1-Page View
2-Page View
Search
Table of Contents
Next
e-version
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
Heres how easy it is to use the e-version of the International Journal of Powder Metallurgy with these built-in navigation buttons Use this button to go to the previous page
Use these buttons to toggle between a 1-page view (shown below) and a view of 2 facing pages
Use this button to go to the next page
Use this button to go to the table of contents of this issue, from where you can go anywhere with a single click
Use this button to access the most powerful feature of the e-version of the Journal: the search capability. In some versions of the Adobe Reader, clicking this button will bring up the following window:
If this is the case, click on the arrow next to Find: and then click on Open Full Reader Search which will bring up the following window:
Type in the term you want to search for, click on the Search button, and the results will include every instance of the term in the current issue.
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
Previous
1-Page View
EDITORIAL REVIEW COMMITTEE P.W. Taubenblat, FAPMI, Chairman I.E. Anderson, FAPMI T. Ando S.G. Caldwell S.C. Deevi D. Dombrowski J.J. Dunkley Z. Fang B.L. Ferguson W. Frazier K. Kulkarni, FAPMI K.S. Kumar T.F. Murphy, FAPMI J.W. Newkirk P.D. Nurthen J.H. Perepezko P.K. Samal H.I. Sanderow, FAPMI D.W. Smith, FAPMI R. Tandon T.A. Tomlin D.T. Whychell, Sr., FAPMI M. Wright, PMT A. Zavaliangos INTERNATIONAL LIAISON COMMITTEE D. Whittaker (UK) Chairman V. Arnhold (Germany) E.C. Barba (Mexico) P. Beiss, FAPMI (Germany) C. Blais (Canada) P. Blanchard (France) G.F. Bocchini (Italy) F. Chagnon (Canada) C-L Chu (Taiwan) O. Coube (Europe) H. Danninger (Austria) U. Engström (Sweden) O. Grinder (Sweden) S. Guo (China) F-L Han (China) K.S. Hwang (Taiwan) Y.D. Kim (Korea) G. L’Espérance, FAPMI (Canada) H. Miura (Japan) C.B. Molins (Spain) R.L. Orban (Romania) T.L. Pecanha (Brazil) F. Petzoldt (Germany) S. Saritas (Turkey) G.B. Schaffer (Australia) L. Sigl (Austria) Y. Takeda (Japan) G.S. Upadhyaya (India) Publisher C. James Trombino, CAE
[email protected] Editor-in-Chief Alan Lawley, FAPMI
[email protected] Managing Editor James P. Adams
[email protected] Contributing Editor Peter K. Johnson
[email protected] Advertising Manager Jessica S. Tamasi
[email protected] Copy Editor Donni Magid
[email protected] Production Assistant Dora Schember
[email protected] President of APMI International Nicholas T. Mares
[email protected] Executive Director/CEO, APMI International C. James Trombino, CAE
[email protected]
2-Page View
Search
Table of Contents
Next
international journal of
powder metallurgy Contents 2 5 9 13 15
44/6 November/December 2008
Editor's Note PM Industry News in Review Company Profile CMW Inc. PMT Spotlight On …John Michiels Consultants’ Corner Howard I. Sanderow
ENGINEERING & TECHNOLOGY 19 Alloy Design and Microstructure of Advanced Permanent Magnets Using Rapid Solidification and Powder Processing I.E. Anderson, R.W. McCallum and W. Tang
RESEARCH & DEVELOPMENT 39 Steel-Sheet Fabrication by Tape Casting M. Rauscher, G. Besendörfer and A. Roosen
49 Transient Liquid-Phase Sintering of Copper–Nickel Powders: In Situ Neutron Diffraction D.M. Turriff, S.F. Corbin, L.M.D. Cranswick and M. Watson
60 61 62 64
DEPARTMENTS Meetings and Conferences PM Bookshelf Table of Contents: Volume 44, Numbers 1–6, 2008 Advertisers’ Index Cover: Lightly milled particulate flake. Photo courtesy Iver E. Anderson, FAPMI, Ames Laboratory, Iowa State University.
The International Journal of Powder Metallurgy (ISSN No. 0888-7462) is a professional publication serving the scientific and technological needs and interests of the powder metallurgist and the metal powder producing and consuming industries. Advertising carried in the Journal is selected so as to meet these needs and interests. Unrelated advertising cannot be accepted. Published bimonthly by APMI International, 105 College Road East, Princeton, N.J. 08540-6692 USA. Telephone (609) 4527700. Periodical postage paid at Princeton, New Jersey, and at additional mailing offices. Copyright © 2008 by APMI International. Subscription rates to non-members; USA, Canada and Mexico: $95.00 individuals, $220.00 institutions; overseas: additional $40.00 postage; single issues $50.00. Printed in USA by Cadmus Communications Corporation, P.O. Box 27367, Richmond, Virginia 23261-7367. Postmaster send address changes to the International Journal of Powder Metallurgy, 105 College Road East, Princeton, New Jersey 08540 USA USPS#267-120 ADVERTISING INFORMATION Jessica Tamasi, APMI International INTERNATIONAL 105 College Road East, Princeton, New Jersey 08540-6692 USA Tel: (609) 452-7700 • Fax: (609) 987-8523 • E-mail:
[email protected]
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
EDITOR’S NOTE
W
ith the first issue of the Journal in 2008, APMI International members gained online access to its contents. Identical to the printed version, the e-version of the Journal offers convenient navigational features coupled with a powerful search capability. The APMI Publications Committee recently surveyed the membership to assess initial reaction to the e-version. Based on more than 230 member responses, overall reaction and acceptance has been positive. More than 70% of the respondents are using the e-version and are experiencing benefits from the attendant search features. On a scale of 1 to 10, 76% of the respondents rated the value of the e-version above 5. A good start, reinforced by individual member comments. Contributing again to the “Consultants’ Corner,” Howard Sanderow homes in on four timely topics of concern to the readership: coating of compacting tools to enhance wear resistance; prospects for wider acceptance of hightemperature sintering in hydrogen; alternative alloying elements to nickel in PM steels to reduce cost without impairing static and dynamic mechanical properties; and the effect of shot peening on the mechanical performance of PM steels, keeping cost in mind. Peter Johnson’s profile on CMW, Inc., clearly identifies the company as a small, successful niche player manufacturing specialized products. Their diverse product mix focuses on electrical contacts, high-density PM tungsten alloys, and resistance welding consumables. Unlike other sectors of the PM industry, CMW’s business outlook for 2009 is positive. A major roadblock to success in electric-car-motor technology is the limited operating temperature—as the temperature increases, the magnets in the motor become weaker, resulting in a drop in power. In the “Engineering & Technology” section, Anderson, McCallum and Tang detail the development and processing of a new magnetic alloy that maintains its magnetic strength at temperatures approaching 200°C. Key to the new alloy’s performance is the replacement of some of the neodymium in the neodymium–iron–boron alloy with a mixed rare earth of yttrium and dysprosium. There are two contributions to the “Research & Development” section. Rauscher, Besendörfer and Roosen take a leaf from the ceramist’s recipe book and describe the application of tape casting to the fabrication of PM stainless steel sheet. A novel joining technique to produce interface-free laminates by sintering of the green tapes is also demonstrated. Turriff et al. utilize in situ neutron diffraction to probe the complexities of liquid-phase sintering in copper–nickel powders. The technique gives insight into the interdiffusion of the two elements and it is concluded that the transient liquid phase is removed primarily by rapid growth of a copper-rich solid solution with a composition dictated by the phase diagram.
Alan Lawley Editor-in-Chief
As in previous years, R&D Magazine’s Annual R&D 100 Awards underscore the importance of technology in our society. These awards recognize worldwide the most technologically significant innovations and products from academe, government, and industry. As a materials engineer, after perusing the awardees, I found the following to be of particular interest and importance: • Meltless production of particulate high-value alloys (e.g., titanium) by a low-cost electrolytic/metallothermic process (Materials Electrochemical Research Corporation, www.mercorp.com).
2
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Next
Table of Contents
EDITOR’S NOTE
• A high-temperature scrubber technology to remove mercury and other trace elements from gases produced by coal gasification utilizing a palladium sorbent instead of activated carbon (National Energy Technology Laboratory, www.netl.doe.gov). • A surface-hardening treatment for stainless steels to enhance protection against corrosion and wear, coupled with improved fatigue properties, without degrading ductility (Swagelok Co., www.swagelok.com). • High-rate, low-cost production of multi-wall carbon nanotube arrays for weaving into high-strength, lightweight fibers (nano-wool). The material can be integrated in products such as aluminum composite disks (heat sinks), resin-bonded diamond grinding wheels, and flexible heater prototypes (Oak Ridge National Laboratory, www.ornl.gov). The listing of awardees is populated predominately by the National Laboratories and relatively small technology companies. It is clear that innovation continues to thrive, notwithstanding the turbulence of the economy.
PURCHASER & PROCESSOR
Powder Metal Scrap (800) 313-9672 Since 1946
Ferrous & Non-Ferrous Metals Green, Sintered, Floor Sweeps, Furnace & Maintenance Scrap
1403 Fourth St. • Kalamazoo, MI 49048 • Tel: 269-342-0183 • Fax: 269-342-0185 Robert Lando E-mail:
[email protected] Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
3
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
International: powder injection molding. If you wish to produce complex ceramic and metal products using the PIM process, then come to the leading international specialists in this field: ARBURG. For you, we have the appropriate ALLROUNDER machine technology and the required know-how from our PIM laboratory. With our expertise, you will be able to manufacture efficiently and to the highest quality, prepare material, injection-mold components, debind and
ARBURG GmbH + Co KG Postfach 11 09 · 72286 Lossburg/Germany Tel.: +49 (0) 74 46 33-0 Fax: +49 (0) 74 46 33 33 65 e-mail:
[email protected] ARBURG, Inc. · 125 Rockwell Road · Newington, CT 06111 · Tel.: +1 (860) 667 6500 · Fax: +1 (860) 667 6522 · e-mail:
[email protected]
www.arburg.com
sinter - finished! You want to find out more about PIM processing? Simply talk to us!
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
PM INDUSTRY NEWS IN REVIEW The following items have appeared in PM Newsbytes since the previous issue of the Journal. To read a fuller treatment of any of these items, go to www.apmiinternational.org, login to the “Members Only” section, and click on “Expanded Stories from PM Newsbytes.”
Former Metaldyne Employees Face Prison Sentences Three ex-Metaldyne Corporation employees have pleaded guilty in Detroit federal court to conspiracy to sell stolen confidential and proprietary information to a competitor in China, Chongquing Huafu Industry Company, Ltd. Based on the results of the pre-sentencing investigation, a preliminary sentence hearing is scheduled for January 15, 2009, reports a court source.
facility in Hamburg, Germany, in September. Production has been transferred to other plants in Europe within the ECKA Granules Group. Sale Completed ALTANA AG, Wesel, Germany, has purchased the pigments business of United States Bronze Powders, Inc., excluding aluminum pigments. Completed on October 1, the sale includes the production and marketing of copper, bronze, and stainless steel pigments and metallic inks.
ESP Acquired Engineered Sinterings & Plastics Inc., Watertown, Conn., and its sister company Alves Precision Engineered Products Inc., have been acquired by Longroad Asset Management, LLC, Stamford, Conn., a private equity firm. The assets of both companies, acquired out of Chapter 11 bankruptcy protection on September 5, are now owned and operated by PM Engineered Solutions Inc.
Outstanding Technical Paper Award Winner The MPIF Technical Board announces that “Development of a Dual-Phase Precipitation-Hardening PM Stainless Steel” by Chris Schade and Tom Murphy, Hoeganaes Corporation, and Alan Lawley and Roger Doherty, Drexel University, is the recipient of the 2008 MPIF Outstanding Technical Paper Award.
Powder Coating Process Decreases Lubricant Need in PM Parts Particle Sciences Inc., Bethlehem, Pa., has received a patent (U.S. 7,419,527, increased density particle molding) for a powder particle coating process it says minimizes or eliminates the lubricant requirement in PM grade powders. The patent was issued on September 2, 2008.
SCM Metal Products Sold Platinum Equity, Los Angeles, Calif., has agreed to purchase SCM Metal Products, Inc., a nonferrous metal powder maker, from Gibraltar Industries, Inc., Buffalo, N.Y., for an undisclosed amount. Subject to regulatory approvals, the transaction should be completed during the fourth quarter of 2008.
Copper Powder Production Shifted ECKA Granulate MicroMet GmbH has closed its copper powder
Company Name Change Superwear Technologies, has changed its name to Ferro-Tic®, a
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
division of PSM Industries, Inc., Los Angeles, Calif. PSM founded Superwear in 2001 after purchasing Alloy Technology International, West Nyack, N.Y. H.C. Starck Invests and Restructures Over the next three years H.C. Starck, Goslar, Germany, expects to invest up to 50 million euros (about $67 million) annually to strengthen its position in growth markets such as tungsten, ceramic and metal powders, surface treatment technology, and ceramic and metal parts. The company says it will not achieve profitability targets for 2008 because of a significant slowdown in the chemical market and substantially higher prices for raw materials and energy. Ultrasonic Screening System Reduces Fine Powder Blinding Reading Alloys Inc., Robesonia, Pa., an Ametek company, reports success using the Russell Finex Vibrasonic 2000 screening system to produce very fine titanium and alloy powders for medical implant applications. Reading’s plasma spray customers require special particlesize distributions that produce very fine coatings to help join bone to the surfaces of implants. Uncertainty Clouds Positive Growth Powder maker Höganäs AB, Sweden, reports net sales increased 7.5 percent to MSEK 4,750 (aboutijpm $606
5
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
PM INDUSTRY NEWS IN REVIEW
million) for the first three quarters of 2008. After-tax income for the nine-month period surged 23 percent to MSEK 398 (about $51 million). However, cautions Höganäs, the financial turbulence that began last autumn has continued into 2008, rendering a highly uncertain future. New Powder-Atomization Unit Erasteel, Paris, France, is investing 17 million euros (about $21.5 million) in a new gas-atomization unit in Söderfors, Sweden. Scheduled to be onstream during the second half of 2010, the unit will double the company’s current atomizing capacity for highalloy powders. GKN Results Third-quarter sales in GKN’s automotive businesses, including powder metallurgy (PM), were slightly lower than the same period in 2007. While overall profits waned, the company’s PM business operated at a break-even level. Tungsten Supply Agreement North American Tungsten Corp. Ltd., Vancouver, BC, Canada, has signed a 12-month strategic supply agreement with Global Tungsten and Powders Corp (GTP), Towanda, Pa. Formerly OSRAM Sylvania’s Tungsten and Powders division, GTP is a division of The Plansee Group, Plansee, Austria. 2009 Conference Draws Heavy Technical-Paper Response The 2009 International Conference on Powder Metallurgy & Particulate Materials (June 28–July 1, Las Vegas) has attracted more than 200 technical paper submissions. The technical program committee meets on November 12 to select the final program. Swedish Powder Maker to Cut Staff Höganäs AB, Sweden, has announced discussions with trade unions to reduce its staff by approximately 160 employees in Sweden, Belgium, the United States, Brazil, and China. In its recent third-quarter financial report, the company lists 1,566 employees. New Nickel Capacity Ahead Vale Inco Ltd. will begin building a commercial hydrometallic processing plant in 2009 in Newfoundland, according to American Metal Market. The plant will process nickel ore mined from the company’s Voisey’s Bay deposit. ijpm
6
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
COMPANY PROFILE
CMW Inc. By Peter K. Johnson* Known for quality, responsiveness, metallurgical finance and manufacturing and saw an opportuniexpertise, application knowledge and helping cus- ty to participate in a company that produces basic tomers solve problems, CMW Inc., Indianapolis, industrial products,” he says. “Our products are Indiana, has been around since the 1920s. The used in many end markets such as aerospace, die company was originally part of P.R. Mallory & casting, automotive, medical, nuclear, defense, Company, then operated as Mallory Metallurgical and electrical. We sell in North America and export Company. “We have a long histo over 40 countries.” CMW’s tory of success,” says Mark B. customer base is diverse, includGramelspacher, president and ing the Department of Defense, CEO, Figure 1. “Our current prime defense contractors product mix includes three (Lockheed Martins, BAE, etc.), major segments—electrical conFortune 500 companies (General tacts, high-density PM tungsten Electric, Halliburton, Boeing, alloys, and resistance welding etc.), U.S. National Laboratories, consumables.” universities and research cenAs a niche player selling speters, NASA, precision machine cialized products, CMW comshops, and many small busipetes against much larger nesses. companies like The Plansee Gramelspacher also recogGroup, H.C. Starck, ATI Firth nized that CMW needed a new Sterling, and Aerojet-General Figure 1. Mark B. Gramelspacher, president focus on strategic planning and Corporation. new investment to meet the and CEO, holding a medical syringe shield Besides metallurgical prod- made from a high-density PM tungsten alloy challenges of the changing global ucts, Mallory was famous for marketplace. He served as chairdeveloping the alkaline dry-cell battery marketed man of the board of directors for five years until under the Duracell® trademark. During the 1970s joining the operation full-time in 2004 as president Mallory achieved annual sales of more than $367 and CEO. “One of my first actions was to reorganmillion and had over 11,000 employees. With the ize the company into three business units (electriexpansion of the Duracell brand, Mallory’s parent cal contacts, high-density metals, and resistance company was sold to Dart Industries. Selected welding consumables) and appointing sales manassets of Mallory Metallurgical Company were agers for each of the units,” he says. “By doing this acquired in 1978 by the divisional management we achieved better focus and better clarity to team that formed CMW Inc. Except for the space understand and meet customer needs.” occupied by CMW, much of the former 101,000 sq. He appointed Steve Jones sales manager for m (25-acre) Mallory manufacturing complex electrical contacts, Jeff Schemel sales manager for remains vacant, Figure 2. resistance welding consumables, and Jennifer In 1999, a holding company established and led Sniderman sales manager for high-density metals. by Gramelspacher, an attorney, purchased CMW Together they manage 12 full-time sales profesfrom the remaining shareholders who wanted to retire. “I have always been interested in corporate *Contributing editor and consultant
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
9
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
COMPANY PROFILE: CMW INC.
Figure 2. CMW is located on the former Mallory manufacturing complex
sionals. In addition, Jeff Schemel manages a large distribution network for the resistance welding consumables. Sniderman’s high-density PM segment is growing domestically and internationally. “Our growth markets include die casting tooling, aerospace and defense, balance weights, inertial guidance components, ballasts, and medical and industrial radiation shielding (syringe shields, collimators, nuclear reactors, geological equipment, cobalt teletherapy machines),” she reports. Gramelspacher is a results-first CEO. He flattened the organization by removing two layers of management and giving his staff the freedom and support to make decisions. Admittedly a non-technical person, he asks probing questions. “My style is hands-off but management by objective,” he says. The company’s 115 employees work in a 13,000 sq. m (140,000 sq. ft.) plant in the former Mallory complex, Figure 3. They produce wrought and PM products via different metal forming processes, Figure 4. Wrought production processes include extrusion, strip rolling and slitting, wire drawing, cold heading, milling and turning, and blanking and coining. Several four -axis CNC milling machines (Figure 5) were recently installed in addition to a new state-of-the-art nondestructive testing laboratory. And CMW’s in-plant shop supplies most of its tooling needs. Representing about 50 percent of CMW’s production, PM parts are fabricated on isostatic and hydraulic compacting presses, and sold as sintered, semi-finished, or finish machined. PM materials include tungsten, nickel, iron, and molybdenum. CMW relies on PM in each of its three major product lines. For example, PM is
10
Figure 3. CMW Indianapolis plant
Figure 4. CMW producing PM products
Figure 5. Four-axis CNC milling machine
used to make high-density tungsten alloys such as Anviloy® 1150 and CMW® 1000 tungsten, nickel, copper material. PM is also used to manufacture Elkonite® silver–tungsten, and Elkonite® copper–tungsten used for Thermkon® materials, Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
COMPANY PROFILE: CMW INC.
resistance welding electrodes, or electrical contacts. “The use of metal powders links these products together,” Sniderman says. PM process equipment includes a cold dry-bag isostatic press, 12 hydraulic compacting presses ranging from 363 to 908 mt (400 to 1,000 st), and 22 pusher and belt sintering furnaces, as well as three high-temperature furnaces capable of temperatures >2,000°C (3,632°F). The company has added three new furnaces this year, including a state-of-the-art high-temperature vacuum sintering furnace. While the bottom line drives his focus, Gramelspacher is revved up about revitalizing the east-side Indianapolis community where CMW is located. It’s a blighted neighborhood with high unemployment and many abandoned homes and buildings. He is collaborating with federal, state, and city officials to redevelop and help clean up
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
the community. About 70 percent of CMW’s employees live within several miles of the plant. Workforce training in basic skills is part of his plan. “We want to redefine quality of life as an organization and provide meaningful jobs while growing our business,” he says. Aiming products at a diverse market mix has protected CMW from the economic storms of 2008. “We are confident about the future, contrary to what you read in the press,” Gramelspacher says. “Our major strengths are our expertise and experience in PM, and we successfully leverage that expertise with modern advanced manufacturing processes.” The company’s Midwest location is another advantage as well as its ability to react quickly to customers’ needs and changes in the marketplace. “The business outlook is positive,” he states. “2009 looks like it will be our best year ever.” ijpm
11
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
SPOTLIGHT ON ...
