0.23226 ± 0.00031 *
0.23272 + 0.00079
-*
0.2324 ± 0 . 0 0 1 2 0.23152 + 0.00017
Average
X2/d.o.f.: 12.8/5
-0.1110'
68.3 95.5 99.5 % CL -0.12 -0.45 -0.44 -0.43 -0.42
-0.41
>
-0.40
CD
o
9Lb
X E
Figure 11. LEP and SLD measurements of gm, versus gu, compared to the Standard Model prediction.
SSSH Actff^ 0.02761 ± 0.00036 as? IYV= 91.1875 ±0.0021 GeV E^mj= 174.3 ± 5 . 1 GeV
10 20.23
0.232
0.234
sin 0 eff sin29^F. as determined from lepton and quark data is presented in Fig. 12. While the lepton data prefer a small value of sin2Ql^fl and thereby a small Higgs mass the quark asymmetries tend to larger sin29leeFf and m # values. Evaluating the average from the lepton data alone yields: sin26leeff(leptons)
= 0.23113 ± 0.00021. (15) The corresponding average from the quark asymmetries is: sin2O^fj {quarks) = 0.23230 ± 0.00029. (16) The two values differ by 3.3 standard deviations. Presently this deviation is unexplained. It could either be due to a statistical fluctuation (the error of the most precise quark asymmetry ApB is completely dominated by statistics), or due to unknown sources of systematic errors (this is unlikely due to the small systematic uncertainty correlated between the different measurements of ApB) or due to completely unexpected new physics. However, one has to keep in
Figure 12. The effective electroweak mixing angle sin26^Fr derived from d a t a depending on lepton couplings only (top) and from data depending on lepton and quark couplings (bottom). Also shown is the prediction of the Standard Model as a function of mn • The band indicates the uncertainty of t h e SM prediction due to the uncertainty of our knowledge on (5) AQS'J, mz,
and
mt.
mind that four of the nine ApB measurements shown in Fig. 9 are still preliminary. One should note that only the average of lepton and quark sin2 9^ measurements is consistent with a Higgs mass of 0(100) GeV. 3
Two Fermion Production above the Z
Two fermion production at high energies provides a beautiful laboratory for searching for new physics. Compared to other processes the cross-section for qq production is still high as shown in Fig. 13 prepared by the L3 Collaboration 23 , where the energy dependences of cross-sections for various final states in e+e~ annihilation are collected. At
356
J. Drees
Review of Final LEP Results or A Tribute to LEP The combined cross-sections and asymmetries and the results on b and c quark production have been used to study models with an additional heavy neutral Z' boson. Limits for the Z' mass have been obtained, for instance, for an E(6) x model mz1 > 0.68 TeV or for the left-right symmetric model mZ' > 0.80 TeV. In both cases the 95% confidence level lower limits are quoted and zero mixing with the Z boson is assumed. It should be remarked that the LEP2 data alone are not sufficient to constrain the mixing angle. But fits including the LEP1 data of a single experiment are consistent with zero mixing, see e.g. 26 .
Figure 13. Energy dependence of cross-sections in e + e _ annihilation. The data are from the L3 Collaboration. The cross-sections for e + e - —> qq are shown for the inclusive sample (full squares) and the non-radiative sample (open squares).
energies above the Z radiative processes are important. Due to the large cross-section for radiative return to the Z resonance only a fraction of the detected events have large s', the square of the centre-of-mass energy transferred to the / / final state. The Electroweak Working Group defines the interesting non-radiative cross-section by y/s'/s > 0.85 24 . For this cut the cross-sections for hadron, fi+fi~, T+T~ , bb, cc production have been combined. Some results are shown in Fig. 14. The lower part of the figure presents the ratio of the data divided by the SM prediction. Obviously the data are in agreement with the prediction but one should notice that the hadronic cross-section is 1.8 a high. The combined measurements of forward-backward asymmetries for /i+/x~ and T+T~ final states are collected in Fig. 15.
