of 42.35 £ 31.76 cm area (97 dpi), i.e. each pixel covers a square area of 0.265 £ 0.265 cm2. A total number of ten different monochrome garments have been selected and 118 RGB images have been acquired. Then a portion of every original image (about 350 £ 450 pixels), including almost straight-line seams, has been cropped. At each new image, ten seam pixels more-or-less uniformly distributed across the seam are manually defined. Finally, the RGB images are converted to grayscale. Experiments The proposed seam detection method has been implemented and evaluated using the Matlab programming language of MathWorks, Inc. and a great number of experiments has been conducted. Moreover, apart from the proposed local standard deviation filter, the moving average filter and the fast Fourier transform (FFT)-filtering pre-processing modules have been alternatively implemented and applied, also in order to decrease the influence of the texture in the proposed seam detection method. A short description of the alternative pre-processing filters follows.

300 250 200 150 100 50 0 800

700 600 500 400 300

200

100

50

100

150

200

250

300

350

Figure 2. Random transform of the binary image showed in Figure 1(c) for 0.5 degrees increment

IJCST 21,5

292

Moving average filter The moving average filter can be described using several equivalent definitions. The most common definition is based on algebraic operations performed on local image neighbourhoods according to a geometric law defined by the window size (Bovik, 2000). The local average is computed over each local neighbourhood of the image, producing a powerful smoothing effect. Since the average is a linear operation, it is usually applied to reduce zero-mean additive noise. In our case though, it is used to smooth the texture. Seam information is also affected by the filter but very slightly with regard to the texture: H MVF ðm; nÞ ¼ Fðm; nÞ* Gðm; nÞ:

ð13Þ

FFT-filtering The reduction of the irrelevant information that lies in the high frequencies (typical textures, salt and pepper noise) can be obtained by using the two-dimensional FFT using the suitable binary mask. In this mask, since the high frequencies are near to matrix center, the values for every pixel closer than a radius to the center are set to zero. The filtering is applied by deriving the point product of the image and the mask. The corresponding image in the spatial domain is derived using the inverse FFT transform. The whole FFT-filtering process can be described by the following equation: H FT ðm; nÞ ¼ FFT21 2 ðFð f 1 ; f 2 Þ · FFT2 ðGðm; nÞÞÞ:

ð14Þ

The accuracy of the seam detection methods is measured by the mean Euclidean distance between the derived seam line and the manually defined pixels belonging to the real seam line. In our experiments, the number of manually defined seam pixels ranges between 5 and 12 according to the image size. The manually defined seam pixels are distributed almost uniformly in the real seam line. If the Euclidean distance is less than or equal to 25 pixels, which correspond to about 6.6 mm in the textile plane, the detection is assumed successful. Various phenomena, not related to the actual seam-detection method, decrease its detection accuracy. Among the most important, the divergence from a perfect straight line of the width of the real seam increases the estimated error. In some low-cost imaging systems, the lenses’ geometric distortion can significantly contribute to the mean Euclidean distance between the manually defined seam pixels and the estimated seam line. The proper camera calibration process decreases this type of image distortion. As a result, the estimated seam line fits the real seam-line, but the corresponding Euclidean error is not zero. Actually, the mean value of the mean Euclidean distance between the manually defined seam pixels and their regression line is 4,167 pixels; in some cases the distance increases to 20 pixels. Since there is a great deal of edge detectors the one that is most suitable for our application has been experimentally selected. In this direction, we tested six different edge detectors, omitting the pre-processing stage, searching for the one that performs best. In our case, it was the Prewitt operator giving the best results as it is illustrated in Table I, and is the one used in the rest of the experiments described below. Owing to the great amount of experiments and the number of different system configurations, each seam-detection method is described by a sequence of abbreviations

in the exact order they have been process the digital image. The complete list is given in Table I. Thus, the name of the method describes the complete seam detection process itself. As an example, consider the method where an FFT low-pass filter is followed by the Prewitt edge detector and the Radon transform is finally used to detect the seam line. We will refer to that procedure as the Fl-Pr-R method (Table II). In the proposed method, a number of manually defined parameters have been derived experimentally: the radius rc of the FFT low-pass mask that cuts high frequencies for an m £ n image is given in the following equation: rﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ m2 n2ﬃ r c ¼ 0:9 : ð15Þ þ 2 2

Illumination invariant seam line detection 293

The size of the window W that is used for the implementation of the moving average and the local standard deviation filters is 9 £ 9 pixels, which corresponds to an area of about 2.5 £ 2.5 mm2. Finally, the increment of the angle in the Radon transform has been set to 0.5 degrees (TH ¼ 360). The Pr-R, Fl-Pr-R, mean-Pr-R and std-Pr-R methods were tested using the original image database and the experiments were repeated by adding uniformly distributed or Gaussian noise to the original images. Both types of noise have zero mean and their variance is adjusted to produce the desirable SNR. In Figure 3, seam images are illustrated in the presence of uniform noise at 25, 20, 15 and 10 dB SNR, respectively. The seam detection methods are evaluated using the noisy images and the results for uniformly distributed noise are shown in Figure 4, while those for additive Gaussian noise are shown in Figure 5. In those plots Inf refers to the original images assuming Local standard deviation Moving average filter FFT lowpass Prewitt Radon transform

Std Mean Fl Pr R

(a)

(b)

(c)

(d)

Notes: (a) 25; (b) 20; (c) 15; (d)10 dBs SNR

Table II. Notation used to build up the seam-line detection method

Figure 3. A seam image in the presence of uniform noise

IJCST 21,5

Uniform noise 95.00

294

Seam detection rate (%)

90.00 85.00 80.00 75.00 70.00 65.00 60.00 55.00 50.00

Figure 4. Seam detection accuracy in clean images and additive uniform noise at 25, 20, 15 and 10 dB SNR

Inf

25

20

15

10

Pr- R

74.58

72.88

69.49

64.41

57.63

Fl-Pr-R

84.75

85.59

83.05

75.42

67.80

Mean-PR-R

88.14

87.29

87.29

87.29

81.36

Std-Pr-R

92.37

91.53

87.29

79.66

76.27

SNR(dB) Normal noise 95.00 90.00

Seam detection rate (%)

85.00 80.00 75.00 70.00 65.00 60.00 55.00 50.00

Figure 5. Seam detection accuracy in clean images and additive Gaussian noisy images at 25, 20, 15 and 10 dB SNR

Pr- R Fl-Pr-R Mean-Pr-R Std-Pr-R

Inf

25

20

15

10

74.58 84.75 88.14 92.37

75.42 81.36 88.98 87.29

72.88 83.05 88.14 84.75

66.10 77.12 83.05 75.42

54.24 66.95 79.66 61.02

SNR(dB)

that any noise during their acquisition is negligible. It becomes evident by these graphs that the best detection rate was achieved by the proposed method std-Pr-R, detecting 109 out of 118 seams (92.37 per cent). In order to get an insight about the accuracy of our method we should mention that the mean error of the correct detected seams is 7.61 pixels, which corresponds to 2 mm. Moreover, in the 82.57 per cent of the correct detected seams the mean error is less than or equal to 10 pixels which is about 2.65 mm. However, this method is severely affected by noise as shown in Figures 4 and 5, especially in the presence of Gaussian noise. On the other hand the mean-Pr-R method, which gives the second best results by detecting 104 out of 118 seams (88.14 per cent), shows remarkable resistance to noise. It is also worth mentioning that Fl-Pr-R method, which gives satisfying seam detection results in noiseless images, radically improves the seam detection rate in the presence of low-level noise. Finally, the evaluation experiments were repeated by performing down-sampling of the original images. The nearest neighbour interpolation reduces the size of each dimension by 75, 50 and 25 per cent. The seam detection rate without the presence of noise is shown in Figure 5, while the whole set of experiments in the case of additive noise and down-sampling is illustrated in Table III. As shown in Figure 6, the down-sampling process has not acted always destructive to our data. In fact, the mean-Pr-R method and 50 per cent image down-sampling, outperforms the other configurations giving 111 out of 118 seams detected (94.07 per cent). The same figure applies to Pr-R method for 25 per cent down-sampling, detecting 100 out of 118 seams (84.75 per cent). Finally, the Fl-Pr-R method is not radically affected by down-sampling, while std-Pr-R is very sensitive and as the down-sampling rate decreases the results are worse, dropping to 64.41 per cent at 25 per cent down-sampling. The positive contribution of the down-sampling process in the high-resolution images is achieved by reducing some texture information that distort the statistical estimators and add noise to the edge detector, while in the case of the std-Pr-R method the new window size

Uniform noise (%) SNR (dB) Downsampling: Pr-R FL-Pr-R Mean-Pr-R Std-Pr-R Downsampling: Pr-R FL-Pr-R Mean-Pr-R Std-Pr-R Downsampling: Pr-R FL-Pr-R Mean-Pr-R Std-Pr-R

25 0.75 8.47 84.75 93.22 85.59 0.5 6.78 79.66 94.07 79.66 0.25 11.86 82.20 83.05 59.32

Illumination invariant seam line detection 295

Normal noise (%)

20

15

10

25

20

15

10

12.71 79.66 90.68 82.20

6.78 74.58 90.68 74.58

11.02 60.17 81.36 67.80

5.08 82.20 91.53 83.90

6.78 81.36 91.53 79.66

11.02 77.12 88.98 72.88

7.63 63.56 78.81 55.08

11.86 77.97 93.22 77.12

8.47 69.49 85.59 66.95

8.47 54.24 77.12 55.08

10.17 77.97 94.92 78.81

10.17 72.88 94.07 72.03

7.63 70.34 86.44 62.71

11.02 51.69 76.27 39.83

9.32 77.12 79.66 55.08

8.47 63.56 75.42 41.53

8.47 49.15 51.69 34.75

9.32 83.90 83.05 54.24

9.32 76.27 75.42 49.15

5.08 62.71 67.80 33.90

14.41 48.31 49.15 26.27

Table III. Correct detection in percentage rates for the proposed seam location method using various image pre-processing algorithms and down-sampling rates in the presence of uniform and Gaussian noise

IJCST 21,5

Down sampling 100.00 95.00

296

Seam detection rate (%)

90.00 85.00 80.00 75.00 70.00 65.00 60.00 55.00 50.00

Figure 6. Seam detection accuracy in different down-sampling rates and pre-processing algorithms

1

3/4

1/2

1/4

Pr- R

74.58

74.58

73.73

84.75

Fl-Pr-R

84.75

83.90

79.66

81.36

Mean-Pr-R

88.14

91.53

94.07

88.14

Std-Pr-R

92.37

88.14

80.51

64.41

reduces to (9 £ 9) (down-sampling rate) and as down-sampling rate decreases the number of samples, the estimation of the statistical features are not reliable. Conclusion In this paper, an illumination invariant method for robust detection of straight seam line segments in Gray-scale digital images is presented and evaluated using the efficient local standard deviation pre-processing module to reduce the texture information and facilitate the edge detection method based on the Prewitt approach. Although cases of more complicated seam lines are outside the scope of this work, an extension of the proposed method to such cases could use the proposed pre-processing and edge detection modules, employing however the Hough transform instead of the Radon transform. The Hough transform can be used for the detection of an object described a priori by an analytical equation (e.g. lines, circles). Moreover, if the curvature cannot be described analytically the generalized Hough transform using the principle of the template matching (Ballard, 1981) can be applied instead. In that case, however, a model describing the seam line is needed. The proposed seam detection method can be easily developed in low-cost hardware in textile processing applications showing satisfactory detection rates as indicated by the experimental results. However, in the presence of severe uniform and Gaussian distributed noise or in case where low-quality (24 dpi) digital images are available, a moving average filter as the pre-processing method is more appropriate. Finally, it is

also shown that the Radon transform, in combination to the appropriate pre-processing method, is an extremely powerful tool for detecting the seam lines even in the presence of uniform and Gaussian distributed noise. References Bahlmann, C., Heidemann, G. and Ritter, H. (1999), “Artificial neural networks for automated quality control of textile seams”, Pattern Recognition, Vol. 32 No. 6, pp. 1049-60. Ballard, D.H. (1981), “Generalizing the Hough transform to detect arbitrary shapes”, Pattern Recognition, Vol. 13 No. 2, pp. 111-22. Bovik, A. (2000), Handbook of Image and Video Processing, Academic Press, New York, NY. Bracewell, R. (1995), Two-dimensional Imaging, Prentice-Hall, Englewood Cliffs, NJ, pp. 505-37. Clapp, T.G., Olson, L.H., Titus, K.J. and Dorrity, J.L. (1995), “The on-line inspection of sewn seams”, National Textile Center Annual Report, available at: http://ntc.tx.ncsu.edu/html/ REPORTS/YEAR-FOLDER/online.html Crispin, A.J., Pokric, B., Rankov, M., Reedman, D. and Taylor, G.E. (2000), “Edge inspection in automatic stitching”, International Journal of Clothing Science & Technology, Vol. 12 No. 4, pp. 265-78. Dorrity, J.L. (1995), “New developments for seam quality monitoring in sewing applications”, IEEE Transactions on Industry Applications, Vol. 31 No. 6, pp. 1371-5. Duda, R.O. and Hart, P.E. (1972), “Use of the Hough transformation to detect lines and curves in pictures”, Communications of the ACM, Vol. 15, pp. 11-15. Gonzales, R.C. and Woods, R.E. (1992), Digital Image Processing, Addison-Wesley, New York, NY. Gotz, W.A. and Druckmuller, H.J. (1996), “A fast digital radon-transform: an efficient means for evaluating the Hough transform”, Pattern Recognition, Vol. 29 No. 4, pp. 711-8. Ja¨hne, B. (1997), Digital Image Processing Concepts, Algorithms, and Scientific Applications, Springer, New York, NY. Mariolis, I. and Dermatas, E. (2004), “Robust detection of seam lines using the Radon transform”, Proceedings of the 1st International Conference “From Scientific Computing to Computational Engineering”, Athens, 8-10 September. Rankov, M. (1999), “Data modeling and analysis for stitch path derivation in automatic stitching”, PhD thesis, Leeds Metropolitan University, Leeds. Stylios, G., Sotomi, O.J., Zhu, R., XU, Y.M. and Deacon, R. (1995), “The mechatronic principles for intelligent sewing environments”, Mechatronics, Vol. 5 Nos 2/3, pp. 309-19. Further reading Lim, J.S. (1990), Two-dimensional Signal and Image Processing, Prentice-Hall, Englewood Cliffs, NJ, pp. 42-5.

Appendix In most industry applications of image processing methods different illumination conditions are met. In typical industrial environment, multiple seam inspection tables are operating in parallel under variable lighting conditions. The illuminant intensity changes through time in most commercially available lamps, and the faulty lamps are not replaced immediately. Moreover, the inspection tables are not illuminated uniformly.

Illumination invariant seam line detection 297

IJCST 21,5

298

Theorem The proposed seam-line detection method, using the local standard deviation pre-processing method, is invariant to the illumination’s intensity. Proof If the lighting conditions changes linearly in the whole visible spectrum, the following equation holds: cðlÞ ¼ a · CðlÞ, where C is the spectral power distribution of the light source and c is spectral power of the new lighting conditions. The light density acquired by the charge-coupled device camera in the new illumination environment can be expressed as: gðm; nÞ ¼ Vðm; nÞ · Nðm; nÞ ¼ Vðm; nÞ · Nðm; nÞ

Z

Sðm; n; lÞ · Rðm; n; lÞ · cðlÞdl

Zl

Sðm; n; lÞ · Rðm; n; lÞ · a · CðlÞdl Z ¼ a · Vðm; nÞ · Nðm; nÞ Sðm; n; lÞ · Rðm; n; lÞ · a · CðlÞdl l

ðA1Þ

l

¼ a · Gðm; nÞ; where, V is the unit vector in the direction of the light source, N is the unit vector corresponding to the object surface, S(m,n,l) is the transfer function of the m, n photodetector, and R(m,n) is the light reflected from the object at wavelength l. According to equations (A1) and (A2) the new standard deviation image h(.,.) becomes: sﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ X 1 ðgðm 2 n0 ; n 2 n0 Þ 2 kglmn Þ2 hðm; nÞ ¼ P 2 1 m0 ;n0 [W sﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ X 1 ¼ ðaGðm 2 n0 ; n 2 n0 Þ 2 akGlmn Þ2 P 2 1 m0 ;n0 [W sﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ X 1 ðGðm 2 n0 ; n 2 n0 Þ 2 kGlmn Þ2 ¼ aH ðm; nÞ: ¼a P 2 1 m0 ;n0 [W

ðA2Þ

The edge magnitudes matrix ie(.,.) under the modified lighting conditions is: qﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ ðDr * hðm; nÞÞ2 þ ðDc * hðm; nÞÞ2 qﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ ¼ ðDr * aH ðm; nÞÞ2 þ ðDc * aH ðm; nÞÞ2 qﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ ¼ a ðDr * H ðm; nÞÞ2 þ ðDc * H ðm; nÞÞ2 ¼ aIeðm; nÞ:

ieðm; nÞ ¼

ðA3Þ

The quantization threshold t of the Prewitt edge detector is:

t¼

vﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ u M N u XX u4 ie 2 ðm; nÞ u t i¼1 j¼1 MN

¼

vﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ u M N u XX u4 a 2 Ie 2 ðm; nÞ u t i¼1 j¼1 MN

vﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ u M N u XX u4 Ie 2 ðm; nÞ u t i¼1 j¼1 ¼a ¼ aT: MN

ðA4Þ

After thresholding the new binary image becomes: ( ( ( 1; 1; aIeðm; nÞ $ aT 1; ieðm; nÞ $ t ¼ ¼ lðm; nÞ ¼ 0; 0; aIeðm; nÞ , aT 0; ieðm; nÞ , t

Ieðm; nÞ $ T Ieðm; nÞ , T

¼ Lðm; nÞ:

Illumination invariant seam line detection

ðA5Þ From equation (A5), the derived binary image in the new illumination conditions l(.,.) is the same to that initially estimated L(.,.).

About the authors Ioannis G. Mariolis was born on 14 January 1979 in New Jersey, USA. He received his degree in Engineering from the Department of Electrical Engineering and Computer Technology, University of Patras in 2002 and ever since he is a PhD student at the same department. His research interests include machine vision, pattern recognition and signal processing. Web site: www.wcl2.ee.upatras.gr. Ioannis G. Mariolis is the corresponding author and can be contacted at: [email protected] Evangelos S. Dermatas is an Assistant Professor in the Department of Electrical Engineering and Computer Technology, University of Patras, Patras, Greece. He received his Diploma and PhD degrees from the Department of Electrical Engineering, University of Patras, Patras, Greece in 1985 and 1991, respectively. His research interest areas include, statistical signal processing, pattern recognition, computer security and information extraction.

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

299

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

IJCST 21,5

A mannequin modeling method based on section templates and silhouette control

300

Jiyun Li and Jiaxun Chen

Received 7 November 2008 Revised 12 January 2009 Accepted 12 January 2009

School of Computer Science and Technology, Donghua University, Shanghai, People’s Republic of China Abstract Purpose – The fast and easy generation of personalized mannequin is the premise for the accurate 3D measurement for the customers in an electronic made to measure (e_MTM) system. The purpose of this paper is to attempt to propose a new virtual human mannequin modeling technique to meet this requirement. Design/methodology/approach – The customized human mannequin is constructed by assemblage of the body parts. The body parts including bust, waist and hip segments are achieved by modification of the standard body section templates, while the silhouette obtained from the front and side photos are used to confine the distortion of the parts. Findings – The main findings are the section template base and assemblage method for mannequin modeling. Originality/value – The proposed method can free the burden of high cost and inconvenience of 3D scanner in the measurement process of e_MTM system, and in the mean time without loss of measurement accuracy. Keywords Modelling, Clothing, Image scanners Paper type Research paper

International Journal of Clothing Science and Technology Vol. 21 No. 5, 2009 pp. 300-310 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910983795

Introduction Nowadays, people are paying more attention to the fitness of what they wear. A faster and easier method for accurate body measurement is the key step of made-to-measure operations in apparel industry. Virtual human mannequin modeling plays an important role in electronic made to measure (e_MTM) systems. But a truly e_MTM system can only be implemented when the measurement of each individual human mannequin can be easily and relatively accurately given out. Many 3D modeling techniques have been proposed for the construction of virtual human body in computer graphics area which can be used to generate human mannequin as well. To our knowledge, former studies in this area can be divided into two categories: 3D-based modeling and 2D photos-based reconstruction. The 3D-based modeling mainly concentrates on feature-based human body surface modeling, parameterized surface modeling, and polygon-based modeling. Most of them use 3D scanner or sensors to achieve body points then generate virtual human body directly. These techniques share the same limits that they all need 3D scanners to acquire the 3D data. Others construct the personalized human body by modification of a standard human body template (Charlie, 2005; Kim and Park, 2004; Seo and Magnenat-Thalmann, 2003; Allen et al., 2003). These techniques usually are mainly focus on animation, so they are either resource demanding and over complicated for apparel industry, or suffering from accuracy lost due to the large error between

personalized data and the unique standard template. 2D photos-based reconstruction techniques free the users from the confinement of 3D scanner by distilling the critical data from one or more 2D photos that is easy to obtain. In this area of study, researchers mainly do the reconstruction by marking the feature points or modification of the body silhouette, and then achieve the 3D points’ coordinates through the combination of 2D information. Seo et al. (2006) suggested using one or more than one 2D photos to construct data driven human body model, the parameterized deformable data driven model comes from real human body range scan; Stylios et al. (2001) devised a parametric and geometric 3D human model by incorporating three 2D photos information with database techniques; Kakadiaris and Metaxas (1998) proposed a modeling technique based on three interleaving vision; Gu et al. (1998) used silhouette and deformable surface as original model to construct parameterized surface model; Lee et al. (2000) used photos from front, side and back to construct vivid 3D virtual human body with no special request for shading and lighting effect of the photos; and Mori and Malik (2006) utilized the front and back relationship to reconstruct the shape of the 3D human body. Only one 2D picture is used to position the joints, reconstruct shape and gesture in 3D. They had to save 2D photo samples from different viewpoints and marked the joint position beforehand. Shen et al. (2002) proposed a shape from silhouette-based way to develop virtual human body from 2D photos by camera calibration, volume generation, surface reconstruction and texture mapping. Here, by combining the idea of template from 3D modeling and silhouette from 2D techniques based on preprocessing of the scanned 3D data of a large amount of individuals from east china area, we proposed a novel virtual human mannequin modeling method mainly for e_MTM oriented applications. The method that we suggested is based on body section templates and silhouette control, which can generate customized human mannequin more quickly and precisely than those techniques which based on one standard body template. The rest of the paper is organized as follows. First, we detail the related concepts and rules of our virtual human mannequin generation method, then describe the system architecture and some effects of each module. Following some tests and results, conclusion and some further research directions are given out at last.

Mannequin modeling method 301

Concepts and rules Virtual human mannequin In this paper, a virtual human mannequin is composed of three parts, namely, bust, waist and hip segments, and each segment consists of some related key sections, as shown in Figure 1:

Bust

Waist

Hip

Figure 1. Body segments and sections

IJCST 21,5

302

(1) Bust segment. As there are more garment-related features in this area, we use six key sections to approach this segment, with bust section as the main section and the other five as subsidiaries. (2) Waist segment. We use two key sections to approach this segment, with waist section as the main section and the other one as subsidiary. (3) Hip segment. We also use two key sections to approach this segment, with hip section as the main section and the other one as subsidiary. Thus, we can generate a virtual human mannequin by assembling the ten key sections, which are chosen from the three segments. The shape of the section can be approached by an ellipse approximately. So, we use the geometric parameters major and minor axis to describe the shape of each section. We can see this in Figure 2. In the middle of Figure 2 is the shape of bust section, the three green points is the datum mark that we used to assemble the sections into virtual human mannequin. The right part of Figure 2 shows the ten sections of one human mannequin. Thus, the 3D human mannequin M can be represented as: M ¼ ððP ij ðx; y; zÞ; Di ðx; y; zÞÞÞ i ¼ 1; 2; . . . ; 10; j ¼ 1; 2; . . . ; 30 ð1Þ where, Pij(x, y, z), stands for the jth points on ith key section; and Di(x, y, z) stands for the barycenter of the triangle formed by the three green points on the ith key section. Body section template base construction As we make the virtual human mannequin by assemblage of ten body sections, which are achieved by modification of standard part templates according to customer’s personal data, the construction of standard parts’ templates base is of vital importance. The body sections used as template are chosen upon the basis of shape analysis and comparison of more than 700 females’ 3D scanning data from eastern China. First, we divide the samples into Y, A, B, C four classes according to the difference between the girth of bust and the girth of waist of the person. The value for class Y, A, B, C,

Figure 2. The bust section and the ten key sections of a virtual human mannequin

respectively, is shown in Table I; then within each class, we use the length ratio of major axis to minor axis of the main section of bust segment, waist segment and hip segment for classification (Figure 3). In this way, each section falls into one of the three categories, namely round, middle and flat. Owing to the anatomy structure of female, the bust section is further divided into three classes namely high, average, and low according to bust height (the definition of bust height is shown in Figure 4). Thus, we get totally 60 section templates for female human mannequin section template base. Parts of the template base values of class Y are shown in Tables II-IV.

Mannequin modeling method 303

Section templates selection and matching algorithm When building a virtual human mannequin, we need to select the most resemble section templates from the section template base, modify and reposition them for assemblage. The rule that we used for section selection can be described as given below. No. 1 2 3 4

Class

Value

Y A B C

24 , 19 18 , 14 13 , 9 8,4

Note: Unit, cm

Major

Table I. Body shape classification table

Minor

Figure 3. Section illustration

Bust height

Figure 4. Definition of bust height

No. Shape class Bust section class Value of bust section Bust height class Value of bust height 1 2 3

Y Y Y

Note: Unit, cm

Round Round Round

[1 , 1.165] [1 , 1.165] [1 , 1.165]

High Mid Low

[15 , 1 ] [8 , 14] [0 , 7]

Table II. Bust segment classification table

IJCST 21,5

304

Take one main section as an example. Let Cx be the length of the major axis of one of the customer’s main section, Txi be the length of the major axis of the ith respective section in the template base, Cy be the length of the minor axis of one of the customer’s main section, Tyi be the length of the minor axis of the ith respective section in the template base. For a customer’s main section with major axis Cx and minor axis Cy, Search in the parts’ template base for its counterpart template section i where minjC x =C y 2 T xi =T yi j satisfied. Using the above rule, we can achieve the ten main section candidates. But if bust section is the case, we put bust height into consideration as well. So the candidate section whose bust height is closest to the customer’s within the selected five minimum candidates will be chosen. In this way, we can get the ten key section templates for modification. Here, we further discuss the matching. We use the principle component analysis to decide the weight of each parameter, the correlative matrix is given in Table V, and the eigenvalues of the matrix and their weights to variation are given in Table VI.

No.

Table III. Waist segment classification table

1 2 3

Table V. The correlative matrix

Table VI. The eigenvalues of the correlative matrix and their weights to variation

Waist section class

Y Y Y

Value [1 , 1.145] (1.145 , 1.26) (1.26 , 1)

Round Mid Flat

Note: Unit, cm

No.

Table IV. Hip segment classification table

Shape class

1 2 3

Shape class Y Y Y

Hip section class

Value [1 , 1.345] (1.345 , 1.475) (1.475 , 1)

Round Mid Flat

Note: Unit, cm

Part

Bust

Waist

Hip

Bust Waist Hip

1.000 0.480 0.243

0.480 1.000 0.339

0.243 0.339 1.000

Factor

Eigenvalues

Weights

Bust Waist Hip

1.717 0.777 0.505

57.249 25.908 16.842

From Table VI, we can see that bust segment contributes to 57.249 percent of the variation, so according to the weights distribution, the match will follow the sequence of bust, waist and hip. The matching process can be described as the following five steps: (1) Input personal shape information, including two photos of the user and her girth difference between the bust and the waist. (2) Decide which class (Y, A, B, C) the user’s measurement falls in according to the user’s girth difference between the bust and the waist. (3) Match the bust section templates. (4) Match the waist section templates. (5) Match the hip section templates.

Mannequin modeling method 305

Section modification rules After selection of the most resemble section templates, we use silhouette control-based template modification rules to modify the chosen template in order to make sure that not only the contour of the template is the same as the customer’s, but also the length of major axis and minor axis for both are the same. For major axis, it is calculated as in equation (2): LK x ¼ LK mx*C x

ð2Þ

where LKx, is the length of the major axis of the section after modification; LKmx, is the length of the major axis of the section before modification; Cx ¼ Csx/Cmx, is the coefficient for x direction. It equals to the length ratio of major axis of customized section to major axis of template section; Csx, is the length of the major axis of customer section; Cmx, is the length of the major axis of template section. The minor axis follows the same practice. Figure 5 shows the shape of the main sections for bust segment, waist segment and hip segment, respectively. Red is for the section template, blue for the section after modification. Rules for section assemblage In order to assemble the sections into human mannequin, we have to study the central trends of human body. From Figure 6, we know that the front view of central line of human body is a straight line, but the side view is a curve. So, we cannot assemble the key sections of the customer that we achieved through modification of the section template only.

Figure 5. The shape of sections and its modification

IJCST 21,5

306

Here, we use the barycenter Di(x, y, z) of the triangle formed by the three datum mark points of the ith key section as the base point. The barycenter of the ith key section should be superposed on the intersection point of the ith section plane and the central line of the human body. Let Ci(x, y, z) be the intersection point of the ith key section plane and the central line of the human body, according to equation (1), for each point Pij(x, y, z) ( j ¼ 1, . . . , 30) of the ith key section, the transformation of the section points can be calculated by equation (3): P ij · x ¼ P ij · x þ di

ð3Þ

where, di ¼ Ci · x–Di · x System architecture Based on the concepts and ideas above, we propose a customized virtual human mannequin generation system, as shown in Figure 7. It consists of four modules: shape information retrieval (SIR) module, parts’ templates selection and matching (PTSM) module, template modification and surface fitting (TMSF) module and silhouette verification (SV) module. First, we use SIR module to get feature parameters of the body section, silhouette of the body and its control information from the two input 2D photos. Then the feature parameters are used to find the appropriate template by PTSM module, the template found by PTSM module is sent to TMSF module for surface fitting while the silhouette control information is used to modify the selected template to produce the customized human mannequin. Finally, in the SV module, the front and side silhouettes of the virtual human mannequin generated by us are compared with the silhouette we obtained from the 2D photos: . Shape information retrieval (SIR). First, we get the girth difference of bust and waist from her and her two 2D photos from front and side view, respectively. Second, we distract the body silhouette by image edge detection and outline tracing. Third, we auto-position the key sections according to the common

Central line (side view)

Central line (front view)

Figure 6. The front and side view of the central line of humanbody

PTSM Input

Figure 7. System architecture

Output SIR

TMSF

SV

.

.

.

proportion relationship of human mannequin, fine-tune the position of the key sections according to the geometric feature of each section. For example, the waist section of young female usually has the smallest width in the front silhouette. Finally, we calculate the length of major axis and minor axis for each key section. Figure 8 shows the auto-positioning of the ten sections on 2D photos from front and side view. Parts’ templates selection and matching (PTSM). The silhouette information getting from SIR including girth difference of bust and waist, length width ratio of the ten key sections are used in this module to choose the resemble section templates. This match procedure starts with body shape classification according to the girth difference of bust and waist, then segment, section accordingly (see the section “Section templates selection and matching algorithm”). Template modification and surface fitting (TMSF). In order to produce the body sections of the specific customer, we need to make some modification to the body section templates according to the silhouette of the customer. This module meets this need and also generates the customized virtual human mannequin by the NURBS surface fitting. The modification process follows the rules in “Section modification rules” and the assemblage process follows the rules given in “Rules for section assemblage.” Figure 9 is one of the virtual human mannequins generated by TMSF. Silhouette verification (SV). To guarantee that the technique we proposed do generate a customized model, we use the projections of the generated 3D model to match the original silhouette which we get from 2D photos.

Mannequin modeling method 307

Tests and results The data we used for testing and verification include two sets: one set of 40 is chosen from the 700 samples from which our templates are deduced; the other set of two females is chosen out of the sample set. Experiments show that in both test sets our system gives out good results with the maximum error occurs at the measurement of the waist segments, but it is less than 2 percent, which is acceptable in apparel industry. We have also compared the garment prototype pattern generated by our virtual human mannequin with that generated by the traditional experimental equations.

Figure 8. Auto-positioning of the ten sections on 2D photos from front and side view

IJCST 21,5

308

Figure 9. The virtual human mannequin generated by TMSF

The traditional garment prototype uses only girth of bust and body height as the input of experimental equations to generate the garment pattern, thus if the two subjects have the same girth of bust and body height, they will surely have the same prototype pattern despite their differences in other parts’ of the body. Using the virtual human mannequin to generate personalized garment prototype pattern can be a good make up for this. Table VII gives out the major measurements of garment prototype pattern from two models of the same height and similar girth of bust. The left is generated from personalized human mannequin generated by our system and the right is generated from traditional garment pattern prototype equations. The traditional garment prototype we used here is Donghua Prototype (a widely adopted garment prototype in apparel industry). The definitions of the parameters are shown in Figure 10. From Table VII, we observed that due to the same height and similar girth of bust, there are little differences in measurements between the two selected models in Donghua garment prototype; but in our personalized prototype, there are many obvious differences in the measurements of these two models. The major differences are:

Parameters (cm)

Table VII. The measurements comparison between personalized human mannequin and traditional garment prototype

Height Girth of bust Length of back Width of front bust Width of back bust Qhk Hhk Qzk Hzk Qfyl Hfyl

Personalized human mannequin Model 1 Model 2 160 90.94 37.66 18.64 18.48 6.44 6.34 7.45 3.82 4.58 1.82

160 89.61 39.36 18.43 16.83 5.39 6.1 8.38 2.92 2.92 1.42

Traditional garment prototype Model 1 Model 2 160 90.94 37.5 17.62 18.82 7.15 7.35 7.85 2.45 4.27 1.77

160 89.61 37.5 17.45 18.65 7.08 7.28 7.78 2.43 4.24 1.74

P8

qhk

22° P7

P6 qfy1 P4

P0:

Length of shoulder

.

.

.

hhk P6

Width of front bust P2 Y 1/4*girth of bust+3

X hzk

P1

P1

P5

P3

P9

X qzk

Length of back

P0

P8 P7 hfy1 P5

Mannequin modeling method 309

P4 Width of back bust

P3

P2 Y 1/4*girth of bust+3

Difference in length of the back. Model 2 has a longer upper-half body compared to model 1, so she has a longer back length. Differences in neckline related parameters. The two models have different neck length and girth of neck. Differences in qfyl and hfyl. The shapes of the bust for the two models are different. The length between the two nipple points of model 1 is relatively longer than that of model 2.

From the comparison above, we can see that garment prototype pattern generated from personalize human mannequin takes more details of human body into consideration, thus can better meet the fitness needs of customers. Conclusion Unlike many other generic 3D human body modeling techniques which are mainly focus on animation, we proposed a section-assemblage based human mannequin generation method which is both easy to implement and useful for fashion industry. With the simple input requirement (only two photos and the girth difference of bust and waist) and less system storage requirements (only 60 section templates need to be stored compares to that of a standard human body template which needs to store all the scanned point data of the body template), it is much easy to access than most of the 3D human body modeling techniques. Sampling, modeling and testing of more than 700 human mannequins have shown that the error of our technique is less than 2 percent, which is within the permission of apparel industry. Two models’ comparison shows that the patterns generated from the virtual human mannequin also have a better personalization than the traditional prototype equations. There are two aspects left for further research: one is the improvement of parts’ template; the other is the application related direction. For the parts’ template, because we have only done experiments on a sample set which is limited in eastern China area, when we use the section templates generated on this sample set to produce virtual human mannequin from larger area, some new feature parameter might need to be taken into consideration, thus brings to the change or modification of the section templates, but the

Figure 10. Definitions for garment prototype patterns

IJCST 21,5

310

idea of section templates assemblage still remain workable; From the application point of view, now we are considering how to make it more practical and effective to e_MTM industry. We are trying to associate the virtual human mannequin with the specific style to give out a customized fashion effect of the style. References Allen, B., Curless, B. and Popovic, Z. (2003), “The space of human body shapes: reconstruction and parameterization from range scans”, ACM Transactions on Graphics, Vol. 22 No. 3, pp. 587-94. Charlie, C.L.W. (2005), “Parameterization and parametric design of mannequins”, Computer-aided Design, Vol. 37 No. 1, pp. 83-98. Gu, J., Chang, T., Mak, I., Gopalsamy, S., Shen, H.C. and Yuen, M.M.F. (1998), “A 3D reconstruction system for human mannequin modeling, modeling and motion capture techniques for virtual environments”, Lecture Notes in Artificial Intelligence, Springer, London, pp. 229-41. Kakadiaris, I.A. and Metaxas, D. (1998), “Three-dimensional human mannequin model acquisition from multiple views”, International Journal of Computer Vision, Vol. 30 No. 3, pp. 191-218. Kim, S. and Park, C. (2004), “Parametric body model generation for garment drape simulation”, Fibers and Polymers, Vol. 5 No. 1, pp. 12-18. Lee, W., Gu, J. and Magnenat-Thalmann, N. (2000), “Generating animatable 3D virtual humans from photographs”, Computer Graphics Forum, Vol. 19 No. 3, p. C1. Mori, G. and Malik, J. (2006), “Recovering 3D human mannequin configurations using shape contexts”, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 28 No. 7, pp. 1052-62. Seo, H. and Magnenat-Thalmann, N. (2003), “An automatic modeling of human bodies from sizing parameters”, Proceedings of the 2003 Symposium on Interactive 3D Graphics, Monterey, CA, pp. 19-26. Seo, H., Yeo, Y.I. and Wohn, K. (2006), “3D body reconstruction from photos based on range scan”, Proceedings of the Technologies for E-learning and Digital Entertainment – First International Conference, Edutainment 2006, Lecture Notes in Computer Science, LNCS, Vol. 3942, pp. 849-60. Shen, J., Sun, S., Huang, Q. and Pan, Y. (2002), “Virtual human body construction based on shape from silhouette”, Journal of China Image and Graphics, Vol. 10, pp. 1089-93. Stylios, G.K., Han, F. and Wan, T.R. (2001), “A remote, on-line 3-D human measurement and reconstruction approach for virtual wearer trials in global retailing”, International Journal of Clothing Science & Technology, Vol. 13 No. 1, pp. 65-75. Corresponding author Jiyun Li can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm

Compression plate buckling behavior of fused fabric composites

Compression plate buckling behavior

B. Namiranian and S. Shaikhzadeh Najar Department of Textile Engineering, Amirkabir University of Technology, Tehran, Iran, and

A. Salehzadeh Nobari Department of Aerospace Engineering, Amirkabir University of Technology, Tehran, Iran

311 Received 21 September 2008 Revised 4 January 2009 Accepted 4 January 2009

Abstract Purpose – The purpose of this paper is to evaluate some important parameters in plate buckling of fused interlining worsted fabric with different weight and laying-up direction. The article compares the formability of fused fabric composite by two different methods (Lindberg’s hypothesis and fabric assurance by simple testing method). Design/methodology/approach – Plate buckling compression behavior of fused fabric composite is investigated using a special designed clamp according to Dahlberg’s test method. Findings – The result shows that fusible interlining lay-up angle significantly influences on buckling parameters. It is indicated that the buckling behavior of fused fabric composite against lay-up interlining direction is in accordance with interlining buckling behavior. The result of research suggests that the formability behavior of fused fabric composite with interlining lay-up direction is predictable according to Lindberg’s method. Research limitations/implications – Experimental design is limited at low speed. Further research works are needed to perform buckling behavior of fused fabric composites at higher speeds as well as under cyclic loading conditions. Originality/value – Compression plate buckling behavior of fused interlining fabrics is predictable against interlining laying-up direction. The result of this research could be used in the area of garment quality serviceability. Keywords Buckling, Mechanical properties of materials, Fabric testing Paper type Research paper

Introduction Mechanical properties of textile fabric influence its performance during actual use and making-up process in garment manufacturing. In making garments, building multilayer structures with fusible interlining is crucial in order to introduce desired shape, bulk, and stiffness in end product. Buckling behavior of the fused interlining fabric is the most significant property which characterizes the shape of cuff and collar. The buckling behavior of the fused fabric also influences the overfeed in stitching of garments and their resultant quality (Lindberg et al., 1960). Compression buckling is important in determining fabric formability (Lindberg et al., 1960). Dahlberg and Eeg-Olefsson (1958) firstly constructed an instrument for The authors would like to thank Mr Emami and Mr Dokhanchi of Iran-Merinos Textile Factory for providing experimental facilities of FAST method for this paper.