JOHN MICHIELS, PMT Education: BS, Ceramic Engineering, Iowa State University, 1991 MS, Materials Science and Engineering, Iowa State University, 1995 MS, Metallurgical Engineering, Illinois Institute of Technology, 2006 Why did you study powder metallurgy/particulate materials? The basic PM processes are integral to ceramics (forming, sintering, or hot pressing powders), and beginning in ceramics gave me a head start. Additionally, this was the area in which employment was available. When did your interest in engineering/ science begin? At about 12 years old. I liked the creative aspects of engineering. What was your first job in PM? What did you do? I had an internship in 1990 at the Argonne National Laboratory developing methods of forming superconducting coils via extrusion and sintering. My first full-time employment was with Micron Display. I performed process development for the production of field emission displays. Describe your career path, companies worked for, and responsibilities. • My first employment in powder materials was the internship at the Argonne National Laboratory in the fall semester of 1990. I prepared yttrium barium copper oxide powders, and tested them via tape casting and sintering, and aided in their extrusion and sintering. • In 1992, I began my master’s degree work. The thesis involved the use of rare earth chalcogenides as hightemperature thermoelectric materials for power applications. Responsibilities included the creation of ther moelectric material samples, ther mal and
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
electrical testing, and arranging third-party analysis, as needed. • In 1996, I began work at Micron Display, Boise, Idaho. My title was faceplate engineer. I performed process development for the production of faceplates and packaging for field emission displays. This involved the mixing of powder-based screen printing, and extrusion pastes, sputtering, and photolithography processes. In addition, I developed two patents covering methods for patterning of electron emitter arrays. • In 1999, Micron Display was bought out by Pixtech. My duties as faceplate engineer remained the same. Pixtech folded in late summer 2001. • In 2002, I made a decision to go back to graduate school at the Illinois Institute of Technology. Thesis work focused on the laser cladding of Inconel onto steel and copper nickel onto steel. I designed experiments, and consulted on the process parameters for cladding, performed by Alion Science and Technology, St. Charles, Illinois. I then analyzed the samples for microstructures and mechanical properties. • In 2005, I began working at American Chemet Corporation as a powder materials engineer and performed product development. My current responsibilities include the creation and development of ideas for new products, experimental design and execution, and analyses of the viability of producing these products. Additionally, I am responsible for the production of customer samples of new products.
Powder Materials Engineer American Chemet Corporation 1 Smelter Road P.O. Box 1160 East Helena, Montana 59635 Phone: 406-227-5302 E-mail:
[email protected]
13
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
SPOTLIGHT ON ...JOHN MICHIELS, PMT
What gives you the most satisfaction in your career? The creative aspect of product development, and the ability to work with a talented and productive team.
14
Why did you choose to pursue PMT certification? To achieve a better understanding of the industry I work in, its products and processes. Additionally, to have it be known that I possess such an understanding. A stronger background helps the R&D process.
List your MPIF/APMI activities. Member of APMI since 2005; attended Basic PM Short Course in 2005; certified as a PMT (Level I) in 2006; attended the MPIF conference in San Diego, 2006; delivered a paper at the MPIF conference in Denver, 2007; delivered a paper at the MPIF world congress in Washington D.C., 2008.
How have you benefited from PMT certification in your career? Study and learning have been of significant benefit to me in understanding the purpose and functionality of the products I help develop. It also affords me the perception of competence in PM.
What major changes/trend(s) in the PM industry have you seen? I have not been in the industry too long, but I have observed a large number of novel processes and materials in the PM industry. These include nanotechnology, exotic alloys, and new uses for existing processes.
What are your current interests, hobbies, and activities outside of work? I have five children, six-years old and under (twins and triplets), which keeps me busy! In addition, I enjoy swimming, and try to do laps three times a week. I also enjoy science fiction, comedy books, and movies. ijpm
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
CONSULTANTS’ CORNER
HOWARD I. SANDEROW, FAPMI* Q A
What coatings have been successfully used on powder compacting tools to combat wear? A wide variety of coating systems and coating methods have been applied to PM compacting tools, typically the punches and core rods. As noted by Trueblood, Gay and Sanderow, 1 tool coating systems can be classified by the method of application. In this paper four systems were identified: plating, diffusion coating, physical vapor deposition (PVD), and lubricious coatings. Plated coatings are typically based on chromium or nickel with additives to the bath to increase hardness and adherence. Several commercial products are in use, some with a lubricious material infused into the coating, such as polytetrafluoroethylene (PTFE). The diffusion coatings are typically applied at high temperature and use specialized equipment such as salt baths or pack cementation systems to diffuse borides or carbides into the surface of the tool steel. These coatings tend to be very expensive to apply, require re-heat treatment after application but are very tenacious and long lasting. Such coatings have been used in stamping and forming tools as well as PM compaction tooling. The PVD coatings are the most widely used and least expensive, and do not require any postapplication treatment. The most popular are TiN, TiAlN, and CrN. Several well-established commercial sources are available with service centers throughout the industrial Midwest. The last group of coatings is applied by several methods and is typically classified in terms of their lubricious nature, not wear resistance. These coating systems include diamond-like coatings (DLCs), molybdenum, or tungsten disulfide and PTFE. Coating longevity tends to be the biggest difficulty facing these coating systems as the powder particles can abrade the coating from the surface. The Center for Powder Metallurgy Technology (CPMT) is now completing a two-phase project to
investigate powder-lubrication systems and tool coatings. That information is only available to CPMTmember companies. Will the future of PM rest on highertemperature sintering in hydrogen? If we limit the response to this question specifically to ferrous materials then the answer is No! A recent survey undertaken by the Technical Board of the Metal Powder Industries Federation (MPIF) confirms that those companies that have embraced high-temperature sintering and find it satisfies specific customer needs will continue to use this technology. But those who have not “seen the light” will remain on the sidelines and continue to use conventional sintering technologies. Unless driven by a unique customer requirement, it appears that the penetration of high-temperature sintering technology has reached an equilibrium position within the ferrous PM parts-making community. As production cost reductions dominate the thinking of most PM parts plants for the foreseeable future, the likelihood of a resurgence in the use of high-temperature sintering looks slim. This concern is further aggravated by the everincreasing cost of hydrogen in a competitive environment where the much-lower cost 90 v/o nitrogen–10 v/o hydrogen atmosphere for conventional sintering is the norm.
Q A
Q A
Nickel is often added to PM steels but is now very expensive. Is it possible to reduce or eliminate nickel without sacrificing mechanical performance (static or dynamic)? Are there any alternatives? The London Metal Exchange (LME) price for nickel topped $53/kg ($24/lb.) in mid-2007,
*President, Management & Engineering Technologies, MET Group Inc., 4337 Oaks Shadow Drive, New Albany, Ohio 43054; Phone: 614-775-0218, Fax: 614-775-0163; E-mail:
[email protected]
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
15
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
CONSULTANTS’ CORNER
but has now fallen precipitously to about $11/kg ($5/lb.), close to the historic levels of this decade. The rush to replace nickel in PM steel mixes was a short-lived aberration of the commodity markets. As most PM metallurgists are well aware, nickel is an excellent alloying additive for PM steels, especially for heat-treated steel products. Nickel additions in the 1–4 w/o range have been widely used in the industry for over 40 years due to the ease of sintering, no detrimental affect on powder compressibility, and its being the only additive that will cause a typical steel powder mix to shrink rather than grow during sintering. During the past two years we have seen a concerted effort to reduce or eliminate nickel from many steel powder mixes. Inco (now Vale Inco) introduced a finer nickel powder compared with the widely accepted 123 grade to help achieve the same sintered strength with less nickel added. The finer powder particles diffuse more readily into the iron matrix,
16
leaving fewer and smaller nickel-rich zones in the microstructure. Other approaches have seen a renewed interest in chromium or manganese steels due to their much lower historic cost as compared with nickel. However, both these alloying additives are sensitive to the sintering process and can oxidize in conventional sintering atmospheres when sintered at conventional temperatures. In addition, both additives reduce compressibility and require more costly heat-treat processes to achieve a carburized case-core microstructure. Parts producers faced the daunting costs of retooling existing parts and having their customer re-PPAP (production part approval process) the new material-process system when a powder mix with lower nickel content is introduced. Patience has been shown to be a virtue in this case as overall commodity prices are now returning to more “normal” levels.
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
CONSULTANTS’ CORNER
Q
Several investigations have shown that shot peening is a powerful process for improving the mechanical performance of PM steels. Is this an expensive process? How much can be achieved in mechanical performance? Could shot peening close the gap in performance relative to wrought steel (if shot peening is used on both materials)? Shot peening is widely used to improve the fatigue strength of wrought as well as PM steel products, most notably connecting rods, gears, and other highly loaded transmission components by increasing the residual compressive stresses at the surface of the part. The cost of the shot-peening process, as in most secondary operations, is highly dependent on whether the equipment is located in-house and performed as part of an inline manufacturing process or the parts are sent out to a service center for shot peening. Part size, quantity, and the parameters of the shot process itself further influence the process cost. Cost estimates are best developed by working directly with a shot-peening equipment manufacturer or a shotpeen service center, such as Metal Improvement Company, Paramus, New Jersey. Published reports vary as to the amount of improvement attributed to shot peening though most data show an increase of 20%–40% over non-shot-peened products. The recent publication by Ilia, Chernenkoff and Tutton 2 clearly demonstrated the marked improvement in fatigue life though shot peening.
A
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
In the case of connecting rods, shot peening has helped to demonstrate PM forged connecting rods to be superior to wrought steel rods processed in the same manner. In the case of transmission gears the answer is not as clear. Surface densification coupled with shot peening is still under development and testing by several research groups. CPMT has programs underway to investigate the influence of the shot-peen process on the performance of PM gear and sprocket materials. Progress reports have been provided to members of the Center. REFERENCES 1. M.E. Trueblood, D. Gay and H.I. Sanderow, “Effect of Tool Coatings and Die Lubrication Techniques on the Warm Compaction Response of Insulated Iron Powder”, Advances in Powder Metallurgy & Particulate Materials—1999, compiled by C.L. Rose and M.H. Thibodeau, Metal Powder Industries Federation, Princeton, NJ, 1999, vol. 1, pp. 2-17–2-32. 2. E. Ilia, R.A. Chernenkoff and K.T. Tutton, “Improvements in Fatigue Performance of Powder Forged Connecting Rods by Enhanced Shot Peening”, Advances in Powder Metallurgy & Particulate Materials—2008, compiled by R. Lawcock, A. Lawley and P.J. McGeehan, Metal Powder Industries Federation, Princeton, NJ, 2008, part 10, pp. 169–179.
Note: Both publications cited above are available from the Publications Department of MPIF at www.mpif.org. ijpm
Readers are invited to send in questions for future issues. Submit your questions to: Consultants’ Corner, APMI International, 105 College Road East, Princeton, NJ 08540-6692; Fax (609) 987-8523; E-mail:
[email protected]
17
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
2009 International Conference on Powder Metallurgy & Particulate Materials June 28–July 1, The Mirage Hotel, Las Vegas
• International Technical Program • Worldwide Trade Exhibition • Special Events
For complete program and registration information contact: INTERNATIONAL
METAL POWDER INDUSTRIES FEDERATION APMI INTERNATIONAL 105 College Road East Princeton, New Jersey 08540 USA Tel: 609-452-7700 Fax: 609-987-8523 www.mpif.org
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ENGINEERING & TECHNOLOGY
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING Iver E. Anderson, FAPMI,* R. William McCallum** and Wei Tang***
INTRODUCTION The ability to control magnetic fields and the forces they produce is a cornerstone of modern technology. Long before magnetic fields or magnetic materials were understood, the magnetic compass was a critical tool for early explorers. More recently, the demands of technology have grown significantly for both time-varying and time-independent magnetic fields. Time-independent fields may be generated either by a steady electric current or by a permanent magnet. Since a permanent magnet generates its field without an external source of electric power, it is attractive for many constant-field applications. However, permanentmagnet usage is constrained by the amount of work each magnet can produce1,2 and the cost and availability3 of the input materials, primarily the rare earth (RE) metals, needed for the strongest known magnets. All permanent-magnet systems function by using the energy stored in the atomic interactions within the magnetic material and the magnetic field that surrounds it. The permanent magnet is trapped in a high-energy state by barriers created in the structure of the magnet itself, and so long as the force exerted against those barriers is not sufficient to overcome them, the magnetic field is maintained without loss.1,2,4 The basis for a permanent magnet is a crystal structure with magnetic moments of the active (ferromagnetic) atoms aligned to produce a large ferromagnetic moment with a strongly preferred direction in relation to the crystal lattice. In a large single crystal of such a material, the magnetic moment will align itself in a series of randomly oriented magnetic domains so as to reduce the amount of energy in the external magnetic field created by the net moment of the crystal. To create a permanent magnet with a high external field, a means must be found to prevent the formation of multiple domains within each crystal by
Current Nd2Fe14B magnet alloys exhibit excellent roomtemperature magnetic properties and they are well suited for applications with operating temperatures ≤120°C, due in part to their low Curie temperature of ~310°C. The poor temperature stability of these rare earth (RE) permanent magnet alloys above 120°C limits their current performance in existing motors and their potential application in advanced drive-motor designs. Consequently, it is necessary to find other compositions to improve the thermal stability of RE2Fe14B magnets. A systematic study was conducted by melt spinning on the magnetic properties of a series of isotropic nanocrystalline magnet alloys where a yttrium (Y)+dysprosium (Dy) mixture replaced neodymium (Nd) or praseodymium (Pr) as the dominant RE constituent in mixed rare earth (MRE)2Fe14B (MRE = Y+Dy+Nd). The most recent results have shown that the Y+Dy–based MRE2Fe14B alloy can result in isotropic bonded magnets with superior magnetic properties in competitive commercial isotropic bonded magnets above ~30°C and well beyond 200°C by a judicious combination of Y, Dy, and Nd, along with a minor cobalt substitution for iron. Gasatomized powders of these advanced magnet alloys have also demonstrated improved magnetic strength over existing commercial spherical powders in a finepowder-size range that is suitable for injection molding of bonded isotropic magnets.
*Senior Metallurgist, Ames Laboratory, Iowa State University, Materials Science & Engineering Department, 222 Metals Development, Ames, Iowa 50011-3020; E-mail:
[email protected], **Senior Scientist, ***Post-Doctoral Associate, MEP Program, Ames Laboratory (USDOE), Iowa State University, 106 Wilhelm Hall, Ames, Iowa 50011-3020
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
19
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
suppressing domain-wall formation.2,4 Since the domain-wall energy is a surface energy, and the magnetic field is associated with the volume of the crystal, reducing the size of the crystal makes it harder to form a domain wall. In practice, if the size of the crystal is reduced to a few microns, domain walls will not form within the crystal. If the crystal has sharp corners, it is possible to nucleate a domain wall at the corners so the crystal should have smooth edges. Unfortunately, a crystal of a few microns in size produces a small magnetic field so it is necessary to utilize a collection of micron-sized crystals (i.e., a material with a fine-grain microstructure). If we want the fields from all the grains to add up to the maximum possible value, the grains must be perfectly aligned producing what is called an anisotropic magnet that can only be magnetized along one preferred direction. If we are willing to settle for half the maximum field, we can produce a random distribution of fine grains, resulting in an isotropic magnet that can be magnetized in any direction.2 Permanent magnetic fields are used to operate many useful devices, including motors, generators, rotating machines, and sensors. In fact, the average use of permanent magnets in domestic applications is well over 50 per household. Automotive applications include over 200 devices per car, with more than 30 DC electric motors used for starting engines, heating and air conditioning blowers, windshield wiper activation, window lifts, fuel pumps, and more. 5 Available hybrid-drive automobiles and experimental fuel cell–powered vehicles also are driving the development of compact and powerful DC traction motors that rely on RE permanent magnets (REPM). Also, a surprising amount of REPMs go into each new large wind-turbine generator unit that is installed, for example, 2,000 kg in a recent model.6 In general, the demand for permanent magnets, particularly for those with a high magnetic-energy density, is steadily increasing. Advanced Permanent-Magnet Materials There are two major alloy families of high-performance permanent magnets that derive their magnetocrystalline anisotropy (magnetic strength) from RE atoms in an ordered intermetallic crystal lattice. They are by far the most powerful permanent magnets available.1,4 The first REPM alloy family to be discovered is based on two inter-
20
metallic compounds of samarium (Sm) and cobalt (Co), SmCo 5 and Sm 2 Co 17 . Sm-Co magnets 4,7 supply the highest coercive fields (µ0HC = 3.5 T at ambient temperature) and the highest Curie point (Tc > 700°C) of any permanent magnets. Also, SmCo magnets have an extremely low temperature coefficient of coercivity, due in part to their high Tc, which makes them well suited for high-temperature applications. However, Sm-Co magnets suffer not only from the high cost of the RE component, Sm, but also from the high cost and price instability of Co. Thus Sm-Co magnets are limited to those applications where a high cost can be justified. The second major REPM alloy family is based on a single intermetallic compound, RE2Fe14B, where nearly all of the rare earth elements can form a stable version of this ordered intermetallic, Figure 1. At room temperature, fully dense magnets based on Nd2Fe14B have the highest maximum energy products, with a (BH)max > 55 MGOe in a well oriented (aligned) sintered magnet and are composed of constituents that are less costly than the Sm-Co magnets. Permanent magnets based on RE 2 Fe 14 B intermetallic compounds, mostly Nd2Fe14B, have had a significant technological impact5 in the nearly 30 years since their discovery8,9 and extensive research has been performed to develop and improve their magnetic
Figure 1. Crystalline unit cell of the general type of RE2Fe14B magnetic compound5,10—schematic
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
properties. Commercially, two classes, aligned sintered microcrystalline and isotropic bonded nanocrystalline, Nd 2Fe 14B magnets have been successfully developed. The aligned sintered microcrystalline magnets are processed by a casting, crushing, aligning/compacting, sintering, and annealing approach that is similar to that used for SmCo5 magnets.4,8 Final magnet shapes typically are machined from large hot-pressed blocks. Also, similar Nd2Fe14B alloys are processed into bonded isotropic magnets by rapid solidification to produce nanocrystalline particulate that is mixed with a polymer binder and molded into netshape magnets before magnetic alignment. 4,2 Because of their isotropic microstructure and the dilution of the magnetic material by the polymer binder, bonded Nd 2 Fe 14 B magnets exhibit a reduced energy product, but can have significant cost advantages for mass-production applications. Both types of current “2-14-1” magnets exhibit excellent room-temperature magnetic properties and are primary candidates for applications with an operating temperature ≤120°C; but their poor temperature stability above this temperature limits their range of use. Aligned Sintered Microcrystalline RE2Fe14B Magnets The highest known energy product for a REPM is obtained with Nd2Fe14B, which has a theoretical maximum of 64 MGOe for a fully dense array of perfectly aligned single-domain particles that are non-interacting.1,2 In order to obtain the highest possible energy product in an aligned sintered microcrystalline magnet, a number of conditions must be met. First, a high volume fraction of the 2-14-1 phase must be obtained in an assembly of ultrafine grains (1–3 µm), which should contain single magnetic domains. Second, a high degree of crystalline texture (alignment) must be achieved in the 2-14-1 phase grains. Third, the magnet must be free of soft magnetic secondary phases, such as α-iron, whose presence will drastically degrade the coercivity of the magnet. From a practical standpoint, there is a conflict between the first and third criteria that creates a processing challenge. From the phase diagram in Figure 2,11 it is clear that the casting of a large ingot of the stoichiometric 2-14-1 composition that solidifies according to the equilibrium peritectic reaction will result in extensive solute segregation (since iron will nucleate within the L+S region on coolVolume 44, Issue 6, 2008 International Journal of Powder Metallurgy
ing) and will require an unacceptably long heat treatment at high temperature (<1,180°C) to produce a homogeneous ingot. In practice, to avoid iron formation and attendant degradation of coercivity, compositions higher in neodymium than the peritectic reaction limit of 33 a/o Nd typically are chosen and strip casting is used to limit segregation.1,8 The neodymium-rich composition is also desirable for magnet consolidation after alignment/compaction, since the presence of some liquid phase (T > 655°C) results in a rapid sintering process that fuses the ultrafine grains metallurgically without time for grain growth. Processing: One of the most significant problems to overcome before RE2Fe14B magnets with enhanced maximum-energy product could be produced was the oxygen content of the RE alloy components. While there is a modest amount of oxygen solubility in the liquid alloy, there is almost none in the solidified phase and any residual oxygen reacts during solidification with the neodymium in the remaining liquid to form Nd2O3, which effectively reduces the neodymium content in an uncontrolled manner and wastes this valuable component. As a result of the demands of the magnet industry, the oxygen content of commercial neodymium metal has been reduced by an order of magnitude in the last 20 years.4 In order to approach the stoichiometric phase composition, the solidification rate of the casting process also has been enhanced either by the use of “thin strip” casting, producing chill cast