357
Many models for physics beyond the SM can be investigated in the general framework of four-fermion contact interactions (analogous to the low energy approximation of the weak force by Fermi theory). Using the combined data, constraints have been placed on the characteristic high energy scale A describing the low energy phenomenology of hypothetical new interactions. Limits for contact interactions between leptons range from V&rA/g > 8.5 to 26 TeV depending on the helicity coupling between initial and final state fermions and on the sign of the interference with the SM. Here g is the coupling of the new interaction. The corresponding limits for contact interactions between leptons and b quarks are \/4nA/g > 2.2 to 15 TeV, for leptons and c quarks ^/4nA/g > 1.4 to 7.2 TeV. Constraints have further been placed on the energy scale of quantum gravity in compactified extra dimensions. Including data from the Bhabha channel the typical result from the analysis of a single experiment is Ms > 1 TeV. Furthermore limits have been set on the masses of leptoquarks. The 7 — Z interference has been investigated in terms of the S-Matrix framework. In all cases no deviations from the SM expectation have been observed. Details on the two fermion analyses can be found m25>26>27>28>29. For a more
J. Drees
Review of Final LEP Results or A Tribute to LEP
preliminary
preliminary
LEP
10 zl
g o CD
10
w CO CO
o
o
a e+e_-^hadrons(Y) * eV-»u + u_ (y)
* e*e_->|iV(y) « e e ->x x (y)
O LL
* e e ^ n (y)
0
1.2 1.1
hh¥\^~
1 0.9
LL.
™ E
CO LL
<
0.2 0-
t*
•Hftjft-
< -0.2-
0.8 120
CO
140
160
180
200
220
120
Vs(GeV)
140
160
180
200
220
Vs(GeV)
Figure 14. Combined LEP measurements of the cross-sections for gc>, fJ.+fi~, T+T~ production. The curves show the SM expectation evaluated with ZFITTER. The lower part shows the ratio data to SM prediction.
Figure 15. Combined L E P results for the forwardbackward asymmetries for n+/J.~ and T+T~ final states. The curves represent the SM expectation. The lower part shows the differences between measurements and SM prediction.
complete recent summary of the two fermion data and their interpretation see 30 .
lected about 10000 W+W~ events which are analysed in terms of five decay classes: fully hadronic events where both W's decay into quarks, three semileptonic decays and fully leptonic decays. In the SM the branching ratio for the four quark class is 45.5%, for each semileptonic class 14.6%, and for the fully leptonic class 10.6%. Powerful tools to separate the four fermion events originating from W production from the background have been developed involving, for instance, neural networks. The efficiency for WW selection is high, typically around 85%, at very high purity.
4
W+W~
Production
Experimental studies of W-pair production have been a focus of the LEP2 physics programme with two main goals: the measurements of the W mass and the investigation of the structure of triple gauge boson couplings. In e+e~ annihilation double resonant W pairs are produced via the so-called CC03 diagrams shown in Fig. 16. Near threshold the cross-section is dominated by the neutrino t-channel exchange. Contributions from the more interesting s-channel exchange of a Z boson or a photon have been measured at centre-of-mass energies from 172 to 209 GeV. Each LEP experiment has finally col-
The total CC03 cross-sections measured by the four collaborations have been combined 31 , the results are summarised in Fig. 17. All experiments have published their final results for centre-of-mass energies up to
358
J. Drees
Review of Final LEP Results or A Tribute to LEP 08/07/2001
Preliminary
LEP
4
•
# '
#
*
Figure 16. CC03 diagrams for W+W production with subsequent decay into ud and /iP^.
189 GeV 32>33.34>35. The results for energies up to 207 GeV are still preliminary 36>37.38>28. Inspection of Fig. 17 immediately shows that all t- and s-channel contributions are needed to understand the data. More subtle is the comparison with predictions of the new four fermion generators Racoon WW 3 9 and YFSWW 40 with improved radiative corrections. The calculations of both programmes are based on the so-called double pole approximation for virtual 0{a) corrections in resonant W-pair production plus all other QED corrections needed for a 0.5% accuracy. It is quite remarkable that for y/s > 180 GeV:
^measured!VRacoonWW
= 1.000 ± 0 . 0 0 9 .