International Journal of Clothing Science and Technology Vol. 21 No. 5, 2009 pp. 311-325 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220910995738

IJCST 21,5

312

measuring plate buckling as an attachment to the Instron machine. Lindberg et al. (1960) evaluated important mechanical properties and problems in garment construction. They measured plate and shell buckling of fabric, according to Dahlberg’s test method. In their studies, buckling curves were obtained and several parameters such as buckling load and ratio of shell buckling load to plate buckling load for different fabrics were discussed. They also introduced formability concept that can be determined with compression buckling curve and bending rigidity. Later, Dahlberg (1961) described an apparatus which can be used for measuring both plate and shell buckling, and evaluated the influence of deformation rate, sample length, and corrugation radius on buckling compressibility of several fabrics in different direction. Grosberg and Swani (1966) also investigated the plate buckling behavior of cotton and worsted fabrics and compared the bending properties obtained by two cantilever and buckling test methods. There are several studies in the field of evaluation of mechanical properties of multilayer textile structures. Shishoo et al. (1971) studied mechanical properties of multilayer textile material and their relation to the properties of its components. Kanayama and Niwa (1982) presented a theoretical model for bending and shear properties of fused fabric used in the front bodies of jackets. Dhingra and Lau (1996) analyzed bending, shear, and tensile deformation/recovery characteristics of fused fabric with a view to predicting the influences of face fabric and interlining on the basic mechanical properties of the composite structure. Fan et al. (1997a, b, c) investigated compatibility of face and fusible interlining fabrics in tailored garment. They established the desirable range of mechanical properties of fused fabric composites and obtained approximated empirical relationship between the mechanical properties of fused fabric composites and those of the face and fusible interlining fabrics. Simona and Gersak (2004) applied finite element method for modeling a fused panel drape. The results for the numerical simulation of a fused panel in comparison with experimental results confirmed the applicability and reasonableness of the finite element method. Recently, Urblis et al. (2005) studied the trends in the redistribution of tensions between the individual components of bi-component textile fabric systems (fused, stitched or simply assembled together fabric) under constant load. The previous researches have mainly focused on the mechanical properties of fused fabrics, but no attention was paid to compressional buckling behavior of fused fabric composite. In this study, we evaluate the buckling behavior and mechanical properties of fused worsted fabric and its components from point view of interlining lay-up direction. The results of this study suggest useful information in garment manufacturing as well as garment appearance. Materials Three different worsted fabrics (A, B, and C) and dot fusing non-woven interlining (I) samples used in this research as listed in Table I. As shown in Figure 1, the fusible interlining fabric was placed over face fabric at five different laying-up directions (warp-warp, warp-22.58, warp-458, warp-67.58, and warp-weft). The fusing process was carried out in a HASHIMA HP-30PS machine according to straight linear creasing press method (temperature ¼ 1858C, pressure ¼ 6 bar, and time ¼ 10 s). All materials were conditioned in a standard atmospheric of 20 ^ 28C, 65 ^ 1 percent RH, before being tested. The fabric thickness measured by fabric assurance by simple testing (FAST) tester (de Boos and Tester, 1994) at a load of 2 gf/cm2.

Sample code IA IB IC A B C I

Weight (g/m2) 317 281 264 253 216 196 70

(2.31) (3.87) (2.15) (2.19) (2.13) (0.53) (5.85)

Thickness (mm)

Weave

Fabric density (cm2 1) Warp Fill

Compression plate buckling behavior

Fiber content

313 0.57 0.46 0.39 0.26

(0.005) (0.007) (0.01) (0.005)

Twill2/1z Twill2/1z Twill2/1z Non-woven (thermo-bounded)

25 31 30

19 25 22

20% Wool/80% polyester 20% Wool/80% polyester 20% Wool/80% polyester 100% Polypropylene Table I. Fabric specification

Note: The data in the brackets are SD values

Interlining

Warp-warp

Warp-22.5°

Warp-45°

Warp-67.5°

Warp-weft

1. Mechanical properties We used the FAST system (de Boos and Tester, 1994) to measure the bending rigidity and extensibility of the worsted fabrics, fusible interlining and fused fabric composites. All tests were conducted in a standard conditions (208C and 65 percent RH) according to the standard procedure specified in the FAST manual (de Boos and Tester, 1994). The reproducibility of the results was similar to that reported in the FAST manual. The bending properties of fused fabric composites were carried out in both face and back views, respectively, with different interlining laying-up directions. Table II lists the mechanical properties of fused fabric composite and its components. 2. Buckling test The experimental load-deflection curves during plate buckling for worsted fabrics, fusible interlining and fused worsted fabric composites were obtained using an Instron tensile tester (Model 5566) employing a designed attachment based on that used by Dahlberg (1961). The materials were cut in a strip of 10 £ 10 cm for plate buckling test. Figure 2 shows principal scheme of buckling tester clamps and experimental region. An upper clamp was suspended at center from the load cell “LC” A special lower clamp was mounted on the base of instrument. Lower clamp could be moved sideways within its bracket. Also, the bracket was capable of an angular movement about the holding base. The position of lower clamp could therefore be adjusted to ensure vertical alignment of specimen using a precision scale attached to the upper clamp. Before starting the test, the lower clamp is aligned with upper one in vertical position. The instrument was calibrated, and then the sample was first positioned in the upper clamp at a pre-determined pre-tension. Then the sample clamped between the jaws initially separated at the required distance (l ¼ 2.5 cm).

Figure 1. Face and fusible interlining fabrics with different lay-up directions

Table II. Fabric mechanical properties Fill

Warp

0.5 (0.00) 1.4 (0.05) 0.9 (0.00) 0.7 (0.05) 1.5 (0.11)

214.65a (4.59) 177.73b (7.79) 0.5 (0.00) 1.4 (0.05) 1 (0.20) 0.7 (0.10) 1.7 (0.1) 234.83a (11.13) 123.33b (19.63) 0.5 (0.05) 1.3 (0.00) 0.9 (0.05) 0.6 (0.05) 1.35 (0.11) 206.40a (21.62) 172.45b (3.64) 1.8 (0.15) – – – 1.6 (0.05) 18.66 (1.16) 2.5 (0.05) – – – 1.2 (0.05) 14.06 (1.82) 1.9 (0.1) – – – 1.2 (0.05) 10.47 (0.94) 0.4 (0.15) 3.3 (0.36) 1.7 (0.41) 0.8 (0.05) 4.5 (0.15) 37.64 (1.33)

Extensibility at 100 gf/cm load (%) 22.58 458 67.58

164.23a (0.00) 177.73b (0.00) 157.31a (2.78) 140.77b (6.82) 148.14a (21.21) 145.84b (36.36) 14.94 (1.70) 8.19 (0.43) 7.86 (0.31) 15.05 (2.28)

206.89a (4.37) 199.32b (33.60) 213.12a (19.16) 163.52b (13.14) 199.28a (11.76) 159.99b (15.83) – – – 31.21 (4.73)

Bending rigidity (mN m) 458 67.58

139.33a (6.99) 161.62b (16.29) 171.19a (10.40) 121.26b (6.08) 131.49a (16.01) 122.13b (10.48) – – – 11.78 (2.23)

22.58

Notes: aFused fabric back side; bfused fabric face side; the data in the brackets are SD values

A B C I

IC

IB

IA

Warp

314

Sample code

122.40a (0.00) 145.30b (3.44) 124.37a (14.46) 113.20b (0.00) 111.30a (2.79) 94.14b (19.03) 16.21 (0.71) 10.77 (1.80) 10.11 (1.06) 9.84 (0.73)

Fill

IJCST 21,5

Compression plate buckling behavior

L.C

Interlining

Face fabric

Upper clamp Lower clamp

315

Bracket Holding base (a) (b) Notes: (a) Schematic of buckling tester clamps; (b) experimental region; (c) buckling shape observation

(c)

Figure 2.

When the upper cross head is moved downwards at a constant speed of 1 cm/min, the buckling force is registered by the load cell (L.C). A recovery graph is obtained by reversing the cross head movement at a given deflection of 40 percent. In this research work, five tests are performed for each specimen. 3. Buckling behavior of fused interlining fabric Figure 3 shows the buckling curves of fused composite, face, and fusible interlining fabrics for one cycle loading. All three buckling curves have some similarities in shape. The slope of the curve at the beginning represents that compressibility of fused fabric composite, fusible interlining, and face fabric increase, respectively. It is shown that interlining as well as face fabric is more compressible than fused fabric composite. All curves after beginning rises up to the highest point or critical buckling load. Obviously, it can be noticed that critical buckling load of fused fabric composite is more than face and fusible interlining. It may be considered that up to 4 percent deflection, a non-linearity trend is observed in post-buckling curve for fused fabric composite. After that the compression load for fused fabric composite increases slowly. Experimental observations have shown that fused fabric composite buckled with interlining on the convex side (Figure 2(c)). However, the compression buckling curves for face and

Load (cN)

Load (cN)

140 90 40 –10 –60 –110

28

IA IB IC

190

0

10

20

30

Deflection (%)

40

28

23

23

18

18

13

A

8

B C

3 –2 0 –7

10

20

30

40

Deflection (%)

–12

(a) (b) Notes: (a) Fused fabric composite; (b) face fabric; (c) fusible interlining

Load (cN)

240

I

13 8 3 –2 0 –7

10

20

30

40

Deflection (%)

–12

(c)

Figure 3. One cycle of buckling test

IJCST 21,5

316

fusible interlining fabrics are different. As shown in Figure 3(b), for fabric A with a higher weight, the load in post buckling zone is increased with compression deflection. For fabrics B and C with a lower weight, the compression load in this zone is suddenly dropped and then gradually increased with compression deflection. In the case of fusible interlining (I), the curve in post buckling zone shows a decrease in load with compression deflection. After 40 percent deflection when the load is removed, recovery curve obtained and a marked hysteresis shows particularly for fused fabric composite. To evaluate buckling behavior of fused fabric composite and its components, some important parameters including critical buckling load, load at 4, 20 and 40 percent deflection, buckling energy, wrinkle remain, and hysteresis obtained from buckling curves. Tables III and IV show buckling parameters for fused fabric composite and its components, respectively. Formability evaluation Formability is a measure of the extent to which a fabric can be compressed in its own plane before it buckles. The formability is obtained and compared by two different methods. We assess formability according to Lindberg et al.’s (1960) hypothesis as equation (1): FormabilityðF C Þ ¼ Bending rigidity £ compressibility £ 10

ð1Þ

With formability in mm2 %, bending rigidity in mN m, and compressibility as inversed slope of the first portion of the buckling curve expressed as %/cN. It is also defined the formability parameter as the product of the bending rigidity and the extensibility of the fabric at a low load (5 and 20 gf/cm) in the FAST system according to equation (2): FormabilityðFÞ ¼ Bending rigidity £

extensionð20Þ 2 extensionð5Þ 14:7

ð2Þ

With formability in mm2 %, bending rigidity in mN m, and extension in percentage. The result of formability parameters for fused fabric composite, interlining, and face fabric based on FAST (de Boos and Tester, 1994) and Lindberg et al. (1960) methods are shown in Table V. It should be notified that, the bending rigidity values of fused fabric composite with interlining on the convex side is substituted in equations (1) and (2) as observed during buckling test. These bending rigidity values have greater validity in clothing construction as mentioned by Kanayama and Niwa (1982). It is also noted that in all cases of fused fabric composite, face fabric was considered in warp direction and interlining fabric was placed over face fabric in different lay-up directions (Figure 1). For this reason, the formability values for face fabric are only presented along warp direction in Table V. Results and discussion In order to evaluate the effects of fusible interlining lay-up direction and fabric weight on plate buckling behavior, the results of experiments were statistically analyzed using analysis of variances (ANOVA) and least square deviation (LSD) test methods at 95 percent confidence limit. The results will be discussed in detail as below.

0.63 0.58 1.02 6.86 1.83 0.43 0.58 14.49 1.34 0.87 0.46 0.55 2.31 1.60 1.80

(0.02) (0.07) (0.27) (0.9) (0.89) (0.02) (0.06) (3.54) (0.00) (0.17) (0.05) (0.12) (0.17) (0.63) (0.16)

Wrinkle remain (%) 137.95 120.14 74.40 54.00 53.00 146.42 81.06 37.16 45.46 50.00 143.60 106.40 68.20 49.32 47.56

(8.54) (7.36) (4.5) (10.03) (2.7) (7.58) (11.42) (2.94) (2.94) (3.0) (22.97) (16.32) (11.83) (12.06) (3.87)

Critical buckling load (cN) 156.20 136.22 91.26 64.16 62.78 158.48 97.86 42.66 50.92 58.50 153.90 117.38 76.00 53.76 52.24

(8.02) (8.78) (9.9) (12.32) (4.38) (8.27) (8.34) (0.8) (2.73) (5.57) (24.12) (17.39) (14.53) (14.13) (7.09)

Load at 4% deflection (cN) 186.65 (7.52) 157.38 (11.60) 106.62 (14.26) 78.90 (15.72) 75.82 (7.64) 176.14 (7.04) 111.98 (6.53) 48.42 (1.98) 59.36 (1.58) 67.90 (8.35) 161.64 (24.39) 128.58 (20.48) 86.42 (19.15) 60.18 (16.11) 60.4 (10.76)

Load at 20% deflection (cN) 201.22 167.86 117.52 90.90 86.10 184.18 120.70 53.60 68.95 74.86 166.16 135.22 93.92 66.30 67.44

(8.66) (14.43) (16.77) (18.76) (9.81) (7.93) (6.35) (2.99) (1.85) (10.14) (25.08) (21.59) (21.81) (17.04) (12.64)

Load at 40% deflection (cN) 7,288.95 6,176.36 4,203.44 3,112.78 2,989.46 6,942.22 4,413.16 1,922.94 2,353.60 2,682.28 6,428.74 5,089.74 3,418.74 2,396.40 2,396.08

31.03 32.86 39.55 56.56 46.83 23.00 33.00 67.56 43.79 38.25 27.07 32.24 45.46 51.33 45.03

(2.44) (4.29) (8.06) (17.32) (7.03) (1.85) (2.86) (13.66) (1.57) (5.59) (4.92) (6.09) (14.24) (13.63) (12.06)

Hysteresis (%)

(294.45) (459.14) (549.38) (612.19) (290.70) (286.63) (260.30) (153.46) (63.46) (321.19) (971.04) (760.59) (740.73) (631.51) (409.02)

Buckling energy (cN %)

Notes: aWarp direction, b22.58, c458, d67.58, efill direction, finterlining lay-up direction; the data in the brackets are SD values

IAa,f IAb IAc IAd IAe IBa IBb IBc IBd IBe ICa ICb ICc ICd ICe

Sample code

Compression plate buckling behavior 317

Table III. Plate buckling characteristics for fused fabric composite

IJCST 21,5 Sample code

318

Table IV. Plate buckling characteristics for face and interlining fabrics

A *a Ab Ac Ad Ae Ba Bb Bc Bd Be Ca Cb Cc Cd Ce Ia Ib Ic Id Ie

Wrinkle remain (%0029 2.94 3.16 0.30 0.21 0.41 7.60 3.73 0.51 0.60 0.63 7.17 5.22 0.55 0.66 0.60 0.53 1.11 4.89 3.04 0.68

(0.72) (1.23) (0.03) (0.04) (0.12) (2.95) (0.73) (0.00) (0.07) (0.29) (0.75) (1.92) (0.04) (0.11) (0.15) (0.07) (0.55) (2.52) (0.81) (0.01)

Critical buckling load (cN)

Load at 4% deflection (cN)

Load at 20% deflection (cN)

4.76 (0.79) 4.52 (0.49) 23.86 (0.83) 29.90 (7.43) 14.56 (4.32) 2.42 (0.82) 2.90 (0.67) 11.84 (1.04) 10.82 (1.66) 8.34 (1.22) 2.37 (0.22) 2.10 (0.23) 8.28 (0.94) 8.62 (0.95) 9.88 (1.72) 22.82 (4.76) 15.12 (4.5) 7.64 (0.63) 6.74 (1.81) 7.12 (1.7)

4.94 (0.78) 4.72 (0.44) 23.18 (1.16) 28.46 (7.57) 13.85 (4.30) 2.39 (0.62) 3.02 (1.17) 11.45 (0.99) 10.43 (1.73) 7.90 (1.03) 2.09 (0.15) 1.93 (0.20) 7.91 (0.89) 8.06 (0.79) 9.25 (1.49) 21.44 (4.63) 13.60 (4.44) 6.78 (0.92) 2.98 (1.76) 6.48 (1.92)

6.91 (0.84) 6.34 (0.43) 24.36 (1.2) 29.32 (7.53) 14.82 (4.17) 3.59 (1.19) 4.12 (1.20) 12.29 (0.94) 11.23 (1.78) 9.01 (0.97) 3.04 (0.15) 2.66 (0.22) 8.56 (0.99) 8.84 (0.77) 10.38 (1.43) 20.22 (4.54) 12.48 (4.78) 6.58 (1.08) 3.00 (1.57) 6.66 (2.1)

Load at 40% deflection (cN)

Buckling energy (cN %)

Hysteresis (%)

9.65 (0.75) 280.11 (33.37) 42.21 (5.36) 8.65 (0.36) 257.53 (16.79) 45.42 (2.99) 26.14 (1.26) 979.74 (140.11) 11.45 (1.45) 30.78 (7.51) 1,178.55 (438.51) 8.21 (2.84) 16.21 (4.16) 596.10 (189.14) 16.63 (4.2) 5.43 (1.21) 147.38 (46.47) 46.38 (8.36) 5.73 (1.21) 168.30 (38.74) 37.94 (8.7) 13.51 (0.95) 494.27 (17.41) 16.48 (2.83) 12.51 (1.81) 453.07 (71.77) 16.82 (4.23) 10.74 (1.41) 364.81 (45.17) 23.30 (5.06) 4.57 (0.24) 125.27 (6.47) 50.64 (0.85) 3.90 (0.20) 110.81 (8.87) 48.11 (3.94) 9.63 (0.96) 345.96 (38.38) 16.66 (2.52) 10.07 (0.81) 357.17 (42.31) 21.21 (4.73) 12.01 (1.56) 418.56 (59.83) 21.70 (5.06) 19.96 (4.56) 821.22 (183.55) 25.27 (9.48) 12.34 (5.03) 511.58 (191.42) 46.05 (25.96) 6.90 (0.96) 269.18 (39.14) 52.57 (14.99) 3.24 (1.77) 123.22 (64.76) 65.43 (22.62) 7.18 (2.25) 270.48 (84.61) 33.42 (3.80)

Notes: * Significant at the 0.05 level; awarp direction, b22.58, c458, d67.58, efill direction; the data in the brackets are SD values

1. Fusible interlining lay-up direction A summary of ANOVA statistical result of buckling parameters for fusible interlining lay-up direction is presented in Table VI. It is shown that the fusible interlining lay-up direction significantly affected all-important buckling parameters. As shown in Figure 4, for fused fabric composite IA and IC, with increase of lay-up angle from 08 to 67.58, buckling load and energy dramatically decreased and after that these buckling parameters reach to a steady condition. However, for fused fabric composite IB, the lowest buckling load and energy is obtained at 458 lay-up direction angle. This result is attributed to the buckling behaviour of interlining component with lay-up direction as also shown in Figure 4. It may be considered that at 08 lay-up angle, the fibre in the interlining is positioned along compression buckling force and hence resists more against compression buckling. However, when interlining is at bias direction, interlining component has less contribution to compression buckling force and as a result buckling force is decreased. In general, the buckling load and energy against lay-up direction angle have almost similar non-linear trends. The LSD test results of buckling parameters for fused fabric composite at different interlining lay-up direction is represented in Table VI. According to the results of LSD test, buckling load and energy in the level of 0 and 22.58 lay-up direction angle is significantly different with all other lay-up direction levels. But the different of these parameters in the levels of 458, 67.58, and 908 lay-up direction are in significant at 5 percent level.

IA IB IC I A B C

Fabric code

1.46 0.79 0.70 0.12 0.57 0.66 0.32

10.73 10.33 9.49 9.03 24.89 37.50 25.65

Warp F (mm2 %) FC (mm2 %) 0.94 2.91 1.78 0.32 – – –

17.41 25.10 17.53 3.92 – – –

1.11 1.07 1.51 0.25 – – –

14.78 28.31 14.81 15.05 – – –

Lay-up direction 22.58 458 F (mm2 %) FC (mm2 %) F (mm2 %) FC (mm2 %) 1.40 1.44 1.35 0.21 – – –

11.63 16.33 12.45 39.01 – – –

67.58 F (mm2 %) FC (mm2 %)

Fill FC (mm2 %) 15.913 16.16 14.84 7.38 – – –

Compression plate buckling behavior

F (mm2 %) 2.49 2.96 1.89 0.40 – – –

319

Table V. Result of formability for fused fabric composite, face and fusible interlining fabric according to FAST (F) and Lindberg (FC) methods

IJCST 21,5

320

As shown in Figure 5(a), un-like to buckling load and energy, the buckling hysteresis is increased with lay-up direction angle up to 67.58 and then decreased at 908 lay-up direction angle. It is shown that the trend of buckling hysteresis with lay-up direction angle is accordance with that of fusible interlining fabric. It is necessary to notified that at 67.58 lay-up direction angle the highest and lowest buckling hysteresis and energy is obtained, respectively. For case of wrinkle remain, no specific trend is found as shown in Figure 5(b). It is interesting to notify that at 458 lay-up direction angle the highest wrinkle remain value is obtained. This result illustrates that a level of lay-up direction angle lower than 458 degree is more suitable and preferable for clothing manufacturing. The LSD test results given in Table VII reveal that the levels of 08 lay-up direction for buckling hysteresis is in-significantly different from 22.58 level at 5 percent confidence limit. Similar findings are obtained for level of 458 and 67.58 lay-up direction angles. However, for wrinkle remain only the level of 458 lay-up direction is significantly different from other levels at 5 percent level. 2. Fabric weight Table VIII shows a summary of ANOVA statistical result of fused fabric buckling parameters for fabric weight factor. It is presented that the fabric weight significantly affected the critical buckling load, buckling load at 4, 20 and 40 percent deflection, buckling energy and wrinkle remain parameters. But, however, buckling hysteresis has not been affected by fabric weight. Table IX summarizes the LSD test results of buckling parameters for fused fabric composite with different weight. As shown in Table VIII and Figure 4, fused fabric composite with a higher weight (IA) exhibits a higher buckling load and energy than other fused fabric composite samples. This difference is however less pronounced at critical buckling load for sample IA and IC. It is also shown that fused fabric composite IB has significant different buckling parameters with other fused fabric composite samples. According to the results of LSD test given in Table VIII, buckling load and buckling energy of IA, IB, and IC fused fabric composite is significantly different in 5 percent level. However, the different of buckling hysteresis for fused fabric composite IA, IB, and IC is not significant. 3. Formability As shown in Figure 6(a), fused fabric composite (IB) has more formability than other fused fabric composite samples (IC and IA). This finding is attributed to lower critical Interlining lay-up direction

Table VI. The ANOVA test result of fused fabric buckling parameters for interlining lay-up direction

Wrinkle remain Critical buckling load Buckling load at 4% deflection Buckling load at 20% deflection Buckling load at 40% deflection Buckling energy Hysteresis Note: P-value ¼ 0.05

0.000 0.000 0.000 0.000 0.000 0.000 0.000

180

Poly. (IA)

140

Poly. (IB)

120

Poly. (IC)

100

Poly. (I)

Poly. (IA)

160 Buckling load 4% (cN)

Critical buckling load (cN)

160

80 60 40 20

Poly. (IB)

140

Poly. (IC)

120

Poly. (I)

100 80

321

60 40 20

0

0 0

22.5 45 67.5 Lay-up direction (degree)

0

90

22.5 45 67.5 Lay-up direction (degree)

(a)

90

(b) 250

200 180 160 140 120 100 80 60 40 20 0

Poly. (IA)

Poly. (IA) Poly. (IB) Poly. (IC) Poly. (I)

Buckling load 40% (cN)

Buckling load 20% (cN)

Compression plate buckling behavior

200

Poly. (IB) Poly. (IC)

150

Poly. (I) 100 50 0

0

22.5

45

67.5

90

0

22.5

45

67.5

Lay-up direction (degree)

Lay-up direction (degree)

(c)

(d)

90

8,000 Poly. (IA)

Buckling energy (cN%)

7,000

Poly. (IB)

6,000

Poly. (IC) 5,000

Poly. (I)

4,000 3,000 2,000 1,000 0 0

22.5 45 67.5 Lay-up direction (degree)

90

(e)

buckling load values obtained from buckling test. However, the formability of fused fabric composite samples in 08 and 908 lay-up direction angle is likely the same since as shown in Figure 4(a), their critical buckling loads at these lay-up direction angles are almost similar. It is interesting to compare the formability values of fused fabric composite obtained by two different FAST and Lindberg methods as depicted in

Figure 4. Effects of interlining lay-up direction on: (a)-(d) buckling load; (e) buckling energy

80

16

70

14

60 50 40 30

Poly. (IA) Poly. (IB) Poly. (IC) Poly. (I)

20

Figure 5. Effects of interlining lay-up direction on hysteresis and wrinkle remain

Wrinkle remain (%)

322

Hysteresis (%)

IJCST 21,5

10 0 0

22.5 45 67.5 Lay-up direction (degree) (a)

Poly. (IA) Poly. (IB) Poly. (IC) Poly. (I)

12 10 8 6 4 2 0 0

90

22.5 45 67.5 Lay-up direction (degree) (b)

90

Critical Buckling Buckling (I) (J) Wrinkle buckling Buckling load load Buckling Direction Direction remain load load 4% 20% 40% energy Hysteresis LSD

0

22.5

45

67.5 Table VII. Summary of LSD test results for buckling parameters of fused fabric composite at different interlining lay-up direction

90

22.5 45 67.5 90 0 45 67.5 90 0 22.5 67.5 90 0 22.5 45 90 0 22.5 45 67.5

0.847 0.000 * 0.052 0.240 0.847 0.000 * 0.067 0.303 0.000 * 0.000 * 0.000 * 0.000 * 0.052 0.067 0.000 * 0.426 0.240 0.303 0.000 * 0.426

0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.162 0.139 0.000 * 0.000 * 0.162 0.932 0.000 * 0.000 * 0.139 0.932

0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.069 0.096 0.000 * 0.000 * 0.069 0.872 0.000 * 0.000 * 0.096 0.872

0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.090 0.147 0.000 * 0.000 * 0.090 0.799 0.000 * 0.000 * 0.147 0.799

0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.203 0.264 0.000 * 0.000 * 0.203 0.872 0.000 * 0.000 * 0.264 0.872

0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.068 0.146 0.000 * 0.000 * 0.068 0.700 0.000 * 0.000 * 0.146 0.700

0.069 0.000 * 0.000 * 0.000 * 0.069 0.000 * 0.000 * 0.005 * 0.000 * 0.000 * 0.555 0.002 * 0.000 * 0.000 * 0.555 0.011 * 0.000 * 0.005 * 0.002 * 0.011 *

Note: *The mean difference is significant at the 0.05 level

Table IV and Figure 6. It is shown that the formability of fused fabric composite obtained according to Lindberg’s method is more than that of obtained by FAST method. As shown in Table IV, the formability of fused fabric composite obtained by Lindberg’s method is much less than face fabric. However, opposite results is obtained by FAST method. This result indicates that the formability of fused fabric composite achieved according to Lindberg’s method is more valid than that of obtained by FAST method. This finding is due to that the bending rigidity and compressibility has an effective role in calculation of formability according to FAST and Lindberg’s methods,

respectively. In addition, the assumption of similarity of extensibility and compressibility for fused fabric composite cannot be considered. Conclusion The aim of this work was to investigate the effects of fusible interlining lay-up and fabric weight on plate buckling behavior of fused fabric composites. In this research, three different worsted fabrics were fused in five different lay up (warp-warp, warp-22.58, warp-458, warp-67.58, and warp-weft) using non woven fusible interlining.

Compression plate buckling behavior 323

Fabric weight Wrinkle remain Critical buckling load Buckling load at 4% deflection Buckling load at 20% deflection Buckling load at 40% deflection Buckling energy Hysteresis

(I) (J) Wrinkle Fabric Fabric remain LSD

IA IB IC

IB IC IA IC IA IB

0.000 * 0.208 0.000 * 0.000 * 0.208 0.000 *

0.000 0.000 0.000 0.000 0.000 0.000 0.783

Critical buckling load 0.000 * 0.042 * 0.000 * 0.000 * 0.042 * 0.000 *

Buckling Buckling Buckling Buckling load 4% load 20% load 40% energy Hysteresis 0.000 * 0.015 * 0.000 * 0.000 * 0.015 * 0.000 *

0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 *

0.000 * 0.000 * 0.000 * 0.007 * 0.000 * 0.007 *

0.000 * 0.000 * 0.000 * 0.002 * 0.000 * 0.002 *

0.798 0.843 0.798 0.646 0.843 0.646

45

Poly. (IA) Poly. (IB) Poly. (IC) Poly. (I)

35 30 25 20 15 10 5

3.5 Fast formability (mm 2.%)

Lindberg's formability (mm 2.%)

Note: *The mean difference is significant at the 0.05 level

40

22.5 45 67.5 90 Interlining lay-up direction (degree) (a)

Notes: (a) Lindberg; (b) FAST methods

Table IX. Summary of LSD test results for buckling parameters of fused fabric composite at different fabric weight

Poly. (IA) Poly. (IB) Poly. (IC) Poly. (I)

3 2.5 2 1.5 1 0.5

Figure 6.

0

0 0

Table VIII. The ANOVA result of fused fabric buckling parameters at 0.05 P-value for fabric weight factor

0

22.5 45 67.5 90 Formability of fused fabric Interlining lay-up direction (degree) composite and interlining with different interlining (b)

lay-up direction

IJCST 21,5

324

The main important fabric mechanical properties including bending rigidity, extensibility, shear rigidity, and formability were measured using FAST method. After designing plate buckling clamps, the buckling test was performed on an Instron machine under one cycle of compression loading. Important buckling parameters including critical buckling load, buckling loads at 4, 20 and 40 percent deflection, buckling energy, wrinkle remain, and hysteresis were obtained and statistical analysis has been performed on buckling parameters using ANOVA and LSD test method. The results of this research show that with increasing interlining lay-up direction, buckling loads and buckling energy parameters significantly decrease, whereas, hysteresis increase. However, the interlining lay-up direction has no significant influence on wrinkle remain. In addition, the result of this work shows that fabric weight has significant effect on buckling loads, energy, and wrinkle remain. But hysteresis had not show difference with increasing fabric weight. The findings of research suggest that the formability behavior of fused fabric composite with interlining lay-up direction is predictable according to Lindberg’s method. Further research works are needed to perform buckling behavior of fused fabric composites at higher speeds and also under cyclic loading conditions. The shell buckling properties of fused fabric composites would be considered in further studies. References Dahlberg, B. (1961), “Mechanical properties of textile fabrics Part II: buckling”, Textile Research Journal, Vol. 31, pp. 94-9. Dahlberg, B. and Eeg-Olefsson, T. (1958), No. 2, Medd Svenska Textileforsknings institutet. de Boos, A. and Tester, D. (1994), “SiroFAST: fabric assurance by simple testing”, Report No. WT92.02, CSIRO Textile and Fibre Technology, Geelong. Dhingra, R.C. and Lau, K.P. (1996), “Mechanical performance behavior of fused composites”, Textile Asia, Vol. 1, pp. 61-4. Fan, J., Leeuwner, W. and Hunter, L. (1997a), “Compatibility of outer and fusible interlining fabrics in tailored garments Part I: desirable range of mechanical properties of fused composites”, Textile Research Journal, Vol. 67, pp. 137-42. Fan, J., Leeuwner, W. and Hunter, L. (1997b), “Compatibility of outer and fusible interlining fabrics in tailored garments Part II: relationship between mechanical properties of fused composites and those of outer and fusible interlining fabrics”, Textile Research Journal, Vol. 67, pp. 194-7. Fan, J., Leeuwner, W. and Hunter, L. (1997c), “Compatibility of outer and fusible interlining fabrics in tailored garments Part III: selecting fusible interlining”, Textile Research Journal, Vol. 67, pp. 258-62. Grosberg, P. and Swani, N.M. (1966), “The mechanical properties of woven fabrics Part IV: the determination of the bending rigidity and frictional restraint in woven fabrics”, Textile Research Journal, Vol. 4, pp. 338-45. Kanayama, M. and Niwa, M. (1982), “Mechanical behavior of the composite fabrics reinforced by fusible interlining”, Proceedings of the Australia-Japan Joint Symposium on Objective Specification of Fabric Quality, Mechanical Properties and Performance, TMSJ, New Hope, MN, pp. 347-70. Lindberg, J., Waesterberg, L. and Svenson, R. (1960), “Wool fabrics as garment construction materials”, Journal of the Textile Institute, Vol. 51, pp. 1475-93.

Shishoo, R., Klevmar, P.H., Cednas, M. and Olofsson, B. (1971), “Relationship between the properties of a textile composite and its components”, Textile Research Journal, Vol. 5, pp. 669-79. Simona, J. and Gersak, J. (2004), “Modeling the fused panel for a numerical simulation of drape”, Fibres & Textiles in Eastern Europe, Vol. 12, pp. 47-52. Urblis, V., Petrauskas, A. and Vitkauskas, A. (2005), “Study into the redistribution of tension on the components of the loaded textile fabric system”, Fibers & Textiles in Eastern Europe, Vol. 13, pp. 38-42. Further reading Anandjiwala, R.D. and Gonsalves, J.W. (2006), “Nonlinear buckling of woven fabrics Part I: elastic and nonelastic cases”, Textile Research Journal, Vol. 76, pp. 160-8. Carr, H. and Latham, B. (2004), Technology of Clothing Manufacture, Blackwell Science, London, p. 188. Corresponding author S. Shaikhzadeh Najar can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

Compression plate buckling behavior 325

International Journal of Clothing Science and Technology

ISSN 0955-6222 Volume 21 Number 6 2009

International textile and clothing research register Editor George K. Stylios

Access this journal online __________________________

2

Editorial advisory board ___________________________

3

Editorial __________________________________________

4

Research register __________________________________

6

CONTENTS

Research index by institution_______________________ 118 Research index by country _________________________ 119 Research index by subject__________________________ 120 Research index by principal investigator ____________ 123

This journal is a member of and subscribes to the principles of the Committee on Publication Ethics

Access this journal electronically The current and past volumes of this journal are available at:

www.emeraldinsight.com/0955-6222.htm Find more articles from this journal and search through Emerald’s 200+ journals at: www.emeraldinsight.com See page following contents for full details of what your access includes

EDITORIAL ADVISORY BOARD Professor Jaffar Amirbayat Amirkabir University of Technology, Tehran, Iran Professor H.J. Barndt Philadelphia College of Textiles & Science, Philadelphia, USA Professor Mario De Araujo Minho University, Portugal Professor Dexiu Fan China Textile University, Shanghai, China Professor P. Grosberg Shankar College of Textile Technology and Fashion, Israel Professor Carl A. Lawrence University of Leeds, UK Professor Gerald A.V. Leaf Heriot-Watt University (Hon), UK

Professor David Lloyd University of Bradford, UK Professor Masako Niwa Nara Women’s University, Japan Professor Isaac Porat School of Textiles, University of Manchester, UK

Editorial advisory board

3

Professor Ron Postle The University of New South Wales, Australia Professor Rosham Shishoo Swedish Institute for Fibre and Polymer Research, Mo¨lndal, Sweden Professor Paul Taylor University of Newcastle, Newcastle upon Tyne, UK Professor Witold Zurek Ło´dz´ Technical University, Poland

Professor Trevor J. Little North Carolina State University, USA

International Journal of Clothing Science and Technology Vol. 21 No. 6, 2009 p. 3 # Emerald Group Publishing Limited 0955-6222

IJCST 21,6

4

International Journal of Clothing Science and Technology Vol. 21 No. 6, 2009 pp. 4-5 q Emerald Group Publishing Limited 0955-6222

Editorial The International Textile and Clothing Research Register (ITCRR) championing the research efforts of the community The International Textile and Clothing Research Register (ITCRR) is now in its 14th year of publishing the research efforts of our community. You can see the breadth of activity, the large funding involved, the hot topics, and the research players in the field of textiles and clothing research. By doing so ITCRR provides a platform of participation and dissemination to those working in our discipline and avoids duplication of effort. Again as you will see in this new edition, textiles and clothing research and practice is increasing in volume, in quality and in diversity. All good news for all of us involved in the field. We try to capture as many projects as possible, but I understand that there maybe projects not having been registered. I will welcome them in our next issue and I invite them to send me their project details anytime during the year for the next ITCRR issue. Research, development and innovation can, without doubt, give us more wisdom, enable our industries to become more competitive, and contribute to our quality of life. I believe that registering research projects will provide the due credit to originators of the research and contribute to the future development of this field. Groups of expertise can be identified in this manner, repetition and re-invention can be avoided leading to best utilisation of time and funding for faster and better-directed research. Textiles and clothing originate from the physiological need to protect ourselves from the environment. New challenges, however, are already upon us with nanotextiles; nanofibres, nanocoatings, with multifunctional and smart textiles and clothing, and with wearable electronics. In this year’s Issue 5 of the IJCST, we published a special issue on technical, functional textiles and clothing. We continue to welcome contributions from textile and clothing aesthetics, design and fashion highlighting our belief that design and technology go hand in hand, and we are working with a special issue in textile and clothing design and technology. IJCST was set up 21 years ago as a platform for the promotion of scientific and technical research at an international level. Our original statement that the manufacture of clothing needs to change to more technologically advanced forms of production and retailing still stands. IJCST has, however, evolved further by also providing opportunities in the new research areas of nanotextiles, SMART textiles and clothing, and in wearable electronics. The journal, now fully indexed in SCI continues with its authoritative style to accredit original technical research, adhering to our refereeing processes however difficult these may prove at times. IJCST will be instrumental in continuing to support conferences and meetings from around the world in its effort to promote the science and technology of clothing. I would like to thank our research community and those authors in particular that have contributed to this volume, our editorial board for their continuous support, and our colleagues who have acted in a refereeing capacity and have given us their free time and expertise to progress our research efforts. I take the liberty to list some of

those names below (apologies in advance if anyone has accidentally been omitted from this list): . Professor Paul Taylor, University of Newcastle (Emeritus) . Professor Isaac Porat, UMIST . Professor R.H. Wardman, Heriot Watt University . Professor R. Christie, Heriot Watt University . Dr Taoruan Wan, University of Bradford . Professor David Lloyd, University of Bradford . Professor G.A.V. Leaf, Heriot-Watt University (Visiting) . Dr David Brook, University of Leeds . Dr Jaffer Amirbayat, UMIST . Professor Jintu Fan, Hong Kong Polytechnic University . Dr Jelka Gersak, University of Maribor . Dr Hua Lin, Nottingham University . Dr Sharon Lam Po Tang, Heriot Watt University . Dr Lisa McIntyre, Heriot Watt University . Dr T. Wan, Heriot Watt University . L. Luo, Heriot Watt University George K. Stylios Editor-in-Chief

Editorial

5

Research register

IJCST 21,6

Arak, Iran 6

Arak Islamic Azad University, Daneshgah Street – Islamic azad University – Arak Branch – Textile Department, 38135/567. Tel: +988613670016; Fax: +988613670016; E-mail: [email protected] Principal investigator(s): Arak Islamic Azad University Research staff: Emadaldin Hezavehi

A new electro-mechanical technique for measurement of stress relaxation of polyester blended fabric with constant torsional strain Other Partners: Academic

Industrial

None None Project start date: 12 January 2009 Project end date: 20 March 2009 Project budget: Private Source of support: Arak Islamic Azad University Keywords: Stress relaxation, Torsional strain, Rotational level, Woven fabric Purpose – The aim of this paper is the analysis of the stress relaxation behavior of the different woven fabrics under constant torsional strain in wrinkled state. For this purpose a new method for determination of stress relaxation behavior of the fabric was used while keeping the torsional strain constant. Design/methodology/approach – In this study, the behavior of stress relaxation of fabric is examined with modification of wrinkle force tester (Shaikhzadeh et al., 2009) sophisticated electro-mechanical method and fabricating a device which uses a computer and microcontroller, with constant torsional strain by a rotational level of 9.1 turn/m in 280 degree, and in 300 seconds. Findings – The results depicts that stress relaxation percentage in fabric in weft alignment is more than warp alignment and the fabrics which tolerate more torsional force, posses less stress relaxation percentages. In this way with increasing polyester percentage in fabric the scale of stress relaxation percentage decreases. Also, adoption of data derived from experiments with Maxwell model, shows that the interlaced model is a suitable model for explaining the stress relaxation decline in fabric. Correlation coefficient of fabrics in weft alignment with Maxwell model is more than warp alignment. Practical implications – This study has practical implications in the clothing as well as in technical textiles structures.

Originality/value – Knowing visco-elastic properties is very important. However, there is no information available to study the stress relaxation of woven fabrics under the combined influences of compression and constant torsional strains.

Research register

Aims and objectives The aim of this paper is the analysis of the stress relaxation behavior of the different woven fabrics under constant torsional strain in wrinkled state. For this purpose a new method for determination of stress relaxation behavior of the fabric was used while keeping the torsional strain constant.

Deliverables This study has practical implications in the clothing as well as in technical textiles structures. Publications and outputs IJCST under reviewing.