Figure 2. Vertical section through the ternary Nd-Fe-B phase diagram11 that starts at pure iron and extends through the Nd2Fe14B compound labeled “1”
21
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
strips of 0.2–0.5 mm in thickness,12,13 or by the use of relatively coarse gas-atomized powders of 0.2–1 mm dia.14 These reduced casting sizes can promote a sufficiently high cooling rate and even modest undercooling (into the L + Nd2Fe14B phase field) to allow the starting neodymium content to closely approach the peritectic limit (33 a/o Nd), enhancing the volume fraction of Nd2Fe14B phase and the resulting energy product in the finished magnets. After preparation of the chill-cast ingot, the processing route for RE2Fe14B microcrystalline aligned magnets is based on the approach developed previously for SmCo5 magnets, namely, liquid-phase sintering, with reference to the ternary Nd-Fe-B phase diagram in Figure 2. Typically, the following procedure is used8 to fabricate sintered magnets of the prototypical Nd2Fe14B compound: (i) chill-cast alloy forms are subject to hydrogen absorption and incomplete desorption to promote initial fracture and particulate decrepitation; (ii) the particles are milled to an average particle size ~3–6 mm in an inert atmosphere, producing single-grain particles; (iii) the tap-densified singlegrain particles are oriented magnetically in a 10–30 kOe field and compacted (under the applied field) by uniaxial die pressing in a direction perpendicular to the alignment direction at a pressure of 200 MPa, producing a highly textured powder compact; (iv) the compact is liquid-phase sintered between 1,090°C and ~1,155°C, under an inert argon atmosphere for ~1 h or less, followed by rapid cooling; (v) a solid-state anneal may be performed for several hours at a temperature approaching but <655°C, the eutectic melting temperature of the neodymium-rich phase to relieve cooling stresses and to help suppress reverse domain nucleation. With reference to Figure 2 as an approximate guide (recognizing that most common compositions are slightly lower in boron than this pseudobinary section), a final sintered magnet microstructure contains the Nd 2 Fe 14 B phase, termed “1,” along with the Nd 1.1 Fe 4 B 4 phase, termed “4,” and the neodymium-rich solid-solution phase, termed Nd. The ideal 2-14-1 magnet of the sintered aligned microcrystalline type would consist of small, uniform single-crystal grains of Nd2Fe14B, all with their easy axis of magnetization perfectly aligned in the direction of the applied field, providing the maximum possible remanence.1,2,4 By minimizing the initial size of
22
the as-milled particles (preferably down to 1–3 µm), the chance of survival of multi-grain 2-14-1 particles is reduced significantly, eliminating internal grain boundaries as one type of active site for reverse domain nucleation to help increase coercivity. As the phase diagram in Figure 2 confirms, liquid-phase sintering at >1,090°C (the Nd2Fe14B peritectic temperature) and subsequent continuous cooling are preferred to minimize the as-solidified content of Nd1.1Fe4B4 phase, which can reduce the final content of 214-1, the sole permanent magnet phase. The neodymium-rich solid solution, which solidifies as the grain boundary-wetting phase from the liquidphase-sintering process, serves to “smooth” the grain boundaries, eliminating more high-activity grain-boundary sites (re-entrants and sharp corners) for reverse domain nucleation. Another effect of the complete wetting of all of the 2-14-1 grain boundaries is to isolate each grain with a non-magnetic phase, predominantly a neodymium-rich solid solution, preventing the spreading of any reverse domains that manage to nucleate. These combined extrinsic characteristics produce an enhanced coercivity in the smooth rounded grains of 2-14-1 with inherently high intrinsic magnetocrystalline anisotropy. The pursuit of these ideal characteristics gives rise to the choices of conventional alloying additions to sintered 2-14-1 magnets to aid in the liquid-phase-sintering process. Minor additions of aluminum, copper, and gallium15,16 are made to Nd2Fe14B to improve wetting of the grains, thereby promoting a complete distribution of the liquid during liquid-phase sintering. Both niobium and vanadium17 are added to promote the formation of small, isolated boride particles on the grain boundaries that stabilize the small grain size during liquid-phase sintering. Also, dysprosium is substituted for neodymium (or praseodymium) to enrich the near-grain-boundary regions to further reduce reverse domain nucleation.18 As a substitute for iron, cobalt is added to raise the Curie temperature (for increased operating temperature) of the resulting 2-14-1 magnetic phase and to form isolated Nd 3 Co particles 19 on the grain boundaries, again to stabilize a small grain size. Because of these alloying efforts and the development of an optimized processing sequence, aligned sintered microcrystalline 2-14-1 magnets based on neodymium have set an extremely high standard for room-temperature magnetic strength Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
and are unsurpassed for applications with an operating temperature ≤120°C. Development of High-Temperature 2-14-1 Permanent Magnets Our recent efforts to increase the high-temperature tolerance of RE2Fe14B magnet alloys toward 200°C are based on substitution for the neodymium with a majority of yttrium and dysprosium, to produce a mixed rare earth (MRE) content of about (Nd0.45, Y0.37, Dy0.18)2Fe14B as one prototype composition that is shifted toward “heavy” RE dominance. The preference for heavy RE elements in the 2-14-1 phase to promote high-temperature permanent-magnetic performance is discussed subsequently. This unique alloying strategy20 makes the assumption that any mixture of RE elements will be tolerated as a solid solution in the 2-14-1 phase, which appears to be supported since all RE elements will form the same 2-14-1 phase.21 However, a major effect of this shift in 2-14-1 magnet alloy composition to MRE 2Fe 14B with neodymium in a minority (by atomic fraction) is a shift in the equilibrium ternary phase diagram from the common type in Figure 2 for Nd-Fe-B to the alternate type, shown for example in Figure 3 for Dy-Fe-B.18 A key difference for the heavy RE metal phase diagrams in the composition region of the 2-14-1 phase (Figure 3) is that the RE2Fe17 intermetallic compound (“2-17”) is the primary phase of highest
Figure 3. Vertical section through the ternary Dy-Fe-B phase diagram18 that starts at pure iron and extends through the Dy2Fe14B compound, labeled “1” line compound
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
stability immediately above the stoichiometric 214-1 compound, instead of iron (Figure 2). Thus, in contrast to the unavoidable formation of iron primary phase (due to rapid nucleation kinetics) at the 2-14-1 composition in the “light” RE (neodymium-rich) phase diagram (Figure 2), a solidification reaction at the 2-14-1 composition in Dy-Fe-B offers the likelihood that a chill-cast ingot can sustain some undercooling18 and will contain only a minor content of primary 2-17 phase, a complex intermetallic with reduced nucleation kinetics, and a majority of 2-14-1 phase. Analysis of the Dy-Fe-B phase diagram (recognizing that most common compositions are slightly lower in boron than this pseudo-binary section) also indicates that it is advantageous to select an alloy composition that is slightly rich in dysprosium to avoid iron formation, which would increase the volume fraction of 2-14-1 phase above that possible in Nd-Fe-B alloys. In terms of sintered aligned microcrystalline magnets, the Dy-Fe-Be phase diagram18 (Figure 3) also shows that it is apparently impossible to completely dissolve any 2-17 (soft magnetic phase) that forms on solidification by high-temperature liquid-phase sintering (similar to step (iii) for Nd-Fe-B) between 1,190°C and ~1,250°C, since the 2-17 phase is within a three-phase equilibrium phase field with L and 2-14-1 phases. However, annealing within a lower-temperature three-phase field that contains DyFe3, for example, may allow liquid-phase sintering with a dysprosium-enriched liquid and 2-17 dissolution, but would dilute unavoidably the 2-14-1 magnetic microstructure with some non-magnetic DyFe3 for slightly dysprosium-rich alloys. Thus, there is no access to a two-phase field with only 2-14-1 and alloy liquid to accomplish the same intrinsic liquid-phase-sintering step8 as in the neodymiumbase REPM alloys for sintered magnets. In spite of the mixed benefits and challenges of the alternative phase equilibria, the potential to pursue hightemperature per for mance benefits in a MRE2Fe14B magnet alloy design with a dominant heavy RE content, provides sufficient motivation for ongoing R&D on aligned sintered microcrystalline 2-14-1 magnets. In order to develop the formulation for improved high-temperature magnetic properties of MRE2Fe14B compounds, the individual intrinsic 214-1 properties of all of the RE types were analyzed and studied. Table I lists the annotated
23
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
TABLE I. SATURATION MAGNETIZATIONS (MS), ANISOTROPY FIELD ENERGY (HA), QUANTUM ORBITAL ANGULAR MOMENTUM NUMBERS (L), AND CURIE TEMPERATURES (TC) FOR THREE 2-14-1 COMPOUNDS5,19, 22–24 Compound
4πMs (kG)
Ha (kOe)
L
Tc (K)
Nd2Fe14B Dy2Fe14B Y2Fe14B
16.0 7.1 14.1
73 ~150 26
6 5 0
585 598 565
properties of the three most promising RE2Fe14B compounds. 5,19,22–24 Of the three compounds, Nd2Fe14B has the highest saturation magnetization Ms, while Dy2Fe14B has the largest anisotropy field Ha. For a solid solution of RE2Fe14B, both the magnitude and temperature dependence of the saturation magnetization (M s), and anisotropy energy (Ha), are expected to follow a rule of mixtures relation in MRE2Fe14B. Thus, it appeared promising to combine all three of these RE elements with their outstanding individual magnetic attributes into a single MRE version of a 2-14-1 magnet alloy. While it has been a standard practice to add minor amounts of dysprosium to neodymium in the Nd-Fe-B magnet alloys, as cited above, 18 this attempt to make neodymium the minority RE element in favor of a mixture of yttrium and dysprosium had not been done before. Figure 4 shows the temperature dependence of anisotropy field energy, H a (T), and magnetic moment, M s(T), of these three RE 2Fe 14B compounds. The anisotropy energy as a function of temperature, Ha(T), of RE2Fe14B for neodymium and dysprosium 24,25 arises from the large RE
moment with an orbital angular moment (L ≠ 0), which results in large crystal field splittings, Figure 4(a). However, Ha for neodymium and dysprosium has a strong temperature dependence and declines monotonically with increasing temperature. In contrast, a smaller Ha at room temperature for Y2Fe14B from the iron lattice exhibits a weak temperature dependence, which initially rises above room temperature before decreasing as the Curie temperature Tc is approached. Thus, the inclusion of yttrium in solid solution with neodymium and dysprosium in the MRE2Fe14B phase may decrease Ha, but should improve the temperature dependence of Ha. The behavior of Ms(T) for Dy2Fe14B (Figure 4(b)) is similar to that of Ha(T) for Y2Fe14B.22,23 This is the result of the anti-ferromagnetic dysprosium–iron coupling that has a widely noted effect on the low Ms of Dy2Fe14B at room temperature. However, as the temperature is increased the anti-ferromagnetic interaction decreases, resulting in an increased Ms for Dy2Fe14B as the temperature is raised above room temperature. The Ms for Dy2Fe14B decreases again as T approaches Tc. Thus, a MRE alloy like (Y1.0Dy1.0)2Fe14B could combine the intrinsic property advantages of Y2Fe14B and Dy2Fe14B, which may result in compensation for the usual loss of Ms and Hci due to heating for a 2-14-1 magnet alloy dominated by Nd2Fe14B. In other words, it is presumed possible to obtain simultaneously a smaller (negative) temperature coefficient for Br and Hci.20 Nanocrystalline RE2Fe14B Magnets In contrast to the limited design space (primari-
Figure 4. Temperature dependence of (a) anisotropy field energy Ha(T) and (b) magnetic moment Ms(T) of three R2Fe14B compounds (After R. Grössinger,19 S. Hirosawa,22 E.B. Boltich23 and D. Givord24)
24
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
Figure 5. Isothermal section at 1,000°C of the general RE-Fe-B phase diagram of the type shown in Figure 2 for Nd-Fe-B, which is accompanied by schematic microstructures for the demagnetized state of the three types of nanocrystalline permanent magnets that can be developed within the three composition ranges indicated11
ly optimization of one type of microstructure/ property relationship) of microcrystalline RE permanent magnets that are well equilibrated, the ability to produce amorphous phases and nanocrystalline grains in RE2Fe14B type magnets by rapid-solidification processing offers several choices. The opportunity is presented by rapid solidification of nanocrystalline magnets to tailor the microstructure and magnetic properties by selection of the base ternary alloy, relative to the stoichiometric 2-14-1 composition, and by minor or major alloying additions, working with highly non-equilibrium processing. The three basic types of nanocrystalline permanent magnets possible in the light RE versions (typically neodymium-rich) of this 2-14-1 alloy system are shown schematically on the right hand side of Figure 5,11 including RE-rich (decoupled nanograins), stoichiometric (single-phase exchange-coupled nanograins), and iron-rich (dual-phase exchangecoupled nanograins). The processing of nanocrystalline RE2Fe14B magnets based on light RE alloys is considered subsequently, focusing primarily on stoichiometric alloys, in ter ms of typical microstructures, processing approaches, and magnetic characteristics. Recent experimental developments in high-temperature MRE 2Fe 14B magnet alloys dominated by heavy RE metals will follow with a review of minor alloying effects and gas-atomized-powder processing efforts. Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Stoichiometric (Single-Phase, ExchangeCoupled) Nanocrystalline Magnets Exchange-coupled nanocrystalline magnets were first reported for melt-spun ribbon of stoichiometric Nd2Fe14B. As the composition of the alloy approaches stoichiometry, it is possible to produce a uniform fine grain (~200 nm) structure with clean grain boundaries, Figure 5. In this case the same interaction that causes the atomic moments to align within a grain will cause magnetization within each grain to rotate away from the easy axis of the grain and to align with neighboring grains. Increases in Mr of the order of 20% have been observed due to this effect and most current melt-spun magnet materials exhibit some degree of remanence enhancement due to interaction between grains. The key to producing exchange-enhanced materials is in the control of the microstructure.4,11 If the coupling between grains (dictated by the grain-boundary area) approaches the magnetocrystalline anisotropy energy of the grains (determined by the volume of the grains), the effect is to average the anisotropy over the coupled grains. For randomly oriented grains this average rapidly approaches zero. In contrast, the exchange coupling affects only those atomic moments within half of a domain-wall width of the grain boundary. As the grain size increases, the volume of the sample within this distance of a grain boundary decreases rapidly so the effect disappears. Thus it is important to create grains in Nd2Fe14B that are ~200 nm across (approximately equiaxed) with a narrow distribution in size. (Note: this critical nanocrystalline grain size can change in different 2-14-1 alloys.) As might be expected, in an interacting system, the interaction enhances the properties of the magnet until the first part of the sample starts to reverse. At this point the interaction causes an avalanche effect resulting in a rapid reversal of the remaining parts of the sample. As a result, the coercivity is determined by the weakest part of the sample rather than the average. Given the stringent nanostructured requirements for stoichiometric exchange-enhanced magnets, a large number of additives have been investigated as grain refining agents. These include refractory metals, their carbides and nitrides, as well as other metallic elements.26 Processing Processing of the nanocrystalline 2-14-1 mag-
25
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
nets of all three basic microstructure types of both light and heavy RE magnet alloy families generally commences with melt spinning to generate the rapidly solidified microstructures. In this process a prealloyed casting is melted inductively under an inert atmosphere in a sealed quartz or alumina tube with a small (1–2 mm dia.) precision-machined orifice. Generally, the surface tension of the melt is sufficiently high to prevent dripping during the rapid melting and superheating cycle. On a laboratory scale, the melt-spinning crucible is located within a larger chamber that is maintained with an inert atmosphere at the same pressure as the crucible during melting. When the molten alloy reaches the designated temperature, an overpressure of the inert gas is introduced into the sealed crucible, forcing the molten alloy out of the orifice and onto the polished outer surface of a rotating metal wheel (surface velocity = 10 to 30 m/s), typically copper. The molten alloy is quenched on the contact surface of the wheel, cooling by conduction, and is transported along the surface, before being flung off the wheel. In free flight it finishes cooling by convection in the stagnant chamber atmosphere on both the wheel and free sides of the resulting ribbon, typically 25–50 µm thick and a few mm wide. Since the liquid film is subject to radial heat extraction from the quenching wheel surface, a preferred solidification-morphology alignment can be established that mirrors this unidirectional heat flow over a significant portion of the ribbon thickness, depending on the wheel surface residence time of the solidifying alloy and whether or not crystallization is permitted to occur.4 The solidification process for melt spun ribbons of Nd-Fe-B (and other light RE elements) can be forced by increased wheel speed to be sufficiently rapid to suppress crystallization of the primary iron phase for iron contents >77 a/o (the peritectic limit in Figure 2), and even for stoichiometric 214-1 phase compositions, if special alloying is employed, for example, with titanium and carbon. 26 In fact, the most effective commercial method for the production of nanocrystalline 2-141 magnets of either RE-rich or stoichiometric compositions is to produce ribbons with a significant amorphous fraction and to promote additional nanocrystalline 2-14-1 grains to crystallize during a subsequent anneal at 700°C–750°C. This method has been found to remove the effects of any variabilities in the melt spinning process,
26
such as trapped porosity on the wheel/ribbon interface, and the magnetic properties that may arise if the nanocrystalline structure is accessed directly from the rapid solidification process.5 In addition, there is a beneficial effect of reduced oxidation and corrosion from an increase in the iron content and a reduction of the RE content in rapidly solidified alloys, especially if a stoichiometric, single-phase alloy can be used. It should be noted that this same “overquenching” melt-spinning approach and ribbon-annealing process at 700°C–750°C is also effective for the RE-Fe-B magnet alloys based on heavy RE elements.20 However, the RE2Fe17 phase (instead of iron) is the soft magnetic phase that is bypassed to form a glassy alloy,18 as represented in the phase diagram of Figure 3. Actually, if the heavy RE–dominated 214-1 magnet alloys are doped with titanium and carbon, ribbon annealing temperatures of 800°C can be used without degrading the near-ideal nanocrystalline grain size, as seen in Figure 6. For light RE versions of 2-14-1 magnet alloys, melt spinning and annealing to promote full crystallization are followed by ribbon chopping or crushing in an inert-atmosphere vessel to produce a flake product that is commercially termed Magnequench “MQP” and is used in bonded isotropic magnets.27 In fact, isotropic nanocrystalline polymer-bonded magnets (Figure 7) can be made from both light RE and heavy RE particulate of 2-14-1 magnet alloys that have been rapidly solidified and annealed. The use of bonded magnets is growing rapidly as mass production of net-shape isotropic 2-14-1 magnets has become commercially attractive for many consumer and industrial applications. Close-coupled gas atomization is an alternative
Figure 6. Melt-spun ribbon of {[Nd0.45 (Y2Dy)0.55]Fe12.5Co1.5B}+Ti0.02+ C0.02 after an 800°C anneal for 15 min, in which the initial ribbon was primarily amorphous
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
rapid-solidification process to melt spinning that also has been utilized in the experimental production of isotropic nanocrystalline RE-Fe-B magnet alloy particulate,28 targeted for bonded magnet fabrication. Observations of the solidification product phases and segregation patterns in 2-141 magnet alloys indicate that the average solidification rate of gas-atomized powders can reach that of melt-spun ribbons (for a moderate wheel speed of 15m/s), if the particle size is reduced below ~15–20 µm.29 It should be noted that atomization with helium gas has a tenfold advantage in convective cooling rate over argon-gas atomization and this pushes the equivalent microstructure of powders to larger sizes (~20–30 µm), compared with ribbons melt spun at 15 m/s. The principal differences between melt spinning and gas atomization are related to the type and directionality of heat transfer and the tendency for melt undercooling. As noted previously, unidirectional conductive cooling dominates initially in melt spinning, followed by mild convective cooling. In contrast, non-directional forced convective gas cooling operates continuously in gas atomization and the outer rim of the spherical droplets is quenched first; it serves to inhibit heat extraction from the particle interior which can solidify at a reduced rate. This difference can be manifested in differences in the continuity and directionality of the resulting solidification microstructure (Figure 8) in spherical gas-atomized particles, which can mirror the isotropic heat flow pattern, especially if cellular or dendritic solidification occurs. 30 At gas-atomized droplet sizes smaller than ~5 µm the RE-Fe-B magnet alloys tend to solidify as an amorphous phase due to high undercooling. Alternatively, melt spinning has a minimal ten-
Figure 7. Schematic cross section transverse to the pressing direction27 of a polymer-bonded or compression-molded permanent magnet made from melt-spun ribbon flake particulate
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
dency for undercooling, and crystallization or glass formation usually occur immediately upon contact of the melt with the metallic wheel surface. However, at high wheel speeds (>25 m/s) the wheel surface results in a highly effective quench that can extend an amorphous phase; for example, throughout the thickness of a 30 µm ribbon of a magnet alloy. Interestingly, gas-atomized droplets ≤5 µm can also experience extreme undercooling, on the order of 0.22Tm or greater,31,32 resulting in rapid adiabatic solidification that can produce amorphous particles. Unfortunately, the dimensions of the gas atomized particles that exhibit this extremely rapid solidification effect are typically about five times smaller than the melt-spun ribbon thickness for common magnet alloys28 and this ultrafine powder has a much higher surface area that is susceptible to oxidation. In spite of the reduced rate of solidification in gas-atomized powder of increased size, the spherical shape of the powder can have important advantages in the molding of bonded magnets. Thus, a second generation of magnet alloy designs have been developed to improve the “quenchability,” or glass-forming tendency of Nd-Fe-B magnet alloys permitting gas atomization,33,34 and other atomization processes35 to produce spherical amorphous powders that can be annealed to achieve improved magnetic properties. These spherical magnet powders are also directly suited to bonded magnet compounding, without any chopping or milling. Bonded Nanocrystalline RE2Fe14B Magnets Isotropic nanocrystalline RE-Fe-B polymer bonded permanent magnets are a fast growing
Figure 8. Gas-atomized spherical powder particle of (Y1.0Dy1.0)2Fe14B from a 10–20 µm-size fraction in the as-atomized condition30
27
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