(17) A very similar result is obtained for the calculation with YFSWW. 4-1
Measurements of the W mass
Even before crossing the W-pair threshold a precise value of the W mass was evaluated from the LEP1 measurement of mz using SM relations. The updated indirect value obtained from a fit to all data excluding the direct W mass measurements but including the measured value of the top mass is mw = 80.368 ±0.023 GeV 4 . The small error sets the scale for all direct measurements. In the SM mw depends on electroweak loop corrections. A recent complete two-loop calculation yields the dependence on the top mass, the Higgs mass, and the QED induced shift of
359
RacoonWW / YFSWW 1.14 no ZWW vertex (Gentle 2.1) only v, exchange (Gentle 2.1)
Figure 17. The W-pair production cross-section as a function of the centre-of-mass energy compared to the predictions of the Monte Carlo generators RacoonWW and Y F S W W .
the fine structure constant Aa as expressed in Eq. (18). In the Eq. (18) only the numerically most important terms are shown, all masses are in GeV. For the complete expression see 41 . An increase of mt will increase, an increase of m # or A a will decrease the SM prediction for mw • A significant deviation of a direct measurement from the indirect value would indicate new physics and the existence of new fundamental particles. At LEP 2 two independent and complementary methods have been used to measure mw- The first is based on the measurement of the cross-section near threshold, which depends strongly on mw- Combining the measurements at a centre-of-mass energy of 161 GeV the LEP groups obtain 4 mw = 80.40±0.22 GeV, where the largest contribution to the total error is due to the low event statistics. One should remark, however, that in principle the threshold method can give a precise result, the estimated error for a GigaZ Linear Collider42 is Amw = 0.006 GeV,
J. Drees
mw
Review of Final LEP Results or A Tribute to LEP
mt = 80.3767 + 0.5235((—j-)^ - 1)
0.05613Zn(^)-1.081(
I300 K
250 200 7 150
-
Aa 0.05924
1)±....
(18)
b— 1 Signal
qqqq
E H Combi
OPAL 183-209 GeV
H i Other 1
I L d t = 677pb"'
100 50
SS^?-V-r.i^rr^.KtesK''«filc?'JSE&;i-iJ.y«£i
50
55
60
65
70
75
80
85
90 M
w
95
Figure 19. Reconstructed W mass distribution for all OPAL W+W~ -> qqqq data from y/s = 183 to 209 GeV. The histogram shows the SM expectation for Mw = 80.42 GeV.
100
(GeV/c2)
Figure 18. Reconstructed invariant mass distribution from the ALEPH experiment for the qqfiV/j, channel.
supposing that radiative corrections are controlled to this level. At higher energies the W mass is directly reconstructed from the invariant mass distribution of the decay products of the two W's. Using constraints set by energy and momentum conservation clean reconstructed mass distributions for the semileptonic and hadronic decay channels are obtained. An example from the semileptonic data taken at y/s > 202 GeV 4 3 is reproduced in Fig. 18. Note that there is practically no background in the [iv^qq channel. This also holds for eveqq channel, the background in the rvrqq and 4q channels is small. The statistical power of the data is illustrated in Fig. 19, where the mass distribution for the fully hadronic channel as reconstructed by the OPAL Collaboration is shown for all data taken at ^fs above 183 GeV 2 8 . The data are compared to the Monte Carlo prediction for mw = 80.42 GeV. From the measured masses in each event the final value of the W mass is extracted by means of sophisticated analysis techniques,
which are somewhat different for the four experiments and in each case require the comparison with a large number of Monte Carlo events. ALEPH, L3, and OPAL use a reweighting technique to determine the W mass, DELPHI uses a convolution technique. Details on the analysis of the four experiments can be found in the final publications for the data taken up to v ^ = 189 GeV 44 - 45 ' 47 or up to ^ = 183 GeV 4 6 and in more recent analyses contributed to this conference43 >48 >49. At present the precision of the combined result is limited by systematic uncertainties. They are smallest for the mass values extracted from semileptonic events. Here the total systematic uncertainty is 29 MeV with the largest contributions due to fragmentation effects, beam energy uncertainty, detector systematics, initial and final state photon radiation. The mass determination from the fully hadronic events contains additional uncertainties due to possible final state interactions between quarks originating from the decay of different W's (colour reconnection) or between hadrons (Bose-Einstein correlations). Both effects may lead to distortions in the invariant mass distribution, they
360
J. Drees
Review of Final LEP Results or A Tribute to LEP the SM prediction.