Arak, Iran Arak Islamic Azad University, Daneshgah Street – Islamic azad University – Arak Branch – Textile Department, 38135/567. Tel: +988613670016; Fax: +988613670016; E-mail: [email protected] Principal investigator(s): Arak Islamic Azad University Research staff: Emadaldin Hezavehi

A new test method to characterize torsional behavior of woven fabrics Other Partners: Academic

Industrial

None None Project start date: 15 March 2008 Project end date: 22 July 2008 Project budget: Private Source of support: Arak Islamic Azad University Keywords: Worsted fabric, Torsional force, Torsional strain, Data acquisition, Micro-controller In this research, an apparatus was designed using data acquisition and micro-controller systems in order to measure torsional force of woven fabrics subjected to combined effects of torsional and compression strains. 6 different worsted wool blended fabric samples were used and then the torsional force of these fabrics were continuously measured along two warp and weft directions using 3 different spiral shafts with 25, 32, and 60 torsional angle degrees, respectively.

7

IJCST 21,6

8

The results showed that the torsional force of worsted fabrics is significantly influenced by used fabric and spiral shaft types. It is shown that with the increase of torsional angle, the torsional force is increased along weft and warp directions. The results indicated that with a spiral shaft of 60 degrees, the torsional force of all fabric samples are almost similar particularly for fabric samples tested along weft direction. The result of this research suggests that using a spiral shaft of 32 degrees is preferable in measuring torsional behavior of worsted fabrics.

Aims and objectives The aim of this research was to investigate the torsional behavior of woven fabrics. An apparatus was designed using data acquisition and micro-controller systems in order to measure torsional force of woven fabrics subjected to combined effects of torsional and compression strains.

Deliverables The results showed that the torsional force of worsted fabrics is significantly influenced by used fabric and spiral shaft types. Publications and outputs Amirkabir Journal is published.

Arak, Iran Arak Islamic Azad University, Daneshgah Street – Islamic azad University – Arak Branch – Textile Department, 38135/567. Tel: +988613670016; Fax: +988613670016; E-mail: [email protected] Principal investigator(s): Research and Science Center, Azad University, Tehran, Iran Research staff: Emadaldin Hezavehi

Investigation into wrinkle behavior of woven fabrics in a cylindrical form by measuring their tangential force Other Partners: Academic

Industrial

None None Project start date: 20 October 2006 Project end date: 8 September 2007 Project budget: Private Source of support: Research and Science Center, Azad University, Tehran, Iran Keywords: Data collection, Wool fabric, Textile technology, Fabric testing, Deformation, Bagging Purpose – The purpose of this paper is to describe a unique approach to investigate the wrinkle force of textile structures in a cylindrical model.

Design/methodology/approach – In this research, an apparatus was designed and constructed in order to investigate the torsional and wrinkle behavior of textile structures in a cylindrical model under a different rotational level using data acquisition and micro-controller systems. Findings – In the light of research results, the fiber and fabric type, fabric physical and mechanical properties and imposed rotational level significantly contributed to wrinkle characteristics of worsted fabrics. It was noticed that with increase of rotational level, the wrinkle force, and energy increased along weft and warp directions. Wrinkle characteristics along warp direction exhibited greater values than in weft direction. Originality/value – The study is aimed at determining wrinkle behavior of worsted fabrics under the combined influences of compression and torsional strains.

Aims and objectives The aim of this paper is to describe a unique approach to investigate the wrinkle parameter of textile structures in a cylindrical model.

Deliverables The results showed that the wrinkle force, energy and hysteresis of worsted fabrics is significantly influenced by fabric physical and mechanical properties and imposed rotational levels. It is shown that with increase of rotational level, the wrinkle force, and energy is increased along weft and warp directions. However, with increasing the rotational levels, the wrinkle hysteresis is significantly decreased both along warp and weft directions. The results indicates that at rotational levels of 25 turn/m value, the wrinkle force of all fabric samples are almost similar particularly for fabric samples tested along weft direction. The results of this research revealed that worsted fabric samples with higher polyester fiber content exhibited higher wrinkle force values. Publications and outputs IJCST is published.

Athens, Greece National Technical University of Athens, Iroon Polytechniou 9, Zografos, GR-15773 Athens, Greece. Tel: +30 210 7721520; Fax: +30 210 7722347; E-mail: [email protected] Principal investigator(s): Christopher Provatidis Research staff: Argyro Kallivretaki

Computational simulation of the mechanical performance of textile products Other Partners: Academic

Industrial

None Project start date: 5 November 2007

None Project end date: 4 November 2009

Research register

9

IJCST 21,6

10

Project budget: e 15,000 Source of support: Committee for the Basic Research of the National Technical University of Athens Keywords: Micromechanical, Macromechanical, Finite Element Method, Simulation The current research project focuses in the mechanical analysis of textile structures implementing the Finite Element Method (FEM). Three stages of analysis are adopted for the integrated modelling of a textile structure. The micromechanical analysis of yarns is implemented for the evaluation of the yarn apparent properties in the first stage. The second stage of modelling corresponds to the mesomechanical modelling of the textile structure for the calculation of the apparent properties of the unit cell. The third stage supports the simulation of complex deformations (drape test) of the macromechanical models of the textile structures.

Aims and objectives The research project aims at the development of a reliable computational method for the evaluation of the mechanical performance of a textile structure considering the fibres’ properties, the yarn and the fabric microstructure.

Deliverables Computational tools for: . The parametric modelling of a typical yarn based on the structural characteristics of the fibre assembly (fibre elastic properties, number of fibres, fibre diameter, yarn diameter). .

The parametric mesoscale modelling for a series of textile structures (basically woven and knitted structures).

The simulation of micro- meso- and macro-scale models in the appropriate deformations. Comparison with experimental data. .

Publications and outputs Vassiliadis, S., Kallivretaki, A., Grancaric, A.M., Giannakis, S. and Provatidis, Ch. (2008), “Computational modelling of twill and satin woven structures”, 8th Autex World Textile Conference, Biella, Italy, June. Vassiliadis, S., Kallivretaki, A., Kavagia, X., Provatidis, Ch., Mecit, D. and Roye, A. (2008), “Computational modelling of spacer fabrics”, 8th Autex World Textile Conference, Biella, Italy, June. Kallivretaki, A., Vassiliadis, S. and Provatidis, Ch. (2008), “Computational modelling of fibrous assemblies”, CIRAT 3 International Conference of Applied Research in Textiles, Sousse, Tunisia, November.

Donghua, China Fumei Wang Team, Donghua University, College of Textiles, Research and developement of higher quality PTT products PTT (Polytrimethylene terephthalate), as one kind of novel polymer, was industrialized from the last century in the late 1990s by Shell Chemical Company and Dupont Company. Our research team devoting to the application of PTT in the field of textile was grouped in the year of 2000. As the technical support of Shell Chemical Company, our team focuses on the characteristic of PTT and its possible applications, and also helps the collaborators develop the PTT fiber and textiles. The research was recognized by many well-known fiber spinning companies and textile mills. The successful work was achieved in the following two fields. PTT shape-memory fiber and fabrics. PTT (Polytrimethylene terephthalate) shape memory fabric is the newly developed advanced fabrics that have the general properties of the shape memory material. The fabrics deriving from low modulus, high recovery and high surface friction of PTT polymer own the prominent characteristics that are shaped at will and readily returned to its initial flat shape and dimension. Such property allows costume designers to design puffy-looking clothing, such as “umbrella” and “tulip” fashions. These clothes can be flattened by simply stroking them with hand and reshaped if the fashion is not to one’s liking. Because of this excellent recovery, clothes made with shape memory fabrics are easy to care, wrinkles can be easily removed, and the fabrics can be flattened and returned to their initial shapes by simply stroking them with hands. PET/PTT Side-by-Side Bicomponent Fibers and its elastic Fabrics. PET/PTT sideby-side bi-component fiber comprises PTT (Polytrimethylene terephthalate) and PET (Polyethylene terephthalate). Due to the high shrinkage and high elastic elongation of PTT polymer, the PTT/PET behaves natural crimp like natural wool with a textured appearance, showing the helix-structured crimp as Figure 1. The crimp brings the fiber excellent extension and recovery properties, and the fabrics made from PTT/PET bicomponent fiber has the elasticity similar as the fabrics made from the spandex, while have the better properties such as stable ability in high temperature than that of spandex fabrics. Our team developed many kinds of novel fabrics with PTT/PET filaments cooperating with fiber spinning companies and textile mills. And meanwhile, we found the main shortcoming of PTT products, such as the knitting fabrics made with PTT/PET bicomponent filament have the random unevenness of striped shadow, due to the unevenness of the spiral structure in the filament yarn which results in the reflective

Research register

11

Figure 1. The structure of PTT/PET bicomponent filament

IJCST 21,6

12

variation of fabric surface. Our team have developed the key technology and patented to overcome this problem. In the past few years, our team has registered four invention patents, and the research was rewarded with the scientific and technological progress second prize in Shanghai, China. Our team now continues to do the research on the basic work such as the structure and elasticity design technology of side-by-side bicomponent fiber, the hand of PTT shape memory fabric at the low temperature. All of our work will promote the applications of novel polymers.

Edinburgh, UK Heriot-Watt University, School of Engineering and Physical Sciences, Riccarton, Edinburgh, Scotland, UK, EH14 4AS. Tel: 00 44 131 451 3034; Fax: 00 44 131 451 3473; E-mail: [email protected]; [email protected] Principal investigator(s): Professor J.I.B. Wilson and Dr R.R. Mather Research staff: Ms. A.H.N. Lind

Solar cells in textiles Other Partners: Academic

Industrial

None None Project start date: 2001 Project end date: 2010 Project budget: N/A Source of support: N/A Keywords: Thin film silicon, Solar energy, Photovoltaics We are developing thin-film silicon solar cells on low cost textile substrates, using chemical vapour deposition (CVD) technology, based on previous thin-film diamond expertise. The CVD technology employs a proprietary microwave plasma system (developed at Heriot-Watt University) with silane/hydrogen/dopant gas mixtures to produce the sequence of layers that forms the active part of these cells. We have shown that relatively low deposition temperatures of 200 degrees C and the active plasma conditions of the process do not affect our textile substrates, whether of woven or non-woven construction. In addition, solutions have been determined to the problem of providing reliable electrical contacts over fibrous, flexible substrates, together with a conventional transparent conducting oxide as the top contact in the cell “sandwich” structure. Effective “first barrier” encapsulation may also use our deposition technology.

Aims and objectives Flexible solar cells for a variety of applications:, e.g. building fac¸ades, use in remote areas, emergency use in disaster relief, camping/leisure industry, portable chargers.

Research register

Deliverables Working prototype. Publications and outputs “Textiles make solar cells that are flexible and lightweight”, Technical Textiles International, December 2002, pp. 5-6. Mather, R.R. and Wilson, J. (2006), “Solar textiles: production and distribution of electricity coming from solar radiation. Applications”, in Mattila, H. (Ed.), Intelligent Textiles and Clothing, Woodhead Publishing Limited, Cambridge.

Galashiels, UK Heriot-Watt University, School of Textiles and Design, Netherdale, Galashiels, Scotland TD1 3HF. Tel: +44 (0) 1896 892140; Fax: +44 (0) 1896 756701; E-mail: [email protected] Principal investigator(s): R.H. Wardman

Verification of algorithm for the numerical specification of standard depths of colour Other Partners: Academic

Industrial

None None Project start date: May 2008 Project end date: May 2011 Project budget: £12,000 Source of support: Society of Dyers and Colourists Keywords: Standard depths, Colour, ISO, Textiles Standard depths of colour are used to determine the fastness properties of dyed materials. The existing standard depths were established over fifty years ago and there is general agreement hat they are not of uniform depth. The project involves the visual assessment of fabric samples dyed to standard depths of shade defined by an algorithm developed by the PI. The visual assessments will be carried out by observers in at least five different countries and their results used to verify the algorithm. The algorithm will then be written into a new proposal for an ISO standard that defines standard depths of colour.

Aims and objectives To prepare samples according to the algorithm previously published and verify its accuracy by visual assessments of a panel of colourists in five different countries.

13

IJCST 21,6

14

Deliverables A new ISO standard for the numerical specification of standard depths of colour. Publications and outputs Chen, C.C., Wardman, R.H. and Smith, K.J. (2002), “The mapping of a surface of constant visual depth in CIELAB colour space”, Coloration Technol., Vol. 118, p. 281. Wardman, R.H., Islam, S. and Smith, K.J. (2006), “Proposal for a numerical definition of standard depths”, Coloration Technol., Vol. 122, p. 350.

Galashiels, UK Heriot-Watt University, School of Textiles and Design, Netherdale, Galashiels, Scotland TD1 3HF. Tel: +44 (0) 1896 892140; Fax: +44 (0) 1896 756701; E-mail: [email protected] Principal investigator(s): R.M. Christie and R.H. Wardman Research staff: R. Shah

Digitial fast patterned microdisposal of fluids for multifunctional protective textiles Other Partners: Academic University of Manchester, Hogeschool Gent, University of Lodz, University of Twente

Project start date: May 2006 Project budget: e12.6million Source of support: EU FP6 Keywords: Inkjet printing, Textiles

Industrial Ten Cate, Grado Zero Espace, B & B Corporate Knitwear, Guantenor S.L., J. Sarens, N.V., Iris D.P. S.r.L., JPC S.P. zoo, Liebaert, Skalmantas, SKA Polska s.p.zoo, Vexed Generation Ltd, Xennia Technology Ltd, D’Appolonia SpA, Lamberti SpA, Xaar PLC, Saxion Project end date: May 2010

To develop breakthrough technology based on digitally microdisposing fluids on textiles enabling high-speed protective functionalisation, continuous processing and customised production. Digital microdisposal has the ability of exact localisation and patterning of functionalities in multilayer textile substrates integrating advanced thermo and hydro regulation, sensorics, actuating and controlled release functions, based on nano-technology and multifunctional materials.

Aims and objectives To realise a generic functionalisation technology for making multifunctional protective textiles on the basis of localised and multilayered compounds.

To use low temperature and low water processessing by inkjet application of functional compounds. Develop innovations in mechatronics and micro-fluids, nanotechnology and multifunctional materials as well as in micro-nano metrology. Set new standards of performance in personal protective equipment.

Deliverables On-line process for the microdisposal of active fluids onto textile fabrics. Monitoring system integrating different inspection techniques. Publications and outputs None in the public domain.

Galashiels and other site partners, UK Research Institute for Flexible Materials, School of Textiles and Design, Heriot-Watt University, Galashiels, Selkirkshire, TD1 3 HF. Tel: 01896 892135; Fax: 01896 758965; E-mail: [email protected] Principal investigator(s): Prof. G.K. Stylios, RIFleX, Heriot-Watt University, Dr K. Lee, Unilever Research (Industrial), Dr R. Potluri, Manchester University and Prof. Long, Nottingham University Research staff: L. Luo plus others in partner universities and industrial partners

Multi-scale integrated modelling for high performance flexible materials Other Partners: Academic University of Nottingham and Manchester University

Industrial Unilever, OCF Plc, TechniTex Faraday Ltd, Crode International Plc, ScotWeave Ltd, Airbags International Ltd, Moxon of Huddersfield Ltd, Carrington Carrer and Workwear Ltd Project end date: 31 December 2010

Project start date: 1 January 2007 Project budget: £1.7 million Source of support: Department of Trade and Industry DTI Keywords: Modelling, Yarn, Fabric, Garment, High performance

This is a flagship proposal for the UK. It is based on integrating micro, meso and macro scale structure/property and deformation models for high performance flexible materials. The outputs will be industry targeted solutions for predicting the properties and behaviour of high performance flexible materials in deformed states during usage including garments. The proposal stems from the modelling achievements of the three academic partners, combining their complimentary work, and after integration, adapting

Research register

15

IJCST 21,6

16

them for industry use, with Unilever, the lead partner, and 12 companies covering diverse applications of the outputs, attempting to represent this sector. OCF and Scotweave will help commercialising the output in the form of a software product or licence. The main aim of the project is to develop the models such that they can be used primarily by the high performance textiles and garment industry, but also by other industries dealing with flexible materials. The key strength of the proposal is that for the first time an attempt is made to model the behaviour of flexible materials in a 3D manner, taking into account the dynamic changes of performance related properties with physical changes during use: (1) Bridging the gap between (a) industry needs for predictive and development tools and (b) academic modelling efforts on two levels: structure/property micro/meso scale and macro-scale whole flexible structures such as high performance clothing. (2) Modelling of overall properties and performance of simple deformed textile structures such as draped/creased/folded fabrics, predicting the dynamic changes of performance related to physical deformations. This will be an innovative engineering tool for industries involved in designing, developing and manufacturing flexible materials, initially targeted at fabrics, but with applications to paper and thin films. (3) Modelling simple high performance whole garments based on structural/geometrical parameters of the flexible materials, to predict the changes in whole garment performance caused by yarn and fabric changes. No model that can perform this function for the high performance clothing industry is yet available. This aspect of the work will be a powerful tool for the high performance clothing and garment industry (high performance medical wear, protective clothing, sportswear, etc.). The project will lead to two way modelling: predicting properties and performance of deformed textile materials and whole garments from structural, mechanical and geometrical parameters, or vice versa, i.e. generating structural, mechanical and geometrical requirements for specific end-product characteristics. Focusing on the high performance clothing sector, the work will provide a first-in-its-kind tool for design, development, engineering and optimisation of high performance flexible materials and exploited by a consortium of diverse companies, three of which are technology providers and the rest users of the technology. Different mechanisms (e.g. interfacial modification) can be exploited in the manufacture, modification, cleaning and care of high performance multi-component textiles. This will permit rational design of these materials and associated products, reducing the need for experimental development and testing. This project will push the boundaries in multi-scale modelling by building on leading edge expertise in fibre to yarn scale modelling at Manchester University, yarn to textile scale modelling at Nottingham University, and textile to whole flexible structures and clothing at Heriot-Watt University. By developing and interfacing these areas closely coupled tools will allow completely predictive models to be developed. It will also be the first time that modelling techniques have been used to understand the effects of interfacial properties on textile performance, leading to increasing innovation in high performance textile products.

Aims and objectives .

.

.

Combining the strong modelling expertise of the four academic partners to create multi-scale integrated models for predicting structural properties and performance of flexible materials. Optimising the integrated models for industry use, focusing in the first instance at high performance clothing and technical textiles. Increasing time and cost efficiency of industry in product design and fabrication, starting with the high performance clothing industry.

The objectives address industry needs for new and improved product design and fabrication aids for high performance materials, e.g. in view of the London 2012 Olympics (high performance sportswear, anti-terrorism protective clothing, etc.). They will enable the prediction of properties and performance before manufacturing, hence accelerating the development stage, reducing costs and increasing competitiveness.

Deliverables A Virtual Testing capability will be developed and transferred into industry, reducing the amount of experimentation required for new product development (new textiles, treatments and laundry products designed and manufactured) and thereby reducing the associated costs and waste generation (improving sustainability of production). Direct beneficiaries of this work will be manufacturers of apparel, textile, technical and allied industries. Indirect beneficiaries will include manufacturers of carpets, non-wovens, composites, paper and structural materials, as well as retailing and software (after commercialising): (1) Bridging the gap between (a) industry needs for predictive and development tools and (b) academic modelling efforts on two levels: structure/property micro/meso scale and macro-scale whole flexible structures such as high performance clothing. (2) Modelling of overall properties and performance of simple deformed textile structures such as draped/creased/folded fabrics, predicting the dynamic changes of performance related to physical deformations. This will be an innovative engineering tool for industries involved in designing, developing and manufacturing flexible materials, initially targeted at fabrics, but with applications to paper and thin films. (3) Modelling simple high performance whole garments based on structural/geometrical parameters of the flexible materials, to predict the changes in whole garment performance caused by yarn and fabric changes. No model that can perform this function for the high performance clothing industry is yet available. This aspect of the work will be a powerful tool for the high performance clothing and garment industry (high performance medical wear, protective clothing, sportswear, etc.). Publications and outputs Too early.

Research register

17

IJCST 21,6

18

Galashiels, UK Research Institute for Flexible Materials, School of Textiles and Design, Heriot-Watt University, Galashiels, Selkirkshire, TD1 3 HF. Tel: 01896 892135; Fax: 01896 758965; E-mail: [email protected] Principal investigator(s): Prof. G.K. Stylios Research staff: X. Zhao

Integration of CFD and CAE for design and performance assessment of protective clothing Other Partners: Academic None

Industrial Tilsatec Ltd, TechniTex Faraday Ltd, Camira Fabrics Ltd, St James’s University Hospital, Remploy Ltd, Pil Membranes Ltd, Altair Engineering Ltd Project end date: 31 May 2010

Project start date: 1 June 2007 Project budget: £600,000 Source of support: Engineering and Physical Science Research Council EPSRC and Department of Trade and Industry DTI Keywords: CFD, CAE, Protective fabric, Garment, Apparel, Modelling high performance This collaborative proposal aims at improving and developing new textiles for protective clothing by integrating Computational Fluid Dynamics (CFD) and Computer Aided Engineering (CAE). A new industrial tool for predicting the diffusion of chemical and/or biological (CB) agents through multilayer, non-homogenous flexible porous materials such as fabrics and whole garments will also be established. Multi-layer textile structures (flat and shaped) and simple garments will be modelled, using equations of mass, heat and momentum balance, integrated with human computational representation. The outcome will be optimisation of current commercial fabrics/ garments and developing new protective clothing whilst, at the same time offering a new objective tool for product designers, engineers and developers to predict and evaluate performance of CB protective products.

Aims and objectives (1) Develop and integrate CAE/CFD for modelling of a clothed human to predict the performance of textiles and garments in CB protective applications, linking with fabric design and manufacturing: (2) Establish an objective measure protocole for CB protection end uses. (3) Improve protection performance of (PPE) and (PC) through new product development for extreme conditions. (4) Develop new materials by better understanding of the complex interactions between the flow of CB agents and textile/materials properties.

Deliverables With a worldwide focus on CB agents, the project is timely for industry, considering the legislative demands for public health, safety and security. Growing concerns from government, the public and private sector about threats of terrorist attacks or epidemic outbreaks have led to increased performance and evaluation requirements. Increasing emphases on security for international events (e.g. London 2012) are also drivers for the above innovations. The proposed project addresses these issues by development key technologies for immediate and long-term use. The consortium experience can achieve the project objectives and lead to a major breakthrough in PC. Publications and outputs Too early.

Galashiels, UK Heriot-Watt University, RIFleX, School of Textiles and Design, Netherdale, Galashiels TD1 3HF, UK. Tel: +44 1896 89 2135; Fax: +44 1896 75 8965; E-mail: [email protected] Principal investigator(s): Prof. George K. Stylios Research staff: Mohamed Basel Bazbouz

Investigating the spinning of yarn from electro-spun nanofibres Other Partners: Academic

Industrial

None None Project start date: November 2004 Project end date: October 2009 Project budget: N/A Source of support: N/A Keywords: Electrospinning, Polymers, Nanofibre, Alignment, Yarn, Composite Our laboratory is using a process called electrospinning which has the ability to produce a wide variety of polymeric fibres with diameters from a couple of micrometer down to the nanometer scale. In this case different structures can be made from electrospun fibers to suit the needs of various industries. Electrospinning, a fibre spinning technique that relies on electrostatic forces to produce fibres in the nanometer to micron diameter range, has been extensively explored as a simple method to prepare fibres from polymer solutions or melts. Under the influence of the electrostatic field, a pendant droplet of the polymer solution at the spinneret is deformed into a conical shape (Taylor cone). If the voltage surpasses a threshold value, electrostatic forces overcome the surface tension, and a fine charged jet is ejected. As these electric static forces increase, the jet will elongate and accelerate

Research register

19

IJCST 21,6

20

by the electric forces. The jet undergoes a variety of instabilities, dries, and deposits on a substrate as a random nanofibre mat. In our work, nonwoven electrospun mats of nylon 6 produced from solutions with formic acid with different concentrations are examined. Each nonwoven mat with average fibre diameters from 200 to 1300 nm was prepared under controlled electrospinning process parameters. Effects of electric field and tip-to-collection plate distances of various nylon 6 concentrations in formic acid on fibre uniformity, morphology and diameters were measured. Processing parameters effects on the morphology such as fibre diameter and its uniformity of electrospun polymer nanofibres was investigated. A process optimization summarized the effects of solutions properties and processing parameters on the electrospun nanofibre morphology was issued. In our work, we control the electrospinning process to move away from just collecting random fibre mesh to enabling the accurate deposit of fibres at any predetermined position. This will be by using a simple method of getting a fibres bundle made of aligned nanofibres between two known points. This collection process has been termed as the “gap method of alignment” involves grounding two circular disks from the spinneret. We have demonstrated that it is possible to produce continuous fiber yarn made out of electrospun nanofibres. The current process has the potential to spin nanofibre at a commercially viable rate.

Aims and objectives .

Understanding of the electrospinning process.

.

Optimizing its process parameters to electrospin polymers into nanofibres with desired morphology.

.

Discussing the models proposed for jet forming, jet travel, processing instabilities.

.

Tensile testing of polymeric nanofibres.

.

Controlling the spatial alignment of electrospun fibres.

.

A technique for making continuous fibre bundle yarns from electrospun fibres. Investigating the other methods for producing continuous fibre bundle yarns, Coreyarn and laminate composite consisting of aligned fibres in different Directions.

.

Deliverables .

Nonwoven electrospun mats of nylon 6 produced from solutions with formic acid with different concentrations are examined.

.

Each nonwoven mat with average fibre diameters from 200 to 1300 nm was prepared under controlled electrospinning process parameters.

.

Effects of electric field and tip-to-collection plate distances of various nylon 6 concentrations in formic acid on fibre uniformity, morphology and diameters were measured.

.

Processing parameters effects on the morphology such as fibre diameter and its uniformity of electrospun polymer nanofibres were investigated.

.

A process optimization summarized the effects of solutions properties and processing parameters on the electrospun nanofibre morphology was issued.

.

The well-known Chauchy’s inequality is applied to prediction the velocity of the end of the jet in electrospinning.

.

A critical relationship between radius r of jet and the axial distance z from nozzle is obtained for the straight jet.

.

Draw ratio between the jet and the final fibres was predicted theoretically.

.

Manufacturing of aligned fibres array was easily achievable. Processing parameters effects on the aligned fibres such as gab distance and collection time were investigated.

.

Publications and outputs In preparation: .

Systematic parameter study for ultra-fine nylon 6 fibre produced by electrospinning technique.

.

Electrospinning of aligned nanofibres with cotrolled deposition.

Conferences Paper and Poster: .

June 2006 (Electrospinning of nanofibres: potential scaffolds for medical applications) a presentation presented in Research in Support of Medicine, Health and Safety Conference, Edinburgh, UK.

.

June 2006 (Systematic parameter study for ultra-fine nylon 6 fibre produced by electrospinning technique), Poster presentation, Research in Support of Medicine, Health and Safety Conference, Edinburgh, UK.

Galashiels, UK Heriot-Watt University, RIFleX, School of Textiles and Design, Netherdale, Galashiels TD1 3HF, UK. Tel: +44 1896 89 2135; Fax: +44 1896 75 8965; E-mail: [email protected] Principal investigator(s): Prof. George K. Stylios Research staff: Mohammad Mahfuzur Rahman Chowdhury

Investigating nano fibre production by the electrospinning process Other Partners: Academic

Industrial

None None Project start date: July 2004 Project end date: December 2009 Project budget: N/A Source of support: N/A Keywords: Electrospinning, Electrospinning Process, Parameters, Polymer, Nanofibre application

Research register

21

IJCST 21,6

22

Electrospinning is a unique way to produce novel polymer nanofibres with diameter typically in the range of 10 nm to 500 nm. Using this process, the polymer nanofibres can be made from a variety of polymer solutions or melt to produce fibres for a wide range of applications. Electrospinning occurs when the electrical force at the surface of a polymer solution or melt overcomes the surface tension and causes an electrically charged jet to be ejected. When the jet dries or solidifies, an electrically charged fibre remains. This charged fibre can be directed or collected or accelerated by electrical forces, then collected in sheets or other geometrical forms. This research project is an investigation of the electrospinning process and the effect of process variables on orientation, crystallinity, microstructure and mechanical properties of the nanofibres produced. Some of the polymeric parameters investigated are polymer type, solvent type, molecular weight, solution properties, viscosity, conductivity and surface tension. In the case of process parameters, the electric potential, flow rate, concentration, distance between capillary and collection screen, ambient parameters are important.

Aims and objectives .

. .

To investigate process-structure-property relationships in polymer fibres with nanosize diameters produced by electrospinning. To investigate the morphology and properties of the polymer nanofibres. To produce fibres at uniform diameters.

Deliverables .

Nanofibres of uniform diameter.

.

Defined mechanical and physical properties. Process-structure-property relationships.

. .

Detailed understanding of the electrospinning process.

.

Nanofibres suitable for applications such as air filtration, protective clothing, fibre reinforced support, and Biomedical.

Publications and outputs “Nano fibre and its medical application”, poster presentation at Research in Support of Medicine, Health and Safety Conference, Heriot-Watt University, Scotland, UK.

Galashiels, UK Heriot-Watt University, RIFleX, School of Textiles and Design, Netherdale, Galashiels TD1 3HF, UK. Tel: +44 1896 89 2135; Fax: +44 1896 75 8965; E-mail: [email protected] Principal investigator(s): Prof. George K. Stylios Research staff: Dr Taoyu Wan

Novel micro-channel membranes for controlled delivery of biopharmaceuticals Other Partners: Academic None

Research register

Industrial Stryker UK Ltd, St James’s University Hospital, Leeds General Infirmary, Camira Fabrics Ltd, Dinsmore Textile Solutions Ltd Project end date: September 2009

Project start date: April 2006 Project budget: £650,000 Source of support: DTI Technology Programme Keywords: Micro-channel, Micro-porous, Membrane, Controlled delivery, Drug release

This project, which stems out of research findings of an EPSRC-funded research, aims at developing micro channel structures (coatings, membranes, foams, etc.) with encapsulated biopharmaceuticals capable of controlled release by changes in temperature, pH, magnetic field or voltage. A technique for encapsulating biopharmaceuticals into micro-channels of a polymer matrix structure and controlling their subsequent release will be developed. Driven by the consortium, the technologies will be veered towards various pay-load bearing applications, e.g. in self-supporting materials for delivering bone growth hormones (INN eptotermin alpha) in bone fractures, or in coated textiles for the personal hygiene, healthcare, treatment and protective clothing industries.

Aims and objectives The main objectives for the projects are to: (1) Establish criteria for channel size and distribution control; engineer the structure and morphology of the porous material to suit the encapsulation of biopharmaceuticals and their slow release. This involves both self-supported materials and coated textiles. (2) Investigate UV-based techniques for the encapsulation of specific biopharmaceuticals, with particular reference to bone growth hormones. (3) Investigate strategies for the controlled release of the biomaterials from the channels within the developed membrane or coating. (4) Characterise and test laboratory and pilot-scale samples – study release rates of the biopharmaceuticals, degradation/absorption rates and efficacy of the system. (5) Investigate alternative exploitation routes as coated textiles with encapsulated biopharmaceuticals. The project will investigate possibilities both for implantable and non-implantable payload bearing materials, both as self-supported materials, and as coatings or membranes supported by a base fabric. Research work on implantable applications will however only be performed in laboratory scale with the aim to prove the concepts.

23

IJCST 21,6

24

Deliverables Pilot-scale production of membranes, coatings, foams, or any other forms of channelcontaining materials with engineered, controllable and tailor-made channel sizes, shapes and distributions: (1) Defined characteristics, properties and performance of the as-produced materials. (2) A technique to encapsulate biopharmaceuticals within the membrane channels using a UV cross-linking technique, which eliminates the need to use high temperature treatments for encapsulation. This enables the biopharmaceuticals to maintain their effectiveness. (3) The final expected result is the development of a system for activated release, e.g. following a change in stimuli or external conditions, such as temperature, pH, magnetic field or voltage. This will enable the controlled and targeted release of the biopharmaceutical as, when and where required for optimum treatments after fractures. Publications and outputs Stylios, G.K., Giannoudis, P.V. and Wan, T. (2005), “Applications of nanotechnologies in medical practice”, Injury, Vol. 36S, pp. S6-S13. Stylios, G.K., Wan, T. and Giannoudis, P.V. (2006), “Present status and future potential in the enhancement of bone healing using nanotechnology”, accepted for publication in Injury.

Galashiels, UK Heriot-Watt University, RIFleX, School of Textiles and Design, Netherdale, Galashiels TD1 3HF, UK. Tel: +44 1896 89 2135; Fax: +44 1896 75 8965; E-mail: [email protected] Principal investigator(s): Prof. George K. Stylios Research staff: Liang Luo

Interactive wireless and smart fabrics for textiles and clothing Other Partners: Academic

Industrial

None None Project start date: September 2002 Project end date: December 2009 Project budget: N/A Source of support: Worshipful Company of Weavers Keywords: Smart, Interactive, Textiles, Garment, Clothing, Sensors, Wireless The last few years have witnessed an increased interest in wearable technologies, smart fabrics and interactive garments. This has come about by certain technological innovations in the areas of sensor-based fabrics, micro devices, wire and wireless

networks. In terms of textiles, most of current developments are towards the fashion markets and have resulted in glorifying garments as gimmicky gadgets. However, some efforts are also being directed in using the technology for improving the quality of life, or even for life saving purposes. Examples of such uses can be found in the military, healthcare, fire fighting, etc. This research project investigates new interdisciplinary technologies in fabrics, sensors and wireless computing, for the development of a prototype interactive garment for monitoring various functions of the wearer.

Aims and objectives The general aim of the project is to develop technologies for use in interactive garments, which can provide monitoring functions for various applications such as the clinical or healthcare sector. More specifically, objectives are: .

Develop suitable wireless sensors for various measurements, including ECG, temperature, breathing, skin conductivity, mobility and movement, humidity, positioning, etc.

.

Develop a Personal Area Network and a Wireless Communication Centre.

.

Optimise suitable wireless technologies such as Bluetooth to enable communication between sensors and a central processing unit.

.

Conceptualise a smart multilayer fabric.

.

Integrate technologies.

Deliverables .

Wireless sensors for physiological and other measurements.

.

Wireless communication centre for relaying information between sensors, wearers, central processing unit and internet.

.

Conceptual multilayer fabric suitable for interactive garments.

Publications and outputs Stylios, G.K. and Luo, L. (2003), “Investigating an interactive wireless textile system for SMART clothing”, 1st International Textile Design and Engineering Conference (INTEDEC 2003), Fibrous Assemblies at the Design and Engineering Interface, Edinburgh, UK, 22-24 September. Stylios, G.K. and Luo, L. (2003), “The concept of interactive, wireless, smart fabrics for textiles and clothing”, 4th International Conference, Innovation and Modelling of Clothing Engineering Processes – IMCEP 2003, Maribor, Slovenia, 9-11 October. Stylios, G.K. and Luo, L. (2004), “A SMART wireless vest system for patient rehabilitation”, Wearable Electronic and Smart Textiles Seminar, Leeds, UK, 11 June. Stylios, G.K., Luo, L., Chan, Y.Y.F. and Lam Po Tang, S. (2005), “The concept of smart textiles at the design/technology interface”, 5th International Istanbul Textile Conference, Recent Advances and Innovations in Textile and Clothing, Istanbul, Turkey, 19-21 May.

Research register

25

IJCST 21,6

26

Galashiels, UK Heriot Watt University, Scottish Borders Campus, Netherdale, Galashiels, TD1 3HF. Tel: 01896 892245; Fax: 01896 758965; E-mail: [email protected] Principal investigator(s): Dr Alex. Fotheringham Research staff: Basel Younes

Optimisation of weaving process for biopolymers Other Partners: Academic

Industrial

None None Project start date: October 2006 Project end date: September 2010 Project budget: N/A Source of support: British Council Keywords: Fibre extrusion, Yarn production, Weaving This is part of a general research programme in the production and use of biopolymers. Current research investigates the optimisation of biopolymer fibre production using experimental design techniques to statistically map processing parameters to mechanical properties. Having produced a model of the fibre and yarn production, the relationship between fibre/yarn characteristics, weaving parameters and fabric properties can then be established. Related work is researching into the use of gas plasma for the pre-treatment of polylactic acid fabric, the printing of knitted biopolymer fabrics and dyeing of such materials in fibre/yarn form.

Aims and objectives Aims: . To establish the relationships which exist between processing and final properties. . To identify the value of gas plasma treatment on biopolymer fabrics for, e.g. dyeing, coating, etc. . To optimise printing on knitted biopolymer fabrics. Objectives: . .

To use experimental designs to create statistical models. To use statistical models as forecasting tools to establish the relationships between processing and properties.

Deliverables . .

A statistical model of fibre, yarn and fabric (weaving) processing. To create similar models for fibre dyeing and printing.

Publications and outputs Not available.

Zwijnaarde (Gent), Belgium Ghent University Department of Textiles, Technologiepark 907 B-9052 Zwijnaarde (Gent) Belgium. Tel: +32 9 264 57 35; Fax: +32 9 264 58 46; E-mail: [email protected] Principal investigator(s): Prof. Dr Ir. Lieva Van Langenhove Research staff: Prof. Dr Ir. Lieva Van Langenhove

LIDWINE – multifunctionalized medical textiles for wound (e.g. decubitus) prevention and improved wound healing (“Lidwine”) Other Partners: Academic

Industrial

None None Project start date: 1 September 2006 Project end date: 31 August 2010 Project budget: e 543.752 Source of support: European Commission – FP6 – IP – SME, FP6 – Integrated Projects – SME Keywords: Decubitus, Electrotherapy, Nanotechnology Several techniques will be developed to prevent decubitus and to stimulate its healing. Examples are passive and active antibacterial action, materials with reduced surface friction, massage and electrotherapy. The latter is the specific task of UGent. The materials will be developed in textile structures. The technologies used will be nanoparticles, encapsulation, brush coatings.

Aims and objectives Not available.

Deliverables Not available. Publications and outputs See http://biblio.ugent.be/input?func ¼ search; http://textiles.ugent.be/docs/AnnualReports/2008.pdf

Zwijnaarde (Gent), Belgium Ghent University Department of Textiles, Technologiepark 907 B-9052 Zwijnaarde (Gent) Belgium. Tel: +32 9 264 57 35; Fax: +32 9 264 58 46; E-mail: [email protected] Principal investigator(s): Prof. Dr Ir. Lieva Van Langenhove Research staff: Prof. Dr Ir. Lieva Van Langenhove

Research register

27

IJCST 21,6

COMPAS – computer based evaluation of aspect change of carpets by wear Other Partners: Academic

28

Industrial

None None Project start date: 1 September 2006 Project end date: 31 August 2009 Project budget: e 415.790,32 Source of support: IWT, VIS Keywords: Carpet wear, Evaluation, Image processing The aim of the project is to develop an objective method to be able to quantitatively measure carpet wear in a univocal and accurate way. To this end, images recorded with a colour CCD camera are being processed. New algorithms developed and in development in the framework of other projects will be tested for the evaluation of carpet wear. A considerable obstacle in the research regarding automatic evaluation of carpets is the lack of an extensive basic library of images of good as well as bad samples, and for a large variety of qualities. Establishing such a library is part of the project as well.

Aims and objectives Not available.

Deliverables Not available. Publications and outputs See http://biblio.ugent.be/input?func ¼ search; http://textiles.ugent.be/docs/AnnualReports/2008.pdf

Zwijnaarde (Gent), Belgium Ghent University Department of Textiles, Technologiepark 907 B-9052 Zwijnaarde (Gent) Belgium. Tel: +32 9 264 57 35; Fax: +32 9 264 58 46; E-mail: [email protected] Principal investigator(s): Prof. Dr Ir. Lieva Van Langenhove Research staff: Prof. Dr Ir. Lieva Van Langenhove

Systex: coordination action for enhancing the breakthrough of intelligent textile systems (e-textiles and wearable microsystems) Other Partners: Academic

Industrial

None None Project start date: 1 May 2008 Project end date: 30 April 2011 Project budget: e 800,000 Source of support: European Commission, FP7 – 1ST – Coordination and Support Action Keywords: Smart textiles, E-textiles, Wearable micro systems, Wearable electronics

The project aims at creating a framework for current and future research activities, technology transfer and education in the area of wearable textile systems and e-textiles. This results in enhancing the cooperation between various entities that contribute to the development and commercialization of the textiles of the future. Technical and non technical information on relevant projects is collected.

Aims and objectives

29

Not available.