market segment, especially in mass-production applications. At this time, the chopped and/or milled form of 2-14-1 melt-spun ribbons (after annealing) is almost universally employed as the magnetic “filler” in a polymer matrix. These nanocrystalline, single-phase, exchange-coupled magnet materials are typically of stoichiometric 214-1 composition and use rapid-solidification processing to generate their isotropically oriented nanograin structure, as described previously, and illustrated in Figure 7. The high (BH)max values (compared with ferrites) and relatively low RE content of these isotropic RE-Fe-B materials makes them ideally suited for loading into bonded magnets, since this magnet form will intrinsically have diluted magnetic strength and is sensitive to the raw-material cost. Bonded magnets are processed by blending or “compounding” the fragmented magnetic material with a polymer (e.g., nylon powder or an epoxy resin solution), followed by injection molding or compression molding, respectively, into a finished magnet shape. Unlike conventional metal injection molding (MIM) of high-per for mance powder metallurgy (PM) parts,34 the polymer binder is retained in the part and serves as a barrier to corrosion or oxidation of the resulting bonded magnets. To some extent, the polymer binder softening temperature serves as an upper limit to the application temperature of bonded magnets, especially for nylon. Efforts to extend the upper temperature limit and high-temperature strength of bonded magnets has led to development efforts with polyphenyl sulfide (PPS) polymers. However, high compounding temperatures (~300°C) and a narrow “working” viscosity temperature range (~5°C–10°C) have been a barrier to progress. Initial efforts to improve the MREFe-B compositions for high-temperature use were described in the section on sintered magnets and will be described more specifically below for nanocrystalline 2-14-1 magnets. The difficulty of high-temperature compounding is related to powder-particle oxidation and corrosion in the blending atmosphere and polymer medium, especially with corrosive additives. Recent efforts to provide oxidation-resistant surface coatings on RE-Fe-B powders may help address particulate surface degradation (oxidation) during powder handling, PPS compounding,30,37 and long-term use. In general, net-shaped bonded magnets avoid expensive sawing or grinding operations and the associated waste “swarf” material. Also, compared
28
with sintered aligned microcrystalline RE-Fe-B magnets, the bonded isotropic magnets are less brittle (reduced breakage losses during handling), have low electrical conductivity, and are often molded into a singular magnet assembly that can be magnetized into a multipole configuration, all advantageous for electric motor applications. The major disadvantage to bonded magnets is the reduction in their magnetic properties, compared with the properties of fully dense (e.g., hotpressed) isotropic nanocrystalline RE-Fe-B magnets, their direct, undiluted counterpart.27 For purposes of comparison of laboratory measurements, an unchopped melt-spun ribbon sample of the same nanocrystalline structure may also serve as a standard, if it has the same isotropic crystallinity. The reduction of magnetic strength can be estimated with the so-called filling factor (f), which reflects the volume fraction of the magnetic powder in the bonded magnet according to the relation: [(BH)max]bonded = (f)2 [(BH)max]isotropic
(1)
To illustrate the importance of maximizing the filling factor, a simple analysis based on equation (1) shows that bonded magnets with loadings that range from 50 v/o to 71 v/o can have a (BH)max that ranges from 25% to 50% of the fully dense value, respectively. In terms of magnetic hysteresis loops, the consequences of bonded magnet dilution are illustrated in Figure 9, which shows the results for isotropic nanocrystalline Pr2Fe14B in polymer bonded and isotropic melt-spun rib-
Figure 9. Hysteresis loops of a polymer-bonded Pr2Fe14B magnet1 Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
bon forms.1 The use of anisotropic magnet powder can be effective for boosting the strength of bonded magnets beyond that predicted by the dilution effect, if the increased processing cost is balanced by the magnetic property benefits. For example, in one approach to accomplish this, hot-pressed and forged magnets made from ribbon flake were milled into anisotropic powder and molded under a magnetic field to produce anisotropic nanocrystalline bonded magnets with an increased energy product,38 but the high processing cost made the resulting magnets non-competitive in the marketplace. An alternative processing approach to anisotropic bonded magnets involving hydrogen decrepitation has been more successful commercially, but has a restricted elevated-temperature range.39 To maximize the filling factor and magnetic strength for a given type of isotropic nanocrystalline bonded magnet, the objective is to make the powder as small and round as possible, much
Figure 10. Effect of particle shape on viscosity vs. solids loading for glass particles36 (a)
like the maximum ideal powder filling in a compound for powder injection molding (PIM).36 This optimum powder filling will minimize the volume fraction of polymer, while still maintaining adequate flow characteristics for molding the compound and sufficient particle bonding for good mechanical properties. From the fundamentals of filled-polymer processing, it is well known that a nonspherical particle shape is detrimental to the viscosity of a powder–binder mixture because of the lower inherent packing density and higher interparticle friction. As illustrated in Figure 10, the infinite viscosity limit is reached at far lower solids loadings for nonspherical particles,36 where chopped magnet alloy flakes could be approximated as a disk shape to relate to this illustration, Figure 11(a). Industrially, injection molded bonded magnets from flake particulate that has been lightly milled, similar to Figure 11(a), can be produced with a volumetric loading limit ~60%. Additional milling of the flake particulate can raise this loading limit to ~70%, in some commercial bonded magnets. Alternatively, spherical gasatomized powders (Figure 11(b)) have been investigated as a promising source of nanocrystalline magnet particulate, especially well suited for bonded magnets.28,30,37 In theory, an ideal dense random packing arrangement of monosized spheres can reach a limit ~63.7% filling, or slightly lower for the optimal molding limit. As a consequence of highly tailored size distributions that provide effective interparticle void filling, spherical-powder packing densities up to 93% can be reached, implying over 90% loading for molding purposes;36 for example, 4–5 size modes and 7:1 dia. ratios. With far more processing simplicity, a (b)
Figure 11. Representative micrographs of (a) flake and (b) gas-atomized powder. SEM
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
29
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
collection of spheres with a broad particle-size distribution, resembling a close-coupled gasatomized powder yield, can reach loadings in excess of 80%.36 In comparison with flake particulate, high-pressure gas (helium or argon)-atomized powders of similar magnet alloys have a near-ideal spherical shape (Figure 11(b)) and a fine-particle-size distribution with a 65% yield <20 µm dia.29,30 The increased loadings via the use of a spherical-powder-size distribution will enhance the effective (BH) max of the resulting isotropic bonded magnets as a consequence of equation (1), if suitable control can be exercised over the nanocrystalline magnetic structure. Alternative Isotropic Nanocrystalline MRE2Fe14B Magnets The goal of the current research is to develop alloys with low-temperature coefficients that may be prepared by gas atomization, particularly for bonded magnet molding. As explained previously, the spherical powder produced by gas atomization is a much more desirable form for MIM than the flake produced by melt spinning. This advantage is tempered by the reduced convective cooling rates and isotropic heat extraction associated with rapid solidification of gas-atomized droplets. Recently, we have conducted a systematic study on the magnetic properties of a series of alloys in which an yttrium/dysprosium mixture replaced neodymium or praseodymium as the primary RE constituent in MRE 2 Fe 14 B (MRE = Y+Dy+Nd). Ribbons and powder were fabricated by melt spinning and gas-atomization techniques.20,30,40 Our results show that the YDy-based MRE2Fe14B alloy can be processed into isotropic nanocrystalline bonded magnets from melt-spun particulate with superior magnetic properties above 115°C by a judicious mixture of yttrium, dysprosium, and neodymium. In addition, a combination of zirconium substitution and ZrC addition has been found to result in adequate microstructural control in both the gas-atomization and melt-spinning techniques, and a new MRE2(Fe, Co)14B alloy with these modifications was developed. Effect of Neodymium Substitution It is known that Nd2Fe14B has the maximum Ms of the three RE2Fe14B compounds, Table I. If yttrium and dysprosium in the (Y1-zDyz)2.2Fe14B ribbons are substituted for by the correct amount of neodymium, a substantial improvement in
30
(BH)max may be obtained without severely degrading high-temperature behavior. (BH)max as a function of temperature up to 300°C is shown in Figure12.20 It is seen that the samples with higher neodymium content exhibit a higher room temperature (BH)max, but their (BH)max also exhibits a strong temperature dependence. Figure 13 shows α and β for annealed [Nd x(YDy) 0.5(1-x)] 2.2Fe 14B ribbons from 27°C to 127°C, as a function of neodymium content. The value for α increases monotonically from -0.045% to -0.106%/°C with increasing x from 0 to 0.8, while β has an essentially constant value of -0.3%/°C when x is below 0.4.20 When x is increased from 0.4 to 0.8, β increases up to -0.38%/°C. These thermal magnetic properties over a wide range of neodymium levels are much better than those of neodymium-based ribbons. Effect of Cobalt Substitution The results cited show that high-performance magnets with superior temperature stability can
Figure 12. (BH)max as a function of temperature for annealed [Ndx(YDy)0.5(1-x)]2.2Fe14B ribbons with different neodymium contents4
Figure 13. Temperature coefficient of coercivity and remanence for annealed ribbons of [Ndx(YDy)0.5(1-x)]2.2Fe14B from 27°C to 127°C as a function of x4 Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
be developed by adjusting the ratio of yttrium to dysprosium and adding the correct amount of neodymium. If higher-temperature performance is required, ther mal stability may be further improved by partial substitution of cobalt for iron in the YDy-based R2Fe14B alloys. Table II lists the magnetic properties of annealed [Nd0.5(YDy)0.25]2.2 Fe14-yCoyB ribbons with different cobalt-substitution levels. The temperature coefficients in the table are in the temperature range of 27°C to 127°C. With increasing cobalt content, the Curie temperature Tc increases rapidly, resulting in a corresponding decrease in α. The (BH)max first increases and then decreases slightly. These trends in the change in magnetic properties as a function of cobalt content are consistent with those in Nd 2 Fe 14-y Co y B pseudo-ternary compounds. 19,41 In Nd 2Fe 14-yCo yB pseudo-ternary compounds, the maximum Ms is achieved when y is about 1.5. Although the Ha of Nd2Fe14-yCoyB compound decreases with cobalt substitution, it is noted from Table II that Hcj declines slightly with increasing cobalt content.40 It indicates that the detrimental effect of cobalt substitution on Ha in YDy-based compounds is less than in the neodymium-based compound. Increasing the cobalt content also leads to an increase in β. For comparison, Figure 14 shows (BH)max as a function of temperature for [Nd x (YDy) 0.5(1x)] 2.2Fe 14-yCo yB ribbons with different compositions. The highest room-temperature (BH)max is obtained in sample A with x = 0.8 and y = 0, but (BH)max exhibits a stronger temperature dependence. When x = 0.4, y = 0, the temperature dependence of (BH)max of sample B is improved although, the room-temperature (BH) max is decreased. If cobalt at y = 1.5 is substituted for iron with x = 0.5 (sample C), both room-temperature and high-temperature (BH) max are improved.40 Thus, at or above 150°C, (BH)max of
Figure 14. (BH)max as a function of temperature for annealed [Ndx(YDy)0.5(1-x)]2.2Fe14-yCoyB ribbon with different neodymium and cobalt contents4
sample C is greater than that of sample A in the YDy-based [Ndx(YDy)0.5(1-x)]2.2Fe14-yCoyB ribbons. Effect of Zirconium Substitution Gas atomization is capable of producing spherical rapidly solidified magnet powders. However, the solidification rate using this process is significantly lower than that characteristic of melt spinning. In order to develop magnet alloys for gas atomization, the effect of zirconium substitution on microstructure and magnetic properties in [Nd0.5(YDy)0.25]2.2-xZrxCo1.5Fe12.5B (x = 0-0.7) ribbons melt spun at a low wheel speed (10 m/s) has been systematically studied. The results show that zirconium substitution for RE can significantly refine the grain size in as-spun [Nd 0.5 (YDy) 0.25] 2.2-xZr xCo 1.5Fe 12.5B ribbon. Figure 15 shows the effect of zirconium substitution on Hc
TABLE II. MAGNETIC PROPERTIES OF ANNEALED [Nd0.5(YDy)0.25]2.2 FE14-YCoYB RIBBON: EFFECT OF COBALT SUBSTITUTION4 y
Br (kG)
Hcj (kOe)
(BH)max (MGOe)
-α (%/°C)
-β (%/°C)
Tc (°C)
0 1.0 1.2 1.5 2.5 3.0
6.0 6.1 6.2 6.6 6.6 6.6
22.6 21.6 21.3 19.8 17.2 16.3
7.9 7.9 8.4 10.0 9.6 9.6
0.09 0.08 0.08 0.07 0.05 0.03
0.37 0.39 0.38 0.39 0.40 0.43
309 374 387 405 455 487
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Figure 15. Effect of zirconium substitution on Hc and (BH)max for asspun [Nd0.5(YDy)0.25]2.2-xZrxCo1.5Fe12.5B ribbon4
31
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
and (BH) max for as-spun [Nd 0.5 (YDy) 0.25 ] 2.240 When x = 0.4, a fine xZrxCo1.5Fe12.5B ribbon. and uniform 2-14-1 type microstructure is formed with an average grain size of 65 nm. The as-spun ribbon has an Hcj value of 10.6 kOe and a (BH)max of 9.6 MGOe, and exhibits enhanced high-temperature magnetic properties in the temperature range of 150°C to 300°C. Larger substitution of zirconium for RE further refines the grain size but also leads to the appearance of the 2-17 phase, which severely degrades the magnetic properties. Gas Atomized Powder Based on these alloy design considerations for rapid solidification by melt spinning, an improved alloy composition [Nd 0.45 (Y 0.66 Dy 0.33 ) 2.2 Zr0.1Co1Fe13B](1-2x)/17.3+Zr0.01C0.01 (termed GA-1114) was developed for gas atomization.42 Ingots of this composition were prepared by plasma arc melting in an argon atmosphere. The ingot was induction heated to 1,550°C and atomized by high-pressure argon gas at a pressure of 5.5 MPa. Powder was sieved into the major ASTM size fractions. As-atomized powders were annealed in argon over the temperature range 700°C–750°C for 15 min. X-ray diffraction (XRD) was performed with Cu Kα radiation on a Philips X-ray diffractometer. Differential thermal analysis (DTA) was performed with a Perkin-Elmer DTA at a heating rate of 10°C/min. Hysteresis loop measurements were performed using a Quantum Design MPMS SQUID magnetometer with a maximum field of 5.0 T. High-temperature hysteresis loops and magnetization vs. temperature curves were obtained utilizing a vibrating sample magnetometer (VSM) with fields 9T and 1T, respectively. Microstructural analysis was carried out using
a JEOL JSM-5910LV scanning electron microscope (SEM). The sieved results for 100 g of powder are given in Table III. The weight of powder <38 µm was 79.4 g, approximately 81.4 w/o of the total powder. A typical result for magnetization as a function of temperature for as-atomized and annealed powder with a particle size of 20–25 µm is shown in Figure 16.42 Analysis reveals the existence of two ferromagnetic ordering transformations at ~280°C and 400°C for the as-atomized powder. These two transformations coincide with the Curie transformations of the 2-17 and 2-14-1 phases, respectively. After annealing at 700°C for 15 min., the M vs. T curve exhibits only the 2-141 phase transition at 410°C, confirming that the phase composition has evolved from a majority 214-1 plus a small amount of 2-17 phases to a single 2-14-1 phase (modified by a cobalt addition to raise the Curie temperature) during the annealing process. Figure 17 shows the X-ray diffraction patterns for powder samples as-atomized and annealed at 700°C for 15 min.42 Both XRD patterns appear to contain reflections mainly from the crystallized 214-1 structure and do not indicate the presence of the 2-17 phase before annealing; however, this evidence does not rule out a small volume fraction of the 2-17 phase. In addition, DTA measurements (Figure 18) of the as-atomized powder show a crystallization (exothermic) transformation around 650°C, indicating that some amorphous phase exists in the powder. Figures 19(a) and 19(b) show typical SEM microstructures of as-
TABLE III. SIEVE RESULTS FOR AS-ATOMIZED POWDER42
32
Sieve Size (um)
Sieve Weight (g)
Sieve Amount (%)
106 90 75 63 53 45 38 32 25 20 <20
0.043 1.187 2.459 4.193 5.76 4.5229 9.587 17.775 25.882 12.545 13.584
0.04 1.22 2.52 4.3 5.91 4.64 9.83 18.22 26.54 12.86 13.93
Figure 16. Magnetization as a function of temperature for as-atomized and annealed powders, 20–25 µm42
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
atomized and annealed powder, respectively, with a particle size of 20–25 µm. It is seen from Figure 19(a) that the as-atomized powder consists predominately of uniform equiaxed grains with an average grain size of 1.5 µm. The amorphous phase is not easily detected by XRD or SEM. After the powder is annealed at 700°C for 15 min, the grains grow slightly and the grain boundaries become narrower, Figure 19(b). By combining the results of the microstructural observations with those of XRD and DTA, it is concluded that the partial substitutions of zirconium and the ZrC additions result in the formation of a uniform, fine-scale microstructure. 42 After annealing at 700°C for 15 min, the powder is fully crystallized, and the phase structure is transformed to the single 2-14-1 phase from the 2-14-1 and 2-17 phases (trace). These microstructure and phase-structure changes are critical to the final magnetic properties.
The magnetic properties of the annealed powder from each size fraction are shown in Table IV.42 It is seen that (BH)max first increases with increasing powder size from 5 µm, reaches the peak value at the size range 20–25 µm, and then decreases with further increasing powder size. It is clear that fine powder (<5 µm), with a high surface area, is readily oxidized, resulting in degraded magnetic properties. The highest energy product is observed when the particle sizes are 20–25 µm. With larger particle sizes, a coarser microstructure is observed, which weakens the reversal field strength, reducing the magnetic properties. Typical demagnetization curves of as-atomized and annealed 20–25 µm sizes powders are shown in Figure 20.42 The curves for the as-atomized powder exhibit a feature consistent with the amorphous phase or magnetically soft phase, which verifies the results revealed by DTA and M vs. T, respectively. Annealing at 700°C for 15 min results in a smoother demagnetization curve and increased “squareness.” The temperature depend-
Figure 17. XRD patterns for as-atomized and annealed powders, 20–25 µm42
Figure 18. DTA heating trace for as-atomized powder, 20–25 µm42
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Figure 19. Representative microstructure of powder, 20–25 µm, (a) as-atomized and (b) annealed at 700°C for 15 min.42 SEM
33
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
TABLE IV. MAGNETIC PROPERTIES OF ANNEALED POWDER AS A FUNCTION OF SIZE FRACTION42 Sieve Size (µm)
Br (kG)
Hcj (kOe)
(BH)max (MGOe)
45-53 38-45 32-38 25-32 20-25 -20 10-15 10-5 -5
4.7 5 5.6 6.3 6.4 6.5 6.3 6.3 5.8
4.7 6 7 9.6 10.5 11.2 10 10.7 10.7
4.8 5.6 7.1 9.0 9.6 10.1 9.8 9.1 7.7
ence of the magnetic properties of the annealed powder was also determined, Figure 21. The results confirm that the temperature coefficients of Br and Hcj are 0.084%/°C and 0.4%/°C in the temperature range of 27°C to 100°C, respectively.
Figure 20. Hysteresis loops for annealed powder, 20–25 µm42
Figure 21. Magnetic energy product (BH)max as a function of temperature for annealed powder <20 µm42
34
(BH)max values at room temperature and 200°C are 9.6 and 5.6 MGOe, respectively. Based on a knowledge of the magnetic behavior of the 2-14-1 compound, it is likely that the ability to produce a fine particulate, in which each particle is a highly aligned cellular (nanocrystalline) structure or a single crystal (microcrystalline), can result in a large gain (four times) in magnetic properties, if the ensemble of particulate can be aligned in a bonding matrix to form a bonded or (extrinsically) sintered magnet.1,2 These concepts are being pursued by focusing on the particulate-production processes as a first step, using a rapid-solidification process as our first choice. To enable the maximum energy product to be realized in high-temperature MRE-Fe-B alloys, development of sintered (full-density) aligned permanent magnets also will be continued. The approach will embrace magnet alloy design and processing of micron-sized (single-crystal) particulate and the development of an extrinsic liquidphase-sintering constituent to lock in the aligned magnetic-particle assembly. SUMMARY A systematic study was conducted by melt spinning to enhance the magnetic properties and to reduce the high-temperature degradation of a series of isotropic nanocrystalline magnet alloys where a Y+Dy mixture replaces neodymium or praseodymium as the dominant RE constituent in MRE2Fe14B (MRE = Y+Dy+Nd). The most recent results have shown that the Y+Dy based MRE2Fe14B alloy can result in isotropic bonded magnets with superior magnetic properties compared with competitive commercial isotropic bonded magnets above about 30°C and well beyond 200°C by a judicious combination of yttrium, dysprosium, and neodymium, along with a minor cobalt substitution for iron. A combination of zirconium substitution and ZrC addition also enhances quenchability and yields a preferred microstructure in MRE 2-14-1 base alloys. In a wide particle-size range, gas-atomized powder exhibits desirable magnetic properties at room and high temperature, superior to commercial spherical magnet alloy powders in the temperature range of 27°C to 277°C. (BH)max values of 9.6 and 5.6 MGOe for gas-atomized 20–25 µm-size powder were obtained at room temperature and 200°C, respectively. The gas-atomized powders with superior magnetic strength were also in a Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
fine-powder-size range that is more suitable for MIM of bonded isotropic magnets than commercial spherical powder. A major remaining barrier to high-performance application of these new high-temperature nanocrystalline MRE 2 Fe 14 B magnet alloys is the development of an aligned magnetic structure in consolidated form (either bonded or sintered) that permits the high level of magnetic strength of aligned sintered microcrystalline magnets to be approached. ACKNOWLEDGEMENTS The authors gratefully acknowledge sustained support from the USDOE Of fice of Energy Ef ficiency and Renewable Energy (EERE), FreedomCar and Vehicle Technologies (FCVT) Program. The work was performed in the facilities of the Department of Energy’s Ames Laboratory at Iowa State University. We also acknowledge the support of the DOE Office of Science (OS), Basic Energy Sciences (BES), Materials Sciences Division through contract no. DE-AC0207CH11358. The research efforts of Kevin Dennis, Yaqaio Wu, Matt Kramer, and Nathaniel Oster are also appreciated, along with the staff of the Materials Preparation Center for production of the gas-atomized powders.
10.
11.
12.
13.
14.
15.
16.
REFERENCES 1. D. Goll and H. Kronmuller, "High-Performance Permanent Magnets", Naturwissenschaften, 2000, vol. 87, pp. 423–438. 2. D. Jiles, Introduction to Magnetism and Magnetic Materials, Second Edition, 1998, Chapman and Hall, New York, NY. 3. Y. Kanazawa and M. Kamitani, "Rare Earth Minerals and Resources in the World", J. Alloys and Compounds, 2006, vol. 408–412, pp. 1,339–1,342. 4. I.E. Anderson, W. Tang and R.W. McCallum, "Particulate Processing and Processing of High Per for mance Permanent Magnets", Int. J. of Powder Met., 2004, vol. 40, no. 6, pp. 37–60. 5. J. F. Herbst, "Rare Earth(R) Iron Boron (R2Fe14B) Materials: Intrinsic Properties and Technological Aspects", Reviews of Modern Physics, 1991, vol. 63, pp. 819–898. 6. G. Hatch, presentation at Magnetics 2008, available at DexterMagnetics.com. 7. K.J. Strnat and R.M.W. Strnat, "Rare Earth-Cobalt Permanent Magnets", Journal of Magnetism and Magnetic Materials, 1991, vol. 100, pp. 38–56. 8. M. Sagawa, S. Fujimura, H. Yamamoto, Y. Matsuura and K. Hiraga, "Permanent Magnet Materials Based on the Rare Earth-Iron-Boron Tetragonal Compounds", IEEE T ransactions on Magnetics, 1984, vol. MAG-20, pp. 1,584–1,589. 9. J.J. Croat, J.F. Herbst, R.W. Lee and F.E. Pinkerton,
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
17.
18.
19.
20.
21.
22.
23.