W-Boson Mass [GeV] —i —
80.454 ± 0.060
4-2
LEP2
—!
80.450 + 0.039
Average
-( i -
80.451 ± 0.033
Measuring the specific form of the nonAbelian triple gauge boson self-coupling •yWW or ZWW has been the second main goal of W physics at LEP. Assuming electromagnetic gauge invariance, charge conjugation and parity conservation and using also constraints from low energy data reduces the number of couplings from 14 in the most general case to three 51 : gf,K 7 ,A 7 which have been most intensively studied. Within the SM model these are given by 1,1,0 at tree level. They are related to the magnetic dipole moment fiw and the electric quadrupole moment qw of the W+:
pp-colliders
X2/DoF: 0.0/1
NuTeV/CCFR
— * ——
LEP-l/SLD/vN/APV
-A-
LEP1/SLD/vN/APV/mt 80
80.2
-A-
80.4
80.25 ±0.11 80.363 + 0.032 80.373 ± 0.023 80.6
m w [GeV ]
Figure 20. Direct and indirect W mass measurements.
are under study. Including such uncertainties in a conservative way, a total systematic uncertainty of 54 MeV is quoted for mw from fully hadronic events. The difference in the masses obtained from the semileptonic and fully hadronic WW decay channels is: Amw(qqqq-qqlD)
= +9 ±44 MeV.
(19)
Combining all LEP measurements 50 yields the nearly final result: mw = 80.450±0.026(stai.)±0.030(s2/st.)GeV. (20) Here the weight of the fully hadronic channel in the combined fit is only 26%. All direct and indirect W mass measurements are summarised in Fig. 20. Since not all LEP data are included yet and studies of the final state interaction effects continue it is hoped that the final LEP error will decrease to about 35 MeV. There is still agreement between the indirect determination from a fit including the measured top mass and the direct measurements of mw, but this year only within 1.9 a. The width of the W boson has also been measured at LEP: Yw = 2.150 ± 0.091 GeV. Within error there is good agreement with 361
Charged Gauge Couplings
Hw =
2mw
- ( 1 + K7
+ A 7 ),
(21) mw A deviation of K 7 or A7 from their SM values would therefore prove the presence of anomalous electromagnetic moments of the W boson and thus indicate completely new physics in the boson sector. Results have been derived using all available information from the total WW production cross-section, the polar angular distribution of the W~, the W± helicities analysed via the fermion decay angles, single W production e + e~ —> evW, and vv~i production. Within errors the measurements agree with the SM expectation with the following precision evaluated from one parameter fits to the combined data 5 2 :
<5#f = ±0.026, O~K7 = ±0.066, <5A7 = ±0.028. (22) Considering higher order effects the SM predicts small deviations from the tree level values, e.g. A K 7 ~ 0.005. Such small effects, however, are outside the scope of present experimental verification. In a more general approach the CP violating couplings have been studied by
J. Drees
Review of Final LEP Results or A Tribute to LEP
ALEPH 53 and OPAL 54 . Within errors no deviation from the SM has been observed. One should mention that limits for the quartic charged gauge couplings have been presented by ALEPH 55 , L3 5 6 , and OPAL 57 albeit with large errors. All results can be summarised by stating: no evidence has been found for any anomalous W boson coupling.