Deliverables Not available. Publications and outputs See http://biblio.ugent.be/input?func ¼ search; http://textiles.ugent.be/docs/AnnualReports/2008.pdf

Zwijnaarde (Gent), Belgium Ghent University Department of Textiles, Technologiepark 907 B-9052 Zwijnaarde (Gent) Belgium. Tel: +32 9 264 57 35; Fax: +32 9 264 58 46; E-mail: [email protected] Principal investigator(s): Prof. Dr P. Kiekens Research staff: ir. Els Van der Burght

MUDRA Learning Network Other Partners: Academic

Industrial

None None Project start date: 1 February 2008 Project end date: 30 September 2009 Project budget: e 43.394,73 Source of support: Vlaamse Gemeenschap, Samenwerkingsprojecten Vlaanderen/ Centraal – EN OOST-EUROPA Keywords: Textiles, Networking, Innovation MUDRA Learning Network is a network project of Flemish business and educational partners with their Croatian and Slovenian counterparts. The mentorship methodology offers the framework to a large group of SMEs and entrepreneurs to exchange expertise and to professionalise their management. The know-how from the universities strengthens the technological capacity and helps these companies to innovate. The target group: textile and design-related companies.

Aims and objectives Not available.

Deliverables Not available.

Research register

IJCST 21,6

Publications and outputs See http://biblio.ugent.be/input?func ¼ search; http://textiles.ugent.be/docs/AnnualReports/2008.pdf

Zwijnaarde (Gent), Belgium 30

Ghent University Department of Textiles, Technologiepark 907 B-9052 Zwijnaarde (Gent) Belgium. Tel: +32 9 264 57 35; Fax: +32 9 264 58 46; E-mail: [email protected] Principal investigator(s): Prof. Dr Ir. Karen De Clerck Research staff: Prof. Dr Ir. Karen De Clerck

Research into new “sensor materials”: pH-sensitive colorants in textile materials Other Partners: Academic

Industrial

None None Project start date: 1 January 2009 Project end date: 31 December 2012 Project budget: e 172,000 Source of support: Universiteit Gent, Bijzonder Onderzoeksfonds 2008 Keywords: pH-sensitive dyes, Textiles, Spectroscopy, Microscopy The aim of the project is to obtain a better understanding of the interactions between pH-sensitive dyes and textiles. Both a macroscopic spectral analysis and a general microscopic evaluation will be performed. The dye-fiber interactions, the local distribution, the impulse sensitivity and spectral variations as a function of time and place will be looked at.

Aims and objectives Not available.

Deliverables Not available. Publications and outputs See http://biblio.ugent.be/input?func ¼ search; http://textiles.ugent.be/docs/AnnualReports/2008.pdf

Zwijnaarde (Gent), Belgium Ghent University Department of Textiles, Technologiepark 907 B-9052 Zwijnaarde (Gent) Belgium. Tel: +32 9 264 57 35; Fax: +32 9 264 58 46; E-mail: [email protected] Principal investigator(s): Prof. Dr P. Kiekens Research staff: Prof. Dr Paul Kiekens; Prof. Dr Ir. K De Clerck

FRONT – flame retardant on textile Other Partners: Academic

Industrial

None None Project start date: 1 November 2008 Project end date: 31 October 2010 Project budget: e 141,000 Source of support: European Commission, FP7 Keywords: Textiles, Flame retardants, Nanoclays The project aim is to introduce finishing products in the European textile market, to produce textile fabrics resistant to fire with high performance and quality, as requested from evolution of legislation and from customer attention. Moreover, flame retardant finishing on textiles would achieve a multifunctional textile composition having not only fire resistance properties, but also an additional functionality. The project proposed by SMEs engaged in the field of polymer processing will contribute to meet needs and to strengthen the competitiveness of the EU manufacturers.

Aims and objectives Not available.

Deliverables Not available. Publications and outputs See http://biblio.ugent.be/input?func ¼ search; http://textiles.ugent.be/docs/AnnualReports/2008.pdf

Zwijnaarde (Gent), Belgium Ghent University Department of Textiles, Technologiepark 907 B-9052 Zwijnaarde (Gent) Belgium Tel: +32 9 264 57 35; Fax: +32 9 264 58 46; E-mail: [email protected] Principal investigator(s): Prof. Dr Ir. Karen De Clerck Research staff: Prof. Dr Ir. Karen De Clerck

Tapijtfabriek ALFA: water absorbing capacity of hollow fibres for filling yarn in artificial turf Other Partners: Academic

Research register

Industrial

None None Project start date: 1 March 2008 Project end date: 28 February 2009 Project budget: e 28,050 Source of support: IWT, KMO-innovatiestudie Type 3 Keywords: Hollow fibre, Water absorption, PP

31

IJCST 21,6

32

This project was conducted together with ALPHA-carpets. The capacity of water absorption of hollow PP-fibres was examined and improved, in order to optimize the quality of actual products such as bath maths and towels, but also to be able to bring new products on the market in the future. In the framework of this project, new methods have been written to quantify the water absorption and the drying of hollow fibres. More understanding was gained in the effect of the extrusion parameters on the fibre diameter and the capacity of water absorption of hollow fibres.

Aims and objectives Not available.

Deliverables Not available. Publications and outputs See http://biblio.ugent.be/input?func ¼ search; http://textiles.ugent.be/docs/AnnualReports/2008.pdf

Zwijnaarde (Gent), Belgium Ghent University Department of Textiles, Technologiepark 907 B-9052 Zwijnaarde (Gent) Belgium. Tel: +32 9 264 57 35; Fax: +32 9 264 58 46; E-mail: [email protected] Principal investigator(s): Prof. Dr P. Kiekens Research staff: Johanna Louwagie

TRITex: transfer of research and innovations in textiles Other Partners: Academic None Project start date: 1 January 2009 Project budget: e 490.948,42 Source of support: INTERREG IV, EFRO Keywords: Research, Innovation, Textiles

Industrial None Project end date: 31 December 2012

Ensait and Ghent University – Department of textiles want to establish common and complementary actions in order to strengthen the cross border cooperation. Four actions are foreseen: multilateral research programmes, development of modules for distance learning, valorisation of digital training modules at industrial partners, organisation of seminars.

Aims and objectives Not available.

Deliverables Not available.

Publications and outputs See http://biblio.ugent.be/input?func ¼ search; http://textiles.ugent.be/docs/AnnualReports/2008.pdf

Zwijnaarde (Gent), Belgium Ghent University Department of Textiles, Technologiepark 907 B-9052 Zwijnaarde (Gent) Belgium Tel: +32 9 264 57 35; Fax: +32 9 264 58 46; E-mail: [email protected] Principal investigator(s): Prof. Dr P. Kiekens Research staff: Dr Ir. Vincent Nierstrasz

BIOTIC: biotechnical functionalization of (bio)polymeric textile surfaces Other Partners: Academic

Industrial

None None Project start date: 1 April 2008 Project end date: 31 March 2010 Project budget: e 223.288,66 Source of support: European Commission, People Marie Curie Actions (Intra-European Fellowships (IEF)) Keywords: Biotechnology, Enzymes, Grafting, Functionalisation, Surface modification, Textiles, Biopolymer, Polymer, Nano-structuring The aim is to functionalise textile materials using biotechnology. The research will be based on a concerted multi-disciplinary approach, thereby creating the possibility to produce functionalised materials with unique properties and functionalities. The research will focus on enzymatic grafting of functional groups on textile fibres, and specific enzymatic surface modification to obtain functional nano-structured surfaces.

Aims and objectives Not available.

Deliverables Not available. Publications and outputs See http://biblio.ugent.be/input?func ¼ search; http://textiles.ugent.be/docs/AnnualReports/2008.pdf

Research register

33

IJCST 21,6

34

Zwijnaarde (Gent), Belgium Ghent University Department of Textiles, Technologiepark 907 B-9052 Zwijnaarde (Gent) Belgium. Tel: +32 9 264 57 35; Fax: +32 9 264 58 46; E-mail: [email protected] Principal investigator(s): Prof. Dr P. Kiekens Research staff: ir. Els Van der Burght

SNAP – production of imidised styrene-malein acid (SMI) nanoparticles and surface interactions with different types of substrates (TOPCHIM) Other Partners: Academic

Industrial

None None Project start date: 1 September 2008 Project end date: 31 August 2011 Project budget: e 249,525 Source of support: IWT, Onderzoeksproject Keywords: SMI nanoparticles, Imidisation, Surface properties This project aims for a better understanding of the chemistry and physics at the nano level of SMI nanoparticles, which are created by imidisation of copolymer styrene-maleic anhydride (SMA). The main aspect is to broaden insight into the physical characteristics of the nanoparticles such as shape, size and uniformity. Interaction with minerals and substances from renewable sources is another focus.

Aims and objectives Not available.

Deliverables Not available. Publications and outputs See http://biblio.ugent.be/input?func ¼ search; http://textiles.ugent.be/docs/AnnualReports/2008.pdf

Zwijnaarde (Gent), Belgium Ghent University Department of Textiles, Technologiepark 907 B-9052 Zwijnaarde (Gent) Belgium. Tel: +32 9 264 57 35; Fax: +32 9 264 58 46; E-mail: [email protected] Principal investigator(s): Prof. Dr Ir. Lieva Van Langenhove Research staff: Ing. Johanna Louwagie

Bexco: Arctic ropes – research into the dynamic behaviour of synthetic ropes in extremely cold circumstances Other Partners: Academic

Industrial

None None Project start date: 1 July 2008 Project end date: 31 October 2009 Project budget: e 21.499,46 Source of support: IWT, KMO-innovatiestudie type 3 Keywords: Dynamic behaviour, Synthetic ropes The purpose of the study is to obtain insight in the phenomena that occur in frozen ropes that are under cyclic load an in the effect of the phenomena on their mechanical properties, fatigue and life span of the rope. The influence of ice will be studied on the macro- and microscopic level on ropes after cyclic loading in frozen conditions.

Aims and objectives Not available.

Deliverables Not available. Publications and outputs See http://biblio.ugent.be/input?func ¼ search; http://textiles.ugent.be/docs/AnnualReports/2008.pdf

Zwijnaarde (Gent), Belgium Ghent University Department of Textiles, Technologiepark 907 B-9052 Zwijnaarde (Gent) Belgium. Tel: +32 9 264 57 35; Fax: +32 9 264 58 46; E-mail: [email protected] Principal investigator(s): Prof. Dr Ir. Lieva Van Langenhove Research staff: Simona Vasile

NIRIS: new insertion rules for a new insertion system (Picanol) Other Partners: Academic

Research register

Industrial

None None Project start date: 1 December 2008 Project end date: 30 November 2011 Project budget: e 840.813,28 Source of support: IWT, Onderzoeksproject Keywords: Weft preparation system, Air jet loom, Speed increase

35

IJCST 21,6

The aim of this project is to ultimately bring a new weft preparation system on the market. This should allow to insert the weft yarn faster on an air jet loom. In concrete figures, a speed increase on the loom of 10 to 15% is set as goal.

Aims and objectives Not available.

36

Deliverables Not available. Publications and outputs See http://biblio.ugent.be/input?func ¼ search; http://textiles.ugent.be/docs/AnnualReports/2008.pdf

Zwijnaarde (Gent), Belgium Ghent University Department of Textiles, Technologiepark 907 B-9052 Zwijnaarde (Gent) Belgium. Tel: +32 9 264 57 35; Fax: +32 9 264 58 46; E-mail: [email protected] Principal investigator(s): Prof. Dr Ir. Karen De Clerck Research staff: Dr Philippe Westbroek

Advanced water filtration with nanofibres (Hogeschool West-Vlaanderen) Other Partners: Academic

Industrial

None None Project start date: 1 October 2008 Project end date: 30 September 2010 Project budget: e 103.000 Source of support: IWT, TETRA-fonds Keywords: Waterfiltration, Nanofibers, MBR The goal of the project is the evaluation of nanofiber nonwovens produced by electro spinning for usage in advanced water filtration. This project is a continuation of a previous project around nanofibers. The focus is put on the research in the MBRtechnology.

Aims and objectives Not available.

Deliverables Not available. Publications and outputs See http://biblio.ugent.be/input?func ¼ search; http://textiles.ugent.be/docs/AnnualReports/2008.pdf

Zwijnaarde (Gent), Belgium Ghent University Department of Textiles, Technologiepark 907 B-9052 Zwijnaarde (Gent) Belgium. Tel: +32 9 264 57 35; Fax: +32 9 264 58 46; E-mail: [email protected] Principal investigator(s): Prof. Dr Ir. Lieva Van langenhove; Prof. Vanfleteren, Prof. Leman Research staff: Hertleer Carla

UGent Mobile Textiles: a travelling compact, attractive and interactive platform demonstrating the multidisciplinary research at Ghent University that contributes to the textiles of the future Other Partners: Academic

Industrial

None None Project start date: 1 January 2009 Project end date: 31 December 2011 Project budget: e 95.100 Source of support: Universiteit Gent, Werkgroep Wetenschapscommunicatie en-popularisering: Projecten Wetenschap en maatschappij Keywords: Smart textiles, Demonstrators, Textile antennas, Music, Stretchable electronics Textiles of the future are smart. This emerging research area has a large economical potential and a huge social relevancy. Four departments of Ghent University join forces to develop demonstrators that exhibit research on smart textiles at UGent. In a mobile stand, the visitor is given the opportunity to acquaint with these new technologies.

Aims and objectives Not available.

Deliverables Not available. Publications and outputs See http://biblio.ugent.be/input?func ¼ search; http://textiles.ugent.be/docs/AnnualReports/2008.pdf

Gent, Belgium Universiteit Gent – Ghent University, Vakgroep Textielkunde – Department of Textiles, Technologiepark 907, 9052 Zwijnaarde (Gent), Belgium. Tel: + 32 9 264 5419; Fax: + 32 9 2645831; E-mail: [email protected]

Research register

37

IJCST 21,6

SysTex – coordination action for enhancing the breakthrough of intelligent textile systems (e-textiles and wearable microsystems) Other Partners: Academic

38

Ghent University, Institut Francais du Textile et de l’Habillement, IMEC, CNR-INFM, University of Pisa, Commissariat a l’Energie Atomique

Industrial Philips Research, Multitel, Smartex, Anne Demoor bvba, Plastic Electronics, IHofmann

Project start date: 1 May 2008 Project end date: 30 April 2011 Project budget: 800.000 Source of support: FP7 programme Keywords: Intelligent textiles, E-textiles, Wearable electronics, Breakthrough Wearable electronics embedded in or transformed into textile systems are a new generation of products that contribute to economy as well as to society. SysTex wants to bring partners involved in European projects in this area together in order to group the results of numerous efforts that are currently going on. It wants to expand to national level and to merge textiles and organic electronics. Inter-project agreements must enable a higher level of exchange of knowledge and materials between linked projects. Information on technical and non-technical aspects of RTD and commercialisation of intelligent textile systems will be collected and made available through a web based tool. Training materials will be collected as well as demonstrators that can be used for specialists as well as for a wider public. The project wants to become a single point of contact for all matters related to intelligent textile systems, linking existing initiatives and completing their activities.

Aims and objectives SysTex aims at developing a framework for current and future actions in research, education and technology transfer in the field of e-textiles and wearable micro systems/electronics in Europe to support the textile industry in the most efficient and effective way to transform into a dynamic, innovative, knowledge-driven competitive and sustainable sector. The objectives are to create an extensive and detailed road map of current and possible future technological developments in e-textiles, to organise and facilitate interproject contacts, to organise training and exchanges and to identify needs, breakthroughs and bottlenecks. The information this will provide will be spread to the appropriate target groups using appropriate tools. The coordination action will also provide a platform for intelligent textiles, for the partners involved in European projects but also for interested companies and users. A road map will be built based on available road maps. An experienced legal counsellor will analyse contracts, co-operation agreements, etc. and prepare agreements that enable

exchange of information and materials between relevant projects without harming protection and exploitation of results. The work programme will include: Analysis of the available and ongoing research activities in e-textiles and wearable microsystems, this will cover activities that should contribute to the progress of smart textile systems. The information should provide an insight in project objectives, status and contacts.

.

.

Production of web tools for disseminating the results of this inventory and analysis.

.

Organisation of contacts and exchange of materials, people and expertise.

.

Building a joint road map.

.

Joint exploitation, dissemination and training of all member parties.

.

Providing legal support for enabling communication and exchanges without hindering protection and exploitation.

Deliverables .

.

.

.

.

. . .

Collection of detailed information on relevant activities in the area of intelligent textile systems on all levels, including technical as well as non technical aspects. Widely accessible website with user-friendly database with full information on activities on intelligent textile systems. Reports on training and education opportunities and collection of information on training material like demonstrators including conditions of use. Organisation of guest lectures, student projects, the annual SysTex Student Award. Distribution of information through the platform like reports on state of the art technology to improve efficiency of research efforts and exploitation of research efforts. Contribution to policy plans of EC and local funding bodies. Templates for inter-project agreements in form of contracts and legal agreements. Increased openness and exchanges between projects and hence faster progress of work.

Publications and outputs The expected outcomes are: .

paper and digital reports/newsletters containing state-of-the-art of research efforts as well as arguments for building future strategies;

.

databases and information systems;

.

inter-project agreements enabling active exchange of project results;

.

organisation of and participation at events;

.

training and education activities;

.

raising awareness in all respective target groups;

.

targeted lectures and training tools;

Research register

39

IJCST 21,6

40

.

materials like powerpoints, flyer for dissemination;

.

alignment of research; and

.

enhancement of research progress.

Hong Kong, China Institute of Textile and Clothing, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong. Tel: 00852-27666470; Fax: 00852-29542521; E-mail: [email protected] Principal investigator(s): Prof. Xiaoming TAO Research staff: Dr Guang-feng Wang, Dr Xiao-hong SUN, Dr Bo Zhu, Dr Yang-yong Wang, Dr Jiang-ming Yu, Mr Zhi-feng Zhang, Mr Wei Zheng

Small sized fiber sensors Other Partners: Academic

Industrial

None None Project start date: 1 March 2008 Project end date: 28 February 2010 Project budget: HK$5,264,000 Source of support: The Hong Kong Research Institute of Textiles and Apparel Limited; Innovation and Technology Commission, The Government of the Hong Kong Special Administrative Region; Hong Kong Tak Ying Trading Company; Pool Heng Company Limited; Best Technology Company Limited; Esquel Enterprises Ltd Keywords: Sensor, Small, Fibers In applications such as medical devices, industry, robotics and wearable electronics, the size of sensors is a very important parameter and sensors with smaller size are required. Fibers of a few microns in diameter are ideal candidates. Based on the previous research conducted at the laboratories of The Hong Kong Polytechnic University, the present project is aimed to further developing small sized fiber sensors with conductive fibers and assemblies, and polymeric photonic fiber. The sensors to be developed should have high sensitivity, long service life, reasonable material and production cost. The proposed project will investigate and develop design and fabrication technologies of the small sized fiber sensors, as well as the package of the fiber sensors, and explore their applications in industries.

Aims and objectives (1) To select and fabricate materials for the small size fiber sensors: (2) To make appropriate fibers. (3) To develop the structural design for the small size fiber sensors.

(4) To develop fabrication technology and equipment for the fiber sensors. (5) To explore package technologies for the fiber sensors.

Research register

Deliverables (1) The optimized design and materials for the small size fiber sensors in term of performance and cost: (2) Fabrication technology and equipment for the small size fiber sensors. (3) Package method and production technology of the small size fiber sensors. (4) A test protocol for performance and reliability of the small size fiber sensors. (5) The theoretical model for the single cell of the fiber sensors.

Publications and outputs Tao, X.M., Yu, J.M. and Tam, H.Y. (2007), “Photosensitive polymeric optical fibres and gratings”, Transactions of the Institute of Measurement and Control, Vol. 29 No. 3, pp. 1-16. Yu, J.M., Tao, X.M. and Tam, H.Y. (2004), “Trans-4-stilbenemethanol-doped photosensitive polymer optical fiber for grating fabrication”, Optics Letters, Vol. 29 No. 2, pp. 156-158, January. Zhang, A.P., Guan, B.O., Tao, X.M. and Tam, H.Y. (2002), “Experimental and theoretical analysis of fiber bragg gratings under lateral compression”, Optics Communication, Vol. 206, pp. 81-87. Zhang, H., Tao, X.M., Yu, T.X. and Wang, S.Y. (2006), “Conductive knitted fabric as large-strain gauge under high temperature”, Sensors and Actuators A, Vol. 126, pp. 129-140 Zhang, H., Tao, X.M., Yu, T.X. and Wang S.Y. (2006), “A novel sensate ‘string’ for large-strain measurement under high temperature”, Measurement Science and Technology, Vol. 17, pp. 450-458. Yang, B., Tao, X.M. and Yu, J.Y. (2006), “A study on the relationship between resistance and strain based on stainless steel fabric”, Rare Metal Materials and Engineering, Vol. 35 No. 1, pp. 96-99. Xue, P. and Tao, X.M. (2005) “Morphological and electromechnical studies of fibers coated with electrically conductive polymers”, Applied Polymer Science, Vol. 98, pp. 1844-1854. Li, Y., Leung, Y.M., Tao, X.M., Chen, X.Y., Tsang, H.Y.J. and Yuen, M.C.W. (2005) “Polypyrrole-coated fabrics as a candidate for strain sensors”, J. Materials Science, Vol. 40 No. 15, pp. 4093-4095. Li, Y., Chen, X.Y., Leung, Y.M., Tao, X.M., Tsang, H.Y.J. and Yuen, M.C.W. (2005), “A flexible strain sensor from polypyrrole coated fabrics”, Synthetic Metals, Vol. 155 No. 1, pp. 89-94. Zhang, H., Tao, X.M., Yu, T.X. and Wang, S.Y. (2005), “Electro-mechanical properties of stainless steel knitted fabric made from multi-filament conductive yarns under unaxial extension”, Text. Res. J., Vol. 75 No. 8, pp. 598-606. Yang, B., Tao, X.M. and Yu, J.Y. (2004), “Fiber Bragg grating sensor for simultaneous measurement of strain and temperature”, J. Industrial Textiles, Vol. 34 No. 2, pp. 97-116. Yu, J.M., Tao, X.M. and Tam, H.Y. (2006), “Fabrication of UV sensitive single-mode polymeric optical fiber”, Optical Materials, Vol. 28 No. 3, pp. 181-188. Yang, B., Tao, X.M., Ho, H.L. and Yu, J.Y. (2004), “Tangential load measurement by optic smart cellular textile composite”, Text. Res. J., Vol. 74 No. 9, pp. 810-818 Yang, B., Tao, X.M. and Yu, J.Y. (2004), “Measurement effectiveness of fiber Bragg grating sensors in textile composites”, J. Text. Res., Vol. 25 No. 3, pp. 48-49. Xue, P., Tao, X.M., Yu, T.X., Kwok, K. and Leung, S. (2004), “Electromechanical behavior and mechanistic analysis of fibers coated with electrically conductive polymer”, Text. Res. J., Vol. 74 No. 10, pp. 929-936.

41

IJCST 21,6

42

Manchester, UK Textiles and Paper, School of Materials, The University of Manchester, Sackville Street, Manchester M60 1QD. Tel: 0161 306 4113; Fax: – E-mail: [email protected] The current 2008/09 projects are: Mathematical modelling of complex 2D/3D weaves. Mechanical modelling of 2D/3D woven composites. 3D compoasites against impact. Riot helmet research. Ballistic body armour. Riot body armours for female officers. 3D weaving technology on conventional looms.

Aims and objectives Not available.

Deliverables Not available. Publications and outputs Not available.

Maribor, Slovenia University of Maribor, Faculty of Mechanical Engineering, Smetanova ulica 17 SI-2000 Maribor, Slovenia. Tel: +386 2 220 7960; Fax: +386 2 220 7996; E-mail: [email protected] Principal investigator(s): Prof. Dr Sc. Jelka Gersˇak Research staff: Research Unit Clothing Engineering

Clothing engineering and textile materials Other Partners: Academic

Industrial

None None Project start date: 1 January 2004 Project end date: 31 December 2009 Project budget: 80.000 ECU for 2008 Source of support: Slovenian Research Agency Keywords: Clothing, Fabric, Fabric mechanics, Behaviour, Comfort, Prediction

The research programme was carried out in the frame of three interdependent and closely thematically connected complexes: (1) basic investigations of mechanics of fabric as complex textile structures; (2) modelling of the behaviour of complex textile structures; and (3) characterisation of the parameters of thermophysiological comfort and design of a model for heat transfer from the human’s body to the environment. In the frame of the first complex of activities we defined the qualitative characteristics of garment appearance. Here, we defined fabric behaviour from the point of view of mechanics during remodelling from a plane into a 3D form of a garment and define the importance of the parameters related to mechanical and physical properties of fabrics considering previously defined factors of garment appearance quality, defined with marks from 1 to 5. Taking into account the defined principles and factors related to the garment appearance quality we designed a system for predicting garment appearance quality. The developed system, entitled InSiNaKVO, represents an innovative and first of its kind tool for objective evaluation and prediction of a level of garment appearance quality. The essence of the system, which is based on defined relationships between the parameters of mechanical and physical properties of fabrics and level of the garment appearance quality, is prediction of garment appearance quality taking into account mechanical and physical properties of fabrics, built-in the garment, which enables: .

objective evaluation of the level of garment appearance quality;

.

prediction of five important factors of the level of garment appearance quality: garment fall, 3D garment form, fit, quality of produced seams and garment appearance quality as a whole;

planning of the required level of garment appearance quality; prediction of the level of garment appearance quality for concrete fabrics in real production systems; and . planning of quality parameters related to mechanical and physical properties of fabrics for desired level of garment appearance quality. In the frame of the second complex of the research, i.e. modelling of the behaviour of complex textile structures, we continued with the researches related to complex deformation of fabrics underlining the study of rheological model. The relaxation phenomena have been studied form the point of view of mechanical multi-component models as tools for describing deformational and relaxation phenomena in the fabrics. Based on extensive research work we have established that there are significant differences between the behaviour of conventional and elastic fabrics. Researches have shown that multi-component models are the most suitable for explaining the relaxation stress in fabrics containing elastane yarns because of greater number of parameters. Special attention was given to simulation of draping using the finite elements method, where the fabric model is based on rheological parameters and mass points method, which is derived form the linear rheological model. Important here is a study of the relationship between the computer based simulation of fabric draping used as material model and parameters of mechanical properties of fabrics. In the continuation we will focus above all on numerical simulation of fabric draping, based on mechanical . .

Research register

43

IJCST 21,6

44

multi-component models. Our expectations are to come closer to real properties of fabrics, which is a starting point for virtual representation of garment models. The third complex of the research is directed to the characterisation of thermophysiological comfort. For this purpose we have carried out extensive researches related to transmission and transition of heat, material properties of textile materials and thermophysiological comfort at garment wearing for business apparel and for an example of technical application (effect of the type of material for automotive seats and type and construction of a bed on thermophysiological comfort of a user). A special attention has been given within the frame of these researches to the study of material properties and specific requirements related to fabrics for men’s business suits (conventional and functional (phase-changing) fabrics). The results of research have shown that the business clothing system with built-in phase-changing material does not influence significantly the thermal regulation of a body. Clothing systems with built-in phase-changing material showed a short-time thermal effect, seen as small increase or decrease of skin temperature, depending on climatic conditions.

Aims and objectives The main goals of the research programme Clothing engineering and textile materials, which is based on complex researches of fabric mechanics, material properties and relationship between the material properties of textile fabrics and parameters of thermophysiological comfort at garment wearing, are as follows: .

to define the non-linear behaviour of flat textile products as complex structures at lower stresses;

.

to design a model for prediction of garment appearance quality;

.

to define the parameterisation of textile materials and products;

.

to enable a virtual representation of different garment styles, based on real properties of textile materials; and

.

to set-up the characterisation of parameters related to thermophysiological comfort.

Deliverables Realisation of the research programme proposed is offering, using the theoretical knowledge obtained by investigating fabric mechanics and the theory resulting from it, as well as numerical and physical experiments, new knowledge than enable prediction of non-linear fabric mechanical properties in the area of low loads, as well as the simulation of fabric behaviour. The theoretical knowledge obtained by studying the relation among textile material properties, heat exchange and human body thermoregulation, will result in the development of a numerical simulation for heat exchange among the body and garment or some other textile product, and the environment, which is a clear trend in further development of textile and garment engineering globally. The development is aimed at the parameterisation of textile materials, engineering design of high-quality and multifunctional textile materials and products, numerical simulation of their behaviour and virtualisation on one hand and ensuring thermal-physiological comfort on the other. The results of the research programme will indirectly help the application of the scientific research results in practice, which has already been made real in the case of the

developed intelligent system for predicting garment appearance quality InSiNaKVO, the main characteristics of which are as follows: . knowledge-based designing and predicting garment appearance quality degree, using known fabric mechanical and physical properties; . planning fabric mechanical and physical properties so as to get the target garment quality level; and . tools to simulate the degree of garment quality. Publications and outputs Tama´s, P., Gersˇak, J. and Hala´sz, M. (2006), “Sylview 3D drape tester – new system for measuring fabric drape”, Tekstil, Vol. 55 No. 10, pp. 497-509. Zavec Pavlinic´, D., Gersˇak, J., Demsˇar, J. and Bratko, I. (2006), “Predicting seam appearance quality”, Textile Research Journal, Vol. 76 No. 3, pp. 235-242. Gersˇak, J. and Marcˇicˇ, M. (2007), “Development of a mathematical model for the heat transfer of the system man – clothing – environment”, International Journal of Clothing Science and Technology, Vol. 19 Nos 3/4, pp. 234-241. Celcar, D., Meinander, H. and Gersˇak, J. (2008), “A study of the influence of different clothing materials on heat and moisture transmission through clothing materials, evaluated using a sweating cylinder”, International Journal of Clothing Science and Technology, Vol. 20 No. 2, pp. 119-130. Celcar, D., Meinander, H. and Gersˇak, J. (2008), “Heat and moisture transmission properties of clothing systems evaluated by using a sweating thermal manikin under different environmental conditions”, International Journal of Clothing Science and Technology, Vol. 20 No. 4, pp. 240-252. Gersˇak, J. and Marcˇicˇ, M. (2008), “Assesment of thermophysiological wear comfort of clothing systems”, Tekstil, Vol. 57, No. 10, pp. 497-515. Gersˇak, J. and Zavec Pavlinicˇ, D. (2005), “System of predicting garment appearance quality”, Book of Proceedings of the 34th Textile Research Symposium at Mt. Fuji, Susono City, Japan, 9-11 August, pp. 51-53. Gersˇak, J. and Marcˇicˇ, M. (2006), “Thermophysiological comfort”, Book of Proceedings of the 1st International Workshop Design – Innovation – Development, Iasi, 24-27 July, pp. 143-147. Gersˇak, J. (2006), “Thermophysiological comfort of the driver”, Book of Papers of 2nd International Material Conference TEXCO, Ruzˇomberok, Slovakia, 17-18 August, pp. 1-8. Gersˇak, J. and Marcˇicˇ, M. (2006), “Modeling the heat transfer of the system man-clothing-environment”, Book of Proceedings of 3rd International Textile, Clothing and Design Conference ITC&DC, Dubrovnik, Croatia, 8-11 October, pp. 420-425. Zavec Pavlinic´, D. and Gersˇak, J. (2007), “The advanced method for the evaluation of garment appearance quality grade”, Proceedings of 7th Annual Textile Conference by AUTEX, Tampere, Finland, 26-28 June, pp. 1-6. Gersˇak, J. and Marcˇicˇ, M. (2007), “A novel approach to improve the thermophysiological clothing comfort”, Book of Proceedings AGILTex Design – 2nd International Workshop Design – Innovation – Development, Ias¸ i, Romania, 31 August-3 September, pp. 102-112. Gersˇak, J.and Zavec Pavlinic´, D. (2007), “Objective evaluation of garment appearance”, Book of Proceedings of 5th International Conference Innovation and Modelling of Clothing Engineering Processes IMCEP 2007, Faculty of Mechanical Engineering, Maribor, Slovenia, 10-12 October, pp. 34-39. Celcar, D., Gersˇak, J. and Meinander, H. (2007), “The influence of environmental conditions on thermophysiological wear comfort of business clothing”, Book of Proceedings of 5th International Conference Innovation and Modelling of Clothing Engineering Processes IMCEP 2007, Faculty of Mechanical Engineering, Maribor, 10-12 October, pp. 141-149.

Research register

45

IJCST 21,6

46

Gersˇak, J. (2008), “Study of the asymmetric behavior of the fabrics in a produced garment”, Book of Proceedings of 4th International Textile, Clothing and Design Conference, Faculty of Textile Technology, University of Zagreb, Zagreb, 5-8 October, pp. 587-592. Gersˇak, J. (2006), Mechanical and Physical Properties of Textile Materials (Mehanske in fizikalne lastnosti tekstilnih materialov), Faculty of Mechanical Engineering, University of Maribor, Maribor, ISBN 86-435-0754-7.

Newark, Delaware, USA University of Delaware, 201 Alison Hall West, Newark, DE 19716, USA. Tel: 1-302-831-6124; Fax: 1-302-831-6081; E-mail: [email protected] Principal investigator(s): Huantian Cao, Rita Chang, Jennifer McCord, Jenna Shaw, Heather Starner, Jo Kallal

Change without buying: an application of adaptable design in apparel Other Partners: Academic

Industrial

None Lauren Heine, Lauren Heine Group LLC Project start date: 15 August 2009 Project end date: 14 August 2010 Project budget: $10,000 Source of support: US Environmental Protection Agency Keywords: Material conservation, Environmentally conscious manufacturing, Inherently benign materials and chemicals, Reuse Excess consumption of apparel is driven by the apparel industry to offer more styles at lower prices in shorter time and the consumers’ desire to change fashion. Environmental problems such as pollution, hazardous waste, and natural resource depletion are related to excess apparel consumption. According to University of Cambridge report, on average, to produce 1 kg of textile and clothing output, about 0.6 kg of oil equivalent primary energy and 60 kg of water are used, and about 2 kg of CO2 equivalent, 45 kg of waste water, and 1 kg of solid waste are generated. The goal of this project is to demonstrate a strategy of collaboration between apparel industry and consumers to decrease discard, increase utilization, retard fast-fashion, and promote longer wear of garments. We will implement adaptable design in apparel and demonstrate that adaptable apparel will meet consumers’ needs to change while reduce overall production and consumption. In this project, the target users are female college students and the apparel adaptability focuses on function, fit, and style.

Aims and objectives The purpose of this project is to apply adaptable design in apparel and demonstrate that this strategy will allow the apparel industry to make a profit with better design and high

quality product rather than large quantity and low quality; and will meet consumers’ desire to change without buying. Our objectives include: . designing and producing adaptable apparel for female college students by using environmentally friendly materials; . evaluating the adaptability, consumers’ acceptance, and cost of our design and product; and . revising the design based on evaluation results and developing educational tools.

Deliverables Design and research are in progress. Publications and outputs Not available.

Nottingham, UK University of Nottingham, School of Mechanical, Materials and Manufacturing Engineering, University Park, Nottingham, NG7 2RD. Tel: 0115 9513779; Fax: 0115 9513800; E-mail: [email protected] Principal investigator(s): C.D. Rudd, A.C. Long, R. Brooks, I.A. Jones, S.J. Pickering, N.A. Warrior, M.J. Clifford, C.A. Scotchford, G.S. Walker Research staff: H. Lin, L. Harper, R.R.H. Naqasha

Platform grant: processing and performance of textile composites Other Partners: Academic

Industrial

None None Project start date: 1 February 2005 Project end date: 31 January 2009 Project budget: £445k Source of support: EPSRC Keywords: Textile composites, Unit cell analysis, TexGen Our current research portfolio is centred on the processing of polymer matrix composites with a growing emphasis on modelling and simulation. Given our high level of interest in textile-based composites and their growing importance in the field, we wish to introduce a common platform for our modelling studies based on our formalised textile generator (TexGen). Textile modelling provides a launchpad for downstream simulation of processing, damage mechanics and environmental performance. The functionality of our existing TexGen software will be extended and coupled to materials models for simulation of each of the above aspects of physical behaviour. Simulation of each of the physical processes will be enhanced by a common, interchangeable geometric definition

Research register

47

IJCST 21,6

48

of the textile structure within the rigid composite. This will enable a rapid understanding of fabric architecture effects to be built and the approach has excellent potential for application to other physical problems which relate to rigid and flexible composites or technical textiles. The platform grant application seeks continuity of support for key workers during the period of this development. Further details are available at: www.textiles.nottingham.ac.uk

Aims and objectives .

Implement an approach based on textile modelling throughout our research portfolio, integrating the multiple streams of processing, energy management, biomedical applications and textile modelling.

.

Develop a series of downstream models relating to: three-dimensional permeability, formability (including shear compliance), static mechanical properties, damage mechanics and residual property estimation, diffusion and environmental degradation. Exploit the potential of the platform grant to raise our international profile, develop strategic links with other leading groups, and enhance our technology transfer activities.

.

Deliverables Not available. Publications and outputs Not available.

Ohtsu, Japan SCI-TEX (Consultant), 12-15, Hanazono-cho, Ohtsu-city, 520-0222 Japan. Tel: 81-77-572-3332; Fax: 81-77-572-3332; E-mail: [email protected] Principal investigator(s): Tatsuki Matsuo

Fiber assembly structures for technical textiles in terms of function and end-use Other Partners: Academic

Industrial

None None Project start date: October 2007 Project end date: 2009 Project budget: 400,000yen Source of support: Own Keywords: Fiber assembly structure, Function, End-use, Technical textiles, Nonwovens, Composites

General criterion for selecting the structure of technical textiles is the ratio (performance/cost). Verasatility in the structure of technical textiles is much wider than apparel textiles. Objective of this project is to find guiding princiiples for designing the optimum structure of technical textile in terms of their required function and cost. Several fundamental analyses on the fiber assembly structure and function of technical textiles are investigated. Directive fiber assembly structures suitable for required functions are analysed based on the analytical results of the investigation. The results are tried to be applyied to nonwovens and composites.

Aims and objectives Not available.

Deliverables Not available. Publications and outputs Not available.

Ohtsu-city, Japan SCI-TEX, 12-15 Hanazono-cho, Ohtsu-city, 520-0222 Japan. Tel: 81-77-572-3332; Fax: 81-77-572-3332; E-mail: [email protected] Principal investigator(s): Tatsuki Matsuo

Propagation of knowledge on new textile science and technology Other Partners: Academic

Industrial

None None Project start date: N/A Project end date: On going Project budget: N/A Source of support: N/A Keywords: Advanced technical textiles, Knowledge propagation The importance of advanced technical textiles has increased in the textile industry of developed countries. In addition, R&D on nano-technologies and electric-textiles are now intensively carried out. In this situation, propagation of knowledge on new textile science and technology must be meaningful. This project is being conducted individually by T. Matsuo through symposium lectures, journal articles and monographic books.

Research register

49

IJCST 21,6

Aims and objectives Not available.

Deliverables Not available.

50

Publications and outputs Not available.

Ohtsu, Japan SCI-TEX, 12-15 Hanazono-cho, Ohtsu-city, 520-0222 Japan. Tel: 81-77-572-3332; Fax: 81-77-572-3332; E-mail: [email protected] Principal investigator(s): Tatsuki Matsuo

Fiber assembly structure science and engineering Other Partners: Academic

Industrial

None None Project start date: 1 February 2009 Project end date: 30 January 2010 Project budget: N/A Source of support: N/A Keywords: Fiber assembly structure, Soft material, Maximum fiber volume fraction, Tranfer properties, Applications The objective of this study is a trial to establish comprehensive concept and systematic knowledge of fiber assembly structure in the relation to soft materials. Major problems to be expolored in this study are; maximum fiber volume fraction, mechanical properties, mass and energy transportation properties in terms of fiber assembly structure.

Aims and objectives Not available.

Deliverables Not available. Publications and outputs Not available.

Pisa, Italy University of Pisa, Via Diotisalvi, 256126 Pisa, Italy. Tel: +39 0502217053; Fax: +39 0502217051; E-mail: [email protected] Principal investigator(s): Prof. Danilo De Rossi Research staff: Prof. Bruno Neri; Ing. Alessandro Tognetti; Ing. Enzo Pasquale Scilingo; Ing. Federico Carpi; Ing. Antonio Lanata`

PROETEX: Protection e-Textiles: micronanostructured fibre systems for emergency-disaster wear Other Partners: Academic Consiglio Nazionale delle Ricerche – INFM, Technical University of Lodz, Ghent University – Department of Textiles, University of Pisa, Dublin City University, Institut National des Sciences Applique´es de Lyon

Industrial Smartex srl, Milior, Sofileta SAS, Thuasne, Commissariat a` l’Energie Atomique – “CEA”, CSEM Centre Suisse d’Electronique et de Microtechnique SA, Sensor Technology and Devices Ltd, Steiger S.A., Philips GmbH, Zweigniederlassung Forschungslaboratorien, Ciba Spezialita¨tenchemie AG, Diadora Invicta SpA, iXscient Ltd, Zarlink Semiconductor Limited, Brunet-Lion SAS, Brigade de Sappeurs Pompiers de Paris, European Centre for Research and Training in Earthquake Engineering, Direction de la De´fense et de la Se´curite´ Civiles

Project start date: February 2006 Project end date: January 2010 Project budget: University of Pisa: e780,443; Total: e 12,792,242 (Requested: e 8.100.000) Source of support: European Commission ProeTEX will develop integrated smart wearables for emergency disaster intervention personnel, improving their safety, coordination and efficiency and for injured civilians, optimising their survival management. This core application area, which is of significant societal importance in itself, will drive a wide range of key technology developments, building on current and past EU and national projects and the commercial activities of partners, to create micro-nano-engineering smart textile systems – integrated systems (fabrics, wearable garments) using specifically fibrebased micronano technologies. These are capable of being combined into diverse products addressing this core application area but also a wide range of other markets from extreme sports, through healthcare to transportation maintenance and building workers. The industrial partners can address these markets.