"High-Energy Product Neodymium-Iron-Boron Permanent Magnets", Applied Physics Letters, 1984, vol. 44, pp. 148–149. K.H.J. Buschow, D.B. De Mooij, J.L.C. Daams and H.M. Van Noort, "Phase Relationships, Magnetic and Crystallographic Properties of Neodymium-Iron-Boron Alloys", Journal of the Less-Common Metals, 1986, vol. 115, pp. 357–366. G. Schneider, E.T. Henig, G. Petzow and H.H. Stadelmaier, "Phase Relations in the System Iron-Neodymium-Boron", Z. Metallkunde, 1986, vol. 77, pp. 755–761. Y. Hirose, H. Hasegawa, S. Sasaki and M. Sagawa, "Microstructure of Strip Cast Alloys for High Performance NdFeB Magnets", Rare-Earth Magnets and Their Applications, Proc. International Workshop on Rare-Earth Magnets and Their Applications, edited by L. Schultz and K-L. Mueller, Werkstof f-Infor mationsgesellschaft, Frankfurt, Germany, 1998, vol. 1, pp. 77–86. W-L. Pei, F-Z. Lian, G-Q. Zhou and M. Fu, "Sintered NdFe-B Magnet Produced by the Strip Casting Technology", Dongbei Daxue Xuebao, Ziran Kexueban, 2003, vol. 24, pp. 64–67. K.S.V.L. Narasimhan and E.J. Dulis, "Rare-EarthElement-Containing Permanent Magnets", application for U.S. Patent, Crucible Material Corp., U.S., 1986. K.G. Knoch, B. Grieb, E.T. Henig, H. Kronmueller and G. Petzow, "Upgraded Neodymium-Iron-Boron-AD (AD = aluminum, gallium) Magnets: Wettability and Microstructure", IEEE Transactions on Magnetics, 1990, vol. 26, pp. 1,951–1,953. O.M. Ragg and I.R. Harris, "A Study of the Effects of Heat T reatment on the Microstructures and Magnetic Properties of Cu-added Nd-Fe-B Type Sintered Magnets", Journal of Alloys and Compounds, 1994, vol. 209, pp. 125–134. X. Song, Y. Yao, Z. Chen, Q. Huang and X. Wang, "Effect of Refractory Elements (Nb, Mo, V) on the Microstructure and Coercivity of NdFeB-Based Magnets", Proc. 2nd Int. Symp. Phys. Magn. Mater., International Academic, Beijing, China, 1992, vol. 2, pp. 742–745. B. Grieb, E-T. Henig, G. Schneider and G. Petzow, "Phase Relations in the Systems Fe-Dy-B and Fe-Tb-B", Z. Metallkde., 1989, vol. 80, pp. 95–100. R. Groessinger, R. Krewenka, X.K. Sun, R. Eibler, H.R. Kirchmayr and K.H.J. Buschow, "Magnetic Phase Transitions and Magnetic Anisotropy in Neodymium Iron Cobalt Boride (Nd2Fe14-xCoxB) Compounds", Journal of the Less-Common Metals, 1986, vol. 124, pp. 165–172. W. Tang, K.W. Dennis, Y.Q. Wu, M.J. Kramer, I.E. Anderson and R.W. McCallum, "Studies of New YDyBased R2Fe14B Magnets for High Temperature Performance (R=Y+Dy+Nd)", IEEE Trans. Magn., 2004, vol. 40, pp. 2,907–2,909. K.H.J. Buschow, "Magnetism and Processing of Permanent Magnet Materials", Handbook of Magnetic Materials, edited by K.H.J. Buschow, Elsevier Amsterdam, Netherlands, 1997, vol. 10, pp. 463–593. S. Hirosawa, Y. Matsuura, H. Yamamoto, S. Fujimura, M. Sagawa and H. Yamauchi, "Magnetization and Magnetic Anisotropy of R2Fe14B Measured on Single Crystals", Journal of Applied Physics, 1986, vol. 59, pp. 873–879. E.B. Boltich, E. Oswald, M.Q. Huang, S. Hirosawa, W.E.
35
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
24.
25.
26.
27. 28.
29.
30.
31.
32.
33.
34.
35.
36
Wallace and E. Burzo, "Magnetic Characteristics of Rare Earth-Iron-Boron (R2Fe14B) Systems Prepared with High Purity Rare Earths (R = Ce, Pr, Dy, and Er)", Journal of Applied Physics, 1985, vol. 57, pp. 4,106–4,108. D. Givord, H.S. Li and J.M. Moreau, "Magnetic Properties and Crystal Structure of Neodymium-Iron-Boron (Nd2Fe14B)", Solid State Communications, 1984, vol. 50, pp. 497–499. M. Boge, J.M.D. Coey, G. Czjzek, D. Givord, C. Jeandey, H.S. Li and J.L. Oddou, "The 3d-4f Magnetic Interactions and Crystalline Electric Field in the R2Fe14B Compounds: Magnetization Measurements and Mossbauer Study of Gadolinium Iron Boride (Gd2Fe14B)", Solid State Communications, 1985, vol. 55, pp. 295–298. D.J. Branagan and R.W. McCallum, "Solubility of Ti with C in the Nd2Fe14B System and Controlled Carbide Precipitation", Journal of Alloys and Compounds, 1995, vol. 218, pp. 143–148. R.W. Lee, "Hot-Pressed Neodymium-Iron-Boron Magnets", Applied Physics Letters, 1985, vol. 46, pp. 790–791. I.E. Anderson, B.K. Lograsso and R.W. McCallum, "High Pressure Gas Atomization of Rare Earth-Iron Alloy Per manent Magnet Powders", First Inter national Conference on Processing Materials for Properties, TMS, Warrendale, PA, 1993, pp. 645-650. N.L. Buelow, I.E. Anderson, R.W. McCallum, M.J. Kramer, W. Tang and K.W. Dennis, "Comparison of Mixed Rare Earth Iron Boride Gas Atomized Powders to Melt Spun Ribbon for Bonded Isotropic Permanent Magnets", Advances in Powder Metallurgy and Particulate Materials— 2004, compiled by R.A. Chernenkoff and W.B. James, Metal Powder Industries Federation, Princeton, NJ, 2004, part 10, pp. 230–243. P.K. Sokolowski, I.E. Anderson, W. Tang, Y.Q. Wu, K.W. Dennis, M.J. Kramer and R.W. McCallum, “Microstructural and Magnetic Studies of Gas Atomized Powder and Melt Spun Ribbon for Improved MRE2Fe14B”, Advances in Powder Metallurgy and Particulate Materials— 2006, compiled by W.R. Gasbarre and J.W. von Arx, Metal Powder Industries Federation, Princeton, NJ, 2006, part 9, pp. 152–167. A.L. Genau, I.E. Anderson and R. Trivedi, “Microstructure Selection During Rapid Solidification of Al-Si Powder", Advances in Powder Metallurgy and Particulate Materials— 2004, compiled by R.A. Chernenkoff and W.B. James, Metal Powder Industries Federation, Princeton, NJ, 2004, part 10, pp. 244–253. I.E. Anderson and M.P. Kemppainen, "Undercooling Effects in Gas Atomized Powders", Undercooled Alloy Phases, edited by E.W. Collings and C.C. Koch, TMSAIME, Warrendale, PA, 1987, pp. 269–288. D.J. Branagan, T.A. Hyde, C.H. Sellers and R.W. McCallum, "Developing Rare Earth Permanent Magnet Alloys for Gas Atomization", Journal of Physics D: Applied Physics, 1996, vol. 29, pp. 2,376–2,385. M.J. Kramer, Y. Xu, K.W. Dennis, I.E. Anderson and R.W. McCallum, "Development of Improved Powder for Bonded Permanent Magnets", IEEE Transactions on Magnetics, 2003, vol. 39, pp. 2,971–2,973. C.O. Bounds, B.M. Ma, W.L. Liu, Y.L. Liang and L. Peeters, "Magnetic Characterization of NdFeB Powder Made by Inert Gas Atomization with Subsequent HDDR
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ALLOY DESIGN AND MICROSTRUCTURE OF ADVANCED PERMANENT MAGNETS USING RAPID SOLIDIFICATION AND POWDER PROCESSING
36. 37.
38.
39.
40.
41.
42.
Treatments", Proc. 12th Int. Workshop Rare-Earth Magnets Their Appl., edited by Hi Perm Laboratory, Rare-Earth Inf. Cent., Ames, IA , 1992, pp. 682–693. R.M. German, Powder Injection Molding, First Edition, 1990, Metal Powder Industries Federation, Princeton, NJ. P.K. Sokolowski, I.E. Anderson, W. Tang, Y.Q. Wu, K.W. Dennis, M.J. Kramer and R.W. McCallum, “In situ Passivation during High Pressure Gas Atomization of Improved MRE2Fe14B for High Performance Permanent Magnet Applications”, 8th Global Innovations Symposium: Trends in Materials and Manufacturing Technologies for Energy Production, The Minerals, Metals, & Materials Society, Warrendale, PA, 2007, pp. 7–17. R.K. Mishra, V. Panchanathan and J.J. Croat, "The Microstructure of Hot Formed Neodymium-Iron-Boron Magnets with Energy Product 48 MGOe", Journal of Applied Physics, 1993, vol. 73, pp. 6,470–6,472. T. Takeshita, R. Nakayama and T. Ogawa, "Rare EarthIron-Boron Magnet Powder and Process for Producing Same", U.S. Patent No. 5,110,374, May 5, 1992. W. Tang, Y.Q. Wu, K.W. Dennis, M.J. Kramer, I.E. Anderson and R.W. McCallum, "Effect of Zr Substitution on Microstructure and Magnetic Properties of New YDybased R2Fe14B Magnets (R = Y+Dy+Nd)", Journal of Applied Physics, 2005, vol. 97, pp. 10H106. C.D. Fuerst, J.F. Herbst and E.A. Alson, "Magnetic Properties of Neodymium-Cobalt-Iron-Boron (Nd2(CoxFe1x)14B) Alloys", Jour nal of Magnetism and Magnetic Materials, 1986, vol. 54–57, pp. 567–569. W. Tang, Y.Q. Wu , K.W. Dennis, N. Oster, M.J. Kramer, I.E. Anderson and R.W. McCallum, "Magnetic Properties and Microstructure of Gas Atomized MRE 2(Fe, Co) 14B Powder with ZrC Addition (MRE=Nd+Y+Dy)", J. Applied Phys., accepted for publication, 2008. ijpm
TRUST must be earned For 90 years, ACuPowder has been delivering the finest quality powders and the most conscientious service. Our customers know that serving their needs and solving their problems is our highest priority. Bring us your toughest assignments. We want to earn your trust, too. The finest powders are from ACuPowder: Copper, Tin, Bronze, Brass, Copper Infiltrant, Bronze Premixes, Antimony, Bismuth, Chromium, Manganese, MnS+, Nickel, Silicon, Graphite and P/M Lubricants.
901 Lehigh Ave., Union, NJ 07083 908-851- 4500, • Fax 908-851- 4597
6621 Hwy. 411 So., Greenback, TN 37742 865- 856- 3021 • Fax 865-856 -3083 e-mail:
[email protected] web: www.acupowder.com ISO 9001 CERTIFIED
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
ISO 14001 CERTIFIED
37
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
RESEARCH & DEVELOPMENT
STEEL-SHEET FABRICATION BY TAPE CASTING Martin Rauscher,* Georg Besendörfer* and Andreas Roosen**
INTRODUCTION PM is a highly developed method for manufacturing ferrous and nonferrous metal parts. Originally developed for the processing of materials with high melting temperatures such as tungsten and molybdenum, PM is now used to process a variety of metals. It is often used when the manufacture of parts by machining, casting, or forging is too expensive. PM parts are used in diverse industries, for example, automotive, aerospace, toolmaking, and armaments.1,2 The process is versatile because it is applicable to simple as well as complex shapes, and a wide range of physical and mechanical properties can be achieved. The basic PM process consists of three steps: (i) powder production, (ii) compaction, and (iii) densification by sintering.1–3 The aim of the present study is to demonstrate that tape-casting technology, an established ceramic forming technique, can be used in PM as a powerful, versatile, and low-cost shaping process, without losing the advantages of PM, in particular: • Processing of hard, brittle materials and refractory alloys • Production of metal parts with a high degree of purity • Production of complex components with high precision • Near-net-shape manufacture, minimizing machining and with high raw-materials utilization • Combination of materials, including dissimilar metals, nonmetallics, and materials with specific, but differing, characteristics • Control of density and porosity Tape-casting technology is commonly used in the ceramics industry for the manufacture of thin, planar ceramic products of large area with uniform surface quality and precise dimensions.4,5 In addition to singlelayer applications for thick- and thin-film circuitry, green ceramic tapes can also be punched, metallized by thick-film techniques, and stacked and laminated to form multilayer components, such as integrated circuits, capacitors, inductors, piezoelectric actuators, and gas sensors.6,7 Using this technology, numerous devices for automotive, communications, and medical applications are produced in large volumes. Figure 1 shows a typical process flow sheet for the manufacture of a ceramic multilayer device. To obtain homogeneous slurries, the
A new method of manufacturing thin, planar metal sheet has been developed by adapting ceramic tapecasting technology to powder metallurgy (PM). By combining the advantages of PM with those of the tape-casting technique, a low-cost process for the production of thin planar metal forms has been demonstrated. A stainless steel 316L powder was used as a model system to form a tape-casting slurry with preset rheological behavior for the subsequent casting process. After drying the cast slurry, a flexible green tape was obtained with sufficient strength and flexibility to withstand further handling by cutting and lamination. The tape was densified by sintering to produce a flexible metal sheet. The mechanical properties and attendant microstructures of the steel sheet were determined and characterized. In addition, a simple, novel joining technique was established by lamination of the green metal sheets, followed by sintering, resulting in a stable homogenous microstructure.
*Research Fellow, **Professor, Department of Materials Science WW3, Martensstr. 5, University of Erlangen-Nürnberg, 91058 Erlangen, Germany; E-mail:
[email protected]
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
39
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
STEEL-SHEET FABRICATION BY TAPE CASTING
Figure 2. Principle design of tape-casting machine (courtesy Ch. Bauer-Lutz)
Figure 1. Process flow in ceramic multilayer fabrication
inorganic powders are dispersed in an aqueous or organic medium containing a dispersing agent by means of ball milling.8 In a second milling step, a binder and plasticizer are added to give the dried tape sufficient strength and flexibility for further processing.9,10 Before casting, residual agglomerates and milling debris are removed by filtration. To prevent the formation of pores in the cast tape, dissolved gases are removed in a degassing step. The slurry is then transferred into the reservoir of a tape-casting head of a casting machine. The principle design of such a tape-casting machine (used in the present work) is shown in Figure 2. Underneath the casting head, which is equipped with the doctor blades, a moving tape-carrier film drags the casting slurry out of the casting head. Passing the doctor blade, the slurry forms a continuous thin film, the thickness of which is controlled by adjusting the gap height of the doctor blade, the speed of the carrier, the hydrostatic pressure in the reservoir, and the viscosity of the slip. After drying, the flexible green sheets are
40
removed from the carrier and shaped by cutting and punching. In the metallization step, vias are filled and circuitry details are printed by a screenprinting technique using conducting metal powder pastes. For the fabrication of multilayer devices, the green ceramic layers are stacked and laminated by thermocompression. During this process, which is performed at temperatures ~50°C to 80°C and pressures of 20 to 150 MPa, the binder/ plasticizer system in the green sheets softens and starts to flow under the applied pressure. This induces a mass flow which moves and rearranges the particle in the interface of the adjacent green tapes (Figure 3). After binder burnout and sintering, a defect-free junction between the stacked tapes is achieved, in which prior interfaces can no longer be detected; thus, a homogenous, stable compound is formed. 11,12 To facilitate particle mobility, a sufficient amount of thermoplastic polymer (e.g., PMMA, PVA, PVB), is used as a binder, together with plasticizers. The plasticizers lower the glass transition temperature (Tg) of the thermoplastic binders. The multilayer structures are then heat treated to remove all the organic compounds, followed by
Figure 3. Schematic view of interface of two green ceramic tapes during lamination12
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Next
Table of Contents
STEEL-SHEET FABRICATION BY TAPE CASTING
densification via sintering at elevated temperatures. In the case of non-oxide materials, the thermal treatment has to be carried out under vacuum, in a neutral-gas atmosphere (nitrogen, argon) or in a reducing-gas atmosphere.6,12,13 The objective of this study was to utilize the ceramic tape-casting technique in PM processing. An austenitic stainless steel powder was used to demonstrate the feasibility of processing metal powders to thin, planar flexible green tapes, which are subsequently sintered to produce metal sheets. The focus of this study was on slurry preparation, tape casting as a forming process, and on lamination and subsequent sintering as a joining technique. EXPERIMENTAL PROCEDURE Raw Materials and Slurry Preparation A spherical chromium–nickel stainless steel powder (EN 1.4404, AISI 316L, Hauner GmbH, Röttenbach, Germany) was processed to a slurry with preset rheological behavior for the tape-casting process. The particle-size distribution of the powder was determined by laser granulometry (Mastersizer 2000, Malvern Instruments, UK) and is shown in Figure 4. The average particle size (d50) of the powder was 8.3 µm, with d10 and d90 values of 3.8 µm and 15.9 µm, respectively. For slurry preparation, a dispersing agent (KD, UniQuema, Belgium) was dissolved in an azeotropic mixture of methylethylketone and ethanol (66:34 w/o). After addition of the metal powder, deagglomeration of the powder particles and homogenization of the slurry was carried out by ball milling for at least 24 h using stainless steel
Figure 4. Frequency plot for particle-size distribution of 316L stainless steel powder
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
TABLE I. SLURRY COMPOSITION (w/o) Slurry Component
Steel Powder Solvent Dispersant Binder Plasticizer 79.7
15.9
0.8
2.4
1.2
grinding balls (8 and 16 mm dia.). Polyvinyl butyral binder (Butvar, Solutia, U.S.) and a plasticizer (Santicizer, Ferro, Belgium) were added, followed by another homogenization step of 24 h. The homogenous slurries were passed through a sieve and degassed in a vacuum of 200 mbar for at least 1 h, and finally tape cast. The slurry composition is given in Table I. The rheological behavior was determined in a rheometer (Physica UDS, Anton Paar Physica Messtechnik GmbH, Austria) using the cone-plate method in a shear-rate-controlled experiment in the range of 0.1 to 400 s-1. Tape Casting Tape casting was performed utilizing the doctor-blade method at room temperature. The slurries were transferred into a double-chamber casting head, equipped with two blades adjusted to gap heights of 1,100 µm and 900 µm. A silicon coated polyethylenterephthalate (PET) film was used as a moving carrier. After drying, green tapes were removed from the carrier film with an average thickness of 350 µm. The tapes were cut into squares (20 mm × 20 mm) by means of a hot knife. Laminates of similar geometry were produced by stacking together two green tapes, followed by lamination at a pressure of 10 MPa for 10 min and a temperature of 70°C; this temperature is above the glass transition temperature (Tg) of the binder-plasticizer system. The average density of the green tapes was determined by measuring the dimensions (hence the volume) of 15 pieces of tape by means of a micrometer, and weighing each piece of tape. Binder Burnout and Sintering The decomposition behavior of the organic constituents was characterized by thermogravimetric analysis (TGA) of the green sheets by heating to 700°C at a heating rate of 5°C/min in a nitrogen atmosphere (model STA 409, Netzsch, Germany). Based on these results, the temperature–time profile of the binder bur nout was established. Damage by uncontrolled gaseous decomposition products was avoided by binder burnout at a slow heating rate (2°C/min) up to 700°C, including an
41
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
STEEL-SHEET FABRICATION BY TAPE CASTING
Figure 5. Sintering time–temperature profiles
intermediate holding step at 300°C for 2 h. For densification by sintering, the temperature was increased to 1,000°C, 1,100°C, 1,150°C, 1,250°C, and 1,300°C. The holding time at each temperature was 180 min. Sintering was performed under a nitrogen atmosphere to prevent oxidation of the metal powders. Figure 5 illustrates the heating schedule for sintering. The tape-cast sheets were put on alumina substrates prior to sintering. The average density of the sintered sheets was determined by the same method used for the green tapes. Porosity was characterized by means of a helium pycnometer (SccuPyc, Micrometrics, U.S.). For microstructural analysis, sintered specimens were polished and etched with a copper etchant (200 ml ethanol, 40 ml HCl, 10 g FeCl3). Analysis was then performed by means of an optical microscope equipped with a microindentation unit (Leitz, Metallux, Germany) to determine the Vickers hardness of the sintered sheets. For tensile tests, specimens were punched out of the green tapes with geometries according to DIN 53504 and sintered utilizing similar profiles to those in Figure 5 up to a maximum temperature of 1,250°C. Tensile tests were carried out in a mechanical testing machine (Instron 4504, Instron, U.S.) with a 100 kN load cell at 10-3 mm/s using a transducer with a range of 12.5 mm. RESULTS AND DISCUSSION Slurry Preparation and Tape Casting The rheological behavior of the tape-cast slurry is shown in Figure 6. The slurry exhibited pseudo-
42
Figure 6. Rheological behavior of metal-filled slurry
plastic behavior, characterized by a decrease in viscosity with increasing shear rate. Upon increasing and decreasing shear rate, the viscosity curve exhibited hysteresis, which is indicative of a small thixotropic effect in the slurry. The slurries were stable and homogeneous and the viscosity was high enough to prevent particle sedimentation. The slurries were cast at a velocity of 0.5 m/min with gap heights of 900 µm and 1,100 µm. This results in a shear rate ~10 s-1 during the casting process, which is delineated as a dashed line in Figure 6. At this shear rate, the viscosity of the slurry is in the range of 2.4 to 2.7 Pa·s which meets the requirement for tape casting (slurry viscosities in the range of 1 to 20 Pa·s).8 After drying, green tapes were obtained with sufficient strength and flexibility for further treatment by cutting and lamination. Lamination parameters were optimized to achieve uniform, defect-free joints between the individual lengths of green tape. After drying, the green tapes exhibited a density of 58% of the porefree density (PFD), namely 7.9 g/cm3 for 316L stainless steel.14 Lamination increased the green density to 63% PFD. Binder Burnout The burnout behavior of the organic additives was analyzed by means of TGA (Figure 7). The decomposition of the organic phases takes place in several steps, analogous to ceramic green tapes, as described by Salam, Matthews and Robertson.15 The first significant weight loss was observed at 170°C, and is attributed to the decomposition of the low-molecular-weight plastiVolume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
STEEL-SHEET FABRICATION BY TAPE CASTING
Figure 7. TGA of 316L stainless steel green tape under nitrogen atmosphere
Figure 8. Densities of steel single sheets and laminates as a function of sintering temperature
cizer. This was followed by a two-step decline in weight, resulting from decomposition of the binder. At 540°C, burnout of the organic constituents is complete, resulting in a total weight loss of 4.8 w/o. To prevent structural damage of the tapes by uncontrolled gas formation during thermal treatment, the organic constituents were removed prior to sintering at a low temperature (550°C) with a low heating rate (2K/min). In comparison with the total amount of organic additives in the starting slurry (Table I), the theoretical weight loss should be 5.2 w/o. The difference of 0.4 w/o is attributed to residual carbon, which results from incomplete decomposition of the organic phase under a non-oxidizing atmosphere. During subsequent sintering the carbon was either dissolved or formed carbides in the grain boundaries, thereby influencing the properties of the steel sheet. No effect of carbon on the densification process was observed.