08/07/2001
LEP
]5
xm ±2.0% uncertainty •--- YFSZZ ZZTO
0.5
4-3
Preliminary
-
ZZ production
Measurements of ZZ production at T/S > 183 GeV allow an investigation of a sector of the SM not tested before. Deviations from the SM production cross-section, which is defined by the NC02 diagrams involving only t- and u-channel electron exchange, would be an indication for the existence of anomalous neutral gauge couplings absent in the SM at tree level. The ability to understand this process is also essential for the Higgs boson search, where ZZ production forms an irreducible background. All experiments have analysed ZZ decays into qqqq (4 jets), qqvv (2 jets plus missing energy), qql+l" (2 jets plus 2 isolated leptons), and l+l~l+l~. New results have been submitted to this conference 58 ' 59 ' 60 ' 28 . Since the cross-section is only about 1 pb, a factor ~ 17 smaller than the WW crosssection, the statistics is very limited. The comparison of the energy dependence of the LEP combined data to the SM prediction in Fig. 21 31 proves the agreement within the large errors of the data. The coupling of a virtual photon or Z boson to ZZ or Z7 final states is not forbidden by fundamental principles. Non SM contributions from the -y*ZZ or Z*ZZ vertex are described by j?'2 (i = 4, 5) couplings, from the 7*^7 or Z*Zy vertex by h]'Z (2 = 1,4) couplings. Experimental tools to search for such anomalous neutral triple gauge couplings are the measurement of the total ZZ or 7Z crosssection (increase at high energies?), the polar angle distribution of the produced Z or 7 (deviations at large #?), and the 7 energy dis-
, 170
\ZZ11^—,—,—I—,—,—,—,—I—,—,—,—,—L_ 180 190 200
Figure 21. LEP combined NC02 cross-sections. The curve shows the SM expectation, the band corresponds to the ± 2 % uncertainty of the prediction.
tribution. New results submitted by all LEP collaborations 61 ' 62,63 ' 64 have been combined by the Electroweak Working Group 5 2 . For CP conserving anomalous amplitudes a large interference with the SM amplitude could arise. However, no evidence for anomalous neutral couplings has been found. To give a few examples, the 95% confidence level limits for the CP conserving couplings f£, hf and /ig a r e :
fi hg hi 4-4
[-0.36, +0.39], [-0.20, +0.07], [-0.049, +0.008].
Consistency test of the SM
A consistency test of the SM can be performed by comparing the indirect and the direct measurements of the W and the top quark masses. In Fig. 22 the indirect contour has been obtained from an SM fit to the data from LEP1, SLD, neutrino nucleon scattering, and from atomic parity violation experiments 4 . Both the direct and the indirect data favour a low Higgs mass. The di-
362
Review of Final LEP Results or A Tribute to LEP
J. Drees
80.6
80.6
i
i
|
i
i
i
r
- L E P 1 . S L D , vN. APVData - L E P 2 , pp'.i?-.!:.
80.5
68% CL
80.4
3 E
>
80.3
CD
S 80.4 5
80.2
E
150
170
190
210
m. [ G e V ]
80.3
80.2
130
150
170
190
210
Figure 23. Same as Fig. 22 but with the data from summer 2000. T h e indirect result is obtained from an SM fit to the L E P 1 , SLD, and neutrino nucleon data.
mt [GeV]
Figure 22. Comparison of the indirect (full line) and the direct (dotted line) measurements of m w and mt- The diagonal band shows the SM prediction for various values of the Higgs mass ranging from 114 GeV to 1000 GeV, mH < 114 GeV has been excluded by direct searches.
rect and the indirect measurements still agree with each other though not as excellently as last year (Fig. 23). The experimental results of the direct searches for the Higgs boson are discussed by G. Hanson 65 , the theoretical aspects by F. Zwirner 66 . With no significant Higgs signal being observed, an indirect mass evaluation becomes again important. Fig. 24 presents the updated version of the traditional plot in form of a Ax 2 versus mu curve. The solid curve shows the result of the SM fit to all data from LEP and SLD, the world data on mw and mt, sin29w from the neutrino experiments CCFR and NUTEV, the measurements of atomic parity violation parameters, and also to the new direct determination of Aa^d(mz) (the contribution of the 5 quarks to the running of the fine structure constant a) from 67 . The fit confirms the preference for a low Higgs mass. The 95% confidence level
363
upper limit for rag is now 196 GeV. The dashed curve in Fig. 24 is the result of a fit with Aa^ 'rf from68 but otherwise unchanged input data and indicates the sensitivity of the rriH prediction; for details see 4 . As discussed before the b quark forwardbackward asymmetry deviates by about 3 a from its SM expectation. One may therefore ask: what is the relative importance of including A0pbB in the SM fit. The answer is given in Fig. 25, where the dotted contour line presents the 68% probability of the SM fit to all data except ApB. The preference for a low Higgs mass is even stronger, the one a contour is then completely excluded by the direct Higgs search. 5
Contributions to the C K M Matrix
LEP was part of the world wide effort to explore the structure of the CabibboKobayashi-Maskawa quark-mixing matrix. From the measurement of the W leptonic branching ratio one can determine Vcs • More important, the determination of the CKM elements Vub,Vcb, and of the ratio Vtd/Vts has been a central part of the LEP B-physics programme. Strong points of the LEP b quark studies are:
Review of Final LEP Results or A Tribute to LEP
J. Drees
- Large statistics, in total about 4 million Z —> bb decays, - fast moving B hadrons, the B hadron decay particles are well separated from the QCD rest, - tools for particle identification including K±, - experience of 12 years of data analysis. In the following only a few examples can be mentioned. A detailed summary of combined B-physics results including the data from the four LEP collaborations, from CDF and from SLD is available 69 .