Research register

51

IJCST 21,6

52

Fiber systems can integrate sensors, actuators, conductors, power management, and the emergency disaster personnel smart garment will, within a wireless ambient planning and managing environment, progressively enhance and integrate fiber systems for: .

continuous monitoring of life signs (biopotentials, breathing movement, cardiac sounds);

.

continuous monitoring biosensors (sweat, dehydration, electrolytes, stress indicators);

.

pose and activity monitoring;

.

low power local wireless communications, including integrated fiber antennae;

.

active visibility enhancement, light emitting fibers;

.

internal temperature monitoring using fiber sensors;

.

external chemical detection, including toxic gases and vapours; and

.

power generation – photovoltaic and thermoelectric and power storage.

The technological base developed will concentrate on smart fibers/e-textiles, but the IP will combine these where appropriate with conventional’ microsystems (such as accelerometers, gyros, microcontrollers and wireless chips).

Aims and objectives The central IP goal is to develop an integrated set of functional garments for emergency disaster personnel, such as firefighters and paramedics, plus systems for injured civilians. These will be produced using both enhanced and novel fibre based micronanosystems, whose development will extend the state of the art in this area. The project will roll out a sequence of progressively more capable integrated wearable systems for emergency disaster intervention personnel and injured civilians. Thus, overall the IP will: .

Progress the fundamentals of fibre-based sensor, processing, communications and power management systems.

.

Integrate these fibre-based capabilities into functional knitted or woven wearable garments.

.

Produce fully capable integrated communicating, sensor wearables, using additional “conventional” systems where necessary.

.

Test their match of user needs and requirements in a lab-based setting.

.

Demonstrate their function in a real-world application in a number of field trials.

Scientific objectives: . Develop a multifunctional garment integrating an increasingly ambitious set of sensors and energy harvesting and storage which is reliable, robust, easy to wear and capable of manufacture. .

Into this garment: Design, test and integrate a bioelectrical heart rate monitor into whole skin contact garment interface; Design, test and Integrate a cardiac sound monitor; Integrate sensor breathing monitor and ensure that signal conditioning and processing results in successful way.

.

Develop fibre and new textile based technological solutions, with reliable functionality, capable of integration into wearable garments covering the following set of technological area capabilities. Monitor bioelectrical potential. Sensing breath movements. Sensing posture and movement. Biochemical sensing, specifically determination of dehydration status. Sensing core temperature. Acting as local communications antennae. Sensing external toxic gases/chemicals, including CO. Generating local energy using thermoelectric generation. Generating local energy using photovoltaic processes. Storing energy using Li-Ion textile batteries.

Technical objectives: . Develop and adapt textile manufacturing processes to these new active fibres and layers (weaving, knitting, coating, laminating) but also innovate in terms of clothes conception to optimise the assembly step regarding interconnection needs for e-textile garment. .

.

.

Develop and test a multifunctional (inner and outer) garment integrating an increasingly ambitious set of sensors and energy harvesting and storage which is reliable, robust, easy to wear and capable of manufacture for both intervention people and injured civilians. The inner and out garment will include an adapted set of functionalities based on the developed technologies. As example first inner garment could integrate bioelectrical heart rate monitor, cardiac sound monitor, strain sensor breathing monitor inner temperature measurement and ensure that signal conditioning and processing results in successful and robust physiological monitoring. Energy generated by the heat (thermoelectricity) and the movement (piezoelectricity) of the of the wearer. Outer garment will typically include toxic gas measurement, external temperature; motion and position monitoring, data transmission system, energy could be provided by photovoltaic external layer and textile Li-Ion batteries. Realize field trial of the instrumented garment for technological validation.

Deliverables The key deliverables will be: . An inner garment for emergency disaster personnel, monitoring the health of the user through vital signs, biochemical parameters, activity and posture, generating and storing own power and communicating locally with other wearables and relaying through (D). . An outer garment for emergency disaster personnel, measuring potential environmental insults (temperature, CO, other toxic gases), sensing posture and movement of the wearer and offering improved visibility, generating and storing its own power communicating locally with other wearables and relaying through (D).

Research register

53

IJCST 21,6

.

.

54

.

An under-garment jerkin or chest band for injured civilians (closely related to (A)) monitoring their health, generating and storing its own power and communicating locally, relaying information via (E). Victims monitoring measures will include: body temperature; cardiac pulse; respiration rate; ECG; percutaneous CO saturation; percutaneous O2. A portable unit for the emergency disaster personnel, communicating with A, B and C, but offering additional “conventional” microsystem, providing both local and long range communication (acting as a relay for A, B and C), including some specific sensors not easily integrated into (B), plus accelerometers, gyros and GPS to enable high accuracy position and movement determination. This device should allow data entry and displays/alarms. A portable unit for injured civilians, to include data relay capability and INS/GPS but no data entry or display. Some kind of integrated alarm or indicator to give the overall civilians health status. A simple user input, such as panic button, may be required.

Publications and outputs Not available.

Seoul, Republic of Korea A-108, iFashion Technology Center, Col. of Eng., Konkuk University, GwangJinGu, Seoul, Korea, 143-701. Tel: +82-2-17-310-6317; Fax: +82-2-452-7865; E-mail: [email protected] Principal investigator(s): InHwan Sul

Automatic shape recognition of garment patterns for digital garment sewing Other Partners: Academic

Industrial

None None Project start date: 1 January 2009 Project end date: N/A Project budget: N/A Source of support: N/A Keywords: Garment pattern, 3D digital garment, Sewing information, Shape recognition, Shock graph The speed of cloth simulation is approaching almost real-time thanks to the algorithmic development and upgraded hardwares. Now 3D apparel CAD systems can generate realistic 3D virtual garment from 2D pattern data. To make a 3D virtual garment data from 2D pattern data, there are two kinds of works the user has to do. The first is to assign sewing information to the patterns so that individuals patterns can be merged into a single

garment. The second is to apply a texture map which will give colors and realistic outlook to the virtual garment. These two steps are bottleneck for the mass production of 3D virtual garment data. In this research, we focus on the automation of the first step, i.e. automation of sewing information assignment. For this to be possible, we need some kind of artificial intelligence to recognize the shape of each pattern. Among various techniques, shock graph is the most viable method to recognize the pattern shapes. We make database of shock graphs of various patterns in advance and find the nearest graph structure if a specific pattern is given. And then we give name based sewing information to each sewing edge. If we repeat this procedure for all the patterns of a garment set, the digital garment can be generated automatically without user intervention.

Aims and objectives There are three components to achieve. (1) Automatic shape recognition using shock graph: the patterns should be converted into shock graph easily and exactly. (2) Construction of pattern shock graphs: to find a sewing edge of anonymous pattern, it is advantageous to have as many standard template patterns as possible. (3) Standardization of sewing edge information: to combine patterns of variuos garment sets, it is essential to make standard format of sewing information. We define a sewing name for each edge and automatize the sewing operation of digital garments.

Deliverables With this technique, there can be several applications: . Automatization of sewing in 3D Apparel CAD System: currently the user should manually assign two sewing edges which will make a sewing line with a mouse action. This can be done automatically without user intervention if the pattern recognition is done perfectly. Even if the recognition has error, the automatic sewing result can be used as a guide for the manual sewing work. . Digitization of garment patterns in apparel industry: although 3D CAD systems are available, there are thousands of garment patterns stored in the apparel companies. It is both time and labor intensive to make them digital. But using this technique, the patterns only have to be photo-taken and image processed to generate digital patterns from the boundary segmentation. Once the boundary shape is known, the 3D garment shape can be acqured with this technique automatically.

Publications and outputs For name based sewing, a paper was submitted: In Hwan Sul (2008), “Style previewing in 3D using name based sewing rules”, International Journal of Clothing Science and Technology, (submitted). Others are in progress.

Research register

55

IJCST 21,6

56

Seoul, Republic of Korea A-108, iFashion Technology Center, Col. of Eng., Konkuk University, GwangJinGu, Seoul, Korea, 143-701. Tel: +82-2-17-310-6317; Fax: +82-2-452-7865; E-mail: [email protected] Principal investigator(s): In Hwan Sul

Automatic generation of personal avatar from 3D body scan data Other Partners: Academic

Industrial

None None Project start date: 1 January 2008 Project end date: N/A Project budget: N/A Source of support: N/A Keywords: Body scan, Avatar generation, Mesh registration, Particle based method The 3D body scan technology has been commercially available and the individuals can get their body shapes in a short time. But the current technology has the limit that the output mesh data has only coordinates and not colors, textures and motion data. Of course some scanners offers vertex colors but they are not satisfactory enough to be used as a avatar. Moreover, the 3D mesh is static, which means that it cannot deform or change pose. We adopt nonrigid mesh registration technique to give the 3D scan data vertex colors and motion skinning weights. To faciliate the calculation, we use particle based method to relax the meshes. The technique is composed of three steps. The first step is to prepare template avatar with perfect colors and motion informations in advance. The second step is to find anthropometric feature points automatically both for template avatar and the target 3D scan data. The last step is to relax the 3D scan mesh elements so that the same feature points have the same texture colors. With this method, an avatar with realistic body colors and pose-changeable skinning weights can be generated from 3D scan data within several minutes. Additionally, the face can be replaced by using commercial face generation softwares.

Aims and objectives .

.

.

Automatic feature points detection of human bodies: to match two different body meshes, we must know the same anthropometric positions. We use the conventional slicing method to find the feature points. Nonrigid mesh registration using particle based method: semi-implicit particle based method is widely used for real-time cloth simulation because of its numerical stability and speed. We adopt this technique to relaxing the body meshes. Changing body parts: some body parts such as hands, feet and head should be changeable. As the template body has fixed number of mesh vertices, we can use external commerical softwares which generates 3D head mesh from face pictures.

Deliverables Any kind of 3D body scan data can be converted to a colorful avatar with body textures and motion skinning weights only if the feature points can be found successfully. Publications and outputs A paper was submitted: Hwan Sul (2008), “Regeneration of 3D body scan data using semi-implicit particle based method”, International Journal of Clothing Science and Technology, (submitted).

Seoul, Republic of Korea A-108, iFashion Technology Center, Col. of Eng., Konkuk University, GwangJinGu, Seoul, Korea, 143-701. Tel: +82-2-17-310-6317; Fax: +82-2-452-7865; E-mail: [email protected] Principal investigator(s): InHwan Sul

Finding the fiber orientation information from SEM image using metaball approximation Other Partners: Academic

Industrial

Pr. ChangKyu Park Dr YoungJun Cho Project start date: 1 January 2008 Project end date: N/A Project budget: N/A Source of support: N/A Keywords: Fiber SEM image, Metaball, Image analysis, Orientation, Radial distribution Scanning electron microscope images are the essential tool for identifying the microstructures of fibrous and polymeric materials. Useful information including fiber volume fraction, orientation and radius distributions can be retrieved from the SEM images. But the image is two dimensional so that manual measurement was inevitable for acquiring the volumetric information. This paper reconstructed the 3D shape of the fibers by approximating the fibers with spheres called as metaballs. Image analysis techniques were used to generate metaballs and the volumetric information were calculated from the size and positions of metaballs. The proposed method was tested with nano-scale polymer SEM images and automatic statistical measurements were compared with manual measurements.

Aims and objectives The objective is to find as many measurements as possible from the SEM image for radius and orientation information. Although not all the fibers in the SEM image can be detected, this method can give much more data than manual measurement using a ruler.

Deliverables SEM image of fiberous material can be analyzed and the volumetric distribution can be known automatically.

Research register

57

IJCST 21,6

58

Publications and outputs A paper was submitted; In Hwan Sul, Young Jun Cho and Chang Kyu Park (2009), “Regeneration and analysis of polymer SEM imageusing image analysis and 3D metaball approximation”, Textile Research Journal, (submitted).

Seoul, Republic of Korea A-108, iFashion Technology Center, Col. of Eng., Konkuk University, GwangJinGu, Seoul, Korea, 143-701. Tel: +82-2-17-310-6317; Fax: +82-2-452-7865; E-mail: [email protected] Principal investigator(s): InHwan Sul

Real-time cloth simulation using particle based method and Pluecker coordinates Other Partners: Academic

Industrial

None None Project start date: 1 January 2009 Project end date: N/A Project budget: N/A Source of support: N/A Keywords: Cloth simulation, Particle based method, Collision detection Particle based method is more suitble method for cloth simulation than finite element method because of its numerical stability and speed. But due to the complex nature of cloth, it is not easy to simulate the cloth in real-time. There are two components to achieve real-time in cloth simulation. The first one is the calculation of velocities (i.e. drape engine) and the second one is the collision detection (collision engine). We use various techniques which are currently known to speed up the calculation of the velocities such as using block symmetry of matrices or modal-analysis based bending force approximation. And for the collision detection, we approximated the meshes to a k-DOP’s and found the intersection among velocity vectors and k-DOP’s. Pluecker coordinate was used for fast ray-BOX intersection test.

Aims and objectives The objective is to make a versatile drape engine which can simulate cloths in any environment.

Deliverables The technique can be readily used in 3D apparel CAD systems and modeling of textile process. Moreover, it can be applied to any othe scientific research such as a medical simulation where calculation speed is more important factor than mechanical accuracy.

Publications and outputs This work is based on the previous works: In Hwan Sul and Tae Jin Kang (2004), “Improvement of drape simulation speed using constrained fabric collision”, International Journal of Clothing Science and Technology, Vol. 16, pp. 43-50. In Hwan Sul, Sung Min Kim, Yong-Seung Chi and Tae Jin Kang (2006), “Simulation of Cusick drapemeter using particle based modeling: stability analysis of explicit integration methods”, Textile Research Journal, Vol. 76 No. 9, pp. 712-19. In Hwan Sul (2008), “Fast cloth collision detection using collision matrix”, International Journal of Clothing Science and Technology (submitted).

Sliven, Bulgaria College of Sliven at TU-Sofia, 59 Bourgasko chausse´e blvd., 8800 Sliven, Bulgaria. Tel: 00359 44 667710; Fax: 00359 44 667505; E-mail: [email protected] Principal investigator(s): Ivelin Rahnev, Associate professor, PhD Research staff: 4 lecturers and a student from College of Sliven

Technology optimum of double fabric weaving cycle for work garment (“Bytoffa09”) Other Partners: Academic

Industrial

None “E. Miroglio”, AD – Sliven Project start date: 15 May 2009 Project end date: 27 November 2009 Project budget: 3,850 euros (7500 bgn) Source of support: TU – Sofia, NIS – Internal competition, project “091ni047-16” Keywords: Cotton textiles, Weaving technology, Double fabric, Technology optimum, Work cloths Fabrics for work garments with general protective properties are well-known textile products. They constitute basic quota in the cotton mills. Two layers fabrics, combining resistance raised with healthful properties improved at aesthetics preserved, are rare in the current range of fabrics for work garments with middle class of protection. At present moment, profound investigations, based on technology optimum, on this problem are not known. Technical conditions to produce two layer fabrics – cotton type, “denim” appearance, for protective garments in accordance to EN 471 and ISO 11092, are not established. The basic idea is to obtain the visible denim relief on the face, which means, that the elementary weave of the face is four-weave twill with warp effects. In the square pattern composed of 14 warp and weft threads there is place for 4 denim rapports; there alternated as follows: 2 along the warp and 2 along the weft. Because the pattern rapport and the denim rapport are not multiples each to other the 7th and 8th warp and weft

Research register

59

IJCST 21,6

60

threads approach each to other in the middle of the whole weave. Thus, the denim is concentrated in the middle of the weave. In the square pattern remains 6 6 warp and weft threads, they are used to distribute the connecting points of the three-twill 4 rapports. On the common fabric back, i.e. on the inner layer face, we obtain twill with weft effect; but concerning the general pattern for the heald shafts movements, this is equivalent to three-weave twill with warp effect. Thus, we have combined on the same fabric, two twills with orthogonal diagonals in order to equilibrate the common construction. The interconnections between the two fabrics are made by means of 4th and 11th ends; they rise alternatively as a common warp covering and attach the third and the 12th picks to the inner layer. The fabrics face shows warp and the back shows weft appearances. The traction, or the feeding, of the warp as well as that of the weft picks are the two controllable and functionally independent machine adjustments. They allow us to control the quantitative factors over a spherical experimentally field. The generalized nonlinear character of the textile materials rheology behaviour needs a model with at least 2nd degree surface response. In our experiment, we want to verify the hypothesis of the viscoelastic reaction of the designed fabric. The working conditions allow us to choose the centrally composed rotatable experiment plan depending on two factors. Our experimental work was realized on the weaving loom Vamatex Leonardo in the weaving mills facilities. The warp tension variations, as well as the weft density variations depend on the simultaneous action of the let-off and the take-up motions.

Aims and objectives The objective of this project is through technology optimimum a fabric “double weaves” consisting of ecological cotton fibres and characterized with increased wear resistance to be designed and introduced in production. The fabric is allocated for the manufacture of work cloths with improved hygienic properties in accordance with European Standard EN471/2. The work aims to determine the influence of the weft density as well as the warp tension on the fabric mechanical properties. The successful solution of this problem includes the consecutive applying of methods as the following: Successive tests of the project thus formulated envisage the consecutive application of complex methods: .

Broad library studying – ergonomic requirements, standards and alternatives.

.

Metrology tests of mechanical resistance of the materials throughout all stages of production of the designed fabric. Metrology selection of simple and plied threads obtained various homogeneous fibrous raw materials.

.

Composition of the weave and design of the principal technical parameters of the load and the adjustment of the weaving loom.

.

Preparation of raw materials for the weaving, warping, sizing, etc.

.

Experimental planning and adjustments of the weaving looms.

.

Statistical processing and composition of optimum models, and technology optimumu of the designed fabric.

.

Development and analysis of photographic images of the fabric samples.

CAD design and 3D simulation with practical application of the textile CADsystems for the designing of plain fabrics in accordance with the standard requirements for wear resistance, mechanical defence and physiological comfort. . Manufacture of the industrial prototype, development of the schedule of conditions. The development of the fabric with protective functions for functional garments will result in the following: . Technology study of the constructive combination of various threads in a plain product with three-dimensional structure. .

.

Application of performed methods for the design of fabrics with functional purposes according to given customers indexes.

.

Accumulation of experimental data and theoretical researches for preliminary preparation and development of a post-graduate thesis.

Deliverables Scientific application The combination of two elementary twills in double woven fabric builds common weave structure of two layers. When the design provides work cloth, the first layer of the three-dimensional weave texture has mostly aesthetical and protective purpose, while the inner layer gives the user’s physiological comfort. Independently of their function, the autonomous layers give common mechanical reaction against the applied external forces. The need of equalizing the internal stresses and their homogenous diffusion over all the fabric elements makes the detailed description of the fabric mechanical behaviour extremely important, and that in the two directions: the long the warp and, the long the weft. The detailed mechanical behaviour considers the current fabric reaction in the case of traction load with arbitrary intensity and duration. That is why the breaking characteristics cannot be used more as universal definitive criterion, obtained from the rheograms of the mechanical reaction when the fabric submits traction load. Subject of this project is the rheology behaviour of double woven fabriccotton type under traction load. The work aims to determine the influence of the weft density as well as the warp tension on the fabric mechanical properties. Combining two primary weaves in the common structure of double-layer tissue is the result of analytical design and arrangement of links between warp and weft threads in the volume of fabric. The rheology behaviour of dual tissue, depending on the structure of the fabric and machine settings of the loom, is new in terms of mechanical testing of surface textile products. Set of optimizing polynomials, to model the mechanical properties and stability of the rub and air-conductivity and drape, build complex assessment of consumer parameters of the designed tissue: durability, security, hygiene and aesthetic properties. Academic application The problem is described with depth in the analytical studies, a large volume of experimental work and can serve in the establishment of a doctoral dissertation topic. The results of development may be used in training of textile engineering staff.

Research register

61

IJCST 21,6

Industrial application The weaving technology optimized of the prototype, as well as the elaborated technical documentation allow the effective promotion of the new surface product into a textile company with conventional industrial plant.

62 Publications and outputs Previous publications: Rahnev, I.R. (2003), “Functional application of CAD systems in weaving mills”, Textile & Clothing, No. 8, pp. 8-10, ISSN 1310-912X. Rahnev, I.R. (2004), “Kinematic comparison of cam and eccentric mechanisms for weft carriage with solid foil”, National Seminar on Analysis and synthesis of Mechanisms, TU-Sofia, IPF Sliven. Mechanics of Machines, Issue 53, Paper 4/2004, pp. 172-5, ISSN 0861-9727. Rahnev, I.R. and Sakli, F. (2004), “Application of the sample system in textile weaver training”, General Textile Conference 2004, in NTS Textiles, Clothing and Leathers, Textile & Clothing, Nos 8/9, pp. 19-22, ISSN 1310-912X. Rahnev, I.R. (2005), “Dynamic equilibrium of lett-off motion with weights”, Mechanics of Machines, No. 59, Paper 4/2005, National Seminar on Analysis and Synthesis of Mechanisms, TU-Sofia, IPF Sliven, 1-3 July, pp. 38-41, ISSN 0861-9727. Rahnev, I.R. (2005), “Rheology behaviour of tissue with different warp and weft threads”, International EMF Conference 2005, TU Varna, 22-24 September, Proceedings, ISSN 1310-9405, pp. 161-6. Rahnev, I.R. (2005), “Experimental plan of the wrap threads tension – part 1”, Textile & Clothing, No. 6, pp. 9-11, ISSN 1310-912X. Rahnev, I.R. (2005), “Experimental plan of the warp threadstension – part 2”, Textile & Clothing, No. 7, pp. 9-11, ISSN 1310-912X. Rahnev, I.R. (2005), “Influence of the warp tension on the mechanical properties of fabric”, Textile & Clothing, Vol. 15, No. 2, p. 10, 2006, ISSN 1310-912X. Rahnev, I.R. (2005), “Technological features in weaving preparation of the fabric ‘denim’”, Textile & Clothing, No. 11, General Textile Conference 2005, in NTS Textiles, Clothing and Leather, Sofia, 13-14 October, pp. 7-11, ISSN 1310-912X. Rahnev, I.R. (2006), “Metrological pattern selection of work clothing”, Textile & Clothing, No. 8, pp. 21 and 27, ISSN 1310-912x. Rahnev, I.R. (2008), “Rheology behaviour of double fabric – type of cotton”, Textile & Clothing, No. 3, pp. 4 and 9, ISSN 1310-912x. Impending publications (subjects): Specialities in the weaving preparation of double cotton fabric with face warp effect. Structure influence on the mechanical resistance of double cotton fabric with face warp effect. Technology optimum generalized of double cotton fabric with face warp effect. Technical conditions documentation of double cotton fabric with face warp effect. Request for invention protection of the double cotton fabric with face warp effect designed under name “Bytoffa09”. Recognitions: Researchers express their gratitude to Dr Gaetano Rimini, executive manager of “E. Miroglio” AD-Sliven for his commitment and support for implementation of the project.

Buca/Izmir, Turkey Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: 00902324127211; Fax: 00902324127210; E-mail: [email protected] Principal investigator(s): Prof. Dr Ays¸ e Okur Research staff: Res. Ass. Musa Kilic¸

Analyses of unevenness and hairiness on blended yarns Other Partners: Academic

Industrial

None None Project start date: May 2008 Project end date: May 2010 Project budget: e 52,000 Source of support: Dokuz Eylu¨l University Keywords: Blended yarns, Hairiness, Unevenness, Modal, Tencel, Promodal Yarns made of regerated cellulosic fibres such as modal, tencel, promodal and blends of these with cotton have wider use especially in recent years. However, when the literature is checked, it is found that there are not adequate scientific researches on unevenness and hairiness of these kinds of yarns. So, in this project it is aimed to analyze the unevenness and hairiness of blended yarns made of cotton/modal, cotton/tencel and cotton/promodal. In the project, the effect of linear density and twist, the effect of blend type and the effect of spinning system on hairiness and unevenness will be analyzed by using yarns made of 100% cotton, 67%-33%, 50%-50%, 33%-67% cotton-regenerated cellulosic fibre and 100% regenerated cellulosic fibre. Also, performance properties of fabrics made of these yarns will be analyzed within the context of the project.

Aims and objectives In this project it is aimed to analyze the unevenness and hairiness of the yarns statistically, determine the relationships between unevenness-hairiness and offer a more realistic formula to calculate the limit irregularity of these blended yarns. It is hoped that the results of the project which has a fairly wide experimental study will be useful for practical uses.

Deliverables The effect of blend ratio, linear density, twist, blend type, spinning system on hairiness and unevenness of blended yarns will be observed at the end of the study. Also, relationships between hairiness and unevenness of these yarns will be analyzed. Within the context of the project, performance properties of the fabrics made of these yarns will be analysed. A new formula is aimed to be derived from the statistical analyses for the limit irregularity of the blended yarns. Furthermore, different measuring principles for the yarn hairiness will be compared. It is thought that this project will contribute to the further studies on hairiness and unevenness of blended yarns made of regenerated cellulosic fibres with cotton.

Research register

63

IJCST 21,6

Publications and outputs Not available.

Bornova/Izmir, Turkey 64

Ege University, Ege Universitesi Muhendislik Fak. Tekstil Muh. Bol. Bornova Izmir Turkey. Tel: 90 232 339 92 22; Fax: 90 232 339 92 22; E-mail: [email protected] Principal investigator(s): Kerim Duran Research staff: Ays¸ egu¨l Ekmekci Ko¨rlu¨, Seher Perincek and M. ˙Ibrahim Bahtiyari

The use of ultrasound and UV for the oxidative treatment of textile materials, for the acceleration of processes and treatment of textile wastewaters Other Partners: Academic

Industrial

None Alenka Majcen Le Marechal, University of Maribor, Faculty of Mechanical Engineering, Textile Department Project start date: 1 December 2006 Project end date: 1 December 2009 Project budget: 41,035 Eur (78,296 YTL) Source of support: Tubitak Keywords: Ultrasound, UV, Pretreatment, Finishing, Ozone The objectives of the proposed project are the wastewater treatment with AOP processes where full decolouration, maximum reduction of surfactants and toxic compounds, maximum recycling of wastewaters, strong reduction of process sludges and extraction of salts will be achieved by combining of different AOP treatment technologies (US, UV, H2O2 and combinations). The use of experimental design and artificial inteligence should be included for the optimization of the wastewater treatment processes (minimum consumption of the energy, water, chemicals, direct toxicity assessment, . . .). So, the second part of the project will focus on the decolorization and cleaning up treatment of wastewaters resulting from the first part of the project. The aim is to use the same AOP processes for the wastewater treatment as for finishing processes (UV/H2O2, thermal/H2O2, US/H2O2 or O3). In the first step we will study the degradation and decolorization of model wastewater. We will study the influence of different operating conditions for each AOP, such as: temperature, intensivity of UV irradiation, the amount of hydrogen peroxide added, treatment duration, initial concentration of the dye, concentration of other chemicals needed for the dyeing process itself, pH, frequency of the sound wave and power. In the second step we would like to degrade or decolorize the wastewater from the Turkish and Slovene textile factories.

Enzymatic treatments supplemented with ultrasonic energy results in shorter processing times, less consumption of expensive enzymes, less fiber damage and better uniformity of treatment to the textiles. The effect of ultrasound power is an important technique increasing the mass transfer towards to the textile material. In the ozonation process, it is expected that there will be an increase in the whiteness degree of fabrics. Because of the combined effects of ultraviolet light in the presence of hydrogen peroxide, the hydroxyl radicals, much more active then the conventional oxidizing agents will be generated. The textile finishing wastewater treatment is expected to achieve the following characteristics: . Maximal decolorization of the wastewater treated (up to 99%). .

Maximal reduction of environmental parameters according to Slovenian and Turkish environmental legislations (TOC, COD, BOD5, toxicity).

.

Reduction of surfactants and toxic compounds (more than 95%).

The combination of different AOP wastewater treatment processes is expected to achieve a complete decolorization of the process waters for every type of wet process (finishing, bleaching, and dyeing).

Aims and objectives In this study, it has been aimed to investigate the effects of ultrasound, UV and ozone on textile pretreatment processes. Conventional methods are compared with alternative methods which are above. The objectives of the proposed project are the wastewater treatment with AOP processes where full decolouration, maximum reduction of surfactants and toxic compounds, maximum recycling of wastewaters, strong reduction of process sludges and extraction of salts will be achieved by combining of different AOP treatment technologies (US, UV, H2O2 and combinations). The use of experimental design and artificial inteligence should be included for the optimization of the wastewater treatment processes (minimum consumption of the energy, water, chemicals, direct toxicity assessment, . . .). So, the second part of the project will focus on the decolouring and cleaning up treatment of wastewaters resulting from the first part of the project. The aim is to use the same AOP processes for the wastewater treatment as for finishing processes (UV/H2O2, thermal/H2O2, US/H2O2 or O3). In the first step we will study the degradation and decolouration of model wastewater. We will study the influence of different operating conditions for each AOP, such as: temperature, intensivity of UV irradiation, the amount of hydrogen peroxide added, treatment duration, initial concentration of the dye, concentration of other chemicals needed for the dyeing process itself, pH, frequency of the sound wave and power. In the second step we would like to degrade or decolorize the wastewater from the Turkish and Slovene textile factories.

Deliverables Experiments are current and in progress.

Research register

65

IJCST 21,6

Publications and outputs Presentation at Autex 2008: World Textile Conference, 24-26 June, 2008, Biella, Italy, Simona Vajnhandl, Alenka Majcen Le Marechal, Darinka Fakin, Kerim Duran, Aysegul E. Korlu, M. ˙Ibrahim Bahtiyari and Seher Perincek, “The use of ultrasound in pretreatment and dyeing processes and wastewater treatment”.

66

Bornova/Izmir, Turkey Ege University, Ege Universitesi Muhendislik Fak. Tekstil Muh. Bol. Bornova Izmir Turkey. Tel: 90 232 339 92 22; Fax*: 90 232 339 92 22; E-mail: [email protected] Principal investigator(s): Ays¸ egu¨l Ko¨rlu¨ Research staff: Kerim Duran, Seher Perincek and M. ˙Ibrahim Bahtiyari

Enzymatic finishing and the effects on celluolsic textile materials Other Partners: Academic

Industrial

None Alina Popescu 16 Lucretiu Patrascanu Street, sector 3, Bucharest, 030508, Romania, The Research Development National Institute for Textile and Leather Project start date: 1 March 2008 Project end date: 1 September 2009 Project budget: 10,487Eur (20,000 YTL) Source of support: Tubitak Keywords: Enzyme, Textile, Finishing, Cellulose Modern biotechnology often utilizes genetically modified micro-organisms which are characterized by enhanced productivity of specific metabolites, enzymes among them. The range of novel applications of cellulolytic enzymes includes the modification of fibres and fabrics from regenerated cellulose, in order to improve their utility properties and wearing comfort, as well as to maintain their mechanical parameters. In the enzymatic modification of fibres and fabrics, the quantitative and qualitative composition of cellulolytic complex, and so the type of micro-organism being produced, plays the key role. This kind of complex, which is a tailored mixture of specific enzymes, may serve as the ideal solution for creating effective, economical and technologically simple methods for modifying cellulosic fibres. Enzymes have been used for over fifty years to remove starch-based sizes in the textile industry. During the last decade, cellulolytic enzymes have replaced the traditional stone-washing of denim garments and found applications in finishing fabrics and clothing from cotton, linen and regenerated cellulose. In the modern textile industry,

finishing processes which are based on biodegradable and environment-friendly enzymes can fully substitute for a wide range of chemical and mechanical operations so far used to improve the quality of textile products, and save on the energy and consumption of chemicals. Within the context of the project, especially cotton fabrics will be treated enzymatically. The aim of project is environmentally textile finishing and the effects of conventional and bio finishing processes on textile material to be compared with each other. During the experiments laccase enzyme, this is not as common as cellulase in textile finishing, will be used. Turkish side will apply enzymatic finishing to fabric and compare with conventional methods. The properties of washed fabrics will be investigated. Romanian partner will elaborate: (i) schemes and plans for enzyme products treating of textile materials of cellulose fibers, in successive or cumulated phases, as alternatives to the classical processing treatments, (ii) experiments in lab and pilot scale on 100% cotton fabrics, (iii) establishing the influence of enzymatic treatments over treated textile materials, and (iv) their efficiency through physical-mechanical, physical-chemical analyses and color measures.

Aims and objectives Especially cotton fabrics will be treated enzymatically. The aim of project is environmentally textile finishing and the effects of conventional and bio finishing processes on textile material to be compared with each other.

Deliverables Experiments are current and in progress. Publications and outputs Not available.

Buca Izmir, Turkey Dokuz Eylu¨l University, Department of Textile Engineering, Tinaztepe Campus 35160, Buca/Izmir/Turkey. Tel: +90 232 4127211; Fax: +90 232 4127210; E-mail: [email protected] Principal investigator(s): Prof. Dr M. Sevil Yesilpinar Research staff: Zeynep Ezgi Senuyar

Development of a Computer aided program on the preparation of garment pattern Other Partners: Academic

Industrial

Yrd. Doc¸. Dr Vecdi Aytac¸ (Ege University) None Project start date: 1 July 2008 Project end date: 1 July 2009 Project budget: e 10,360

Research register

67

IJCST 21,6

68

Source of support: The Scientific and Technological Research Council of Turkey (TUBiTAK) Keywords: Garment, Garment simulation, E-learning It is so difficult to imagine for a student what the lecturer explains on the board about pattern and the view of the sewn product. The student should understand first the shape of 2D patterns and then the constitution of the garment. In somebody’s opinion understanding the subjects explained on 2D plane is easy but for the other it has different difficulties. The goal of this project is to develop an education program that provides to combine the patterns prepared on 2D plane as if they are sewn. The aim of this program is to support the use of e-learning, to visualize the garment patterns and to help students and professionals who begin pattern learning. Another purpose of this program is to explain and let cognize the basics of garment pattern in an easier and in a faster way by simulations which are still 3D as this is the case with most of detail avoiding computer drawings and photographs used in this field. On the Project different skirt and trouser models are chosen as material. The 2D patterns of the skirt and trouser models will be obtained on 2D CAD system. The most appropriate softwares are chosen to compose the 3D materials for the program and to design the interface to constitute the target educational garment pattern CD. In this context, the programs 3DS Max, Cult3D, Adobew Flashw CS3 Professional which work on Windows Vista operating system will be used. In this project with the help of the technological developments effective, easy and fast teaching of the preparation of garment patterns is aimed. Therefore, realistic objects, animations and visual materials will be taken. So the students will have the advantage of viewing on computer the 3D view of the 2D patterns as if they are sewn. For example, on a skirt model they will be able to understand the order of the parts like pocket flap, pocket bag, front pattern. Consequently, they will have a thorough knowledge of the sewing process. Furthermore, they will watch the combining of garment patterns on a 3D and interactive computer environment. They will see the front, back and side views of the garment. They will be able to watch the virtual garment alone or on a mannequin with the fit measures. And this will accelerate the learning period.

Aims and objectives The 3D systems used in the product development projects became an international research field in clothing industry for the last 10 years. Companies have used 2D CAD systems in all processes from creating the garment pattern to preparing the cutting layout. However, there are just a few “accurate” 2D CAD systems which are able to prepare cutting layout by considering the kind of material. These programs have a feature that allows to make graphical presentations of samples, colors and surfaces. In this context to be in need of being used of these garment simulations in educational area cannot negate. That means these programs provide to learn easily and consequently to explane in a shorter time. As is known, the future of a country and industry depends on the skilled workers and personnel. The better people are improved during the education the sooner they would take an active and important part in the business life. This improving hangs on not only the learning person but also the teaching person and the type of education.

The aim of this project is to develop an education program that provides to combine the patterns prepared on 2D plane as if they are sewn. The aim of this program is to support the use of e-learning, to visualize the garment patterns and to help students and professionals who begin pattern learning. Another purpose of this program is to explain and let cognize the basics of garment pattern in an easier and in a faster way by simulations which are still 3D as this is the case with most of detail avoiding computer drawings and photographs used in this field.

Deliverables Not available. Publications and outputs Not available.

Izmir, Turkey Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: 0090 232 4127211; Fax: 0090 232 4127210; E-mail: [email protected] Principal investigator(s): Aysun Cireli Aks¸ it Research staff: Bengi Kutlu, Nurhan Onar, Mu¨ge Ko¨se

Usage of plasma technology for development of conductivity properties of textile materials Other Partners: Academic

Industrial

None None Project start date: May 2008 Project end date: May 2010 Project budget: e 40,000 Source of support: T.C. Dokuz Eylu¨l U¨niersity (BAP) Keywords: Plasma polymerization, Textile, Conductivity The surface structure of fibres is very important in processing and use, since friction, abrasion, wetting, adhesion, adsorption and penetration phenomena as well as antistatic behaviour are involved. In order to obtain textile materials with a desired property, the fibre surface is often modified with polymer, inorganic or hybrid organic/inorganic layers before use. The demand for electrically conductive fibres and textiles has increased in recent years because of applications as antistatic materials, sensors, materials for electromagnetic shielding and biomedical use. However, an ideal method for modification remains to be found for the preparation of stable conductive textiles. We want to give conductive feature on the textile materials with the use of plasma technology. In plasma processing technology, it is well established that exposure to

Research register

69

IJCST 21,6

70

plasmas generated in inert gases and/or reactive gases can clean the surface of materials and modify their characteristics, particularly their surface energy. Active species from the plasma bombard and/or react with monolayers on the surface of materials and change their surface properties either temporarily or permanently. Such work includes metals or polymers of interest in many industries and less commonly, textiles. Plasma technology applied to the treatment of textiles has developed markedly during the past decade, due to its potential environmental and energy conservation benefits, in developing highperformance materials for the world market and. In practice, the surface properties of natural and synthetic fibres or filaments can be modified using plasma treatment. This can lead to processes such as polymerisation, grafting, cross-linking, etc. with concomitant effects on wetting and wicking, dyeing, printing, surface adhesion, electrical conductivity and other characteristics of interest in the textile industry. Since adhesion is a surfacedependent property, mediated at a molecular scale, plasma technology can effectively achieve modification of this near-surface region without affecting the bulk properties of the materials of interest. Like polyaniline, polypyrrole conductive chemicals which compose a conductive thin film on textile’s surface is covered in this project. For furnising conductivity on textiles this project needs plasma machine which work with low pressure, have RF (radio frequency) plasma (10-100W and 30-360s working power and time) with heat system (25-95 C).

Aims and objectives Aim of this project is, composing conductive thin film on textile’s surface with conductive chemicals like polyaniline, polypyrrole by plasma technology.

Deliverables To obtain conductivity on textile materials. Publications and outputs Not available.

Izmir, Turkey Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: +902324127211; Fax: +902324127210; E-mail: [email protected] Principal investigator(s): Prof. Dr Arif Kurbak Research staff: Tuba Alpyıldız

Studies on textile composites Other Partners: Academic

Industrial

None Project start date: December 2007

None Project end date: December 2010

Project budget: e 30,000 Source of support: Dokuz Eylul Univesity Keywords: Knit, Composite, Impact, Reinforcement Textile preforms are to be investigated and improved as reinforcements in the structural composite materials. Among the textile preforms, knitted fabrics can be formed, with considerably low costs, into almost every possible shape by making use of their extensional deformability offering the advantage that a more homogeneous fibre content is achieved over the entire surface of the part, and also at points of strong curvature. In the composite materials reinforced by knitted fabrics, fibre orientation distribution in the composite is determined by the knit structure and does not change significantly during the production of the composite. In this study different knitted structures will be investigated and the structure with sufiicient mechanical properties will be indicated.

Aims and objectives The major aim of this project is to carry on the studies of the improvements of the textile preforms and to manufacture a knitted reinforcement which is light in weight and has adequate resistance against impact.

Deliverables Not available Publications and outputs

Not available

Izmir, Turkey Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: +90-2324127211; Fax: +90-232-4127210; E-mail: [email protected] Principal investigator(s): Prof. Dr Arif Kurbak Research staff: Dr Ozlem Kayacan

Investigations and improvements on the properties of medical textiles Other Partners: Academic None Project start date: 18 April 2006 Project budget: e 22,000

Industrial None Project end date: 18 April 2009

Research register

71

IJCST 21,6

72

Source of support: Dokuz Eylul University Keywords: Microclimate cooling garment, Water cooling garment, Thermal manikin, Heat stress, Knitted garment, Medical textiles The body temperatures of individuals could increase when they are working in hot conditions, when they have special illness, etc. In order to decrease this temperature, special garments are needed. These garments are called microclimate cooling garments and can be classified as water cooling, air cooling and phase changing materials cooling systems. They are used in military clothes, space suits, protective clothes, in surgical clothes and in medical field to relieve the symptoms of special diseases like multiple sclerosis and ectodermal dysplasias. In this work, four types of water cooling garments, which are different from the other researches, are designed. In order to investigate the cooling effects of these garments, a test method is developed. For this aim a cooling device (chiller) which is pumping the water into these garments and a thermal manikin are designed and manufactured. The effects of water inlet temperature and flow rate on cooling effect and the differences of manikin temperature are investigated.