The reason for this is twofold: First, organic phase degradation during binder burnout leads to a weight loss, and the temperature is too low for densification to occur by sintering. Second, the steel sheets exhibited a small volume expansion. The first significant increase in density was observed after sintering at 1,150°C, while maximum values of sintered density for both single and laminated steel sheets were achieved at 1,250°C. A further rise in sintering temperature to 1,300°C led to inhomogeneous surfaces and warping of both the sheets and the laminates. Figure 9 illustrates a warped sheet in comparison with a planar, homogenous sample sintered at 1,250°C. Single-sheet densities of 7.26 g/cm 3 (92% PFD) and laminate densities of 7.51 g/cm3 (95% PFD) resulted from sintering at 1,250°C. Density levels comparable with those achieved by conventional PM processing (≥90% PFD) were reached. 2 Since these samples were processed
Shrinkage and Densification Sintering at different temperatures resulted in ductile steel sheet of varying density. As demonstrated in Figure 8, sheet densities increased with sintering temperature. In general the density of the laminate sheets exceeded those of the single sheets. This is attributed to the lamination process acting as a pre-compaction step. For temperatures of 1,000°C and 1,100°C, the densities of the sintered sheets and laminates remained in the range of the initial green densities, namely 58% PFD and 63% PFD, respectively (delineated by the dashed lines in Figure 8). In some cases densities below the PFD initial green state were obtained. Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Figure 9. Sintered steel sheets: (a) 1,250°C—no warping, (b)1,300°C—warped
43
Previous
1-Page View
2-Page View
Search
Next
Table of Contents
STEEL-SHEET FABRICATION BY TAPE CASTING
without a post-sintering treatment, the laminated green sheets reached sintered densities similar to those realized only by powder injection molding (PIM). By further increasing the sintering temperature, higher densities can be achieved, but with attendant warpage. To avoid the latter, additional optimization of the sintering profile is necessary. The level of porosity in the steel sheets decreased with increasing sintering temperature (T S), Table II. Characterization of the porosity showed a predominantly open (interconnected) pore structure in the sheets after binder burnout. With an increase in sintering temperature from 1,000°C to 1,150°C the fraction of open porosity decreased from 40 to <30 v/o in single sheets, and from 35 to <20 v/o in laminate sheets. This marks the transition from the initial to the intermediate stage of sintering. Concurrently, the level of closed porosity remained essentially constant at these sintering temperatures. After sintering at 1,250°C, a total porosity of <9 v/o for single steel and <5 v/o for laminated sheets was achieved, corresponding to the final sintering stage. In this condition the open pore structure was eliminated and only closed pores remained. In Figure 10 the average volume shrinkage of the sintered single sheets and laminated sheets is plotted as a function of sintering temperature. Analogous to the densification results (Figure 8), the sheets sintered at 1,000°C and 1,100°C exhibit no shrinkage compared with the initial geometry of the green sheets. In contrast, after sintering, a small volume increase was observed, delineated by the negative volume change in Figure 10. A possible explanation for the volume expansion is the thermal expansion of the steel powder, which is not compensated for by densification or contraction during cooling. The first significant shrinkage occurred at a sintering temperature of 1,150°C. After sintering at 1,250°C, at which maximum sintering densities were obtained, the shrinkage reached values of 38 v/o for single sheets and 43 v/o for laminated sheets. Thus, the laminates with higher sintered densities also showed larger shrinkage levels than the single tapes. Shrinkage in thickness (z-direction) generally exceeded the in-plane shrinkage (x and y directions), as shown in Table III. This shrinkage anisotropy is typical of tape-cast products.16,17 Sheet Microstructure The microstructures of the sintered single and
44
TABLE II. CHARACTERIZATION OF POROSITY IN SINTERED SHEETS MEASURED GEOMETRICALLY AND BY HELIUM PYCNOMETRY TS [°C]
Sheet Type
Density (g/cm3)
Open Porosity (v/o)
Closed Porosity (v/o)
1,000
Single Laminate
4.56 4.86
39 35
4 3
1,100
Single Laminate
4.56 5.19
38 30
4 4
1,150
Single Laminate
5.33 6.22
29 19
3 4
1,250
Single Laminate
7.26 7.51
<1 <1
8 5
Figure 10. Volume shrinkage of steel sheets as a function of sintering temperature
TABLE III. SHRINKAGE OF STEEL SHEETS AND LAMINATED SHEETS AS A FUNCTION OF SINTERING TEMPERATURE TS (°C)
Type
z-shrinkage (%)
x-shrinkage (%)
y-shrinkage (%)
1,000
Single Laminate
-2 -3
-2 -2
-1 -2
1,100
Single Laminate
0 1
0 0
0 0
1,150
Single Laminate
7 8
4 5
4 5
1,250
Single Laminate
19 26
13 13
13 12
laminated sheets were similar for a given sintering temperature. No thickness effect was observed in relation to the microstructures. In general, the laminated sheets exhibited microstructures free of delaminations at the initial interfaces. Thus, a novel and simple method of joining metal sheets Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
STEEL-SHEET FABRICATION BY TAPE CASTING
was developed. Representative scanning electron micrographs (SEM) of laminated steel sheets, sintered at 1,150°C, 1,250°C, and 1,300°C, are shown in Figures 11 through 13. Sheet microstructures of specimens sintered at 1,000°C, 1,100°C, and 1,150°C exhibited identical phases and were characterized by a high level of porosity, reflecting the initial and intermediate stages of sintering. With increasing sintering temperature a decrease in porosity was observed, accompanied by grain growth. As shown in Figure 11, the grain shape of the original steel powder was still recognizable after sintering at 1,150°C. Between the grains, interconnecting sintering necks developed and partial grain coarsening could be observed. Dark areas around the grains show the highly porous structure of the sheet, whereas the grains per se exhibit an internal structure consisting primarily of austenite and carbides. In those alloys, carbides typically precipitate when cooled to temperatures below the solubility line at 1,100°C. In addition, some carbides remained from binder burnout. In Figure 12 the microstructure of the laminated sheets sintered at 1,250°C exhibits dense, homogenous austenite of uniform grain size. Reaching the final stage of sintering with densities >95% PFD, the open-pore structure was geometrically unstable and consequently collapsed into closed pores of spherical shape; these pores are located predominantly in the grain boundaries. In addition, some pores were generated during the etching process, in which segregated carbon was dissolved. In principle, prolonged holding time below the solubility line, as well as slow cooling, lead to carbide segregation. The carbides migrate to the grain boundaries where they coalesce, especially at triple points.18 As a result of over-sintering, the microstructure of the laminated sheets sintered at 1,300°C exhibited significant grain coarsening due to Ostwald ripening, Figure 13. As a consequence, the carbides and pores, initially located at the grain boundaries, are now incorporated in the body of the grains. Additionally, the grains exhibit an inhomogeneous structure due to further segregation of the alloying constituents. Thus, between the grains, chromium and nickel-enriched regions were developed. This depletion of the alloying constituents led to the formation of needle-shaped structures of δ-ferrite and martensite.16 Alloying effects from the residual carbon of the decomVolume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Figure 11. Representative micrograph of polished cross section of laminated steel sheet, sintered at 1,150°C. SEM
Figure 12. Representative micrograph of polished cross section of laminated steel sheet, sintered at 1,250°C. SEM
Figure 13. Representative micrograph of polished cross section of laminated steel sheet, sintered at 1,300°C. SEM
posed organic compounds and nitrogen in the sintering atmosphere were not evaluated in this study.
45
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
STEEL-SHEET FABRICATION BY TAPE CASTING
Mechanical Properties The tensile strength of 316L stainless steel typically ranges from 490 MPa to 690 MPa with an average elongation at fracture of 40%, accompanied by a lateral contraction of 60%. 14 The Young’s modulus is ~200 GPa and Brinell hardness values of 120 HB30 to 180 HB30 have been reported, which correspond to Vickers hardness levels of 126 HV1 to 187 HV1, respectively. In light of the porosity in the sintered sheets, microindention hardness tests were only carried out on sheets sintered at 1,250°C and 1,300°C. The tests were performed across the sheet cross sections and revealed no hardness gradients across the sample thickness. Thus, no hardening due to slow cooling was evident. Further, as shown in Table IV, for single sheets and laminated sheets at a given sintering temperature, similar hardness levels existed. Sheet hardness was also independent of sample thickness. With values around 180 HV1, sheets sintered at 1,250°C exhibited hardness levels representative of 316L sheet. As expected, for sheets sintered at 1,300°C, a hardness >180 HV1 was measured due to the formation of hard, brittle carbides and martensite. The latter also showed a comparably high deviation, which is attributed to the inhomogeneous microstructure in the stainless steel sheets. In this context, additional microindentation hardness testing was performed to determine TABLE IV. MICROINDENTATION HARDNESS OF SINTERED STEEL SHEETS (HV 1/10)* TS [°C]: Sheet Type:
1250
1300
Single Laminate
Single Laminate
Average Hardness:
182
185
280
265
Standard Deviation:
7
12
37
25
*DIN EN ISO6507: HV 1/10 = 1 kilopond load for 10 s 1 pond = 9.80665 mN TABLE V. MICROINDENTATION HARDNESS OF SINTERED STEEL SHEETS BY LOCATION TS [°C]:
1,150
1,250
1,250
1,300
1,300
Position:
Grain
Grain
Boundary
Grain
Boundary
Structure:
γ-phase/ carbide
γ-phase
γ-phase/ carbide
Martensite & γ/δ-phase
Cr-/Ni-ric phase
Hardness: (HV 1/10)
180
137
196
179
165
46
individual phases and their hardness, as shown in Table V. As noted previously, grains in the sheets sintered at 1,150°C exhibited a heterogeneous microstructure with fine-scale austenite and carbides; indented areas were too large to isolate single phases. However, the measured average hardness of 180 HV1 is in the range of values reported in the literature. At 1,250°C, where the carbide phase was found predominately in the grain boundaries, both phases were clearly distinguishable. In contrast to their comparably highhardness values, the microindentation hardness values for sheets sintered at 1,300°C were lower, namely 165 HV1 in the chromium-rich and nickel-rich grain boundaries and 179 HV1 in the grains. However, the hardness of the grains was still higher than that of the homogenous austenite after sintering at 1,250°C. This can be ascribed to the development of brittle phases most likely δ-ferrite and martensite. The tensile tests were carried out on sheets sintered at 1,250°C. For an investigation of mechanical property anisotropy after sintering, samples in both the transverse and tape-casting direction were analyzed. Figure 14 shows a representative stress–strain curve of a tape-cast sintered 316L stainless steel sheet. For stresses up to ~200 MPa, the stress–strain response is linear, characteristic of elastic behavior. With increasing stress, the elastic behavior is followed by plastic deformation. Finally, fracture occurs at comparatively small elongations (~14%), and no lateral contraction was observed. In the absence of necking, even at elevated stresses, the true stress–strain behavior could be calculated on the basis of constant volume. In Figure 15, the true tensile strength (σtr) for the longitudinal and transverse samples are plotted against the strain at fracture. For samples in the tape-casting direction (longitudinal), an average true tensile strength of 495 ± 37 MPa was measured at an average breaking strain of 13 ± 5%. Fracture of the transverse specimens occurred at an average stress of 475 ± 9 MPa with an average elongation at fracture of 13 ± 2%. Young’s moludi were determined from the initial linear portion of the stress–strain curve with an average value of 139 ± 9 GPa for samples in the tape-casting (longitudinal) direction and 140 ± 10 GPa in the transverse direction. Comparing both series of samples, no significant dependency of mechanical properties on direction was evident, Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
STEEL-SHEET FABRICATION BY TAPE CASTING
Figure 14. Representative engineering stress–strain curve for steel sheet sintered at 1,250°C
Figure 15. True tensile strength and strain at fracture in longitudinal and transverse directions of steel sheets
hence the mechanical properties are isotropic. The resulting tensile strengths of both series of samples were in the range of values reported in the literature, whereas strains at fracture and Young’s moduli were lower. As described by German et al.,2 the mechanical properties of PM products correlate exponentially with bulk density. As shown in the sintering experiments, the levels of porosity in the sheets after sintering at 1,250°C were approximately 8 v/o. Thus, both tensile strength and ductility were reduced. Besides porosity, the segregation of phases within the microstructure was also influential in relation to tensile properties. The carVolume 44, Issue 6, 2008 International Journal of Powder Metallurgy
bides contributed to a further decline in ductility, while concurrently the tensile strength increased. Also, the potential alloying of the austenite with nitrogen would lead to a further rise in strength. From a combination of these factors, low elongation values resulted, even though the tensile strength reached values representative of annealed sheet. The intent of this study was to show the potential and viability of the application of ceramic tape-casting technology to metal powders. With emphases on slurry preparation, tape casting, and the lamination process, the sheet-making process was not optimized in ter ms of the mechanical properties of the stainless steel. The mechanical properties of the sheets produced before casting are not comparable with those of rolled steel sheet, but the primary aim of adopting a new, simple low-cost forming method for PM processing was demonstrated. For future applications of tape casting as a PM process, the potential is evident in its capability to fabricate metals, as well as metal/ceramic composites to thin, planar structures, with a high throughput. To this end, the process has to be optimized. Processing agents have to be identified for which complete decomposition is achieved during binder burnout. This is necessary in order to avoid residual carbon and the formation of brittle compounds. CONCLUSIONS Based on the present work a new low-cost method for the manufacture of thin planar metal structures has been demonstrated by adopting ceramic tape-casting technology to PM. Steel sheets were produced by tape casting a 316L stainless steel powder–filled slurry with preset rheological behavior to form a green tape, which was subsequently sintered under nitrogen to prevent oxidation. By raising the sintering temperature, the level of porosity in the microstructure of the sheet can be controlled. Optimum sheet densities were obtained at a sintering temperature of 1,250°C with densities of 92% PFD for single sheets. With slow cooling from 1,250°C an austenitic microstructure with carbides in grain boundaries was obtained. Notwithstanding residual sheet porosity, the sintered sheets achieve a tensile strength similar to that of rolled annealed 316L stainless sheet, but at the expense of ductility. Although the mechanical properties of the
47
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
STEEL-SHEET FABRICATION BY TAPE CASTING
steel sheets were not comparable with those of rolled stainless steel sheet, the primary aim of introducing the tape-casting technique as a new forming method for PM processing was achieved. A novel joining technique for metals was established by the lamination of green metal tapes and subsequent sintering. This resulted in sintered laminated sheets of homogenous, interface-free microstructures with sinter densities >95% PFD. For non-post-treated PM products these values can only be reached by PIM. Another potential of the lamination process is the formation of complex three-dimensional structures and composites, and in rapid prototyping.
8.
9.
10.
11.
12.
REFERENCES 1. L.F. Pease and R.J. Sansoucy, Advances in Powder Metallurgy—1991: Aerospace, Refractory and Advanced Materials, 1991, Metal Powder Industries Federation, Princeton, NJ. 2. R.M. German, Powder Metallurgy Science, Second Edition, 1994, Metal Powder Industries Federation, Princeton, NJ. 3. L.F. Pease and W.G. West, Fundamentals of Powder Metallurgy, 2002, Metal Powder Industries Federation, Princeton, NJ. 4. R.E. Mistler and E.R. Twiname, Tape Casting, 2000, The American Ceramic Society, Westerville, OH. 5. A. Roosen, “Basic Requirements for Tape Casting of Ceramic Powders”, Ceramic Powder Science II: Ceramic Transactions, compiled by G.L. Messing, E.R. Fuller and H. Hausner, The American Ceramic Society, Westerville, OH, 1988, vol. 1, part B, pp. 675–692. 6. A. Roosen, “Tape Casting of Ceramic Green Tapes for Mulitlayer Device Processing”, Ceramic Trans., 1999, vol. 97, pp. 103–121. 7. A. Roosen, “3-D Structures via Tape Casting and
48
13.
14. 15.
16.
17.
18.
Lamination”, Adv. in Sci. and Techn., 2006, vol. 45, pp. 397–406. R. Moreno, “The Role of Slip Additives in Tape-Casting Technology: Part I – Solvent and Dispersant”, Am. Ceram. Soc. Bull., 1992, vol. 71, no. 10, pp. 1,521–1,531. R. Moreno, “The Role of Slip Additives in Tape-Casting Technology: Part II-Binders and Plasticizers” Am. Ceram. Soc. Bull., 1992, vol. 71, no. 11, pp. 1,647–1,657. M. Descamps, G. Ringuet and D. Leger, “Tape Casting: Relationship between Organic Constituents and the Physical and Mechanical Properties of Tapes”, J. Europ. Ceram. Soc., 1995, vol. 15, pp. 357–362. A. Roosen, “New Lamination Technique to Join Ceramic Green Tapes for the Manufacturing of Multilayer Devices”, J. Europ. Ceram. Soc., 2001, vol. 21, pp. 1,993–1,996. H. Hellebrand, “Tape Casting“, Materials Science and Technology, Vol. 17, Processing of Ceramics, compiled by R.J. Brook, VCH Verlagsgesellschaft, Weinheim, Germany, 1996, pp. 189–265. K. Utsumi, “Development of Multilayer Ceramic Components Using Green Sheet Technology”, Ceram. Bull., 1991, vol. 70, pp. 1,050–1,055. C.W. Wegst, Stahlschlüssel, 17th Edition, Stahlschlüssel Wegst, Marbach, Germany, 1995. L.A. Salam, R.D. Matthews and H. Robertson, ”Pyrolysis of Polyvinyl Butyral (PVB) Binder in Thermoelectric Green Tapes”, J. Europ. Ceram Soc., 2000, vol. 20, no. 9, pp. 1,375–1,383. M. Rauscher and A. Roosen, “Influence of LowTemperature Co-Fired Ceramics Green Tape Characteristics on Shrinkage Behavior”, Int. J. Appl. Ceram Techol., 2007, vol. 4, no. 5, pp. 387–397. I.O. Ozer, E. Suvaci, B. Karademir, J.M. Missiaen and C.P. Carry, “Anisotropic Sintering Shrinkage in Alumina Ceramics Containing Oriented Platelets”, J. Am. Ceram. Soc., 2006, vol. 89, no. 6, pp. 1,972–1,976. H. Schumann, Metallographie, 13th Edition, VEB Deutscher Verlag für Grundstoffindustrie, Leipzig, Germany, 1990. ijpm
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
RESEARCH & DEVELOPMENT
TRANSIENT LIQUIDPHASE SINTERING OF COPPER–NICKEL POWDERS: IN SITU NEUTRON DIFFRACTION Dennis M. Turriff,* Stephen F. Corbin,** Lachlan M.D. Cranswick,*** and Michael J. Watson****
INTRODUCTION Transient liquid-phase sintering (TLPS) is a unique powder metallurgy (PM) processing technique typically used to form near-net-shape parts1 and more recently, as in this study, as a means of developing liquid-rich low-temperature solders2,3 and brazing filler materials4 that exhibit variable melting point (VMP) characteristics. In TLPS, starting mixtures normally consist of a low-melting-point additive powder and a higher-melting-point base-metal powder. The transient liquid phase that forms during heat-up past the additive’s melting point aids in rapid densification of the mixture.1,5–8 This liquid alloys with the basemetal powder during sintering and can lead to complete isothermal, or diffusional, solidification and a shift in melting point for the bulk powder mixture.4 In order to achieve the maximum melting-point shift for VMP brazing applications the isothermally solidified phase should have a completely homogenized composition. The consequence of diffusional solidification and incomplete homogenization was previously studied4 via differential scanning calorimetry (DSC). DSC results for nickel and copper powder mixtures (65 w/o Cu) showed quantitatively that a hold time of 150 min at 1,140°C enabled complete isothermal solidification of the copper-rich liquid during TLPS. Upon reheating, the TLP-sintered specimens exhibited a measurable melting-point increase, and an enhanced melting range due to incomplete homogenization of the isothermally solidified microstructure. Metallographic characterization of the post-sintered DSC specimens revealed that significant compositional gradients remained between the nickel-rich particle cores and the surrounding copper-rich solid-solution regions, even after holding for 150 min at 1,140°C.4
The initial melting behavior and solidification kinetics during transient liquidphase sintering (TLPS) of elemental copper–nickel powder mixtures have been investigated via in situ neutron diffraction (ND). By conducting ND experiments at various isothermal sintering temperatures and durations above the copper melting point, partial liquation of the powder mixtures was induced. By tracking the post-melt evolution of the (200) diffraction peak profiles during prolonged sintering cycles, insight has been obtained regarding interdiffusion of copper and nickel within the unmelted nickel powder, which is responsible for the diffusional, or isothermal, solidification of the surrounding liquid phase. It is concluded that the transient liquid is removed primarily by the rapid growth of a copperrich solid solution having a composition given by the phase diagram solidus.