CM
< 2-
Excluded
Preliminary
mH
5.1
10 [GeV]
Figure 24. A x 2 = X2 ~Xmin a s f u n c t i ° n of the Higgs mass. The solid curve presents the result of the SM fit, the band indicates the theoretical uncertainty. Also shown is the Higgs mass 95% CL exclusion limit from the direct search. The dashed curve shows an SM fit assuming a lower value of Aa^d(mz).
0.11
-
. - - All except Aj, 68% CL
0.105
\VCS\ from BR(W -> lv)
The leptonic branching fraction of the W boson is directly related to the squares of the six CKM matrix elements not depending on the t quark: = 1+[1+Mmvl}
1 3BR{W
^
lVij]2
->W) j=
d,a,b
(23) Taking the LEP average branching fraction as determined under the assumption of lepton universality yields 31 :
J2\Vij\2 = 2.039 ±0.025 ••-..
_ consistent with the value of 2 expected from unitarity. With the world average values for the other five CKM elements:
0.1
BmmmmKMttm 0.095
Excluded 10
10
10
mH [GeV] Figure 25. 68% probability contour curve in the (ApB, mu) plane obtained from an SM fit to all data except ApB. The direct measurement of A£B is shown as horizontal band of width ± 1 a. Also shown is the exclusion limit from the direct Higgs search.
(24)
V,. 1=0.996 ±0.013.
Preliminary
5.2
Inclusive measurement of \Vub
At LEP the measurement of \Vut\ relies on the inclusive reconstruction of the b —> ulv fraction:
2
\Vub\
_ BR(B -> XUW) —
> ibn
364
l^OJ
J. Drees
Review of Final LEP Results or A Tribute to LEP
r
ALEPH DELPHI
44+
L3
Here the error includes all theoretical uncertainties. The LEP value of Eq. (26) agrees very well with the most recent measurement of the CLEO Collaboration 75 . It should be mentioned that the accuracy of \Vut,\ achieved at LEP is far beyond of what was originally hoped for.
1.73 ±0.56 ±0.55
l.S7± 0.51 ±0.49 3 3 + 1 . 3 + 1.5
1
1
i
OPAL
1.63+0.57 ±0.52
i
ft
LEP Average 1
I . I2
1.67 ±0.31 ±0.42 3
4
5.3
5
B R ( B ^ X„ 1 l>) x 103
Figure 26. Measurements of the branching ratio B —> Xulv by the four LEP experiments and the resulting average. The first error is due to statistics and experimental systematics uncorrelated between experiments, the second due to all other systematic uncertainties.
where u is the average b lifetime and jb includes QCD corrections and b quark mass effects. Much progress has been made during the last years as a consequence of both, improved understanding of the theoretical uncertainties of 7t and improved experimental analysis techniques 69 . Obviously it is very difficult to separate charmless b decays from the dominant b -> c background. Several techniques have been applied in earlier publications 70 ' 71,72 based, for instance, on inclusive analysis of semileptonic decays. In a new analysis submitted to this conference the OPAL Collaboration uses 7 kinematic variables as neural net input in order to enrich the B -> Xuli> sample 73 . All measurements of the four collaborations are collected in Fig. 26. With the average branching ratio as determined by the LEP Vu\, Group: BR(B ->• Xun>i)
= (1.67 ± 0.52) x 1(T 3
and taking the world average B hadron lifetime n = (1.564 ± 0.014) ps one finds: |K*| = (4.04 t ° J 9 ) x KT 3 .