Aims and objectives In this study cooling effects of different liquid cooling garments are investigated. For this aim four different types of liquid cooling garments are designed in order to alleviate the heat stress of patients in hospitals such as who have special illnesses like multiple sclerosis and hypohydrotic ectodermal dysplasia, etc. The effects of water inlet temperature and flow rate on cooling effect and the differences of manikin temperature are investigated. The most effective garment design is chosen according to the cooling temperatures of the manikin and the rate of cooling of the chiller. Deliverables Publications and outputs Not available.

Izmir, Turkey Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: +90 232 4127211; Fax: +90 232 4127210; E-mail: [email protected]; [email protected] Principal investigator(s): Prof. Ays¸ e Okur, Sibel Kaplan Research staff: Prof. Ays¸ e Okur, Assoc. Prof. Serhan Ku¨cuka, Sibel Kaplan

Development of a method to determine garment thermal comfort Other Partners: Academic

Research register

Industrial

None None Project start date: 1 July 2007 Project end date: 1 December 2008 Project budget: e 25,000 Source of support: Dokuz Eylul University, The Scientific and Technological Research Council of Turkey (TUBITAK) Keywords: Clothing thermal comfort, Dynamic sweating hotplate system, Thermal manikin system, Subjective wear trials Clothing comfort is one of the parameters affecting purchase decisions of people especially in recent years as it has a decisive influence on the daily life and work performance of people. Thermophysiological comfort is one of the important dimensions of comfort and affected by psychological and physiological state of the body and the physical mechanisms occurring in body-clothing-environment system. It affects the thermoregulation mechanism of the body, hence the physiological and psychological status of an individual. In this study, a method including objective and subjective thermal comfort measurements was developed to determine the thermal comfort performances of fabrics/garment systems. Knitted fabrics having different physical/constructional properties were tested by the developed dynamic sweating hotplate system to determine their thermal and water vapor resistance values. This device was used also for determination of water vapor permeability performance of garments under dynamic transfer conditions by a new calculation method put forward in this study. Thermal and water vapor resistance values of garments produced from some of these fabrics were determined by thermal manikin system and subjective wear trials were also conducted by using these garments with a group of professional sportsmen. Dynamic sweating hotplate and thermal manikin systems which are the devices used for objective thermal comfort measurements were developed and produced within this study. Some physiological parameters (skin temperature, microclimate temperature and relative humidity, heart rate, etc.) were measured and subjective coolness-dampness-comfort evaluations were gathered by psychological scales from the subjects during subjective wear trials carried out under determined environmental conditions and activity program. Forearm test, a method used for determination of coolness-dampness to touch sensations during skinfabric contact was also carried out with the selected sportsmen as a preparation stage for the subjective wear trials. Relationships between objective comfort parameters, subjective and physiological data were investigated to find out how much the objectively measured parameters can simulate real wear conditions. Moreover, thermal comfort performances of different fabrics were determined in the light of their physical/constructional/permeability properties. Conclusions were put forward about comfort performances of different fabrics and relationships between objectively and subjectively measured parameters.

Aims and objectives Throughout the history, clothing has changed from being a medium protecting human from unsuitable physical environments to a multi-dimensional concept affecting social

73

IJCST 21,6

74

life and psychology of a human. According to the consumer research studies, comfort has been identified as one of the key attributes taken into account in garment preferences. Comfort can be classified into three basic categories: sensorial comfort related to the psychological and physiological changes arising from skin-textile contact, body movement comfort and thermal comfort including heat and mass transfer mechanisms occuring in body-clothing-environment system. Thermal comfort performance of a garment has a decisive influence on thermoregulation mechanism of a body. As comfort is a sensation affected from the interrelations among body-clothing-environment system parameters, it is a nebulous and complex subject to define. Therefore, it is not possible to measure comfort property of a fabric/garment by using only objective measurement devices; it needs a more comprehensive study. In studies about determination of thermal comfort properties of garments, objective and subjective evaluation methods were developed and results were put forward about the relations between parameters measured with these methods and fabric/garment properties. Studies about objective comfort evaluation methods have generally focused on system developments for measuring thermal and water vapour resistance properties of fabrics/garments as these properties have a significant influence on thermal comfort. It is not possible to measure the sensations of a person about his/her garment objectively, so subjective wear trial method is used to evaluate comfort perceptions. It has been stated by many researchers that the most realistic results for clothing comfort evaluations were obtained by using both objective and subjective methods. This study aims to put forward a method to determine thermal comfort performances of fabrics/garment systems including objective and subjective measurement systems or methods. To carry out this aim, two objective thermal comfort measurement systems, a dynamic sweating hotplate and a thermal manikin system were developed and produced to conduct thermal and water vapor resistance measurements on fabrics and garment systems. A group of knitted fabrics used for sportive garment production were investigated in the study. Subjective wear trials were also conducted with the garments measured by the thermal manikin system on a group of professional sportmen. There are two objectives for this study; one of them is to determine the relations between the objective measurement system results and the other is to put forward conclusions about the thermal comfort performances of the investigated fabric group by taking into account both objective and subjective results. The more the correlation between the objective comfort parameters and subjective comfort evaluations, the better the performance of a measurement system to simulate the real wear condidions. The relations between the resistance measurement results of fabrics and garment systems are also important. If a prediction model for garment can be developed by considering the thickness of microclimate air layer from the resistance values of fabrics, thermal comfort may be determined by a simpler and cheaper system. And if the thermal comfort performances of garments determined by objective and subjective measurements is significantly correlated, it can be concluded that the objective measurement systems developed can be a useful tool to determine the thermal comfort performances of garments.

Deliverables Objective and subjective measurement results obtained were evaluated to investigate the interrelations between them and to put forward conclusions about the thermal comfort

performances of fabrics/garment systems. According to the results, thermal and dampness sensations arise during skin-fabric contact are related to the surface and permeability properties of fabrics and the effect of fabric material is smaller. Among the physiological data obtained during subjective wear trials, temperature and relative humidity of the microclimate can reflect the differences between different fabrics better. Relative humidity of the microclimate is also significantly correlated with the subjective coolness and dampness sensations. According to the subjective evaluation and thermal resistance results, fabrics introduced as having higher water absorption (coolmax) and insulation (thermolite) capacities did not show their advantages about thermal comfort under the selected conditions. Significant relations could not be found out between methods (BS, ASTM, B and water vapor resistance values determined by the dynamic sweating hotplate system) used for determination of water vapor transmittance characteristics of fabrics. However, it was observed that water vapor permeability values determined by the ASTM Standard and resistance values determined by the dynamic sweating hotplate system differentiated fabric permeability performances better than the other methods. Insignificant differences between the thermal resistance values of different garments determined by the developed thermal manikin system may be attributed to the fact that selected environmental conditions was not sufficient to create the necessary temperature difference. It is thought that results obtained from this study can contribute to the literature about thermal comfort performances of different fabrics and relationships between objective-subjective thermal comfort parameters. Publications and outputs Kaplan, S. and Okur, A., “The meaning and importance of clothing comfort and sensory descriptors used to describe discomfort: a case study for Turkey”, Journal of Sensory Studies (in press). Kaplan, S. and Okur, A. (2008), “Investigating the relations between fabric properties and coolness to touch sensation with forearm test”, AUTEX 2008 World Textile Conference, Biella, Italy, 24-26 June. Kaplan, S. and Okur, A. (2007), “Determination of the product attributes and sensory descriptors related to clothing comfort: a case study for Turkey”, AUTEX 7th Annual Textile Conference, Tampere, Finland, 26-28 June. Kaplan, S. and Okur, A. (2006), “Effects of heat and mass transfer mechanisms in textile materials on clothing thermal comfort”, Tekstil ve Mu¨hendis, Vol. 62/63, pp. 28-36 (in Turkish). Kaplan, S. and Okur, A. (2005), “Effects of permeability-conductivity properties of fabrics on clothing thermal comfort”, Tekstil Maraton, pp. 56-65, March/April (in Turkish).

Izmir, Turkey Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: 0090 232 4127211; Fax: 0090 232 4127210; E-mail: [email protected] Principal investigator(s): Aysun Cireli Aks¸ it Research staff: Nurhan Onar, Umit Halis Erdog˘an, Mu¨ge Ko¨se

Research register

75

IJCST 21,6

Producing conductive fibre and developing of their properties Other Partners: Academic

76

Industrial

None None Project start date: July 2008 Project end date: July 2010 Project budget: e 30,000 Source of support: T.C. Dokuz Eylul University (BAP, Scientific Research Project) Keywords: Fibre, Textile, Conductivity The multifunctional textiles for leisure clothing are required while progressing of technology. Increasing of functional properties of textile materials provided useable of textile materials in various areas. For example, intelligent textiles intensively attract the interest in the world. Conductivity textiles are the part of intelligent textiles. Conductive textiles can be used from data transfering to electromagnetic shielding, antistatic properties, heating element and sensors, so on. In the project, we will produce the powder of polyaniline and polypyrrole polymer as conductive polymer. Then we will add the powders to the melting during spinning. What’s more, we will coat the fabric with polyaniline and polypyrrole film.

Aims and objectives In the project, we aimed to produce conductive textiles by coating with conductive polymers by chemical oxidative process and by melt spinning the textile fibers by doping of conductive polymer powders produced by chemical oxidative polymerization. To the aim, we wil produce polyaniline, polypyrrole polymers. What’s more, we will characterized the samples by using FTIR, DTA-TG, XRD and SEM.

Deliverables To occur conductivity on textile materials by chemical oxidative polymerisation method. Publications and outputs Aysun Cireli Aks¸it, Nurhan Onar, M. Faruk Ebeoglugil, Isil Kayatekin, Erdal Celik and Ismail O¨zdemir (2008), “Electromagnetic and electrical properties of coated cotton fabric with barium ferrite doped polyaniline film, APP-2008-06-1917”, Journal of Applied Polymer Science (submitted). ¨ zdemir, I. (2007), “Conductivity and Onar, N., Aks ¸ it, A., Ebeoglugil, M.F., Birlik, I., C¸elik, E. and O magnetic properties of coated fabrics with barium ferrite doped aniline solution, III”, International Technical Textiles Congress, Istanbul Fair Center, Yesilkoy/Istanbul, 1-2 December, pp. 198-206 (oral presentation). Onar, N., Aksit, A., Avgin, I., Celik, E., Ebeoglugil, M.F., Kayatekin, I. and Ozdemir, I., (2007) “Magnetic properties of coated fabrics with barium ferrite doped silica sol”, 10th International Conference and Exhibition of the European Ceramic Society, Berlin, 17-21 June (oral presentation).

Izmir, Turkey Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: 0090 232 4127211; Fax: 0090 232 4127210; E-mail: [email protected] Principal investigator(s): Aysun Cireli Aks¸ it Research staff: Bengi Kutlu

Development of flame retardant and durable press properties of synthetic and natural fibers using plasma polymerization technology Other Partners: Academic

Industrial

Mehmet Mutlu None Project start date: June 2006 Project end date: December 2008 Project budget: e 30,000 Source of support: T.C. Dokuz Eylu¨l U¨niersitesi (BAP) Keywords: Plasma polymerization, Textile, Flame retardancy, Durable-press, Fiber Development of performance properties of textiles is possible by application of chemical agents in an aqueous medium. These methods are known as finishing treatments. They include flame retardancy and durable-press. In spite of a wide variety of treatments and agents, to date, it is still very difficult to durable press and fire retard textiles in a durable manner, i.e. the protection should be resistant to the atmospheric condition for the first application, and for both to several cycles of washing. There are two different ways to confer these properties, either at the beginning of their manufacture by modifying the bulk (macromolecule or polymer) or by surface modification of the fibers/fabrics with active compounds. It is worth considering this later approach interms of durability only if the finishing agent is linked covalently to the fiber/fabric. Among all the different kinds of surface treatments (wet or dry), the cold plasma technique is one of the processes allowing to graft covalently small functional groups as well as macromolecular compounds. Plasma is the fourth phase of materials and has the highest energy. Plasma is occurred when energy which is higher than ionization energy is given. In this work, plasma polimerisation is applied on the synthetic and natural fibers for durable press and flame retardancy. Nitrogen methylol composite, silicone composite, phosphonium salts, composite with halogene and phosphorus nitrogen composites ar used for flame retardancy and durable-press.

Aims and objectives Aims of the project is to prevent any wastewater and to save water by development of flame retardant properties of synthetic and natural fibers using non-aqueous-plasma polymerization technology, to eliminate the negative effects originated by chemicals used and to introduce an ecological production method which is harmless for environment

Research register

77

IJCST 21,6

and human health by decreasing the amount of chemicals used, to decrease process time and the step of process, to balk the changes on handle properties of textilematerials and to compose ecological product method.

Deliverables

78

Synthetic and natural fibers will have flame retardant and durable press properties at the end of this work. Waste water, changes on handle properties of fabric, decline on endurance value and waste gases harmful for human health won’t be included in this project. Publications and outputs Cireli, A., Kutlu, B. and Mutlu, M. “Surface modification of polyester and polyamide fabrics by low frequency plasma polymerization of acrylic acid”, Journal of Applied Polymer Science, Vol. 104, No. 4, pp. 2318-2322. Aysun Cireli, Bengi Kilic¸ (Kutlu), Mehmet Mutlu, “Surface modification of cotton fabrics using low pressure nitrogen plasma”, 6th International Conference TEXSCI 2007 – Book of Abstracts, pp. 273-274. Kılıc¸ (Kutlu), B. and Cireli, A., (2007). “Plazma Teknolojisinin Tekstillerde Kullanımı Ve Tekstil ¨ zerindeki Etkileri, II”, Tekstil Teknolojileri Ve Tekstil Makinalari Kongresi, Materyalleri U Gaziantep-Tu¨rkiye, 19-20 Ekim. Kilic (Kutlu), B., Aksit, A. and Mutlu, M. (2007), “Surface modification and characterization of cotton and polyamide fabrics by plasma polymerization of hexamethyldisilane and hexamethyldisiloxane, III”, International Technical Textiles Congress, Istanbul, 1-2 December.

I˙zmir, Turkey Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: +90-2324127211; Fax: +90-232-4127210; E-mail: [email protected] Principal investigator(s): Dr Vildan Sular Research staff: Gonca Balci

Factors affecting yarn friction Other Partners: Academic

Industrial

None None Project start date: May 2008 Project end date: May 2010 Project budget: e 47,000 Source of support: Dokuz Eylul University Keywords: Yarn friction, Normal load, Contact area, Friction coefficient, Friction force

Yarn friction properties is one of the important properties affecting production stages such as yarn production, fabric and garment formation and also the importance of friction properties continues till the end of the life of a textile product. Sometimes friction property is needed although it is undesirable for some processes. Yarn-to-different surfaces (metal, ceramic, etc.) friction, friction between and within yarns play a great role in winding, weaving, knitting and also sewing. Furthermore, the magnitude of friction of textile materials affects most of textile processing. For this reason to investigate yarn friction properties is very important. In this research, frictional properties of yarns produced having different structural properties (fibre composition, linear density, twist, production system) and different test parameters (normal load, contact area, etc.) will be investigated. Friction coefficients and frictional forces of the yarns will be compared for different conditions and factors affecting these properties will be examined in a systematic way.

Aims and objectives Friction is an important property affecting quality, productivity and performance of a product and it is also important for textiles from fiber to finished product. Although there are a lot of researches on frictional properties of textiles about the nature of friction, its impact on textile processing and its role in determining yarn and fabric properties, studies are still going on this topic because there are a lot of parameters affecting friction and to get under control these properties is difficult. This research aims to determine yarn frictional properties of the yarns having different characterictics and espeacially yarns made of new fibers are planned to examine. In the context of this research, also test parameters will be examined.

Deliverables A large database containing different yarn characteristics and yarn frictional properties will be developed at the end of study. This database may be useful for researchers and also friction results may be useful for yarn producers, woven or knitted fabric producers. Relationships between yarn parameters and friction results will be examined and furthermore frictional properties of new fibers and comparison between different yarn types will be given for different test conditions. Publications and outputs The research began May 2008.

Wuxi, China Jiangnan University, Lihu Road 1800, Wuxi, China, 214122. Tel: +86-510-85327307; Fax: +86-510-85327307; E-mail: [email protected] Principal investigator(s): Liu Jihong Research staff: Gao Weidong, Lu Yuzheng, Pan ruru, Yang Ruihua

Research register

79

IJCST 21,6

80

A wettability evaluation system for woven fabric Other Partners: Academic

Industrial

Jiangsu Information College Jiangying No. 4 Textile Machinery Factory Project start date: 1 January 2009 Project end date: 31 December 2010 Project budget: $15,000 Source of support: Jiangying No. 4 Textile Machinery Factory Keywords: Wettability, Contact angle, Wetting time, Woven fabric, Image processing The research provides an evaluation method for the wettability property of woven fabric. The method is based on video recording dropping process of droplet on the surface of fabric by drop shape analysis apparatus and measuring continuously the surface contact angle. Three kinds of different wettability fabrics were selected to be wet by distilled water droplet. The results have revealed that wetting time can be divided into four periods according to the curve of contact angle with wetting time: dropping time that droplet transforms caused by the gravity, spread time caused by wettability of the fabric, wick time that the contact angle keep same almost, and absorption time that the droplet is absorbed by fabric completely. There is no relationship between dropping time and wettability. The shorter spread time, the better wettability is. The contact angle during wick time can be used to evaluate the bad wettability of fabric.

Aims and objectives . . .

Research on the wettability property of woven fabric. Research on the process of wettabity. Build a system for evaluation the wettability of woven fabric.

Deliverables .

An evaluation mechanical equipment.

Publications and outputs A Wettability Evaluation for Woven Fabric Based on Continuous Contact Angle: About Wettability Anisotropic and Omnidirectional Property of Woven Fabric.

Wuxi, China Jiangnan University, Lihu Road 1800, Wuxi, China, 214122. Tel: +86-510-85327307; Fax: +86-510-85327307; E-mail: [email protected] Principal investigator(s): Liu Jihong Research staff: Qian Kun, Panruru, Yang Ruihua

Mechanical property and application of “8” shape 3D woven enhancing composite

Other Partners: Academic

Industrial

Jiangsu Information College Nanjing Composite Company of China Project start date: 1 January 2008 Project end date: 31 December 2011 Project budget: $20,000 Source of support: Nanjing Composite Company of China Keywords: 3D woven enhancing fabric, Composites, Binder yarn, Mechanical property, Model, “8” Shape Three-dimensional (3D) woven enhancing fabric and its composite was produced on a modified rapier loom. Weaving parameters were studied and restructuring method was researched. After that mechanical property including tension, compress, and so on will be modelled an research. The results express that the property has relationship with the direction of fabric and layers.

Aims and objectives . .

Research on the parameters of producing woven fabric. Build a model and research on mechanical properties.

Deliverables . .

A reconstruction loom for producing “8” shape woven fabric. A kind of method for producing composite of “8” shape woven fabric.

Publications and outputs . .

Weaving Thickness Parameters of “8” Shape 3D Woven Enhancing Fabric. Mechanical Property and Model of “8” Shape 3D Woven Enhancing Composite.

Zagreb, Croatia University of Zagreb, Faculty of Textile Technology, Prilaz baruna Filipovic´a 30, HR-10000 Zagreb, Croatia. Tel: +385 1 37 12 521; Fax: +385 1 37 12 599; E-mail: [email protected] Principal investigator(s): Assoc. Prof. Emira Pezelj, PhD Research staff: Prof. Ruzˇica Cˇunko, PhD; Prof. Maja Andrassy, PhD; Assist. Prof. Edita Vujasinovic, Prof. Vili Bukosˇek, PhD; Antoneta Tomljenovic´, PhD; Sanja Ercegovic´, MSc; Maja Somogyi, BSc; Dubravka Gordosˇ, MSc

Mulifunctional human protective textile materials Other Partners: Academic

Industrial

None

None

Research register

81

IJCST 21,6

82

Project start date: 1 January 2007 Project end date: 31 December 2011 Project budget: N/A Source of support: Croatian Ministry of Science, Education and Sport, Republic of Croatia Keywords: Protective textiles, Multifunctionality, Smart textiles, Ceramic coatings, Sol-gel process The investigations proposed have been motivated by the fact that people are more and more exposed to various influences from the environment, which can be harmful to their health. Such harmful influences are, for example, UV irradiation, electromagnetic smog, high temperature, fire, etc. Contemporary textile materials for personal protection are required to offer high efficiency, in most cases multifunctionality, as well as a necessary level of comfort. The fabrics used are high-performance ones and interdisciplinary approach is necessary in research dealing with their development and manufacture. The thesis we propose is that the application of contemporary research results in the field of materials can be used to offer a new contribution to the development of multifunctional protective textile materials. The accent will be given to a purposeful surface modification of fabrics, using environmentally friendly agents and processes, which is in accordance with contemporary European trends of research in the field of materials. Special attention will be paid to investigating modifications using the new sol/gel process, combined with preceding ultrasound, laser and plasma treatment of textile surfaces. New possibilities of manufacturing efficient protective layers will be investigated, using various inorganic substances, including functional layers of nano-dimension made of hybrid inorganicorganic polymers. The aim is to optimise modification parameters of achieving efficient protection from UV and EM irradiation, as well as to increase resistance to abrasion, cutting and heat in particular materials, establishing antimicrobial properties at the same time. Adequate testing procedures will be established to evaluate the newly created materials. New levels of knowledge is expected to be achieved regarding correlation of protective properties and textile fabric composition, as well as the development of practical processes of obtaining aimed fabric modifications and the development of the methods of new material evaluation. New knowledge will contribute to the quality of education in the field of textile materials. Transfer of knowledge into actual industrial production is also expected. The results will be presented on international conferences and will be published in relevant international publications. The obtained results to be obtained could be used to stimulate manufacture of new high-performance textile materials for special purposes in Croatia.

Aims and objectives The purpose of the investigations is to obtain new knowledge in the field of material development, especially regarding the new composites with textiles as a basic component. The knowledge should be directly applicable in practice, and simultaneously used to improve the quality of education, of both students, young researchers and experts from the industry. The new knowledge is expected to further the development of the Department of textile materials, where the investigations are organised. Based on the knowledge of high-performance materials, that has resulted in the development of the composites, and the role of textile component in them, the possibility will be investigated of obtaining high-performance composites for protection, in which textiles are the basic component.

These are new textile materials to be used as protection from harmful influences of the general and working environment in high-risk industrial processes and other activities where people are exposed to risks of mechanical, thermal or chemical injuries, of infection by micro-organisms and even fatal risks from the causes. This is why protective materials are expected to offer high efficiency under various conditions, while the best solutions are aimed at obtaining multi-functional protection by a single material. The purpose of the research is to investigate the solutions that could be applied in textile industry, which could stimulate the introduction of knowledge-based and new-technology-based production in the industry, through adapting the industry to manufacture high-performance composite materials for special purposes. The aim of the investigation is to determine the procedures of obtaining multi-functional textiles for personal protection, simple to manufacture and use. The protective properties will be obtained by modifying the surfaces of the fabrics of various constructions, with the aim to establish optimal modification procedures and processing parameters which could offer efficient protection from individual influences, or, otherwise, protection from more influences. The investigations are supposed to result in solutions for objective evaluation of the effect achieved and the durability of protection as well, but also in the evaluation of the adequacy of the materials for a particular purpose. Adequate testing methods and procedures will be developed, appropriate indicators defined and the correlation of the modification parameters and properties achieved established.

Deliverables The purpose of the investigations is to obtain new knowledge in the field of material development, especially regarding the new composites with textiles as a basic component. The knowledge should be directly applicable in practice, and simultaneously used to improve the quality of education, of both students, young researchers and experts from the industry. The new knowledge is expected to further the development of the Department of textile materials, where the investigations are organised. Based on the knowledge of high-performance materials, that has resulted in the development of the composites, and the role of textile component in them, the possibility will be investigated of obtaining high-performance composites for protection, in which textiles are the basic component. These are new textile materials to be used as protection from harmful influences of the general and working environment in high-risk industrial processes and other activities where people are exposed to risks of mechanical, thermal or chemical injuries, of infection by micro-organisms and even fatal risks from the causes. This is why protective materials are expected to offer high efficiency under various conditions, while the best solutions are aimed at obtaining multi-functional protection by a single material. The purpose of the research is to investigate the solutions that could be applied in textile industry, which could stimulate the introduction of knowledge-based and new-technology-based production in the industry, through adapting the industry to manufacture high-performance composite materials for special purposes. The aim of the investigation is to determine the procedures of obtaining multi-functional textiles for personal protection, simple to manufacture and use. The protective properties will be obtained by modifying the surfaces of the fabrics of various constructions, with the aim to establish optimal modification procedures and processing parameters which could offer efficient protection from individual influences, or, otherwise, protection from more influences. The investigations are supposed to result in solutions for objective evaluation of the effect achieved and the durability of protection as well, but also in the evaluation of the adequacy of the materials for a particular purpose. Adequate testing

Research register

83

IJCST 21,6

84

methods and procedures will be developed, appropriate indicators defined and the correlation of the modification parameters and properties achieved established. Publications and outputs Cˇunko, R., Ercegovic´, S., Gordosˇ, D. and Pezelj, E. (2006), “Influence of ultrasound on physical properties of wool fibres”, Tekstil, Vol. 55, pp. 1-9. Tomljenovic´, A., Pezelj, E. and Sluga, F. (2007), “Application of TiO2 nanoparticles for UV protective shade textile materials”, Proceedings of 38th Symposium of Textile Novelity, Ljubljana, Slovenija, 21 June. Vujasinovic, E., Jankovic, Z., Dragcevic, Z., Petrunic, I. and Rogale, D. (2007), “Investigation of the strength of ultrasonically welded sails”, International Journal of Clothing Science and Technology, Vol. 19 Nos 3/4, pp. 204-214, ISSN: 0955-6222. Vujasinovic, E., Dragcevic, Z. and Bezic, Z. (2007), “Descriptors for the objective evaluation of sailcloth weather resistance”, Proceedings of 7th Autex Conference 2007, Tampere, Finland, 26-28 June, ISBN: 978-952-15-1794-5. Surina, R. and Somogyi, M. (2006), “Biodegradable polymers for biomedical purpose”, Tekstil, Vol. 55 No. 12, pp. 642-645. Surina, R. and Somogyi, M. (2007), “Resistance of lignocellulosic fibers to microorganisms”, XX. hrvatski skup kemicˇara i kemijskih inzˇenjera, knjiga sazˇetaka, posvec´en Lavoslavu Ruzˇicˇki i Vladimiru Prelogu, hrvatskim nobelovcima u kemiji, Zagreb, 26. veljacˇa-01. ozˇujka, p. 286. Surina, R. and Somogyi, M. (2007), “Intelligent manufacturing & automation: focus on creativity, responsibility and ethics of engineers”, Quality of Modified Flax Fibers, The 18th International DAAAM Symposium, 24-27 October.

Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipovic´a 30, HR-10000 Zagreb, Croatia. Tel: +385 1 48 77 352; Fax: +385 1 48 77 352; E-mail: [email protected] Principal investigator(s): Prof. Drago Katovic´, PhD Research staff: Asoc. Prof. Sandra Bischof Vukusˇic´, PhD; Prof. emeritus Ivo Soljacˇic´, PhD; Dubravka Dosˇen Sˇver, PhD; Sandra Flincˇec Grgac, BSc; Asoc. Prof. Radovan Despot, PhD; Asist. Prof. Jelena Trajkovic´, PhD; Asist. Prof. Branka Lozo, PhD; Luka Cˇavara, MSc; Bozˇo Tomic´, Mc; Prof. Charles Yang, PhD; Prof. Christian Schram, PhD.

Alternative eco-friendly processing and methods of cellulose chemical modification Other Partners: Academic

Industrial

Faculty of Forestry, Croatia; Faculty Cˇateks, d.d., www.cateks.hr of Graphic Art, Croatia; University of Georgia, USA; University of Innsbruck, Austria

Project start date: 1 January 2007 Project end date: 31 December 2011 Project budget: N/A Source of support: Ministy of Science, Education and Sports, Republic of Croatia Keywords: Multifunctional eco-friendly textile finishing, Polycarboxylic acids, Protective functionalities, Chemical modification of cellulose, Microwave treatment of cellulose materials One of the requests of European Union for higher competiteveness of European market is rebuilding and reconstruction of traditional industrial sectors, especialy textile and wood industry. According to the strategical goals of the Republic of Croatia the project emphasizes the use of highly sofisticated production processes and treatments of cellulose materials, i.e. obtaining additional and improved characteristics of wooden and paper matherials which can be acchieved by using high-tech processes and by introduction of nano- micro- and biotechnologies. One of the alternative methods for replaciong the conventional reactants containing formaldehyde which were used in textile and wood treatments so far, would be the modification with eco-friendly agents such as polycarboxylic acids. Efficiency of these treatments will be determined quantitatively by ester crosslinking analytical methods or by means of isocratic HPLC and spectrophotometric FTIR method. Standard methods of textile, wood and paper material testing would be used for examining their protective performance and resistance to weathering conditions. Part of the proposed project will be development of optional multifunctional treatment that would provide better protection of cellulose materials against microorganisms, UV, electromagnetic rays, flame, oil or water. Therefore, a particular attention will be payed to development and application of the agents which will not only improve the characteristics of textile matherials but also give it permanent freshness and provide additional care and protection, i.e. medical characteristics. Optimisation of alternative processing and methods will provide ecologically and economicaly favorable characteristics of treated matherials. Further process optimisation in order to improve processing quality could be obtained with new alternative method using microwave energy. Improved characteristics obtained with this method in our previous research confirm its usability in textile finishing processes as well as in chemical modification of wood. Previous research in this field represent worlwide novelty which should be by all means continued.

Aims and objectives The purpose and aim of the proposed project is to obtain highly valuable and multifunctional treated textile materials that will acquire analogous price on the demanding market. This is the basic condition for the survival of Croatian textile, wood and paper industry on EU market. In textile area experiments will be conducted to obtain multifunctional environmentally friendly textile material which will simultaneously offer dimensional stability, flame retardancy, crease and antimicrobial resistance and will have no effects on human health. Further goal is to obtain chemicaly modified wood that will have reduced shrinking and water absorption as well as to obtain flame retardancy on wood and paper products. One of the equally important goals is construction of a semi industrial microwave device for continuous planar treatment of cellulose materials. The results obtained would be presented in the world best known papers in the relevant field.

Research register

85

IJCST 21,6

86

The most important goal of the project is affirmation of Croatian science in Europe and rest of the World, by presenting the results in international papers so as on International Conferences. It is important to stress that established cooperation with EU and USA experts, so as with their scientific institutions will be continued and expanded. In this project, where will scientists from abroad have an active contribution with their work, further contribution to development of high quality products will be added. We certainly hope it will affect development of Croatian industry and economy.

Deliverables Not available. Publications and outputs Katovic´, D., Bischof Vukusˇic´, S. and Flincˇec Grgac, S. (2007), “Crosslinking cotton with citric acid and organophosphorus agent for the purpose of flame retardant finishing”, 85th Textile Institute Conference, Colombo, Sri Lanka, pp. 820-824. Bischof Vukusˇic´, S., Flincˇec Grgac, S. and Katovic´, D. (2007), “Catalyst influence in low formaldehyde flame retardant finishing system”, 7th AUTEX Conference, Tampere, pp. 60-61. Flincˇec Grgac, S., Katovic´, D. and Bischof Vukusˇic´, S. (2007), “Combination of organophosphorus agent and citric acid in durable press finishing of cellulose fabrics”, XX Croatian Society of Chemical Engineers, Zagreb, Croatia, pp. 281. Bischof Vukusˇic´, S., Flinecˇ Grgac, S. and Katovic´, D. (2007), “Antimicrobial textile treatment and problems of testing methods”, Tekstil, Vol. 56 (accepted for publication).

Zagreb, Croatia University of Zagreb, Faculty of Textile Technology, Prilaz bazuna Filipovic´a 30, HR-10000 Zagreb, Croatia. Tel: +385 1 48 77 359; Fax: +385 1 48 77 355; E-mail: [email protected] Principal investigator(s): Prof. Durdica Parac-Osterman, PhD Research staff: Martinia Ira Glogar, PhD; Assist. Prof. Darko Golob, PhD; Assoc. Prof. Marija Gorensˇek, PhD; Assoc. Prof. Darko Grundler, PhD; Assoc. Prof. Nina Knesˇaurek, PhD; Prof. Nina Rezˇek-Wilson; Assist. Prof. Tomislav Rolich, PhD; Ana Sutlovic´, MSc; Dedran Durasˇevic´ BSc

Colour and dyestuff in processes of ecologically acceptable sustainable development Other Partners: Academic

Industrial

University of Ljubljana and Maribor, Jadran Stockings Factory Slovenia Project start date: 1 January 2007 Project end date: 31 December 2011 Project budget: N/A Source of support: Ministry of Science, Education and Sports, Republic of Croatia

Keywords: Dyestuff selection, Nano-technology, Optimizing dyeing process, Purifying and decolouring wastewaters, Colour management, Fuzzy logic Scientific contribution to sustainable development relies on unlimited support of basic, developing and employable research. Therefore, selection of multi-functional dyes (UV protection, antibacterial protection, micro capsules of multi-functional performance), applying nano-technology in the dyeing processes with the aim of preventing water contamination, development of new methods as well as purifying dyed wastewaters contribute to sustainable development. Both input and output parameters of water will be controlled throughout the entire dyeing process: amount of residual dye in dye-bath using Lamber-Beer absorption model; X, Y, Z standard spectral characteristics of colour defined by specific absorption coefficient SAC and water quality defined by BOD5, COD, TOC, AOX, electrical conductivity and other defining values. System of control comprising advance models of control such as fuzzy logic (model based on rules) and model based on physical and chemical processes will be developed and applied. Capital area of research will involve models of dyeing processes, colour control and its correlation to dye as well as the interactive system of dye control. Models should describe and predict kinetics, reactivity, affinity, exhaustion, fixation and interaction of solutions containing various dyestuffs. Prediction of output process result as well as definition of both physical and chemical parameters crucial for controlling the process will be conducted based on afore-mentioned models. These models encompass kinetic models (according to Nernst and Langmuir) modified for interactions between dyestuff on fibre and in the solution. Interdisciplinarity of dye within the system of sustainable development is based on spectral characteristics of colour as the fundament model dependent of the employment conditions. Instrumental measurement of colour is involved in all industrial production processes: textile technology, design, graphic industry and elsewhere which enables implementing control and colour harmonization. Application of evolutionary algorithms for modeling computer aided design of textiles based on principals of examinee’s subjective evaluation. Methods of descriptive statistics as well as methods of statistic reasoning will be applied within the frame of statistic analysis. Scientific affirmation of research results will be computer simulation as well as in vivo confirmation.

Aims and objectives Contribution to sustainable development relies on unlimited support of basic, developing and employable research. Aim of the project is to contribute to humane ecology (regarding UV, antibacterial and other protective properties), through use of multifunctional dyes and selection of appropriate waste water discoloration and purifying methods, all in order of obtaining biological quality of water (free of toxic, aromatic components which may form in the process of dye degradation). Cognition of structure and use of thermo sensible nano sized dyes will enable their use on fibres for special purposes. Advance dyeing technologies, with the overview on pretreatment of textile substrates (enzymatic, plasmatic and other) in order of preserving environment and saving energy, will be applied. Base of the project is application of dyeing process control, including advance models of control; fuzzy logic and model based on physico-chemical processes.

Research register

87

IJCST 21,6

88

Colour used as constant value will be applied for formation of fuzzy logic model, used for complex colour designing, automatic dye selection, direct transfer of colour coordinates data into the dyeing recipe setup system, advance recipe correction, as well as control and colour matching. Project research will enable use of new instrumental methods and development of researcher’s creativity, while graduate students and potential Ph.D. students will be given a chance to get acquainted with scientific methodology, development of experimental skills and writing scientific papers.

Deliverables Influence of dye’s chemical constitution and mode of dye-fibre bond onto antibacterial (e.g. Staphiloccoci, Escherichia), UVA and UVB protection properties. Influence of additives (electrolytes and surfactants) on dyeing process and degree of water pollution. Control of, in dye-bath and wastewaters, present electrolytes – elaboration of mathematical model. Further results considering influence of dye onto protection properties are expected. Application of thermo sensible dyes on children clothing. Influence of textile substrate’s pre-treatment (enzymatic, plasmatic pre-treatment, etc.) on dyeing kinetics and energy saving. Results of wastewater purifying and discoloration methods, with the emphasis on salt removal using physico-chemical methods, nano filtration and reverse osmosis. Colour as constant value of monitoring process, dye properties and colour matching in design applying evolutionary algorithms. Application of nano size particles. Influence of surfactants onto reactivity, affinity, exhaustion and fixation degree of reactive dyes. Advantages and disadvantages of physicochemical decolouring methods. Dye degradation products and their toxicity (considering aromatic components) in wastewaters. Selection of dyestuff and its interaction with in the dye-bath present additives. Mathematical model based on measured values will be elaborated, while control system including a model based on physico-chemical processes will be applied. Fuzzy logic model, based on colour as constant value within control system will be worked out. Application of capsulated dyestuff and nano particles of zink and silver for special use (medical textiles). Pre-treatment of hydrophobic, synthetic fibres in the aim of increasing hydrophility and applicability of, in water soluble, dyes. From the economical and ecological aspect, a more acceptable system of purifying and decolouring wastewaters ii expected. Mathematical model based on the analyses of input and output measured values, considering coloured waters, will be elaborated, while a control system using advance models, such as fuzzy logic (model based on rules) and based on physicochemical processes model. Evolutionary algorithms for modelling computer design of fabrics based on principals of subjective asessment. Model must comply with standards, flexible, stabile, precise, and easily applicable. It includes complete process modelling: dyeing, colour control and its relation to dye. These models involve kinetics models (Nernst, Langmuir) modified for dye – dissolved dye. Applying CCM (computer colour matching) methods based on Kubelka-Munk theory, spectral characteristics and colour parameters according to CIEL*a*b* system, a model of fuzzy logic for complex design by colour, automatic dye selection, direct transfer of colour coordinates data into the dyeing recipe setup system, advance recipe correction, as well as control and colour matching, will be elaborated. Evolutionary algorithms for modelling computer design of fabrics based on principles of subjective assessment.

Publications and outputs Durdica Parac-Ostreman, Ana Sutlovic´, Vedran Durasˇevic´ and Tjasa Griessler Bulc, “Use of wetland for dye-house waste waters purifying purposes”, Asian Journal of Water, Environment and Pollution, Vol. 4 No. 1, pp. 101-106. Durdica Parac-Osterman, Vedran Durasˇevic´ and Ana Sutlovic´, “Comparison of some chemical and physical-chemical waste water discoloring methods”, Chemistry in Industry (in press). Martinia Ira Glogar, Darko Grundler, Durdica Parac-Osterman and Tomislav Rolich, “Fuzzy logic based approach to textile surface structure influence in colour matching”, AATCC Review (in press). Vesna Tralic´-Kulenovic´, Livio Racane, Ana Sutlovic´ and Vedran Durasˇevic´ (2007), “Dyeing properties of new benzothiazol disperse dyestuff”, XX Croatian Meeting of Chemists and Chemical Engineers, Zagreb, Croatia, 26 February-1 March. Durdica Parac-Ostreman, Nevenka Tkalec Makovec, Ana Sutlovic´ and Ljerka Dugan (2007), “Staphylococcus Aureus and Escherichia Coli Behavior on Undyed and Dyed Wool”, XX. Croatian Meeting of Chemists and Chemical Engineers, Zagreb, Croatia, 26 February-1 March. Durdica Parac-Osterman, Ana Sutlovic´ and Vedran Durasˇevic´ (2007), “Application of wetland system”, Textile Dyes Zagreb 2007, Zagreb, Croatia, 9 March. Durdica Parac-Osterman, Ana Sutlovic´ and Martinia Ira Glogar (2007), “Dyeing wool with natural dyes in light of the technological heritage”, 7th Annual Textile Conference by Autex, “From emerging innovations to global business”, Tampere, Finland, 26-28 June. Durdica Parac-Osterman, Ana Sutlovic´ and Vedran Durasˇevic´, “Application of wool, CA and PP fibers as filters in wetland pretreatment media formation”, University of Zagreb, Faculty of Textile Technology International Conference on Multi Functions of Wetland Systems, Legnaro (Padova), Italy, 26-29 June.