*Instructor/Research Associate, **Professor and Associate Chair of Research, Materials Processing and Engineering Group, Department of Mechanical & Mechatronics Engineering, University of Waterloo, 200 University Avenue W, Bldg. E2-4404, Waterloo, Ontario, Canada, N2L 3G1; E-mail:
[email protected], ***Research Scientist, ****Research Scientist, Retired, Canadian Neutron Beam Centre, National Research Council Canada, Bldg. 459, Chalk River Laboratories, Chalk River Ontario, Canada, K0J 1J0
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
49
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
TRANSIENT LIQUID-PHASE SINTERING OF COPPER-NICKEL POWDERS: IN-SITU NEUTRON DIFFRACTION
From this discussion, it is clear that interdiffusion between the copper and nickel phases during the isothermal solidification and homogenization stages of TLPS is important in determining the rate of liquid removal at 1,140°C, as well as the extent of melting-point shift. Therefore, the primary objective of this investigation was to develop an experimental method using neutron diffraction (ND) to measure in situ the extent of interdiffusion taking place during TLPS of copper–nickel powders. Rudman and Fischer 9 and Delhez et al.10 have shown that X-ray diffraction techniques (XRD) can be used to investigate the increasing degree of interdiffusion of copper–nickel during solid-state sintering. This was done by tracking the evolution of the (220) diffraction profiles for copper and nickel after isothermal sintering of copper and nickel powder mixtures below 1,085°C. The gradual formation of alloyed regions within the compacts due to sintering and interdiffusion gave rise to XRD peak broadening and the generation of a wide diffraction signal at all intermediate 2θ angles, since nickel and copper constitute an isomorphous alloy system. It should be noted that the studies cited focused on solid-state sintering and XRD measurements were done after the sintering treatment (i.e., ex situ). The authors are not aware of any in situ studies on sintering of powder mixtures in the presence of a transient liquid, which inherently complicates the process. The large penetration depth and large representative-sample volumes (grams) offered by monochromatic neutron beams makes them an ideal probe for such investigations. MATERIALS AND METHODS Table I lists the copper and nickel powders used and their relevant characteristics. The copper powder represents the low-melting-point additive phase (Tm = 1,085°C) and source of liquid during sintering, whereas the nickel powder represents the high-melting-point base-metal phase
TABLE I. POWDER DATA Average Mesh Size Particle Radius
Powder
Purity
Shape
Supplier
Copper
99.9 w/o <1,000 ppm O2
-170 + 400
23.09 µm**
Spherical* Alfa Aesar
Nickel
99.90 w/o
-48 + 150
80.50 µm*
Spherical* Alfa Aesar
Verified via optical microscopy*, SEM* and a Horiba CAPA–700 particle size analyzer**
50
(Tm = 1,455°C). Specimens were prepared by mixing the loose elemental powders such that the bulk mixture composition, (CO) was 65 w/o Cu. The specimens were placed in large cylindrical Al2O3 crucibles (6 mm dia. × 42 mm tall), which could accommodate large sample sizes in order to provide sufficiently strong diffraction signals. Neutron diffraction experiments were conducted in the C2 powder diffractometer located at the National Research Universal (NRU) research reactor at the Chalk River Laboratory, Chalk River, Canada.11,12 The crucibles were placed into cylindrical vanadium containment canisters, which were centered/aligned in the body of the vacuum furnace and the diffractometer. A planar silicon single crystal monochromator was used via the 531 reflection and 92.7° takeoff angle to generate a monochromatic incident beam having a wavelength of 0.133069 (7) nm. Calibration and alignment of the instrument were performed prior to experimentation via Rietveld analysis using the General Structure Analysis System (GSAS) code on an external powder standard at room temperature (Si 640c; National Insitute of Standards & Technology (NIST)). Diffraction patterns spanning 20°–100° in 2θ (with a 0.1° step size) were collected during 1 min time sequences/steps (or data sets) during the sintering cycles. These data sets could be summed over sequential 5 min periods to improve signal quality at the expense of time resolution. Multiple experiments were conducted at different heating rates (10°C–40°C/min), processing temperatures TP (1,080°C–1,200°C) and isothermal hold durations in a 99.998 v/o dynamic nitrogen atmosphere (Table II). Sample no. 1 consisted of a preliminary experiment performed on a pure copper powder sample. Sample no. 2 consisted of a preliminary experiment in which an alumina powder barrier layer was used to divide two separate copper and nickel powder layers within the crucible, thus preventing any copper–nickel interactions. Slow heating rates were used to allow complete temperature equilibration of the sample and to initially characterize the evolving, interaction-free nickel and copper peaks within the ND patterns using GSAS. Sample temperatures were measured using redundant thermocouples. These measurements were verified by tracking the Al2O3 peak positions, which inevitably contributed to the diffraction signal, but without overlap or interaction issues with Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
TRANSIENT LIQUID-PHASE SINTERING OF COPPER-NICKEL POWDERS: IN-SITU NEUTRON DIFFRACTION
the (200) face-centered cubic (FCC) copper/nickel peaks. The resulting shifts of the Al2O3 reflections due to thermal expansion of the inert alumina crucible thus served as a an internal temperature reference.13 It should be noted that the temperature variation from the top to the bottom of the crucible was 50°C. 14 The temperature quoted throughout this paper is that determined from the ND signal collected from the entire Al2O3 crucible and therefore represents the average temperature of the specimen. Following the ND experiments, samples were prepared using standard metallographic techniques. RESULTS AND DISCUSSION Following a presentation of preliminary experiments on the pure copper and non-interacting copper–nickel samples, the ND results will be presented according to the metallurgical stages of TLPS, namely, (1) solid-state sintering during heat-up, (2) melting/dissolution, (3) diffusional or isothermal solidification, and (4) homogenization. A description of the important TLPS stages in liquid-rich systems is described elsewhere.1,4,15
terns was collected at room temperature, all the peaks shifted to lower diffraction angles due to the thermal expansion of the copper and Al2O3 as temperature increased. Once a temperature above 1,085°C is reached, the complete disappearance of the copper (200) diffraction peak (and the entire copper diffraction pattern not shown in Figure 1) confirms that the copper phase has undergone melting. As such, the in situ ND technique is effective in detecting the copper melting event. Figure 2 illustrates ND patterns for the noninteracting copper–nickel specimen (no. 2) at selected temperatures below and above the melting point of pure copper. As in Figure 1, both the copper (200) and nickel (200) diffraction peaks shift to lower angles due to thermal expansion until the copper (200) peak disappears above the melting point of copper. Upon cooling from the high-temperature segments, the liquid copper appeared to have solidified directionally. This is evidenced by the reappearance of larger copper (220) peaks, generating a significant degree of solidification texture only in this experiment employing the Al2O3 powder barrier. The nickel (200) peak is clearly stable over the entire temperature range up to 1,178°C. In addi-
Preliminary Experiments Figure 1 illustrates a series of ND patterns, or data sets, acquired/summed over 5 min time intervals for pure copper powder heated to 1,112°C and then cooled. After a series of pat-
Figure 1. Three-dimensional plot of diffraction pattern evolution at 5 min intervals for pure copper powder heated to 1,112°C and then cooled (sample 1)
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Figure 2. 5 min ND patterns collected in situ for a Ni-65 w/o Cu non-interacting mixture (sample 2). The nickel and copper powders were separated by an alumina barrier. The mixture was slowly heated to 1,096°C, isothermally held at 1,145°C, and 1,196°C, and cooled to 27°C
51
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
TRANSIENT LIQUID-PHASE SINTERING OF COPPER-NICKEL POWDERS: IN-SITU NEUTRON DIFFRACTION
tion, upon cooling back to room temperature, the nickel (200) peak returns to its original 2θ position, as indicated by the dashed vertical line. This confirms that the alumina barrier was effective in preventing interdiffusion between the copper and nickel and that the heating and cooling of the nickel powder does not alter its 2θ diffraction line positions, other than reversible shifting due to thermal expansion. As will be shown subsequently, interaction between the nickel and copper powders during TLPS creates an evolving copper/nickel (200) diffraction profile that is significantly different from the behavior exhibited in Figure 2. Solid-State Sintering and Interdiffusion Prior to Melting Before an analysis of ND patterns from interacting powders is presented, it is important to note that the lattice parameter of copper–nickel solid solutions at a given temperature (a alloy ) varies linearly from aCu to aNi (the elemental lattice parameters);16 it is given by: aalloy = aCuCalloy + aNi (1–Calloy)
(1)
where Calloy is the fractional copper concentration in the alloy. For cubic systems such as FCC nickel and copper, the peak positions q of a given hkl reflection are given by: λ2 a2 = ———— (h2+ k2 + l2) 4sin2 θ
(2)
Substitution of equation (2) in (1) yields:
(
Calloy 1–Calloy sinθalloy = ———— + ———— sinθNi sinθCu
)
-1
(3)
In equation (3), θCu and θNi are the diffraction angles of the pure copper and pure nickel hkl reflections respectively. θalloy is the diffraction angle of the same hkl reflection corresponding to a given homogeneous alloy composition. Consequently, the 2θ axis in diffraction patterns for isomorphous copper–nickel alloys is analogous to composition (i.e., decreasing copper concentration from the copper peak position to the nickel peak position).9 Equation (3) can be used to estimate the expected (200) reflection position for a 65 w/o copper specimen (i.e., CO). Figure 3(a) shows an isolated nickel–copper contact from a 65 w/o copper mixture after solidstate sintering during heat up to 1,075°C and immediate cooling. The energy dispersive spectroscopy (EDS) line scans for copper and nickel clearly indicate that interdiffusion between pure copper and nickel particles has occurred in the neck regions. This typical interdiffusion profile illustrates the isomorphous nature of the copper–nickel binary system, where solid solutions of any intermediate composition can form. Figure 4 is a “film plot” showing the evolution of diffraction patterns over time during prolonged solid-state sintering of a 65 w/o copper powder mixture (sample 3). This graphical presentation technique17 provides a useful means of tracking the evolution of peaks (i.e., dark lines) over numerous sequential 1 min diffraction patterns (i.e., “data sets”) along the ordinate axis. This specimen was heated to 1,080°C at 40°C/min, (just below the copper melting point), held for 630 min, and cooled. The initial diffraction patterns
Figure 3. Backscattered electron images (BSE) and EDS line scans showing copper–nickel interdiffusion profiles between contacting solid-statesintered particles: (a) during heat-up to 1,075°C, and (b) after 630 min at 1,080°C
52
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
TRANSIENT LIQUID-PHASE SINTERING OF COPPER-NICKEL POWDERS: IN-SITU NEUTRON DIFFRACTION
collected at room temperature show stable copper and nickel (200) peaks at 42.24° and 44.42°, respectively, as well as Al 2O 3 peaks ((20-4) at 44.99° and (116) at 49.11°).
Figure 4. Film plot of in situ diffraction pattern evolution collected at 1 min intervals during solid-state sintering of a Ni-65 w/o Cu powder mixture heated to 1,080°C for 630 min (sample 3)
Figure 5. 5 min ND patterns collected in situ during the isothermal segment of a solid-state-sintered Ni-65 w/o Cu powder mixture at 1,080°C (sample 3). Bottom plot directly compares presintered and postsintered diffraction patterns at room temperature
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
During heat-up, all peaks shift to lower angles due to thermal expansion of the respective lattices. Vertical dashed lines have been included to mark the positions of key high-temperature peaks. Once the isothermal processing temperature is reached (TP = 1,080°C), the alumina peaks cease to shift. However, as the sample is held at this temperature, ongoing copper–nickel interdiffusion causes noticeable asymmetric broadening effects in both the copper and nickel peaks. The developing alloy regions within the sample cause inward broadening of the initial elemental copper and nickel diffraction peaks, as well as increasing diffraction intensities at intermediate angles between them over time. Upon cooling after the isothermal hold, a broad diffraction profile still exists at room temperature. However, it is clear that the most intense portion of this profile has a diffraction angle between that of the original pure copper and pure nickel peaks. As the micrograph in Figure 3(b) indicates, this broad peak is attributed to the significant compositional gradients that still remain between the nickel particle core (point c) and the sintered copper particles (point d) at the end of this experiment. Figure 5 shows time–temperature resolved patterns acquired over 5 min sequences (summations of five 1 min data sets) at discrete points during the sintering cycle to more clearly show the evolution of two dimensional patterns. The pre-sintered and post-sintered patterns (bottom) show that solid-state interdiffusion for over 600 min has generated a broad alloy peak indicative of a partially homogenized sample, Figure 1. The centroid of the alloy peak is in agreement with that expected for the copper-rich bulk composition of the sample (dashed line at CO = 65 w/o copper). While the centroid of the post-sintered peak corresponds closely to the lattice parameter for a 65 w/o copper mixture, it is clear that a significant “tail” in the peak exists toward higher 2θ angles, indicating that some nickel-rich regions are still present in the sample after 600 min at temperature. Close examination of Figure 5 reveals that even after 120 min at 1,080°C, there exists near-pure copper and nickel regions in the sample. Note that the furnace temperature rose slightly after this 2 h hold but enough interdiffusion occurred prior to this point such that liquid should not have formed up to 1,096°C (i.e., the sample still constitutes a solid-state-sintered case). At this
53
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
TRANSIENT LIQUID-PHASE SINTERING OF COPPER-NICKEL POWDERS: IN-SITU NEUTRON DIFFRACTION
slightly higher temperature, and for times ranging from 180 to 600 min, the maximum peak intensity begins to shift away from the pure-copper position and the distinct peak at the pure nickel 2θ position disappears. This is consistent with the EDS line scan of Figure 3(b) which shows a measurable copper concentration at the core of the nickel particle. Figure 6 shows diffraction patterns collected immediately prior to melting for specimens heated below 1,085°C at different heating rates (10°C and 40°C/min). In comparison, a diffraction pattern for sample 2 is included, which contained an alumina powder barrier preventing any copper and nickel interdiffusion from taking place. Sample 5, which was slowly heated at 10°C/min, appears to
Figure 6. Comparison of peak profiles for diffraction patterns collected immediately prior to melting in samples heated at different heating rates
Figure 7. Film plot of in situ diffraction pattern evolution collected at 1 min intervals during TLP sintering of a Ni-65 w/o Cu powder mixture heated to 1,128°C for 900 min (sample 5)
54
have higher diffraction intensities at intermediate angles between the pure copper and pure nickel peak positions, which would indicate a higher degree of interdiffusion and alloying. However, the difference between the 10°C and 40°C/min data is minor. This can be attributed to the limited sensitivity of this technique, and to the low degree of interdiffusion that has occurred during these short heating regimes. Melting and Dissolution Figure 7 shows the evolution of diffraction patterns during TLP sintering of sample 5. This specimen was heated to 1,128°C at 40°C/min to form a copper-rich liquid phase, held for 900 min, and cooled. This plot illustrates different sintering and interdiffusion behavior in contrast to the solidstate-sintered specimen, Figure 4. In this case, a clear melting event is observed by the removal of copper peaks, evidenced by a break in the copper (200) curve at 1,085°C (circled). The nickel peaks persist since a portion of the mixture remains solid/crystalline at the sintering temperature (i.e., the nickel particles). Immediately after melting, a slightly shifted (higher 2θ) copper-rich diffraction peak appears to grow in intensity as the purenickel intensity decreases over time. Concurrently, diffraction intensities gradually increase at intermediate 2θ angles due to interdiffusion, or solute uptake, by the solid-nickel particles. Due to reactor flux variations, 63 min of data could not be collected during the isothermal segment as well as the cool-down segment. Nonetheless, the interdiffusion process (i.e., solute uptake by solid-nickel particles) appears to be much more rapid during liquid-phase sintering vs. the solid-state sintering. This is clear when comparing the film plots of specimens 2 and 5 (Figures 4 and 7). The presence of the liquid phase during sintering at 1,128°C generates a much more uniform distribution of solute over the wetted nickel particles, which will accelerate the solute-uptake process. In order to further analyze ND pattern evolution during TLPS, it is necessary to examine time–temperature resolved data for sample 5, Figure 8. At 1,075°C, distinct copper and nickel peaks are present during what is referred to as the “premelt” condition during heat-up. At the average sample temperature of 1,078°C, the onset of copper melting is first detected and this is designated as the start of the liquid-phase regime (post-melt time = 0 Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
TRANSIENT LIQUID-PHASE SINTERING OF COPPER-NICKEL POWDERS: IN-SITU NEUTRON DIFFRACTION
min). At this point, copper regions at the top of the crucible (the hottest furnace region) have begun to melt and this causes an abrupt decrease in the pure-copper peak intensity. It should be noted that the copper peak in the pattern at t = 0 exhibits a clear shoulder at intermediate angles. This diffraction signal originates from sintered neck alloy regions of the specimen having copper-rich compositions less than or equal to the phase diagram solidus composition. Therefore, they will not melt at 1,085°C since they constitute stable solid solutions. In the 1,095°C, t = 5 min pattern, only this shifted residual shoulder peak remains, which will be referred to as the solidus peak (or CS). Isothermal Solidification After the melting event in Figure 8, the copper-
Figure 8. ND patterns collected at 5 min intervals during the isothermal solidification stage of a Ni-65 w/o Cu powder mixture sintered at 1,128°C (sample 5)
rich C S (200) peak appears to grow rapidly in intensity while the pure-nickel peak intensity decreases slowly (time segments 5 min through 60 min). This indicates that a copper-rich solid solution is rapidly forming within the two-phase liquid–solid specimen during the isother mal segment. This supports nonisothermal DSC and metallographic results, where rapid liquid removal rates at short times via isothermal solidification would leave behind a copper-rich layer (composition near CS) that surrounds the still solute-deficient nickel particles.4 As the heavy dashed line in Figure 8 indicates, the 2θ peak location for the maximum intensity of the copper -rich peak remains fixed at the CS location up to the 60 min diffraction pattern. This is due to the continued presence of liquid phase which dictates, through interface equilibrium considerations, that the solid adjacent to the liquid–solid interface remain at CS. Once all the liquid has isothermally solidified, this restriction on the interfacial solid composition is removed. Beginning with the 145 min diffraction pattern, the 2θ peak location for the maximum intensity of the copper peak begins to shift to higher 2θ until it reaches a position that corresponds to the bulk composition of the mixture, CO. This indicates that complete isothermal solidification occurred near 145 min, which is consistent with a previous analysis using DSC.4 Over the same time frame, solute diffusion deeper into the nickel particles eventually causes the nickel peaks to be completely removed after 145 min. After 900 min at 1,128°C, a sharp single peak is formed, indicating that a homogeneous composition has been established, which is in good agreement with the bulk mixture composition (CO = 65 w/o copper). As Table II indicates, multiple TLPS experiments were conducted at different isothermal processing temperatures to investigate the effects on liquid formation, dissolution, and the general ND profile
TABLE II. NEUTRON DIFFRACTION EXPERIMENTS Sample No.
Powder Mass (g)
CO (w/o Cu)
Atmosphere
1 2 3 4 5 6 7 8
2.522 2.140 2.515 2.523 2.529 2.495 2.502 2.490
100.0 64.0 65.2 65.0 65.3 65.0 65.0 64.8
N2 N2 N2 N2 N2 N3 N2 N2
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Rh (°C/min)
TP (°C)
Hold Time at TP (min)
Experiment
40 N/A 40 40 10 40 40 40
1,112 1,096–1,196 1,080 1,091 1,128 1,162 1,178 1,194
30 60 630 1,040 900 10 720 30
Pure-copper melt characterization Non-interacting, slow heating Solid-state sintering TLPS TLPS TLPS TLPS TLPS
55
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
TRANSIENT LIQUID-PHASE SINTERING OF COPPER-NICKEL POWDERS: IN-SITU NEUTRON DIFFRACTION
evolution behavior. Figure 9 shows diffraction patterns collected immediately after the onset of melting (5 min) for six samples heated to different peak temperatures. A diffraction pattern for specimen 3 is also shown, which was solid-state sintered at 1,080°C and is included for relative comparisons with the other LPS specimens. Note: the diffraction patterns for the 1,194°C and 1,178°C specimens contain molybdenum (200) peaks originating from the molybdenum sample canisters used for these particular higher -temperature experiments. Vertical dashed lines have been included to identify the pure-nickel and -copper peak positions at 1,085°C, where copper melting occurs and the pure-copper lattice becomes unstable. These lines are useful in interpreting small peak shifts during sintering at different temperatures. In Figure 9, the nickel and Al2O3 peaks shift to lower angles at higher temperatures due to thermal expansion. Conversely, the residual copperrich C S peaks (post-melt) show a distinctively different trend since they are shifted to higher 2θ angles. This is due to superimposed compositional shifting effects and the fact the pure copper is no longer stable at these temperatures. As the sintering temperature increases, the copper–nickel phase-diagram solidus indicates that the solubility of copper in nickel (C S ) decreases with increasing temperature (becoming more nickel rich). For the diffraction pattern at 1,091°C, the unmelted, residual interdiffusion regions can have very high copper contents close to pure copper. Accordingly, Figure 9 shows that the postmelt C S peak at 1,091°C is shifted minimally relative to the pure copper peak at 1,080°C. Higher temperatures show increased composition-
Figure 9. ND patterns collected 5 min after the melting event for Ni-65 w/o Cu powder mixtures isothermally held at different process temperatures
56
al shifting due to decreasing CS (or increasing nickel solubility). However, after only 5 min there are only minor differences in the patterns for each sample and a considerable fraction of the nickelparticle core remains unalloyed. Figure 10 shows diffraction patterns collected 15 min after melting and temperature stabilization. At this point the residual C S peaks have already grown significantly. At higher temperatures, the CS peaks are smaller (due to increased melt-back of solid-state alloyed regions) and the nickel peaks are also smaller due to nickel dissolution. This is most clear when comparing the total diffraction profile areas with the fully solid 1,080°C case (areas above the horizontal dashed lines). This is indicative of the decreasing total volume fraction of solid-solution material within the liquid–solid mixtures due to dissolution (increased liquid formation) at higher TP . It is also clear that there is a reduction of the intermediate alloy range ∆θ between the copper-rich and nickel peaks since CS is becoming increasingly nickel-rich (shifted to higher 2θ angles) and smaller compositional gradients are possible within the solid particles. The isothermal solidification mechanism in TLPS can be more clearly elucidated by plotting the 2θ position of the rapidly growing copper-rich solid-solution peaks at the various processing temperatures. Figure 11 shows the nickel and (CS) peak positions measured at half maximum for all specimens shortly after the melting event. The shifting peak positions of each sample are plotted according to their respective peak processing temperature (TP) on the ordinate axis. It should be noted that at short times (0 to 10 min) many of the samples were still equilibrating to their peak
Figure 10. ND patterns collected 15 min after the melting event for Ni-65 w/o Cu powder mixtures isothermally held at different process temperatures
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
TRANSIENT LIQUID-PHASE SINTERING OF COPPER-NICKEL POWDERS: IN-SITU NEUTRON DIFFRACTION
isothermal processing temperature. The presentation format in Figure 11 is somewhat similar to a film plot, but these data summarize the melting behavior of all specimens as they are heated to their respective TP from 1,080°C to 1,194°C. The peak positions at t = 5 min and 10 min have been connected for the different specimens to trace the peak positions at similar times when sintering at different TP. The nickel peaks after 5 min and 10 min shift to lower 2θ angles due to thermal expansion at higher temperature. The degree of expansion is in agreement with calculations based on the known temperature-dependent expansion of the nickel lattice (dark dashed line in Figure 11).18 The copper-rich peaks at 5 min and 10 min are also plotted shortly after the melting event. At these times, the copper peaks show a distinctly different behavior relative to the nickel peaks. Rather than shifting to lower angles due to thermal expansion, the residual peaks shift to higher angles at increasing T P due to compositional effects. A theoretical curve for the expected expansion of pure copper (if it were stable) is also included. This theoretical curve is calculated by extrapolating copper-expansion data18 and illustrates the divergent behavior of the copper peaks upon melting (thermal shifting vs. compositional shifting effects). As the typical film plot in Figure 7 shows, as well as Figure 9, all specimens exhibited this characteristic copper melting behavior above 1,085°C, namely, melting of unstable pure copper regions leaving behind copper-rich solidsolution regions responsible for the C S peaks, which grow rapidly immediately after melting. Based on the known temperature-dependent thermal-expansion coefficients of the copper and nickel lattices,18 equation 3 can be used to estimate the composition of the solid-solution alloy regions responsible for generating the CS peaks. These data are plotted vs. sintering temperature and superimposed on the copper–nickel phase diagram in Figure 12.19 The in situ results show good agreement with the solidus line of the phase diagram. Discrepancies are likely due to temperature gradients within the samples as they equilibrate to their respective TP. However, this datum affirms, in situ, that the isothermal solidification mechanism responsible for liquid removal during copper–nickel TLPS occurs via the epitaxial growth of a copper-rich solid solution (at CS) surrounding the solid base-metal particles, and limited long-range diffusion into the nickel core. Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Figure 11. Cu, Ni, and CS peak positions at t = 5 min (after melt onset) and 10 min for samples 3–8 (i.e., 1,080°C–1,094°C)
Figure 12. Calculated CS diffraction peak compositions in comparison with the copper–nickel phase-diagram solidus line19
Figure 13. ND patterns collected 180 min after the melting event in Ni-65 w/o Cu powder mixtures isothermally held at different process temperatures
Homogenization Figure 13 shows diffraction patterns for samples that were held at different T P for a much
57
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
TRANSIENT LIQUID-PHASE SINTERING OF COPPER-NICKEL POWDERS: IN-SITU NEUTRON DIFFRACTION
longer duration (180 min), where DSC results4 indicate that the liquid phase should be fully solidified. The solid-state-sintered specimen at 1,080°C still exhibits a bimodal ND profile, indicating a much less homogenous composition. In comparison, the profiles of the specimens TLP sintered above 1,085°C clearly show that the presence of a liquid phase has accelerated the evolution of a more refined single peak and therefore a more homogeneous specimen composition. The liquid phase effectively increases the net rate of interdiffusion and solute uptake into and surrounding the nickel solid-solution particles. However, Figure 13 indicates that, even after 180 min, the TLPS samples are not completely homogeneous since the peaks at each hold temperature are still somewhat broad and asymmetric. Also, the peak maxima positions are to the left (lower 2θ angles) of the expected positions for homogeneous alloys at CO. Compositional inhomogeneity is also evidenced by the nickel-rich “tail” at the right of each diffraction profile, which decreases with increasing TP. This is consistent with DSC results4 where samples held at 1,140°C for similar times (i.e., 150 min) still exhibited a broad melting event and therefore a broad compositional range. Diffraction patterns collected at longer times (as in Figure 8) show that complete homogenization occurs with further sintering at each TP . CONCLUSIONS This study has shown that ND is capable of identifying the melting event during sintering of copper and nickel powder mixtures as well as the ongoing interdiffusion process taking place during TLPS at elevated temperatures. By monitoring the evolution of copper–nickel solid-solution diffraction peak profiles in situ, this technique has shown that isothermal solidification of the liquid phase occurs primarily by the epitaxial growth of a copper-rich “layer” surrounding the nickel-rich base-metal particles. This is evidenced by the rapid growth of residual “CS” diffraction peaks immediately after the melting event, whose 2θ position is coincident with that predicted by the phase-diagram solidus. With increasing sintering times, homogenization of the powder mixtures eventually causes a sharp, single-peak profile to be generated. This characteristic profile evolution is significantly accelerated when a liquid phase is formed (i.e., TLPS vs. solid-state sintering).