(26) 365
B° - B°
oscillations
Much progress has also been made recently in the search for £?° oscillations. The main impact on the determination of the CKM elements is explained in Eq. (27): Am. Amd
Wu mB, mBd " \Vtd\2'
e
(27)
In the ratio of the £?° and B°d mass differences Ams to Arrid many uncertainties cancel (see e.g.76) and the remaining nonperturbative quantity £2 is well known from lattice gauge theory: £2 = 1.16 ± 0.05 7 7 . No measurement of Ams has been performed yet, but upper limits have been set by each experiment applying the so-called amplitude method. The idea of the method is to replace the expression for the time dependent probability that a produced 5 ° is detected as B° by P(B°S -> B°s) = - (1 - Acos{Am s t)) e
-t/r
o
(28) and then fit the amplitude A to the data for various fixed values of Ams. Fig. 27 shows the amplitude spectrum resulting from the combination of the spectra of all LEP The combined spectrum experiments 69,78 includes the new results from DELPHI 79 ' 80 and from OPAL 81 . From the LEP data in Fig. 27 a 95% confidence level lower limit of Ams > 14.3 p s - 1 is derived. Including the data from SLD and CDF the present world limit increases to 8 2 : Ams > 14.6ps" 1 at 95% CL.
(29)
Review of Final LEP Results or A Tribute to LEP
J. Drees
Figure 27. Combined B° oscillation amplitude A as a function of A m , . The 95% CL limit derived from this spectrum is marked by the small solid triangle.
With the measured B® and B° masses and the world average value of A m j the limit for the ratio of the CKM elements is now: \Vtd\/\Vu\ < 0.22. 6
Contributions to Q C D
An important point to remember is that electroweak precision quantities depend on the strong coupling as. One of the best known examples is the ratio of the Z partial decay widths Riept, which is known to O(al) as given in Eq.(30). With the final value R°lept = 20.767 ± 0.025 (derived by assuming lepton universality) one gets the result of Eq. (31). The advantage of evaluating as from Eq. (30) is that nonperturbative corrections are suppressed and the dependence on the renormalization scale /i (which is often responsible for the dominant uncertainty of as measurements) is small. All theoretical uncertainties including the renormalization scale uncertainty amount to only +0.003, —0.001, for details see 83 . Varying mt within ± 5 GeV and ran from 100 to 1000 GeV leads to the additional small uncertainty of ±0.002. A fit
to all electroweak Z pole data from LEP and SLD and to the direct measurements of mt and mw yields: as(mz) = 0.1183 ±0.0027 4 . One may wonder whether these are the most reliable evaluations of as(mz) using the LEP data. The problem is, however, that the quoted results fully rely on the validity of the electroweak sector of the SM. Small deviations can lead to large changes. It is therefore necessary to measure as from infrared safe hadronic event shape variables like jet rates, thrust, jet mass, jet broadenings, etc. not depending on the electroweak theory. Such studies have been performed by all LEP experiments, for more recent publications see84>85>86>87. Measurements extracted by using resummed calculations in next-to-leading logarithmic approximation (NLLA) matched to 0{a2s) calculations have been combined by the LEP QCD Working Group 8 8 . As an example Fig. 28 shows as values from fits to event shape distributions at all LEP energies including measurements of the JADE Collaboration at lower energies. A fit to the combined data results in as{mz) = 0.1195 ±0.0047, where the error is almost entirely due to theoretical uncertainties (renormalization scale). The figure also indicates to which extent the running of as can be tested. All LEP as measurements using a multitude of analysis methods are collected in Fig. 29 8 9 . The three entries at the top present inclusive measurements for which perturbative calculations are known in 0(a3s). One of the most precise measurements is obtained from the ratio of the r partial decay widths RT = T(T —» hadrons + VT)/T(T —> ei>evT), the quoted value is from 90 . The figure also includes the average values from each of five different methods to extract as from hadronic event shape distributions: four jet rates, 3 jet like observables analysed in 0{a2s) using either power corrections or hadronic Monte Carlo generators for evaluating hadronisation effects, three jet like observables analysed in
366
J. Drees
Review of Final LEP Results or A Tribute to LEP
i?!L, = TJ:adrons- = 19.934 {1 + 1 . 0 4.<Xs 5^ + 0.94(^)2 - 1 5 ( ^ ) 3 } Hept Ti eptons
as{mz)
6 0.15 8
7T
= 0.124 ± 0.004(ea;p.) ± 0.002{mH,mt)
LEP/Jade 35...207GeV
7T
(30)
IT
ig;gg?(QC7D).