Zagreb, Croatia Faculty of Textile technology, University of Zagreb, Prilaz baruna Filipovica 28a, 10000 Zagreb, Croatia. Tel: ++385 1 3712 540; Fax: ++385 1 37 12 599; E-mail: [email protected] Principal investigator(s): Dubravko Rogale Research staff: Zvonko Dragcˇevic´, Gojko Nikolic´, Maja Vinkovic´, Snjezˇana Firsˇt Rogale, Slavenka Petrak, Goran Cˇubric´

Intelligent garment and environment Other Partners: Academic

Industrial

None None Project start date: 2007 Project end date: 2012 Project budget: e 100,000 Source of support: Ministry of Science, Education and Sports Keywords: Intelligent garment, Thermal protection, Environment Investigations, construction and development of intelligent article of clothing related to its direct environment by developing an adaptable bed, adaptable ironing machine and

Research register

89

IJCST 21,6

90

measuring instrument for multiaxial testing physical-mechanical properties of technical textile and joined parts. The purpose of the project is that a research team makes researches resulting in a construction and realization of the first intelligent garment whose basic function is active thermal protection. It contains a sensor system for monitoring the values of air temperature inside and outside of the garment, data bus for data transfer, microcomputer and micro controller, and execution devices for the automatic regulation of thermal protection value. Controlling conduction and convection of the heat of the human body regulates thermal protection in such a way that based on anthropometric measurements several types of various air thermo insulation elastic chambers are constructed which are integrated into the construction of the garment between the outer shell and lining. Thermoinsulation chambers consist of several segments and have a twofold function so that by inflating sealing properties are assumed, and the heat loss of the human body by convection can be regulated and the thickness of the air chambers can be changed by program, whereby the heat loss of the human body by conduction can to be regulated. Micropneumatic elements and the chambers would be equipped with sensors of air pressure integrated into them, because depending on air pressure values in the chambers there will be defined chamber forms, their sealing properties and thickness on which thermal resistance depends. Investigations would prove that the integration and efficient joint operation of the integrated sensors, microcomputers with associated algorithms of intelligent behavior and actuators so that an independent action of the garment is realized with the aim of thermal protection whereby the garment would have the attribute of active, adaptable and intelligent behavior in variable temperature conditions. Communication possibilities of intelligent garment with the environment would be examined and an intelligent sick bed, adaptable ironing-machine and an instrument for testing load would be developed. They would practically use the same or very similar sensor, computer and micropneumatic actuator systems, connection techniques, constructions and design as well as intelligent garment.

Aims and objectives The basic aim of the proposed research project is to investigate the possible construction and practical realization of an intelligent article of clothing with thermal protection, adaptable bed, ironing machine for the technological manufacturing process and necessary measuring instruments. The purpose of the investigation is to investigate characteristics of all elements of the system and behavior of the system as a whole and the communication between intelligent garment and environment. In addition to the basic aim of all investigations it is necessary to point out other aims too emerging as the result of the said investigation. Establishment of the leading European and world scientific role in investigations and development of intelligent garment. Writing four doctoral dissertations in the mentioned field (two dissertations in the mentioned field have been registered and approved by the Senate of the University of Zagreb. To prove that the Croatian clothing industry possesses a strong scientific research basis that guaranties it a technological excellence on demanding foreign markets.

Deliverables The final assumed investigation results are practically applicable immediately upon completion of the project. It may be expected that the industrial production of intelligent garment with active thermal protection could be commenced immediately upon completion of the project. The examined and realized intelligent article of clothing would be a world unique item and might become an original Croatian product, especially from the point of view that original production principles have been protected by patent so that the future production rights are unquestionable. The developed intelligent article of clothing is very interesting for all the people being in extreme climatic conditions (soldiers, policemen, security services agents, construction workers, sailors, maintenance of roads, buildings and industrial facilities, drivers of trucks and construction machinery, athletes, recreationists, and other persons who wish to have such an article of clothing). In addition to the clothing industry, other industry branches such as mechanical engineers, electronics engineers, and programmers, participate in the production of intelligent garment, so that benefits can be expected for them too. Publications and outputs Rogale, D., Firsˇt Rogale, S., Dragcˇevic´, Z. and Nikolic´, G., “Intelligent article of clothing with an active thermal protection”, European Patent Office, Munich, Germany, No. PCT/HR2004/000026. Firsˇt Rogale, S., Nikolic´, G., Dragcˇevic´, Z., Rogale, D. and Bartosˇ, M. (2005), “Architecture of clothing with an active thermal protection”, in Katalinic´, Branko (Ed.), Proceedings of the 16th DAAAM International Symposium: Intelligent Manufacturing & Automation: Focus on Young Researchers and Scientists, DAAAM International Vienna, Vienna, pp. 121-2. Rogale, D., Firsˇt Rogale, S., Dragcˇevic´, Z., Nikolic´, G. and Bartosˇ, M. (2006), “Development of intelligent clothing with an active thermal protection”, 6th World Textile Conference, North Carolina, 11-14 June, pp. 106-112. Petrak, S. and Rogale, D. (2001), “Methods of automatic computerised cutting pattern construction”, International Journal of Clothing Science and Technology, Vol. 13 Nos 3/4, pp. 228-39. Firsˇt Rogale, S., Dragcˇevic´, Z. and Rogale, D. (2003), “Determining reaction abilities of sewing machine operators in joining curved seams”, International Journal of Clothing Science and Technology, Vol. 15 Nos 3/4, pp. 179-88. Rogale, D., Petrunic´, I., Dragcˇevic´, Z. and Firsˇt Rogale, S. (2005), “Equipment and methods used to investigate energy processing parameters of sewing technology operations”, International Journal of Clothing Science and Technology, Vol. 17 Nos 3/4, str. 179-87. Petrak, S. and Rogale, D. (2006), “Systematic representation and application of a 3D computer-aided garment construction method, part I 3D garment basic cut construction on a virtual body model”, International Journal of Clothing Science and Technology, Vol. 18 No. 3, pp. 179-87. Petrak, S., Rogale, D. and Mandekic´-Botteri, V. (2006), “Systematic representation and application of a 3D computer-aided garment construction method, part II spatial transformation of 3D garment cut segments”, International Journal of Clothing Science and Technology, Vol. 18 No. 3, pp. 188-99. Firsˇt Rogale, S., Rogale, D., Dragcˇevic´, Z., Nikolic´, G. and Bartosˇ, Milivoj (2007), “Technical systems in intelligent clothing with active thermal protection”, International Journal of Clothing Science and Technology, Vol. 19 Nos 3/4, pp. 222-33. Firsˇt Rogale, S., Rogale, D., Dragcˇevic´, Z., Nikolic´, G. and Runkas, Martin (2008), “Intelligent clothing whit programmabile insulation”, in Katalinic´, B. (Ed.), DAAAM International Scientific Book 2008, DAAAM International, Vienna, Str. 273-86. Firsˇt Rogale, S., Rogale, D., Dragcˇevic´, Z., Nikolic´, G. and Bartosˇ, M. (2007), “Technical systems in intelligent clothing with active thermal protection”, in Kniewland, Z. (Ed.), Annual 2007 of Croatian Academy of Engineering, Croatian Academy of Engineering, Zagreb, pp. 301-17.

Research register

91

IJCST 21,6

92

Firsˇt Rogale, S., Rogale, D., Dragcˇevic´ , Z. and Nikolic´ , G. (2007), “Realization of the prototype of intelligent article of clothing with active thermal protection”, Tekstil, Vol. 56 No. 10, pp. 610-26. Firsˇt Rogale, S., Rogale, D., Nikolic´, G., Dragcˇevic´, Z. and Bartosˇ, M. (2007), “Chambers in the intelligent clothing with active thermal protection”, in Gersˇak, G. (Ed.), Proceedings of 5th International Conference IMCEP 2007, Faculty of Mechanical Engineering, University of Maribor, Maribor, pp. 23-33. Nikolic´, G., Firsˇt Rogale, S., Rogale, D., Dragcˇevic´, Z. and Bartosˇ, M. (2008), “Pneumatic system of the intelligent article of clothing with active thermal protection”, Ventil, Vol. 14 No. 6, pp. 552-6. Firsˇt Rogale, S., Rogale, D., Dragcˇevic´, Z., Nikolic´, G. and Runkas, Martin (2008), “Intelligent clothing with programmabile insulation”, in Katalinic´, B. (Ed.), DAAAM International Scientific Book 2008, DAAAM International, Vienna, pp. 273-86. Firsˇt Rogale, S., Rogale, D., Nikolic´, G. and Dragcˇevic´, Z. (2009), Controllable Ribbed Thermoinsulative Chamber of ContinuallyAdjustable Thickness and its Application, European Patent Office, Munich, Germany, No. PCT/HR2009/000008.

Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipovic´a 30, HR-10 000 Zagreb, Croatia. Tel: +385 1 4877 360; Fax: +385 1 4877 355; E-mail: [email protected] Principal investigator(s): Prof. Ana Marija Grancaric´, PhD Research staff: Assoc. Prof. Tanja Pusˇic´, PhD; Assist. Prof. Zˇeljko Penava, PhD; Anita Tarbuk, MSc, Lea Markovic´, BSc, Assist. Prof. Jasenka Bisˇc´an, PhD; Sonja Besˇenski, MSc, Ivancˇica Kovacˇek, PhD, D. Med., Prof. Djamal Akbarov, PhD, Prof. Emil Chibowski, PhD, Prof. Rybicki Edward, PhD, Prof. Eckhard Schollmeyer, PhD, Prof. M.M.C.G. Warmoeskerken, PhD.

Interface phenomena of active multifunctional textile materials Other Partners: Academic Croatian National Institute of Public Health, Zagreb; Tashkent Institute of Textile and Light Industry, Uzbekistan; Maria Curie-Skłodowska University, Lublin, Poland; Technical University of Lodz, Poland; Deutsches Textilforschungsinstitut Nord-West eV; Institut der Universitat Duisburg Essen; University of Twente, The Netherlands Project start date: 1 January 2007 Project budget: N/A

Industrial Pamucˇna industrija Duga Resa, Duga Resa

Project end date: 31 December 2011

Source of support: Ministry of Science, Education and Sports, Republic of Croatia Keywords: Textile material, Interface phenomena, Surface modification and finishing, Multifunctionality The goal of the project is synergistic effects of some compounds on modified textile surfaces for achieving multifunctionality of textiles. Interface phenomena of textile surfaces with special accent on surface free energy, zeta potential, electroconductivity, adsorption and desorption of surfactants and other compounds usually used in textile finishing will give a great contribution to multifunctionality of textile. The mechanism of adsorption and desorption of surfactants and other finishing agents on modified textile surfaces is expected to be clarified in the present project. Different surface modifications, pretreatment and finishing of textile, especially cotton and polyester, will be performed according to European Technology Platform for the future of textile and clothing. For such purpose advance processes like mercerization, cationization, alkali, EDTA, other compounds and enzymes for surface hydrolysis of PET fabric, optical bleaching, implementation of nano antimicrobial active silver ions and mineral delivery mechanism, zeolite and others will be performed. Aminofunctional and other compounds will be added to azalides for the synergistic high antimicrobial effects. In cotton pretreatment enzymatic scouring will be applied using enzymes pectinase and the newest cutinase, for removal of pectins and bioplymers from cotton impurities with lipophylic character, instead of ecologically unfavorable alkali scouring. The goal of the project is synergistic effects of some compounds on modified textile surface. Interface phenomena of the new textile materials produced from electroconductive, low electro resistance fibers will be investigated for the purpose of static electricity and electromagnetic protection and for its implementation as sensors or other electronic devices in intelligent textiles. Traditional protection and aesthetic role of textile will be spread in active textile multifunctionality. Project will deal with elektrokinetic phenomena (zeta potential, isoelectric point, IEP, point of zero charge, PZC, surface electrical charge, surface free energy), hydrophility and hydrophobilicity, whiteness, fluorescence and phosphorescence, friction, fabric cover factor, elasticity, air and water vapor permeability of textile materials and their protection on UV radiation, microbes and fungi, coldness, heat and flame, static electricity and electromagnetic field.

Aims and objectives Project will continue researching on assignments from previous project (0117012). Purpose of these investigations is based on lightening of interface phenomena on textile which effect directly to its adsorption and interaction intensity between textile fibers and chemical compounds. Almost all possibilities in modification during manufacturing high performance synthetic fibers are used, therefore nowadays attention and research is on textile surface modification. Procedures and compounds for that modification varies, as their effect varies, but the purpose and aim are directed to synergism of two or more components for accomplishing hydrophob or hydrophil textile, textile highly resistant to atmospheric condition, bacteria, microbe and fungi, UV radiation and open flame. Furthermore, important aim of the project is cotton high level of purity by unconventional agents and material pretreatment procedures for mercerization and cationization. Pectinase in previous project investigation showed good elimination of pectine from

Research register

93

IJCST 21,6

94

primary cotton layer, but hydrophility was not so high like alkali scoured cotton. Chemical composition of cotton cuticula has lypophilic polymers, biopolyesters, which can be degraded by cutinase, new enzymes for degradation of waxes for better hydrophility. Cotton cationization during mercerization is the most important innovation of previous project and the patent for it was asked. Electronegative cotton surface charge, of which anionic substances adsorption depends, is lower after cationization in harsh mercerization conditions. The aim of this project is antibacterial, UV and flame protection by nanoparticle implementation (Ag) using mineral delivery compound (zeolite and others) as well. Electroconductive fibers implementation in yarns of textile materials should result in static electricity removal, and hopefully other effects. The aim of polyester surface modification, optical bleaching, other compounds treatments is well-known aesthetic, as well as high UV protection, high material elasticity as a result of changes in fiber microstructure. Interface phenomena research on wide range possible fabric knitted and woven construction will givethe solution of problems of fabric construction influence to high effect in this project.

Deliverables Interface phenomena of textile materials surface in wet medium results in textile electric surface charge cognition and surface free energy as well on which adsorption depends. Important application of this project results is in ecological enzymatic scouring with pectinases. Enzymatic scouring with new enzymes, cutinase, will remove biopolyester cuticula and improve cotton hydrophility, and therefore replace harsh conventional alkali scouring entirely. Important application will have, patent requested cotton cationization during mercerization. By this pretreatment electropositive cotton is achieved, with great anion adsorption on its surface in all textile finishing processes. These anions enclose all low and high molecular compounds for textile finishing and all pricondensates. Implementation of nanoparticles (Ag and others) is predicted during mercerization and cationization processes, therefore it is important to emphasize rational component of these procedures which gives cotton multifunctionality in all textile usage. The next important application is antibacterial textile accomplished with azalide treatment especially in synergism with aminofunctional and other compounds and systems. It is well-known that fluorescence of optically bleached increases whitening of textiles. Optical brighteners and other compounds researching will be of great importance in UV protection with textile material. Heavy metals are toxic and their research is of great importance in human health protection. Furthermore, in nowadays growing demands on life safety from external influences especially UV radiation, research of differently structured textile material interface phenomena will find application in textile for summer clothing. Publications and outputs Grancaric´, Anamarija, Pusˇic´, Tanja and Tarbuk, Anita (2006), “Enzymatic scouring for better textile properties of knitted cotton fabrics”, in Guebitz, Georg, Cavaco-Paulo, Artur and Kozlowski, Rysard (Eds), Biotechnology in Textile Processing, The Haworth Press, Inc., New York, NY. Grancaric´, Ana Marija, Tarbuk, Anita, Dumitrescu, Iuliana and Bisˇc´an, Jasenka (2006), “UV protection of pretreated cotton – influence of FWA’s fluorescence”, AATCC Review, Vol. 6 No. 4, pp. 2-6. Anita Tarbuk, Ana Marija Grancaric´ and Volker Ribitsch (2007), “Electrokinetic phenomena of textile fibers”, Book of Abstracts: XX Croatian Meeting of Chemists and Chemical Engineers, p. 301.

Ana Marija Grancaric´, Lea Markovic´, Anita Tarbuk and Eckhard Schollmeyer (2007), “Properties of multifunctional cotton in accordance with international standards”, Conference of Textile Days, Zagreb. Ana Marija Grancaric´, Anita Tarbuk and Ivancˇica Kovacˇek (2007), “Micro and nanoparticles of zeolite for the protective textiles”, Book of Proceedings of 7th Annual AUTEX Conference, AUTEX, Tampere, Finland, p. 1123. Anita Tarbuk, Ana Marija Grancaric´ and Mirela Leskovac (2007), “Surface free energy of pretreated and modified cotton woven fabric”, Book of Proceedings of 7th Annual AUTEX Conference, AUTEX, Tampere, Finland, p. 1104.

Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipovic´a 30, HR-10000 Zagreb, Croatia. Tel: +385 1 48 77 352; Fax: +385 1 48 77 352; E-mail: [email protected] Principal investigator(s): Prof. Drago Katovic´, PhD Research staff: Asoc. Prof. Sandra Bischof Vukusˇic´, PhD; Prof. emeritus Ivo Soljacˇic´, PhD; Dubravka Dosˇen Sˇver, PhD; Sandra Flincˇec Grgac, BSc; Asoc. Prof. Radovan Despot, PhD; Asist. Prof. Jelena Trajkovic´, PhD; Asist. Prof. Branka Lozo, PhD; Luka Cˇavara, MSc; Bozˇo Tomic´, Mc; Prof. Charles Yang, PhD; Prof. Christian Schram, PhD

Alternative eco-friendly processing and methods of cellulose chemical modification Other Partners: Academic

Industrial

Cˇateks, d.d., www.cateks.hr Faculty of Forestry, Croatia; Faculty of Graphic Art, Croatia; University of Georgia, USA; University of Innsbruck, Austria Project start date: 1 January 2007 Project end date: 31 December 2011 Project budget: N/A Source of support: Ministy of Science, Education and Sports, Republic of Croatia Keywords: Multifunctional eco-friendly textile finishing, Polycarboxylic acids, Protective functionalities, Chemical modification of cellulose, Microvawe treatment of cellulose materials One of the requests of European Union for higher competiteveness of European market is rebuilding and reconstruction of traditional industrial sectors, especialy textile and wood industry. According to the strategical goals of the Republic of Croatia the project emphasizes the use of highly sofisticated production processes and treatments of cellulose materials, i.e. obtaining additional and improved characteristics of wooden and paper materials which can

Research register

95

IJCST 21,6

96

be acchieved by using high-tech processes and by introduction of nano- micro- and biotechnologies. One of the alternative methods for replaciong the conventional reactants containing formaldehyde which were used in textile and wood treatments so far, would be the modification with eco-friendly agents such as polycarboxylic acids. Efficiency of these treatments will be determined quantitatively by ester crosslinking analytical methods or by means of isocratic HPLC and spectrophotometric FTIR method. Standard methods of textile, wood and paper material testing would be used for examining their protective performance and resistance to weathering conditions. Part of the proposed project will be development of optional multifunctional treatment that would provide better protection of cellulose materials against microorganisms, UV, electromagnetic rays, flame, oil or water. Therefore, a particular attention will be paid to development and application of the agents which will not only improve the characteristics of textile matherials but also give it permanent freshness and provide additional care and protection, i.e. medical characteristics. Optimisation of alternative processing and methods will provide ecologically and economicaly favorable characteristics of treated matherials. Further process optimisation in order to improve processing quality could be obtained with new alternative method using microwave energy. Improved characteristics obtained with this method in our previous research confirm its usability in textile finishing processes as well as in chemical modification of wood. Previous research in this field represent worlwide novelty which should be by all means continued.

Aims and objectives The purpose and aim of the proposed project is to obtain highly valuable and multifunctional treated textile materials that will acquire analogous price on the demanding market. This is the basic condition for the survival of Croatian textile, wood and paper industry on EU market. In textile area experiments will be conducted to obtain multifunctional environmentally friendly textile material which will simultaneously offer dimensional stability, flame retardancy, crease and antimicrobial resistance and will have no effects on human health. Further goal is to obtain chemicaly modified wood that will have reduced shrinking and water absorption as well as to obtain flame retardancy on wood and paper products. One of the equally important goals is construction of a semi industrial microwave device for continuous planar treatment of cellulose materials. The results obtained would be presented in the world best known papers in the relevant field. The most important goal of the project is affirmation of Croatian science in Europe and rest of the World, by presenting the results in international papers so as on International Conferences. It is important to stress that established cooperation with EU and USA experts, so as with their scientific institutions will be continued and expanded. In this project, where will scientists from abroad have an active contribution with their work, further contribution to development of high quality products will be added. We certainly hope it will affect development of Croatian industry and economy.

Deliverables Not available.

Publications and outputs Katovic´, D., Bischof Vukusˇic´, S. and Flincˇec Grgac, S. (2007), “Crosslinking cotton with citric acid and organophosphorus agent for the purpose of flame retardant finishing”, 85th Textile Institute Conference, Colombo, Sri Lanka, pp. 820-4. Bischof Vukusˇic´, S., Flincˇec Grgac, S. and Katovic´, D. (2007), “Catalyst influence in low formaldehyde flame retardant finishing system”, 7th AUTEX Conference, Tampere, pp. 60-1. Flincˇec Grgac, S., Katovic´, D. and Bischof Vukusˇic´, S. (2007), “Combination of organophosphorus agent and citric acid in durable press finishing of cellulose fabrics”, XX Croatian Society of Chemical Engineers, Zagreb, Croatia, p. 281. Bischof Vukusˇic´, S., Flinecˇ Grgac, S. and Katovic´, D. (2007), “Antimicrobial textile treatment and problems of testing methods”, Tekstil, Vol. 56, accepted for publication.

Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipovica 30, HR-10000 Zagreb, Croatia. Tel: ++385 1 37 12 557; Fax: ++385 1 37 12 591; E-mail: [email protected] Principal investigator(s): Prof. Budimir Mijovic, PhD Research staff: Prof. Miroslav Skoko PhD, in retirement; Prof. Dragutin Taborsak PhD, Profesor Emeritus; Prof. Salah-Eldien Omer, PhD; Prof. Jovan Vucinic, PhD; Nenad Mustapic, Mr Sc; Jasenka Pivac, Mr Sc; Zlatko Jurac, Mr Sc

Ergonomic design of the worker-furniture-environment system Other Partners: Academic

Industrial

Faculty of Forestry, University of Zagreb Tvin, Virovitica, Croatia Project start date: 1 January 2007 Project end date: 31 December 2009 Project budget: N/A Source of support: Ministry of Science, Education and Sports, Republic of Croatia Keywords: Sitting, Furniture, 3D model of workplace, Work environment, Virtual reality Sitting furniture should enable the worker to take an optimal bodily sitting posture, ensuring active and dynamic sitting. A long-lasting and non-ergonomic bodily posture in this position causes uncomfortable sitting. Defining optimal working postures and strains makes a contribution to the reduction of necessary energy and facilitates working and circulation functions. The total work space should be designed in compliance with all criteria of the working posture and technical requirements. It is necessary to know the worker well, his working capabilities, work place and work methods to ensure an optimal working environment. Furniture dimensions and workplace, surrounding the furniture, regarding its optimal utilization, should be harmonized with the worker’s anthropometric sizes. Research methods are experimental, theoretical and numerical. Functional

Research register

97

IJCST 21,6

98

dependences of the worker-furniture-environment system will be investigated, based on ergonomic postulates in order to find optimal conditions between work humanization and productivity. Investigations are determined by measuring and recording typical working postures as well as conditions of excessive workload. Using digitally scanned 3D anthropometric characteristics of the human body, a digital 3D biomechanical model is obtained, taking account of the appropriate kinematic-dynamic motion rules and the construction of the inner skeleton. 3D program applications with advanced automated defined anthropometric and ergonomic features of biomechanical models and digital figures will be recorded. A 3D visualization of the workplace by using a computer-based 3D model of furniture and computer-based character animations of workers will be performed. By using computer 3D program solutions, the prototype is substituted by 3D models on which all necessary designs and changes in real time have been carried out interactively. Computer visualization will be used to perform a biomechanical analysis of movements based on the real correlation within the space of the interaction of workers and belonging working environment on the obtained 3D models of workers and workspace. It is necessary to analyze the workspace and time studies of motion accurately. The 3D virtual model will enable a detailed biomechanical analysis of motion, speed and acceleration and more designer’s solutions of furniture with biomechanical and ergonomic parameters. Detrimental impact of too a high noise on workers as well as efficient procedures of noise reduction will be investigated. Special attention will be focused on detrimental action of microclimatic conditions regarding technological requirements of the industry. The optimization of work energy during the performance of the work by the worker will be performed to lessen fatigue and to remove excessive workload and to reduce sick-leaves. The performance of these investigations will result in ergonomic technicaleconomic design of the interactive work-furniture-environment system which is of great importance for the development of the Republic of Croatia and elsewhere in the world.

Aims and objectives The purpose of these investigations is to achieve the reduction of work energy when workers work in sitting position and to enable the reduction of fatigue and falling ill. When performing work, a better economic performance with a lower energy consumption and operator’s fatigue should be obtained. In testing the level of detrimental sound effects their efficacious reduction could be achieved. The action of noise and vibrations could be prevented by investigating efficient noise and vibration dampers, damping the transfer of vibrations on workers by using pads and personal protective equipment. The aim of these investigations is to optimize work movements in sitting position with less operator’s fatige and reduction of sick-leaves in the industry. Thereby, working conditions and environment as well as safety in technological processes are to be especially stressed. They should be used to improve economic characteristics substantially. Special purposes of the investigation are the optimization of working postures and the confirmation of the knowledge about defining new criteria for the right ergonomic design of working pieces of furniture.

Deliverables The application of investigations means work simplification, investigation and determination of manufacture time from the point of view of ergonomic starting points. When performing work, a better economic effect is to be achieved with less energy

consumption and operator’s fatigue. Based on the investigated causes of the increase of sound pressure level of machines, it is necessary to find out qualitative solutions to reduce the level into permitted and tolerable limits. To obtain satisfactory microclimatic conditions, work spaces, machinery and air-conditioning instruments with better economic and other characteristics will be defined, whereby current energy consumption will be reduced. Furniture dimensions are determined by computer based visualization of virtual 3D character in interaction with digitized and really designed furniture and environment. Virtual simulation offers the opportunity to create optimal constructive solutions of furniture which enables active and dynamic sitting. It is a relaxation bodily posture that is thereby obtained when sitting. Ergonomic, anthropometric, functional and technical requirements of workers are thereby satisfied. A 3D simulation model of comfortable and safe bodily posture in sitting position is created, together with all biomechanical characteristics, and interactive effects with existing and new materials used for making sitting furniture are attained. Thereby, design and construction of sitting furniture is supplemented, which, because of its designer and constructive solutions and design observance, reduces the possibilities of low-quality designs and problems of body illness and causing inability to work. Publications and outputs Agic, A., Nikolic, V. and Mijovic, B. (2006), “Foot anthropometry and morphology phenomena”, Collegium Antropologicum, Vol. 30 No. 4, pp. 815-21. Agic, A. and Mijovic, B. (2006), “Planar model of the deformation behaviour of electrospun fibrous nanocomposites”, Tekstil, Vol. 55, pp. 606-12. Agic, A., Mijovic, B. and Nikolic, T. (2007), “Blood flow multiscale phenomena”, Collegium Antropologicum, Vol. 31 No. 2, pp. 523-9. Reischl, U., Nandikolla, V., Colby, C., Mijovic, B. and Wei, H.C. (2007), “Ergonomics consequence of Chinese footbinding: a case study”, in Mijovic, B. (Ed.), Ergonomics 2007, Croatian Society of Ergonomics, pp. 1-6. Novak, D. and Mijovic, B. (2007), “Applying cognitive ergonomics to teaching math”, in Mijovic, B. (Ed.), Ergonomics 2007, Croatian Society of Ergonomics, pp. 15-20. Mustapic, N. and Mijovic, B. (2007), “Assessing the slip resistence of flooring”, in Mijovic, B. (Ed.), Ergonomics 2007, Croatian Society of Ergonomics, pp. 155-163.

Zagreb, Croatia Faculty of Textile Technology University of Zagreb, Prilaz baruna Filipovic´a 30, HR-10000 Zagreb, Croatia. Tel: ++38513712500; Fax: ++38513712599; E-mail: [email protected] Principal investigator(s): Prof. Darko Ujevic´, PhD Research staff: Jadranka Akalovic´, BSc; Prof. Jadranka Bacˇic´; Prof. Zoltan Baracˇkai, PhD; Vinko Barisˇic´, BSc; Ing. Iva Berket; Bajro Bolic´, BSc; Blazˇenka Brlobasˇic´ Sˇajatovic´, BSc; Ksenija Dolezˇal, BSc; Mirko Drenovac, PhD; Prof. Milan Galovic´, PhD; Marijan Hrastinski, BSc; Renata Hrzˇenjak, BSc; M.D. Natasˇa Kaleboti; Prof. Isak Karabegovic´, PhD; Ivan Klanac, BSc; M.D. Irena Kos-Topic´; Prof. Tonc´i Lazibat, PhD; Nikol Margetic´, BSc; Prof. Zlatka Mencl-Bajs; M.D. Zˇeljko Mimica;

Research register

99

IJCST 21,6

100

Prof. Gojko Nikolic´, PhD; Alem Orlic´, BSc; PhD, M.D. Vedrana Petrovecˇki; BSc, M.E. Zˇeljko Petrovic´; Prof. Dubravko Rogale, PhD; Prof. Andrea Russo; Igor Sutlovic´, PhD; Prof. Vlasta Szirovicza, PhD; Irena Sˇabaric´, BSc; MSc, M.D. Nadica SˇkrebRakijasˇic´, Marija Sˇutina, BSc; Prof. Larry C. Wadsworth, PhD

Anthropometric measurements and adaptation of garment size system Other Partners: Academic

Industrial

None None Project start date: 1 January 2007 Project end date: 31 December 2011 Project budget: N/A Source of support: Ministry of Science, Education and Sports, Republic of Croatia Keywords: Anthropometric measurements, Garment size system Systematic anthropometric surveys have been conducted since 1901 with the aim of developing and improving systems for clothing and footwear sizes. The measurement results show how a national population changes over a period of several decades in physical build and size due to a series of factors (food habits, sports development, genetic predispositions, population migrations, climatic conditions, etc.). Based on the results of anthropometric measurements in the Republic of Croatia (2004/2005) on the sample of 30,866 test persons aged between 1 and 82 a statistical analysis of body measurements was performed, a database including 5 basic studies of sex and age as well as a new standard for clothing and footwear was built. These results enable a significant and stimulating continuation of scientific research and a comparison to other national standards and their contributions to the creation of systems for clothing and footwear sizes. Elements common for national standards of garment sizing by an exact approach will be investigated and analyzed, in particular because the presumptions of national systems and starting elements, respectively, are not universally founded like intersize intervals which differ in sizes since the conformity of individual starting places is missing. Data will be provided for a common base with methods of body measuring and size designation of clothes according to the recommendations of the Technical Committee TC133 within ISO and EN standards as well as the design and development of a sophisticated computer system (DOV-KO) for unifying all body measurements and basic garment construction based on one or all other sizes. Within the scope of this project and based on experience, a very important cycle of anthropometric measurements of the sporting population in football, water polo, rowing, basketball and handball will be performed. 4,000 test persons from Zagreb, Osijek, Rijeka, Split and Dubrovnik will be measured, whereby specific body differences and deformations of muscles caused by longstanding training will be analyzed. A comparative analysis of the representative sample of the anthropometric measurements of sportsmen and other population as well as the investigation of other trends of body measurements will be performed. This will enable an exceptional insight

into the anthropometric dimensions which reflect body shape, proportionality, composition and elements of success in sports, respectively. Stadiometar or a new measuring instrument for continuous measuring body height, foot length and width will be designed too.

Research register

Aims and objectives Problems of garment sizing and fit affect the market globally, and a consequence of bad predictions of the quantity of necessary stocks for manufacturers and dealers poses a risk of high costs. In the case of domestic manufacturers samples may be additionally divided. Particular solutions may be considered more efficiently by interpreting the data from the anthropometric database connecting 5 studies according to sex and age and the system of sizing. Therefore, using the results of the anthropometric measurements taken and the basic projection of the new standard for clothing and footwear, one of the directives of the project is to investigate other national standards of Europe and the world and to create size intervals and a new Croatian standard. The study of body differences and specific deformities of the body muscles during the longstanding practice of athletes such as leg circumference, chest circumference, torso, shoulder width, arm and leg length, body height, palm length is an additional aim of this project which will show the body elements affecting success level in sport. Besides a greater adaptation of clothing and footwear to the home market, it would be advisable to ensure the continuity of investigating the national size standard by creating a sustainable Croatian system of clothing and footwear sizes in conformity with anthropometric surveys that are conducted periodically and systematically in developed countries in which a change in the morphology of the human body occurred over the last decades. By way of proof, systematic anthropometric measurements and sizing in France showed a tendency of average height growth. In Great Britain in female population a growth of bust circumference was recognized, whereas in USA studies point to the tendency of an increase in obesity (besides an aesthetic also a health problem). The interest of the scientific and professional public, manufacturers, tradespeople and consumers in sizing will continue to grow, since a faster change in established body proportions may be expected thanks to changes in living and food habits of the population, an unavoidable mingling of ethnic groups, increase in the number of older consumers of clothing and footwear which will be doubled in the next two decades, etc. Thus, it is additionally stressed how much it essential at the moment to ensure a valid starting point or a Croatian standard to pursue next movements of body dimensions in order to avoid a discrepancy and imposition of the specificities of domicile consumers to home and foreign manufacturers.

Deliverables Anthropometry is the study of the measurement of the human body, but Pheasant has expanded it as “applied anthropometry” including quantitative data of size, forms and other physical characteristics of people that can be used in garment design. Since the form of the human body changed through time, the problem of ageing proved to make a contribution to perceived changes in body shape and size more than any other individual factor, such as for example improved nutrition and prolonged life, in particular the knowledge that the number of older consumers will be doubled by 2003.

101

IJCST 21,6

102

Therefore, the systems of sizing shall be updated periodically to ensure a correct fit of ready-made clothing. On the other hand, the home industry of clothing, fashion wear and footwear disposes of modest and aged data based on the out-of-date anthropometric measurements from 1962. It was therefore necessary to conduct a new cycle of anthropometric measurements and to use the obtained results. World fashion industry shows a special interest in measuring anthropometric characteristics of the population so that it gathers such data permanently, motivated by the wish for designing articles of clothing for all population groups, including the persons with pronounced specificities (higher stature, higher body weight, etc.). The use of investigations will contribute to creating a new and modern Croatian standard for clothing and footwear harmonized with ISO and EN standards. Besides the clothing and footwear industry, pediatricians, specialists of occupational and sports medicine, experts in wood processing industry, automotive industry, in the army and police will benefit from the investigation results. Teachers and students in undergraduate and graduate studies as well as teachers and pupils at technical schools will benefit from the development results of the computer system based on the selection of garment sizes. By using the investigations of the sporting population, one can get an insight into tendencies of diversities of body measurements and changes in muscles as a result of longstanding practice. Various specialists of sports medicine, orthopedists, garment and footwear designers will benefit form the results of this investigation because based on previous experience it is evident that mass customization is necessary for athletes. Knowing dimensional characteristics, this method would be considerably promoted and improved.

Publications and outputs Ujevic´, D., Rogale, D., Hrastinski, M., Drenovac, M., Szirovicza, L., Lazibat, T., Bacˇic´, J., Prebeg, Zˇ., MenclBajs, Z., Mujkic´, A., Sˇutina, M., Klanac, I., Brlobasˇic´ Sˇajatovic´, B., Dolezˇal, K. and Hrzˇenjak, R. (2006), “Normizacija, antropometrijski pregledi i Hrvatski antropometrijski sustav”, Tekstil, Vol. 55 No. 10, pp. 516-26. Ujevic´, D., Firsˇt-Rogale, S., Nikolic´, G. and Rogale, D. (2006), “Pregled razvojnih dostignuc´a u tehnologiji sˇivanja – IMB 2006”, Tekstil, Vol. 55 No. 12, pp. 624-31. Ujevic´, D., Dolezˇal, K. and Lesˇina, M. (2007), “Analiza antropometrijskih izmjera za obuc´arsku industriju”, Poslovna izvrsnost, Vol. 1 No. 1, pp. 171-83. Ujevic´, D., Hrzˇenjak, R., Dolezˇal, K. and Brlobasˇic´ Sˇajatovic´, B. (2007), “Hrvatski antropometrijski sustav – jucˇer, danas, sutra”, HZN Glasilo, Vol. 3 No. 1, pp. 5-10. Hrzˇenjak, R., Ujevic´, D., Dolezˇal, K. and Brlobasˇic´ Sˇajatovic´, B. (2007), “Investigation of anthropometric characteristics and body proportions in the Republic of Croatia”, Proceedings of 7th Annual Textile Conference by Autex, Tampere, Finland, 25-28 June, pp. 1191-8. Ujevic´, D., Brlobasˇic´ Sˇajatovic´, B., Dolezˇal, K., Hrzˇenjak, R. and Mujkic´, A. (2007), “Rezultati prvog antropometrijskog mjerenja stanovnisˇtva Republike Hrvatske”, Drugi kongres hrvatskih znanstvenika iz domovine i inozemstva, Split, Croatia, 5-10 May. Nikolic´, G. and Ujevic´, D. (2007), “Protractor for measuring shoulder slope”, Patent. Ujevic´, D. (2007), “One-arm and/or two-arm anthropometer”, Patent.

Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipivic´a 30, HR-10000 Zagreb, Croatia. Tel: +38514877351; Fax: +38514877357; E-mail: [email protected] Principal investigator(s): Prof. emeritus, Ivo Soljacˇic´, PhD Research staff: Asoc. Prof. Tanja Pusˇic´, PhD; Prof. Ljerka Bokic´, PhD; Asst. Prof. Branka Vojnovic´, PhD; Iva Rezic´, PhD; Prof. Jelena Macan, PhD; Asoc. Prof. Barbara Simoncˇic´, PhD, Prof. Sonja Sˇostar-Turk, PhD; Asist. Prof. Sabina Fijan, PhD; Mila Nuber, MSc; Ivan Sˇimic´, MSc; Dinko Pezelj, PhD, Versˇec Josip, MSc

Ethics and ecology in textile finishing and care Other Partners: Academic

Industrial

Labud, d.d. Zagreb and Vodovod, Zagreb University of Maribor and University of Ljubljana, Slovenia Project start date: 1 January 2007 Project end date: 31 December 2011 Project budget: N/A Source of support: Ministry of Science, Education and Sports, Republic of Croatia Keywords: Wellness finishing of textiles, Determination of harmful substances on textiles, Toxicological and alergenic properties, Environmental protection, Hygiene and effects of textile care, Textile material sample preparation. Modern textile finishing processes have to fulfill high demands due to the expectations of new textile materials properties and their persistence during care. Especially interesting in this respect are the new production processes of socks which include implementation of microcapsules that can release active materials for skin moisturizing. Their primal role is prevention of dryness, dandruff and allergenic reactions of the skin. The most suitable analytical methods for determination of durability to washing, friction and sweat will be tested. Durability to washing of products with special properties will be tested with different amounts of anionic and cationic surfactants in liquid detergents. The mechanism of adsorption and desorption, their influence on primary effect of the treatment, and the influence of the pH value and the mechanical way of treatment will be tested. On the ground of the obtained results, analytical methods for determination of micro components in the macro components of textile materials should be proposed, without regards to the specifications of the materials or the method of the treatment. The testing will involve a review of the analytical method of each individual analytical procedure as well as its impact on the obtained information. The parameters of the analytical procedure will be worked out with the purpose of restoration of historical textile by destructive and non-destructive methods for the preservation of national heritage. European controlling methods of new materials have ethical demands involving the human population health which demands an environmental friendly process. For this

Research register

103

IJCST 21,6

104

purpose the processes of textile finishing and care will be optimized. The possibility of obtaining new preventive properties, which were not previously present on the textile material or improvement of present protection, will be tested. The impact of washing cycles with detergent and UV absorber on pastel colored textile materials made of cotton, polyester and their mixtures on UPF and the shade change will be investigated. The quality control of water and effluents will be based on the determination of micro quantities of potential allergens, heavy metals, pesticides, dyes, and surfactants. The traces of solvents will be controlled on the clothing material and in the air during the chemical cleaning and further treatment processing.

Aims and objectives The main goal of this investigation is to stimulate ethical ecological demands on the production processes, care processes, and thereby on the utilization properties of the textile materials wherewith it would be possible to get the optimal properties of materials regarding their functional properties by avoiding all possible harmful allergenic and toxicological influences of textile materials to consumers. Elaboration of production and textile finishing processes for the optimal effects (wellness finishing, protection from unwanted changes of utilization properties in texcare, elaboration of pastel dyed textiles laundering in detergent with UV absorber, additional laundering quality – UV protection), formulation of new compositions for laundering for the purpose of avoiding secondary harmful effects in modern conditions with maximal saving of water and energy, more safe treatment with solvents during dry cleaning. By monitoring of harmful inorganic and organic substances that are present in micro quantities on the textile materials, textile accessories, textile wastewaters and finished textile products, new analytical methods would be determined. Sampling procedures, sampling preparation steps, selection of appropriate analytical method and the processing of the obtained result will be optimized. In this investigation the mathematical modes for guiding of analytical procedure will be applied, what is economically justified because the time spend for investigation is much shorter, and the consumption of chemical reagents, energy and emission of harmful substances to the environment reduced. Special contribution will be in development of analytical methods for determination of components present on the historical textile, for the purpose of avoiding the damaging of the textile material during restoration conservation treatments.