58
ACKNOWLEDGMENTS This work was supported by the Canadian Neutron Beam Centre (CNBC) and the National Sciences and Engineering Research Council of Canada (NSERC). The authors would also like to thank Erik Szakaly for contributions regarding metallographic work. REFERENCES 1. R.M. German, Sintering Theory and Practice, 1996, WileyInterscience Publications, New York, NY. 2. S.F. Corbin and D.J. McIsaac, “Differential Scanning Calorimetry of the Stages of Transient Liquid Phase Sintering”, Mater. Sci. Eng. A, 2003, vol. A346, no. 1–2, pp. 132–140. 3. S.F. Corbin and P. Lucier, “Thermal Analysis of Isothermal Solidification Kinetics During Transient Liquid-Phase Sintering”, Metall. Mater. Trans. A, 2001, vol. 32A, no. 4, pp. 971–976. 4. D.M Turriff and S.F. Corbin, “Quantitative Thermal Analysis of T ransient Liquid-Phase-Sintered Cu-Ni Powders”, Metall. Mater. Trans. A, 2008, vol. 39, no. 1, pp. 28–38. 5. F.J. Puckert, W.A. Kaysser and G. Petzow, “Transient Liquid Phase Sintering of Ni-Cu”, Z. Metallkde, 1983, vol. 74, no. 11, pp. 737–743. 6. R.M. German and J.W. Dunlap, “Processing of IronTitanium Powder Mixtures by Transient Liquid Phase Sintering”, Metall. Trans. A, 1986, vol. 17A, no. 2, pp. 205–213. 7. W.H. Baek and R.M. German, “Transient Liquid Phase Sintering in the Fe-Fe2Ti System”, Int. J. Powder Metall., 1986, vol. 22, no. 4, pp. 235–244. 8. R.N. Lumley and G.B. Schaffer, “The Effect of Solubility and Particle Size on Liquid Phase Sintering”, Scripta Mater., 1996, vol. 35, no. 5, pp. 589–595. 9. B. Fisher and P.S. Rudman, “X-ray Diffraction Study of Interdiffusion of Cu-Ni Powder Compacts”, J. Applied Physics, 1961, vol. 32, no. 8, pp. 1,604–1,612. 10. R. Delhez, E.J. Mittemeijer and E.A. van den Bergen, “ Xray Diffraction Line Profile Analysis of Diffusional Homogenization in Powder Blends“, J. Mater. Sci., 1978, vol. 13, no. 8, pp. 1,671–1,679. 11. M. Potter, H. Fritzsche, D.H. Ryan and L.M.D. Cranswick, “Low-Background Single-Crystal Silicon Sample Holders for Neutron Powder Diffraction”, J. of Appl. Crystallogr., 2007, vol. 40, no. 3, pp. 489–495. 12. L.M.D. Cranswick, R. Donaberger, I.P. Swainson and Z. Tun, “Convenient Off-Line Error Quantification and Characterization of Concentricity of Two Circles of Rotation for Diffractometer Alignment”, J. of Appl. Crystallogr., 2008, vol. 41, no. 2, pp. 373–376. 13. I. Bull, P. Lightfoot, L.A. Villaescusa, L.M. Bull, R.K.B. Gover, J.S.O. Evans and R.E. Morris, “An X-ray Dif fraction and MAS NMR Study of the Ther mal Expansion Properties of Calcined Siliceous Ferrierite”, J. Am. Chem. Soc., 2003, vol. 125, pp. 4,342–4,349. 14. D.M. Turriff, “Process Kinetics of Transient Liquid Phase Sintering in a Binary-Isomorphous Alloy System”, 2007, PhD Thesis, University of Waterloo, Waterloo, ON, Canada. 15. D.M. Turriff and S.F. Corbin, “Modelling the Influences of
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
TRANSIENT LIQUID-PHASE SINTERING OF COPPER-NICKEL POWDERS: IN-SITU NEUTRON DIFFRACTION
Solid-State Interdiffusion and Dissolution on Transient Liquid Phase Sintering Kinetics in a Binary Isomorphous System”, Metall. Mater. Trans. A, 2006, vol. 37A, no. 5, pp. 1,645–1,655. 16. C.S. Barrett, Structure of Metals, Second Edition, 1952, McGraw-Hill, New York, NY. 17. B. Hinrichsen, R.E. Dinnebier and M. Jansen, “Powder3D: An Easy to Use Program for Data Reduction and Graphical Presentation of Large Numbers of Powder
Diffraction Patterns”, Z. Kristallogr. Suppl., 2006, vol. 23, pp. 231–236, 18. Y.S. Touloukian, R.K. Kirby and P.E. Taylor, Thermophysical Properties of Matter—TPRC Data Series, Vol. 12, Thermal Expansion Metallic Elements and Alloys, 1979, IFI/Plenum Press, New York, NY. 19. T.B. Massalski, editor, “Ni-Cu Binary Alloy Phase Diagram”, Binary Alloy Phase Diagrams, 1986, vol. 1, American Society for Metals, Metals Park, OH. ijpm
ERRATUM In the article “Hot Isostatic Pressing Simulation for Titanium Alloys” (Int. J. Powder Metall., 2008, vol. 44, no. 5, pp. 57–61), equation 4 should read:
{
dε11P – dε22P β = (2/3) ––––––––––– dεPij
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
}
–0.5
59
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
MEETINGS AND CONFERENCES
2009 PM-09 5TH INTERNATIONAL CONFERENCE & EXHIBITION February 16–18 Goa, India www.pmai.in/ PIM2009 INTERNATIONAL CONFERENCE ON POWDER INJECTION MOLDING & WORKSHOP ON MEDICAL APPLICATIONS OF MICRO POWDER INJECTION MOLDING March 2–5 Lake Buena Vista (Orlando), FL MPIF* LASER ADDITIVE MANUFACTURING CONFERENCE March 3–4 San Antonio, TX www.laserinstitute.org BIOMATERIALS ASIA 2009 April 5–8 Hong Kong www.biomaterialsasia.com POWDER WORLD 2009— 2009 CHINA INTERNATIONAL POWDER TECHNOLOGY & EQUIPMENT EXHIBITION AND CHINA INTERNATIONAL INDUSTRIAL POWDER RAW MATERIALS EXHIBITION April 1–3 Beijing, China www.powderworld.org/en/ MATERIAIS 2009 5TH INTERNATIONAL MATERIALS SYMPOSIUM April 5–8 Lisbon, Portugal http://www.demat.ist.utl.pt/ materiais2009/
60
17TH PLANSEE SEMINAR ON HIGH-PERFORMANCE PM MATERIALS May 25–29 Reutte, Austria www.plansee.com AEROMAT 2009 June 7–11 Dayton, OH asmcommunity.asminternational.org/content/Events/ aeromat09/ TOOL 09—TOOL STEELS June 2–4 Aachen, Germany www.tool09.rwth-aachen.de POWDERMET2009: MPIF/APMI INTERNATIONAL CONFERENCE ON POWDER METALLURGY & PARTICULATE MATERIALS June 28–July 1 Las Vegas, NV MPIF* BASIC PM SHORT COURSE July 27–29 State College, PA MPIF* THERMEC 2009: SIXTH INTERNATIONAL CONFERENCE ON ADVANCED MATERIALS AND PROCESSES August 25–29 Berlin, Germany SDMA 2009/ICSF VII—4TH INTERNATIONAL CONFERENCE ON SPRAY DEPOSITION AND MELT ATOMIZATION/7TH INTERNATIONAL CONFERENCE ON SPRAY FORMING September 7–9 Bremen, Germany www.sdma-conference.de/
EURO PM2009 INTERNATIONAL POWDER METALLURGY CONGRESS & EXHIBITION October 12–14 Copenhagen, Denmark www.epma.com/pm2009 CERAMITEC 2009 11TH INTERNATIONAL TRADE FAIR FOR MACHINERY, EQUIPMENT, PLAN, PROCESSES AND RAW MATERIALS FOR CERAMICS AND POWDER METALLURGY October 20–23 Munich, Germany www.ceramitec.de
2010 POWDERMET2010: MPIF/APMI INTERNATIONAL CONFERENCE ON POWDER METALLURGY & PARTICULATE MATERIALS June 27–30 Hollywood (Ft. Lauderdale), FL MPIF* PM2010 WORLD CONGRESS October 10–14 Florence, Italy
*Metal Powder Industries Federation 105 College Road East Princeton, New Jersey 08540-6692 USA (609) 452-7700 Fax (609) 987-8523 Visit www.mpif.org for updates and registration. Dates and locations may change
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Search
Table of Contents
Next
61
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
YEARLY CONTENTS INTERNATIONAL JOURNAL OF POWDER METALLURGY TABLE OF CONTENTS FOR VOLUME 44, NUMBERS 1–6, 2008 44/1 JANUARY/ FEBRUARY 2008 2 5 7 9
Editor's Note PM Industry News in Review PMT Spotlight On … David Rector Consultants’ Corner James G. Marsden, FAPMI
GLOBAL REVIEW 15 Powder Metallurgy in Italy O. Morandi and E. Mosca RESEARCH & DEVELOPMENT 22 Effect of Die Filling on Powder Compaction D. Korachkin, D.T. Gethin, R.W. Lewis and J.H. Tweed 35 High-Density Inconel 718: Three-Dimensional Printing Coupled with Hot Isostatic Pressing J. Sicre-Artalejo, F. Petzoldt, M. Campos and J.M. Torralba ENGINEERING & TECHNOLOGY 44 Economics of Processing Nanoscale Powders J.L. Johnson OUTSTANDING TECHNICAL PAPER FROM POWDERMET2007 55 Close-Coupled Gas Atomization: High-Frame-Rate Analysis of Spray-Cone Geometry A.M. Mullis, N.J.E. Adkins, Z. Aslam, I. McCarthy and R.F. Cochrane 65 78 79 80
DEPARTMENTS Web Site Directory Meetings and Conferences APMI Membership Application Advertisers’ Index
44/2 MARCH/APRIL 2008 2 5 7 9
FOCUS: PM Machinability 13 Machining of PM Materials: A Secondary Shaping Operation of Primary Concern C. Blais 15 Characterization of PM Machinability: Practical Approach and Analysis D. Christopherson 21 Machining of PM Steels: Effect of Additives and Sinter Hardening B. Lindsley 33 Effect of Prealloyed MnS Content and Sintered Density on Machinability and Mechanical Properties P. Boilard, G. L’Espérance and C. Blais 41 Green Machining: Parameters, Applications, and Sintered Properties É. Robert-Perron and C. Blais 49 Face Turning of PM Steels: Effect of Porosity and Carbon Level A˘. Salak, M. Selecká, K. Vasilko and H. Danninger DEPARTMENTS 62 Meetings and Conferences 63 PM Bookshelf 64 Advertisers’ Index
44/4 JULY/AUGUST 2008
43/4 MAY/JUNE 2008 2 5 9 13 19
Editor's Note PM Industry News in Review PMT Spotlight On … Stephen P. Madill Consultants’ Corner John A. Shields, Jr. Innovations Drive PM's Growth Prospects Peter K. Johnson 25 Exhibitor Showcase: PM2008 World Congress GLOBAL REVIEW 41 Powder Metallurgy in Denmark, Finland, and Sweden O. Grinder and J. Tengzelius ENGINEERING & TECHNOLOGY 57 Stainless Steel AISI Grades for PM Applications C.T. Schade, J.W. Schaberl and A. Lawley 69 Control of Defects in Powder Injection Molded Aluminum Matrix Composites F. Ahmad RESEARCH & DEVELOPMENT 77 Universal Hardness Test to Characterize PM Steels G.F. Bocchini, B. Rivolta and R. Gerosa 85 86 87 88
62
DEPARTMENTS Meetings and Conferences APMI Membership Application PM Bookshelf Advertisers’ Index
Editor's Note PM Industry News in Review PMT Spotlight On … Rajendra Kelkar Consultants’ Corner B. Pittenger
2 5 9 11 15
Editor's Note PM Industry News in Review PMT Spotlight On …Luis Bernardo Zambrano Merino Consultants’ Corner Harb S. Nayar, FAPMI 2008 APMI Fellow Awards Paul Beiss and Pierre Taubenblat 16 2008 Poster Awards H. Jorge and A.M. Cunha J. Martz, C. Braun and S.C. Johnson 20 Kempton H. Roll Powder Metallurgy Lifetime Achievement Award Arlan J. Clayton 21 2008 PM Design Excellence Awards Competition Winners P.K. Johnson RESEARCH & DEVELOPMENT 27 Consolidation of Aluminum Powder During Extrusion V.V. Dabhade, P. Kansuwan and W.Z. Misiolek GLOBAL REVIEW 37 Powder Metallurgy in India G.S. Upadhyaya HISTORICAL PROFILE 43 Tungsten Filaments—The First Modern PM Product P.K. Johnson ENGINEERING & TECHNOLOGY 49 State of the PM Industry in North America—2008 M. Paullin DEPARTMENTS 53 Book Review 55 Meetings and Conferences 56 Advertisers’ Index
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
YEARLY CONTENTS 44/5 SEPTEMBER/OCTOBER 2008 2 5 9 11 15 18 23 27 33
41 49 57 62 63 64
Editor's Note Newsmaker Joseph Tunick Strauss PM Industry News in Review PMT Spotlight On …Christopher Hammond Consultants’ Corner Myron I. Jaffe 2008 POSTER AWARDS K. Songsiri, A. Manonukul, P. Chalermkarnnon, H. Nakayama and M. Fujiwara M.N. Chikhradze and G.S. Oniashvili Axel Madsen/CPMT Scholar Reports M. Boisvert; E.M. Byrne; J. Martz; and N. Oster FOCUS: Hot Isostatic Pressing Hot Isostatic Pressing: More than a Niche Technology S.J. Mashl Diversification of Hot Isostatic Pressing Equipment Technology K. Watanabe, K. Suzuki, S. Kofune, N. Nakai, M. Yoneda, Y. Manabe and T. Fujikawa Applications for Large-Scale Prealloyed Hot Isostatically Pressed Powder Metallurgy Materials B. McTiernan Cladding of Briquetting Tools by Hot Isostatic Pressing for Wear Resistance C. Broeckmann, A. Höfter and A. Packeisen Hot Isostatic Pressing Simulation for Titanium Alloys T. Teraoku DEPARTMENTS Meetings and Conferences APMI Membership Application Advertisers’ Index
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
44/6 NOVEMBER/DECEMBER 2008 2 5 9 13 15
Editor's Note PM Industry News in Review Company Profile CMW Inc. PMT Spotlight On …John Michiels Consultants’ Corner Howard I. Sanderow
ENGINEERING & TECHNOLOGY 19 Alloy Design and Microstructure of Advanced Permanent Magnets Using Rapid Solidification and Powder Processing I.E. Anderson, R.W. McCallum and W. Tang RESEARCH & DEVELOPMENT 39 Steel-Sheet Fabrication by Tape Casting M. Rauscher, G. Besendörfer and A. Roosen 49 Transient Liquid-Phase Sintering of Copper–Nickel Powders: In Situ Neutron Diffraction D.M. Turriff, S.F. Corbin, L.M.D. Cranswick and M. Watson 60 61 62 64
DEPARTMENTS Meetings and Conferences PM Bookshelf Table of Contents: Volume 44, Numbers 1–6, 2008 Advertisers’ Index
63
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
ADVERTISERS’ INDEX
ADVERTISER
FAX
WEB SITE
PAGE
ACE IRON & METAL CO. INC.___________________(269) 342-0185 _________________________________________________________3 ACUPOWDER INTERNATIONAL, LLC _____________(908) 851-4597 ___________www.acupowder.com ___________________________37 AMERICAN CHEMET __________________________(847) 948-0811 ___________www.chemet.com ______________________________14 ARBURG GmbH + Co KG ______________________(860) 667-6522 ___________www.arburg.com _______________________________4 ASBURY CARBONS___________________________(908) 537-2908 ___________www.asbury.com _______________________________6 BÖHLER UDDEHOLM _________________________(603) 883-3101 ___________www.bucorp.com ______________________________11 ELNIK SYSTEMS _____________________________(973) 239-6066 ___________www.elnik.com ________________________________36 HOEGANAES CORPORATION ___________________(856) 786-2574 ___________www.hoeganaes.com ___________INSIDE FRONT COVER INCO SPECIAL PRODUCTS _____________________(201) 848-1022 ___________www.incosp.com _______________________________7 NORILSK NICKEL ____________________________(+ 7 495) 785 58 08 _______www.norilsknickel.com _________________________12 NORTH AMERICAN HÖGANÄS INC. ______________(814) 479-2636 ___________www.nah.com __________________INSIDE BACK COVER SCM METAL PRODUCTS, INC. __________________(919) 544-7996 ___________www.scmmetals.com ____________________________8 QMP ______________________________________(734) 953-0082 ___________www.qmp-powders.com ________________BACK COVER TIMCAL ____________________________________+41 91 873 2009 __________www.timcal.com_______________________________16
ADVERTISER’S REQUEST FOR INFORMATION FAX FORM Need more information on products or services seen in this issue? Complete the form below and fax to the advertiser(s) of your choice. Fax numbers are listed in the advertisers’ index above.
international journal of
powder metallurgy
To:___________________________________ Fax #: ____________________________________________________________________ Company: _______________________________________________________________________________________________________ Please send me more information on: __________________________________________________________________________ __________________________________________________________________________________________________________________ as advertised in the __________ issue of the International Journal of Powder Metallurgy. Please send information to: Name: Title:______________________________________________________________________________________________________ Company: _______________________________________________________________________________________________________ Address: _________________________________________________________________________________________________________ City:____________________________ State:_______________ Postal Code: ____________________________________________ Country: _________________________________________________________________________________________________________ Phone:___________________ Fax:___________________ E-Mail: ______________________________________________________
64
Volume 44, Issue 6, 2008 International Journal of Powder Metallurgy
Previous
1-Page View
2-Page View
Search
Table of Contents
Next
Previous
1-Page View
2-Page View
Search
Table of Contents