Deliverables The project is scheduled over three years. Eco problems and human ecology, especially presence of heavy metal traces in textile processes and fibres will be investigated and some results will be published. Analytical methods for qualitative and quantitative determination will be developed. The influence of sweat on the heavy metal emission will be tested from colored textile materials. Possibility and durability of wellness finishing effects, particularly on PA pantyhoses, as well as methods will be established. UPF and change in shade of white and pastel colored textiles made from cotton, PET, PA and their blend with cotton during laundering with addition fluorescent compounds in detergent will be researched, too. Hygienic laundering with chemothermic and chemical treatments in order to destroy micro-organisms in compliance with existing recommendations will be done. Potentially irritations of the skin

caused by textiles, finishing agents and inadequate rinsing during laundering will be studied. Investigation of anionic, cationic and nonionic surfactant adsorption and desorption influenced by different composition of textile fibres, pH and temperature will be performed. The adsorption and desorption will be studied in order to establish a correlation between zeta potential and swelling capacity of textile fibres.

Research register

Publications and outputs Rezic´, Iva, and Steffan, Ilse (2007), “ICP-OES determination of metals present in textile materials”, Microchemical Journal, Vol. 85 No. 1, pp. 46-51 (scientific paper). Fijan, Sabina, Pusˇic´, Tanja, Sˇostar-Turk, Sonja and Neral, Branko (2007), “The influence of industrial laundering of hospital textiles on the properties of cotton fabrics”, Textile Research Journal (in press). Pusˇic´, Tanja, Jelicˇic´, Jasenka, Nuber, Mila and Soljacˇic´, Ivo (2007), “Istrazˇivanje sredstava za kemijsko bijeljenje u pranju”, Tekstil (in press). Pusˇic´, Tanja and Soljacˇic´, Ivo (2007), “Changes in shade of cotton fabrics during laundering with detergents containing fluorescent brightening agent and UV absorber”, AATCC Review (in press). Vojnovic´, Branka, Bokic´, Ljerka, Kozina, Maja and Kozina, Ana (2007), “Optimization of analytical procedure for phosphate determination in detergent powders and in loundry wastewater”, Tekstil (in press).

105

Zagreb, Croatia University of Zagreb, Faculty of Textile Technology, Prilaz baruna Filipovica 30, HR-10000 Zagreb, Croatia. Tel: +385 1 37 12 566; Fax: +385 1 37 12 599; E-mail: [email protected] Principal investigator(s): Prof. Maja Andrassy, PhD Research staff: Prof. Zvonko Dragcevic, PhD; Assoc. Prof. Emira Pezelj, PhD; Prof. Dubravka Raffaelli, PhD; Assist. Prof. Edita Vujasinovic, PhD; Zvonko Orehovec, PhD; Vera Friscic, MSc; Ruzica Surina, BSc; Prof. Majda Sfiligoj Smole, PhD

High performance textile materials and added-value fibres Other Partners: Academic

Industrial

None University of Maribor Faculty of Mechanical Engineering, Maribor, Slovenia Project start date: 1 January 2007 Project end date: 31 December 2011 Project budget: N/A Source of support: Ministry of Science, Education and Sport, Republic of Croatia Keywords: HP materials, Textile fibers, Textile design Contemporary global trends of development in the field of textile fibres and fibrous materials have led to their increased use in various fields of industry and technique. Increase in consumption of these types of materials has constantly been recorded and

IJCST 21,6

106

by the beginning of the 21st century technical fibres account for half of all the fibres manufactured. The requirements imposed on fibres and materials in particular fields of application and extraordinary high and specific. These requirements have been met through fibre engineering, i.e. development of new generic types of fibres. It can be assumed that innovative manufacturing and finishing processes, applied to conventional fibres, will result in their added value, so that they can be used to design new fabrics of pre-determined end-use properties. Such improvements in fibre properties and their use in fabric manufacture of added market value and broader scope of application are completely in accordance with the intentions of the European technological platform for future textile and garment, but it can also strongly stimulate the development of Croatian textile industry and its comptetitiveness in the global market. This is supported by the fact that there are considerable research, industrial and raw-material potentials in Croatia, necessary to accomplish the goals. Although domestic production is mostly based on imported fibres, clearly defined modifications of fibre structure and propertties, even for domestic fibres, such as wool, flax, textile regenerates and fibres made from recycled PET, that have been used in Croatian textile industry insufficiently until now, could be used as a starting raw material for the manufacture of high-performance textiles. In this manner, domestic fibrous raw materials would cease being waste material and would become strategic Croatian raw material, as well as a basis of future rational management of natural resources and a step in approaching sustainable development trends, recommended by the European Union and United Nations. The investigations proposed aim at establishing the possibilities of modifying conventional fibres, as well as developing the methods and procedures of objective measurement and evaluation of unconventional textile materials, in accordance with specific rules and requirements for individual types of high-performance textiles, including composites reinforced with fibres of modified properties. The results obtained will offer the construction of high-performance materials based on conventional fibres of added value, as well as the design and optimisation in accordance with the properties of the fibres used and pre-determined high end-use properties.

Aims and objectives The main purpose of the project proposed is to determine possible interventions and modifications of conventional fibres and textiles, so as to obtain added value and to broaden the scope of their application. This improvement of fibres and their application in the manufacture of new, knowledge-based innovative textiles of added market value is in accordance with the short term (energy and materials) and long-term (nano-science, new materials, constructions and production processes) strategic trends of research in the Republic of Croatia, strategies of the European technological platform for the future of textiles and garment in the XXI century, as well as with the trends of rational management of raw material resources and the concept of sustainable development, as proposed by the EU and UN recommendations. Valuable results and new knowledge are expected, especially regarding ecologically friendly and economically feasible production of high-performance textiles through the usage of domestic raw materials in their manufacture. There is a broad diversity of fibres and constructions present in the area of non-conventional textile materials and structures, which makes objective characterisation of their quality a difficult task, we propose to develop new methods,

procedures and equipment for testing, so as to enable higher degree of objectivity in quality evaluation. The investigations are planned to initiate the development and optimising of the manufacture of high-performance textiles, matched with their increased and more and more specified areas of application. The results expected to be obtained will enhance the scope of knowledge in the field of textile fibres, materials and textile testing, also valuable in the education of young researchers, knowledge transfer and preparation of future textile engineers for the European labour market.

Deliverables The results of the investigation will be directly applicable in Croatian textile industry, since new solutions for designing high-performance materials, based on conventional fibres of added value, will be proposed as based on the results of the investigations proposed. As there are some processing capacities still in Croatia (Regeneracija, Kelteks, Vrbenka, Konoplja, LIO, Feniks), working with imported fibres, the system of objective description and evaluation of domestic fibrous raw materials (especially flax and wool), as well as some instructions regarding their environmentally acceptable manufacture, processing and modifications, with the aim of enhancing end-use properties, are expected to create adequate conditions for econbomically feasible manufacture of textiles. Higher content of domestic fibres and raw materials in manufacture of textile would be a sound basis for new development and growth of the Croatian textile industry and its competitiveness in the European and global markets, as well as for realising the principles of sustainable development and rational raw material resource management. Developments and innovations in testing metodology and objective evaluation of relenat properties of the modified fibres and new high-performance textiles will enhance objectivity of testing and evaluation of high-performance technical textiles in general, which is, not only in Croatia but globally as well, a problem with no acceptable solution on the horizon. Some testing methods are expected to be used in production monitoring and control, which could contribute to more reliable and stable manufacture and realising pre-planned levels of quality. The investigations proposed open the way to scientific and professional collaboration with other institutions and with the industry. Publications and outputs Ruzˇica Sˇurina i Maja Somogyi (2006), “Biodegradable polymers for biomedical purpose”, Tekstil, Vol. 55 No. 12, pp. 642-5. Ruzˇica Sˇurina i Maja Andrassy (2007) “Resistance of lignocellulosic fibers to microorganisms”, XX. hrvatski skup kemicˇara i kemijskih inzˇenjera, knjiga sazˇetaka, posvec´en Lavoslavu Ruzˇicˇki i Vladimiru Prelogu, hrvatskim nobelovcima u kemiji, Zagreb, 26. veljacˇa-01. ozˇujka, p. 286. Cindric´, Jasna (2007), “Improvements properties of flax fibers, diploma work”, Tekstilno-tehnolosˇki fakultet, Zagreb, Voditelj, Andrassy, Maja, 25 April, 56 str. Klasic´, Sanja (2007), “Usable properties of modified flax fabric, diploma work”, Tekstilno-tehnolosˇki fakultet, Zagreb, Voditelj, Andrassy, Maja, 25 April, 53 str. Sˇurina Ruzˇica i Andrassy Maja (2007), “Quality of modified flax fibers”, Proceedings: 18th International DAAAM Symposium on Intelligent Manufacturing & Automation: Focus on Creativity, Responsibility and Ethics of Engineers, 24-27 October (in press). Vujasinovic, E., Jankovic, Z., Dragcevic, Z., Petrunic, I. and Rogale, D. (2007), “Investigation of the strength of ultrasonically welded sails”, International Journal of Clothing Science and Technology, Vol. 19 Nos 3/4, pp. 204-14, ISSN: 0955-6222.

Research register

107

IJCST 21,6

108

Vujasinovic, E., Dragcevic, Z. and Bezic, Z. (2007), “Descriptors for the objective evaluation of sailcloth weather resistance”, Proceedings of 7th Autex Conference 2007, Tampere, Finland, 26-28 June, ISBN: 978-952-15-1794-5.

Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipovic´a 30, HR-10000 Zagreb, Croatia. Tel: +385 1 37 12 552; Fax: +385 1 37 12 599; E-mail: [email protected] Principal investigator(s): Assist. Prof. Zˇeljko Sˇomodi, PhD Research staff: Assist. Prof. Ana Kunsˇtek, PhD; Slavica Bogovic´, MSc; Anica Hursa, MSc, Igor Petrunic´, MSc; Assist. Prof. Simona Jevsˇnik, PhD; Daniela Zavec-Pavlinic´, PhD

Computational modelling in engineering analysis of textiles and garment Other Partners: Academic

Industrial Kamensko d.d., Zagreb

University of Maribor, Faculty of Mechanical Engineering, Maribor, Slovenia Project start date: 1 January 2007 Project end date: 31 December 2011 Project budget: N/A Source of support: Ministry of Science, Education and Sports, Republic of Croatia Keywords: Clothing technology, Numerical methods, Optimisation, Reinforcements on clothing The intention of this project is to give a contribution to advanced application of the methods of engineering analysis in the field of textiles and garment. This goal will be achieved by introduction, adaptation, elaboration and application of up-todate computational methods in the analyses of problems relevant for the field of textile and clothing engineering. Considering the existing experience and an overview of questions and problems actual for the engineering science in the field, the research is to be undertaken in a number of areas, such as: optimal design of structural reinforcements in garment based on the finite element analysis; three parameter model of tensile nonlinearity of textiles; computational evaluation of post-buckling stable state in prediction and simulation of fabric drape; general numerical solution of thin plate bending with application to optimal grip geometry in automated work piece manipulation. Depending on the timing and realization of these researches, there is a possibility of opening further research areas from the field of computational modelling in mechanics of textiles and garment, including the spatial modelling and design of clothing items. The methods of research to be applied primarily consist of derivation and elaboration of numerical models

suitable for application in the problems under consideration, and the development and application of computer programmes based on these models. At the same time, the plan is to acquire and apply some of the existing software applicable in the problems to be considered, as well as to prepare and conduct experimental verification of results obtained by computations.

Aims and objectives Aim and scope of the proposed research is to give a contribution in the improvement of the level of engineering and technological know-how in the field of textile and garment. The research is expected to result in computer programmes or engineering data collected in tables, diagrams, etc. that will be useful for the problem solution in the area of expertise covered by the research. The knowledge and methods developed by the research will be on offer for the interested subjects, from Croatia or elsewhere, primarily from the branch of textiles and garment. It can also be expected that after some time the collected knowledge and methods will be included in the teaching process at the Faculty of Textile Technology, primarily as parts of the subjects at the doctoral or diploma levels.

Deliverables The principal user of the results will be Faculty of Textile Technology, as the institution for production and transfer of knowledge in the field of textile and garment. Further users will be the firms, institutions and individuals from among the designers and manufacturers of textiles and garments, who already have the co-operation with the faculty, or shall have that co-operation in the future. The results and findings of the research shall be offered to these subjects by means of the Centre for development and transfer of textile and clothing technology and fashion design, as a unit in the structure of the faculty. The specific applications are expected in the expert analyses related to engineering in preparation of production processes in which the problems from the field of research appear. Publications and outputs Hursa, Anica, Rogale, D. and Sˇomodi, Zˇeljko (2006), “Application of numerical methods in the textile and clothing technology”, Tekstil, Vol. 55 No. 12, pp. 613-23. Akrap-Kotevski, Visˇnja and Kunsˇtek, Ana (2007), “Writing ability after brain damage”, in Mijovic´, B. (Ed.), Proceedings of 3rd International Ergonomics Conference, Ergonomics 2007, Croatian Society of Ergonomics, Zagreb, pp. 279-85. Sˇomodi, Zˇeljko, Hursa, Anica, Rolich, Tomislav and Rogale, D. (2007), “Numerical analysis and optimisation of mechanical reinforcement on clothing”, in Cˇanadija, M. (Ed.), Book of Proceedings of the 1st Meeting of Croatian Society of Mechanics, Croatian Society of Mechanics, Rijeka, pp. 173-8. Sˇomodi, Zˇeljko, Hursa, Anica and Rogale, D. (2007), “A minimisation algorithm with application to optimal design of reinforcements in textiles and garments”, Internationl Journal of Clothing Science and Technology, Vol. 19 Nos 3/4, pp. 159-66.

Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipovica 30, 10000 Zagreb, Croatia. Tel: +385 1 37 12 577; Fax: +385 1 37 12 533; E-mail: [email protected]

Research register

109

IJCST 21,6

110

Principal investigator(s): Prof. Zenun Skenderi, PhD Research staff: Prof. Miroslav Srdjak, PhD; Prof. Momir Nikolic´, PhD; Prof. Alka Mihelic´-Bogdanic´, PhD; Bozˇo Tomic´, MSc; Vesna Marija Potocˇic´ Matkovic´, MSc; Ivana Salopek, MSc; Dragana Kopitar, BSc

Multifunctional technical nonwoven and knitted textiles, composites and yarns Other Partners: Academic None

Industrial Cˇateks d.d. Cˇakovec, Croatia, Regeneracija non-woven and carpets j.s.c., Zabok, Croatia Project end date: 1 January 2010

Project start date: 1 January 2007 Project budget: N/A Source of support: Ministry of Science, Education and Sport, Republic of Croatia Keywords: Yarns, Knitted and nonwoven fabrics, Structures, Properties, Thermal and water-vapour resistance properties Further dislocation of the textile production from developed countries into Asia is a basic characteristic for the world textile industry today. In the field of technical textiles profound resistance is felt against relocation. An increase in the production of technical textiles is recorded due to a permanent expansion of the application range. It is used in: transportation, industry, medicine, hygiene, household, garment industry, agriculture, fishing trade, civil engineering, sport, safety, ecology, etc. Nonwovens make the most significant contribution to the development of technical textiles. Over last decades the technology of nonwovens production has experienced a rapid development, and the production of late years has registered an annual increase of approx. 10%. A significant application range for technical textiles or geotextiles is civil engineering, in particular road building. In addition to woven, knitted and similar structures, nonwovens play a predominant role with a share of approx. 75% in 2005. The most important functions of geotextiles are: separation of weak soil, reinforcement of soil or elements of building structures, filtration and drainage. Geotextile properties are: stability, uniform structure, small thickness, high strength and stretching, porosity, small surface mass and water permeability. Various applications require a more or less marked particular structure and characteristic. The first part of the project will deal with various structures and properties of technical textiles based on nonwoven and knitted structures, in particular on geotextiles. Moreover, manufacturing technologies of technical textiles and knitted materials as well as their controlling parameters will be discussed. Conventional technologies such as: spinning, weaving, knitting and clothing technology will probably not withstand the competitiveness coming from Asia. Besides, relocation of the manufacture of man-made fibres into the Far East is taking place. It is undoubtedly the case that only those disposing of raw materials and enough knowledge to produce and sell high-quality products will have the chances of survival on the market.

The investigation of possibilities of manufacturing from coarser sorts of wool which have similar fineness as domestic wool and the investigation of their possible use for products such as carpets and several articles of clothing will be within the scope of this project. The limit of fibre spinnability, typical stress-strain curves, yarn behavior in cyclic examinations of elongation properties, surface friction and yarn hairiness.

Aims and objectives There are two dominant reasons why the field of technical textiles is dealt with in the project: (1) Intensive development of technical textiles because of an increase in the application (technical textiles on average of approx. 5.5%, nonwoven fabric approx. 10%). (2) Resistance to the relocation of the manufacture of technical textiles to the Far East. As a result the manufacture of technical textiles (quantity) in the developed countries accounts for more than 40% of the total production of textiles. The state of the Croatian textile industry should be noted where the classic textile industry has almost vanished. Two companies Regeneracija and Cˇateks manufacture technical textiles which have their markets. Regeneracija and Cˇateks confirmed their participation in the project. Moreover, the use of domestic wool for various products of higher value is a challenge and obligation. However, today domestic wool is sold as a raw material, and in some regions it is not bought off which is an ecological problem. Based on the above mentioned facts, it is reasonable to deal with the subject matter of technical textiles and yarns and products, respectively, such as carpets within the scope of the project. In this way production of higher-quality products is promoted which is the purpose of this investigation. The aims of the investigation are as follows: (1) Definition of the interdependence, process parameters and physical-mechanical as wel as other relevant properties: . technical nonwoven textile; . technical textile based on nonwoven fabric coated with polyurethane (PUR); and . technical textile based on knitted fabric from PET and PA coated with polyurethane (PUR) and other technical textiles based on knitted fabric. (2) Definition of the interdependence of raw materials, process parameters and physical-mechanical yarn properties, primarily wool, and the behavior of carpets in dynamic investigations (new instrument required for the project). (3) It will be attempted in case of interest of Regeneracija or other interested parties to investigate thermal properties of wool insulation materials.

Deliverables Obtaining the new understandings of: . Thermal and vapor resistance properties of the knitted and nonwoven fabrics. . Spun yarns, primarily coarser wool yarns: procedures of manufacturing and defined the controlling parameters of the processes.

Research register

111

IJCST 21,6

.

.

.

112 .

Structure and properties of the coarser, industrially manufactured wool yarns should be emphasized that can be similar to the yarn spun from domestic wool as far as their properties. Influence of fibres parameteres (finenness, length,. . .) on limit of spinnability and spun yarn properties. Possibilities of manufacturing carpets from coarser wool fibres will be investigate as well as compressibility of carpets on the new instrument purchased from funds of the project. Processes of manufacturing: technical nonwoven fabrics intended for use in civil engineering, technical nonwoven fabric coated with polyurethane (PUR), technical textiles based on knitted fabric coated with polyurethane (PUR), as well as their structures and properties.

Publications and outputs Salopek, Ivana, Skenderi, Zenun and Srdjak, Miroslav (2007), “Stoffgriff – ein Aspekt des Tragekomforts von Strickware”, Melliand Textilberichte, Vol. 88, pp. 426-8. Salopek, Ivana and Skenderi, Zenun (2007), “Thermophysiological comfort of knitted fabrics in moderate and hot environment”, in Mijovic´, B. (ur.), Proceedings of the 3rd International Ergonomics Conference, Croatian Society of Ergonomics, Zagreb, pp. 287-93. Potocˇic´ Matkovic´, Vesna Marija and Salopek, Ivana (2007), “Computer assisted study of knitted structures”, Proceedings Vol. IV, CE Computers in Education, pp. 99-102. Kopitar, Dragana and Skenderi, Zenun (2006), “Prsteni i trkacˇi – glavni elementi prstenaste predilice”, Tekstil, Vol. 55, pp. 543-01.

Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipovic´a 30, HR-10000 Zagreb, Croatia. Tel: + 385 1 37 12 500; Fax: + 385 1 37 12 595; E-mail: [email protected] Principal investigator(s): Assoc. Prof. Zlatko Vrljicak, PhD Research staff: Kresimir Hajdarovic, PhD; Valent Strmecki, MSc; Tomislav Koren, MSc; Ivan Basnec, MSc

Design and manufacture of nets for the protection of fruit and vegetables against hail Other Partners: Academic None

Industrial Tvornica mreza i ambalaze, Biograd n/m, Croatia Project end date: 31 December 2009

Project start date: 1 January 2007 Project budget: N/A Source of support: Ministry of Science, Education and Sport, Republic of Croatia

Keywords: Fruit production, Nets, Hail, Viticulture, Plant material, Economic profit Approximately 7% of Croatian fruit is present on the Croatian market. The production and sales of Croatian fruit can be quadrupled and sold at present prices on the Croatian market, but as first class fruit. Over the last ten years The Ministry of Finance received damage reports worth more than 100 million kuna which were caused by hail. Fruit, vegetables, plants, flowers, nursery-gardens, and animals, material resources: houses, agricultural machinery, automobiles and the like get damaged. Up to now rockets have been used to ensure hail protection of fruit and vegetables. On account of a rapid increase in the volume of air transport this technique is less used and is substituted by using protection nets. Several more developed and neighboring European countries have started using nets for the protection of fruit and vegetables against hail. Within the scope of this project systems of applying protection nets in European countries and their use in Croatia would be studied. The emphasis here would be on the safety of orchards, new plants or crops and how to pay compensation for damages by hail. Appropriate protection nets would be designed and manufactured for particular agricultural products and then installed on plantations. Net construction depends on the application so that protection nets of various widths, shapes, colors and structures with special emphasis on the raw material for the production of nets and for shadowing the area to be covered. Across Croatia nets would be offered to the registered fruit growers for use. During the first year of the project a fruit grower would be offered 1000 m2 of nets without charge with the aim that he buys the same quantity (ratio 1:1) and that he should cover only one part of his plantation. By continuous monitoring of orchards all changes under the nets would be analyzed and then compared to the results obtained outside the nets. Over the period of five years relevant conclusions about the use of nets for the protection of fruit and vegetables against hail can be made. It is to be emphasized that the nets, which protect agricultural products against hail, can protect against sun, birds, animals, etc. By adequate use of the above mentioned nets it is expected that yield per hectar will be increased as well as fruit quality. Up to now our fruit growers have collected less than 50% of first class fruit, but using nets it can be expected to increase this limit over 80%. In this way, when Croatia enters into the European Union, we can sell our quality fruits on our market and compete on international markets.

Aims and objectives The purpose of the investigation is to provide assistance to fruit growers to protect their orchards, both trees and fruits against hail. The aim of the project is to design and make nets that will be helpful and commercially acceptable in the protection of fruits and vegetables against hail. Manufactured nets would be offered to fruit growers for use. During the duration of the project several types of nets for the protection of fruit and vegetables would be developed. The application aim of these protection nets is primarily to protect trees and fruits. In this way fruit growers would increase the quality of collected fruit and yield quantity per unit area. In the long term, when Croatia enters into the European Union, our fruit growers will be able to sell their quality fruits on the Croatian market and compete on the international markets. If we do not do it, the European fruit producers will sell their fruits on the Croatian market without competition. The Croatian fruit growers will not be able to enter into this market because

Research register

113

IJCST 21,6

of minor fruit quality. This action will prevent the breakthrough of export fruit on the Croatian market.

Deliverables

114

The whole plan and protocol is mentioned in section 9.2. Therefore, only basic characteristics are mentioned here. After 1st year more than 20,000 m2 of nets will be installed on different fruit plantations across Croatia. We would collect data on necessary properties of nets for the application in a particular sector or for a particular orchard. Based on the gathered data we would design and make new nets with different physical-mechanical properties. We would come to basic data on the influence of sun beams and weathering on the changes in net characteristics. We would publish approximately five papers. During previous three years we would keep records of all important influences on the fruit quantity and quality. We should come to data on the influence of hail, sun, wind, thrash, birds, animals and the like on the fruit quality and quantity. We would publish a newsletter about the application of protection nets in the cultivation of fruit and vegetables. Publications and outputs Not available.

Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipovic´a 30, HR-10000 Zagreb, Croatia. Tel: +385 1 3712575; Fax: +385 1 3712599; E-mail: [email protected] Principal investigator(s): Assoc. Prof. Stana Kovacˇevic´, PhD Research staff: Assist. Prof. Zˇeljko Penava, PhD; Josip Hadina, MSc; Ivana Schwarz, BSc Assist. Prof. Andrea Pavetic´, Nikol Margetic´, BSc; Irena Sˇabaric´, BSc; Valent Strmecˇki, MSc; Dubravka Gordosˇ, MSc; Biserka Vuljanic´, MSc; Prof. Vladimir Oresˇkovic´, PhD, in retirement; Assoc. Prof. Krste Dimitrovski, PhD, Blago Brkic´, PhD in retirement, Diana Franulic´ Sˇaric´, MSc

Advanced technical woven fabrics and processes Other Partners: Academic University of Ljubljana, Slovenia

Project start date: 1 January 2007

Industrial Varteks d.d., Varazˇdin; Cˇateks d.d., Cˇakovec; Kelteks d.o.o., Karlovac; Lola Ribar d.d., Karlovac; Uriho, Zagreb; Peng d.o.o., Zagreb Project end date: 31 December 2009

Source of support: Croatian Ministry of Science, Education and Sport, Republic of Croatia Keywords: Technical woven fabrics, Composites, Wool, Flax, Tapestry, Ethno heritage

Research register

The subject of this project is advanced technical woven fabrics and processes. They are intended for the use in interior decoration, transportation, industrial and medical purposes, tapestry and the like. These fabrics contain raw materials in common, and domestic wool and linen yarn as well as glass and carbon yarn will be preferred, but other natural raw materials will be used too, yarns of chemical fibers from synthetic polymers and “smart” yarns. The aim of this research is to find the most optimal raw material and fabric construction and to make a commercially acceptable, qualitative, healthy, comfortable and smart technical fabric. Basic investigations will include: physical-mechanical, thermal, relaxation and elongation properties, dimensional stability, abrasion, effect of sun rays, inflammability, air permeability, water repellency, degradation and the investigation of these properties depending on fabric application. The scope of investigation will include technical fabrics intended for use in civil engineering, transportation and household (3D fabrics for composites, fabrics for seat covers, furnishing fabrics, etc.) on which high requirements are set, such as: safety, resistance, comfort and aesthetics. Technical fabrics for industrial purposes such as filter fabrics and fabrics for composites which are of great importance for better utilization and productivity, and still more important in terms of ecological protection of environment, will be investigated. Healthy fabrics in medical terms from natural raw, and generally fabrics with various properties and applications subjected to additional treatments according to health standards. Part of the project will be directed at the investigation and revival of Croatian eco and ethno heritage, including the manufacture of tapestry, blankets and mats interwoven with art and skill of weaving, using domestic raw materials. The aim of this research is that tapestry authenticity and originality of work of art represent a unique value. The significance of this project is to revive the processing of domestic wool and flax in parallel with the investigation of new constructions and forms of glass technical fabrics, and new materials processed by new technologies. Several technical fabrics replicated in this project will serve as an encouragement for processing domestic wool and flax in smaller batches in karts regions of Croatia.

115

Aims and objectives .

To design and manufacture technical fabrics from natural and new materials, and with their combination, with new methods and manufacturing procedures.

.

To investigate qualitative properties of technical fabrics from natural textile raw materials and to promote the use of primarily domestic wool and linen.

.

Within the basic aim it is necessary to define a numerical model of the manufacturing process and development of new composites so that chemical and physical processes affecting the quality and applicability of composites are investigated and analyzed experimentally. Thereby the influence of particular

IJCST 21,6

116

processes (manufacturing processes, temperatures, outside influences) on the changes of composite properties as well as the interrelationships will be determined. .

The knowledge gained will be used to define a computer model for the construction of new types of technical fabrics and accordingly new composite materials.

.

To systematize utility and quality values of wool yarn from domestic sheep breeds and to investigate the possibility of the industrial process in blending with other raw materials.

.

To optimize the production of applicable products of domestic raw materials and to promote the revival of the textile ethno heritage in Croatia.

.

To investigate the most cost-effective process and the most optimal shares of individual raw materials for the target product with new properties.

.

To include other raw materials with domestic materials with the aim of higherquality, more valuable and attractive products.

.

To manufacture healthy, high-quality, “smart”, comfortable, unique and commercially acceptable textile materials by integrating “smart” materials and new materials dyed with natural dyes.

Deliverables Manufacture of technical fabrics from different composites with improved properties such as: more healthy (natural fibers), more comfortable (better thermal properties), and more attractive (uneven surface structure made by weaving various densities, fineness, colors and raw materials with different weaving techniques and different finishing treatments). It may be expected that the results obtained will make a considerable contribution to further development of manufacturing technical fabrics and composites based on them, and that their application range will be expanded. Manufacture of medical fabrics based on natural raw materials and new composites which will be used as part of orthopedic and surgical aids that will dispose of improved properties such as: strength, thermal properties, liquid absorption, elasticity, stability and other properties. Manufacture of tapestries, blankets and mats with natural and new materials dyed with natural dyes whereby modern, high-quality and unique products can be produced. Moreover, by mastering these skills the possibility of a quality restoration of the textile ethno heritage will be provided. By applying the results obtained in various economic branches, the development of new construction materials as a substitution for classic materials will be possible, which will make a direct contribution environmental protection and entire ecological development of the country. By realizing such results the project will make a further contribution of the development of science in the field of developing new materials which according to long-term strategic directions of research in Croatia. They emphasize the need for the investigation of new materials, constructions and manufacturing processes. At the same time, the project is on the track of short-term strategic directions of scientific researches which direct investigations at natural, glass and organic-inorganic hybrids, making intelligent materials and polymer research.

Publications and outputs Kovacˇevic´, S. and Schwarz, I. (2007), “Hand weaving – tradition of the future”, 7th Annual Textile Conference by Autex: From Emerging Innovations to Global Business, Tampere, Finland, 26-28 June. Schwarz, I., Flincˇec Grgac, S., Kovacˇevic´, S., Katovic´, D. and Bischof Vukusˇic´, S. (2007), “The efect of drying methods on sized yarn characteristics”, in Katalinic, B. (Ed.), The 18th International DAAAM Symposium – Intelligent Manufacturing & Automation: Focus on Creativity, Responsibility, and Ethics of Engineers, Zadar, Croatia, 24-27 October 2007, DAAAM International, Vienna (in press). Ujevic´, D., Kovacˇevic´, S., Schwarz, I. and Brlobasˇic´ Sˇajatovic´, B. (2007), “Novi visˇeslojni tekstilni plosˇni proizvodi”, in Karabegovic, I., Dolecek, V. and Jurkovic, M. (Eds), 6th International Scientific Conference on Production Engineering: Development and Modernization of Production, 24-26 October, RIM 2007 (in press). Kovacˇevic´, S., Ujevic´, D., Schwarz, I., Brlobasˇic´ Sˇajatovic´, B. and Brnada, S. (2007), “Analysis of motor vehicle fabrics”, Fibres & Textiles in Eastern Europe (in press).

Research register

117

Research index by institution

IJCST 21,6 Institution

118

Arak Islamic University Csi-Tex Dokuz Eylul University Donghua University Ege University

Page 6 48, 50 63, 67, 69, 70 11 64,66

Ghent University

27, 28, 29, 30

Heriot-Watt University

12, 13, 14, 15, 18, 19, 21, 22, 25, 26

The Hong Kong Polytechnic University

40

Jiangnan University

79, 80

Konkuk University

54, 56, 57, 58

National University of Athens

9

TU-Sofia

59

University of Delaware

46

University of Manchester

42

University of Maribor

42

University of Nottingham

47

University of Pisa

51

University of Zagreb

81, 84, 86, 89, 92, 95, 97, 99, 102, 105, 108, 109, 112, 114

Research index by country Country

Page

Belgium

27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37,

Bulgaria

59

China

11, 40, 79, 80

Croatia

81, 84, 86, 89, 92, 95, 97, 99, 102, 105, 108, 109, 112, 114

Greece

9

Hong Kong

40

Iran

6

Italy

51

Japan

48, 50

Korea, Republic

54, 56, 57, 58

Scotland

12, 13, 14, 15, 18, 19, 21, 22, 25, 26

Slovenia

42

Turkey

63, 64, 66, 67, 69, 70, 71, 72, 75, 77, 78

UK

42, 47

USA

46

Index by country

119

IJCST 21,6

120

Research index by subject Biotechnology Dynamic behaviour, Synthetic ropes Enzymes, Grafting, Functionalisation, Surface modification, Carpets Carpet wear, Evaluation, Image processing Colour Dyestuff selection, Nano-technology, Optimizing dyeing process, Purifying and decolouring wastewaters, Colour management, Fuzzy logic pH-sensitive dyes, Textiles, Spectroscopy, Microscopy Standard depths, Colour, ISO, Textiles, Inkjet printing Comfort Clothing, Fabric, Fabric mechanics, Behaviour, Comfort, Prediction Clothing thermal comfort, Dynamic sweating hotplate system, Thermal manikin system, Subjective wear trials Heat stress, Knitted garment, Medical textiles Microclimate cooling garment, Water cooling garment, Thermal manikin,

34

28

86

30 13 14

43 72

71

Plasma polymerization, Textile, Conductivity Electrospinning Electrospinning Process, Parameters, Polymer, Nanofibre application Micro-channel, Micro-porous, Membrane, Controlled delivery, Drug release Polymers, Nanofibre, Alignment, Yarn, Composite

69

21 23 19

Environmental Intelligent garment, 89 Thermal protection, Environment 46 Material conservation, Environmentally conscious manufacturing, Inherently benign materials and chemicals, Reuse Multifunctional eco-friendly textile 84 finishing, Polycarboxylic acids, Protective functionalities, Chemical modification of cellulose, Microvawe treatment of cellulose materials 102 Multifunctional eco-friendly textile finishing, Polycarboxylic acids, Wellness finishing of textiles, Determination of harmful substances on textiles, Toxicological and alergenic properties, Environmental protection, Hygiene and effects of textile care, Textile material sample preparation.

49

Composites Fiber assembly structure, Soft material, Maximum fiber volume fraction Fiber SEM image, Metaball, Image analysis, Orientation, Radial distribution Tranfer properties, Applications

50

Conductivity Fibre, Textile, Conductivity

75

Ethno heritage, tapestry, 58

114

Fibre and Yarn and properties Blended yarns, Hairiness, 63 Unevenness, Modal, Tencel, Promodal Fibre extrusion, Yarn production, 26 Weaving Hollow fibre, Water absorption, PP 31 HP materials, Textile fibers, 105 Textile design

Plasma polymerization, Textile, Flame retardancy, Durable-press, Fiber Wettability, Contact angle, Wetting time, Woven fabric, Image processing Yarn friction, Normal load, Contact area, Friction coefficient, Friction force

77 79 78

Finishing Enzyme, Textile, Finishing, Cellulose 66 Multifunctional eco-friendly textile finishing, Polycarboxylic acids, Toxicological and alergenic properties 102 Environmental protection, Hygiene and effects of textile care, Textile material sample preparation. Ultrasound, UV, Pre-treatment, 64 Finishing, Ozone Wellness finishing of textiles, Determination of harmful substances on textiles, Inkjet printing, textiles Medical textiles Drug Release Multifunctional eco-friendly textile finishing, Polycarboxylic acids, Wellness finishing of textiles, Determination of harmful substances on textiles, Toxicological and alergenic properties, Environmental protection, Hygiene and effects of textile care, Textile material sample preparation. Modelling Anthropometric measurements, Garment size system Body scan, Avatar generation, Mesh registration, Particle based method CFD, CAE, Protective fabric, Garment, Apparel, Modelling high performance Cloth simulation, Particle based method, Collision detection Garment, Garment simulation, E-learning

14

23 102

Garment pattern, 3D digital garment, Sewing information, Shape recognition, Shock graph Mathematical modelling of complex 2D/3D weaves. Mechanical modelling of 2D/3D woven composites. Micromechanical, Macromechanical, Finite Element Method, Simulation Modelling, Yarn, Fabric, Garment, High performance Sitting, Furniture, 3D model of workplace, Work environment, Virtual reality Textile composites, Unit cell analysis, TexGen

54

42 42 10 15 97 47

Nanotextiles Decubitus, Electrotherapy, Nanotechnology SMI nanoparticles, Imidisation, Surface properties MicroNanoStructured, fibre systems for Emergency-Disaster Wear Textiles, Biopolymer, Polymer, Nano-structuring Textiles, Flame retardants, Nanoclays Waterfiltration, Nanofibers, MBR

31 36

Networking Textiles, Networking, Innovation

29

27 34 51 33

Photovolteics, Thin film silicon, Solar energy

12 18 42 89

58

Protective fabrics, garments Ballistic body armour Intelligent garment, Thermal protection, Environment MicroNanoStructured, fibre systems for emergency-disaster wear Protective functionalities, Chemical modification of cellulose, Microwave treatment of cellulose materials Protective textiles, Multifunctionality, Smart textiles, Ceramic coatings Riot body armours for female officers. Riot helmet research Textiles, Flame retardants, Nanoclays

42 42 31

67

Shape Memory fabric, PTT

11

Sol-gel process

81

99 56 18

Index by subject

51 95

81

121

IJCST 21,6

122

Smart Interactive, Textiles, Garment, Clothing, Sensors, Wireless Smart textiles, Demonstrators, Textile antennas, Music, Stretchable electronics Smart textiles, E-textiles, Wearable micro systems, Wearable electronics Sensor, Fibers Stress Bagging Clothing, Fabric, Fabric mechanics, Behaviour, Comfort, Prediction Micro-controller Relaxation, torsional strain, Rotational level, Woven fabric Stress relaxation Wool fabric, Textile technology, Fabric testing, Deformation, Worsted fabric, Torsional force, Torsional strain, Data acquisition Technical textiles Advanced technical textiles, Knowledge propagation Fiber assembly structure, Function, End-use, Technical textiles, Nonwovens, Multifunctional eco-friendly textile finishing, Polycarboxylic acids, Microwave Protective functionalities, Chemical modification of cellulose, Microvawe treatment of cellulose materials Textile material, Interface phenomena, Surface modification and finishing, Multifunctionality

24 37 28 40 8 43 8 6 6

49 84

92

Yarns, Knitted and nonwoven fabrics, Structures, Properties, Thermal and water-vapour resistance properties Textile composites Knit, Composite, Impact, Reinforcement Unit cell analysis, TexGen Viticulture Fruit production, Nets, Hail, Plant material, Economic profit Wearable electronics Intelligent textiles, E-textiles, Wearable electronics, Breakthrough Sensor, Fibers Smart textiles, Demonstrators, Textile antennas, Music, Stretchable electronics Weaving 3D compoasites against impact 3-D weaving technology on conventional looms Weft preparation system, Air jet loom, Speed increase

110

70 47

112

28 38 40 37

42 42 35

Workwear Clothing technology, Numerical 108 methods, Optimisation, Reinforcements on Clothing 59 Cotton textiles, Weaving technology, Double fabric, Technology optimum, Work cloths

Research index by principal investigator Principal investigator Aksit, A.C.

Page 69, 75, 77

Index by principal investigator

Matsuo,T.

48, 50

Mijovic, B.

97

Andrassy, M.

105

Okur, A.

63, 72

Brooks, R.

47

Parac-Osterman, D.

Cao, H.

46

Pezel, E.

81

Chang, R.

46

Pickering, S.J.

47

Chen, X.

42

Provatidis, G.

9

Christie, R.M.

14

Rahnev, I.

59

Clifford, M.J.

47

Rogale, D.

89

Crancaric, A.M.

92

Rudd, C.D.

47

Scotchford, C.A.

47

86

DeClerk, K.

30, 31, 36

DeRossi, D.

51

Shaw, J.

46

Duran, K.

64

Skenderi, Z.

109

Fotheringham, A.

26

Soljacic, I.

102

Gersak, J.

42

Somodi, Z.

108

Hezavehi, E.

6

Starner, H.

Sul InHwan

54, 56, 57, 58

Stylios, G.K.

46 15, 18, 19, 21, 22, 25

Jihong, L.

79, 80

Sular, V.

78

Jones, I.A.

47

Tao, X.

40

Kallal, J.

46

Ujevic, D.

99

Kaplan, S.

72

Urljicak, Z.

112

Katovic, D.

84, 95

Vanfleteren

37

Kiekens, P.

29, 30, 32, 33, 34

Van Langenhove, L.

27, 28, 34, 35

Korlu, A.

66

Walker, G.S.

47

Kovacevic, S.

114

Wang, F.

11

Kurbak, A.

70, 71

Wardman, R.H.

13, 14

Leman

37

Warrior, N.A.

47

Long, A.C.

47

Wilson, J.I.B.

12

McCord, J.

46

Yesilpinar, S.

67

Mather, R.R.

12

123

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

Recommend Documents