(31)
EWfit(5 Param. to LEP data)
NLLA+0(a2) log(R)
Miaif'lepc
R
0.14
,
4-jets O(o^) 3-jet like 0(a^)+PC
0.13
3-jet like O(a^)+Models 3-jet like NLLA i 3-jet like NLLA+0(a^) log(R)
0.12 PDG 2001
0.11
0.1
0.11 0.115 0.12 0.125 0.13 afM 7 )
A Jade T L3
• DLO (preliminary) • ADLO (preliminary) 50
0.1181±0.OO2O
100
150
Figure 29. Summary of as measurements at L E P compared t o the world average. T h e theoretical uncertainty for all 5 measurements from event shapes (ES) is evaluated by changing the renormalization scale /j, by a factor of 2.
200
E cm [GeV] Figure 28. Energy dependence of as- The d a t a are extracted from the analysis of infrared safe hadronic event shape distributions in the next-to-leading logarithmic approximation. T h e dotted curve presents the expected running of as.
pure NLLA and in matched NLLA as mentioned above. All measurements agree well with each other and with the world average. Studying QCD at LEP has several advantages: the centre-of-mass energy is high and well denned, jets are collimated, the environment is clean, statistics is high enough to investigate even rare topologies. In consequence more than 200 QCD papers have been published till now including detailed investigations of perturbation theory, hadronisation models, power corrections, quark and gluon jet fragmentation, local parton-hadron duality, soft gluon coherence etc. The exper-
367
imental aspects are reviewed, e.g. in91>92>93. Of the many new QCD studies contributed by the LEP Collaborations to this conference only few can be briefly mentioned, for instance, measurements of the colour factors and/or of as based on 4-jet events 94 ' 95 ' 96 , studies of the energy evolution of event shape distributions and of inclusive charged particle production including measurements at the highest energies compared to the prediction of hadronisation models 97 ' 98 ' 99 ' 100 ' 28 , measurements of the b quark mass at the Z mass scale 101 . As the outcome of the work at LEP one can conclude that the understanding of QCD phenomenology has much improved and even rather subtle measurements are all consistent with QCD predictions.
Review of Final LEP Results or A Tribute to LEP
J. Drees 7
Conclusion and Reflection
Table 1. Expected and achieved precision at LEP.
It is appropriate now to recall what was known in summer 1989, when LEP started and what was expected from LEP for the future. Some examples of what was known are given below:
Quantity
mz = 91.12 ±0.16 GeV, mw = 80.0 ± 0.36 GeV, sin2ew = 0.227 ± 0.006, Nv = 3.0 ± 0.9.
Achieved
50 to 20 MeV
2.1 MeV
mw
100 MeV
39 MeV
Nv
0.3
0.008
FB
0.0035
0.0013
A°'b FB
0.0050
0.0017
AT
0.0110
0.0043
A
Why was LEP so successful? Many fortunate facts had to come together: - A highly dedicated machine group responsible for the excellent performance of LEP, - low background in the detectors, - good performance of all detectors from the pilot run in August 1989 till the end of data taking, - effective division of work between CERN and the outside laboratories, - close cooperation between the 4 collaborations and also between LEP and SLD (without avoiding competition), - close cooperation between experiments and the machine group, - and, very important, close cooperation with theory groups. Many analyses are continuing and still more can be expected in the future.
error
mz
A
It was expected, of course, that LEP would improve the accuracy substantially. Looking back at the review talks presented by G. Altarelli 102 at the Lepton Photon Symposium 1989 in Stanford and by R. Barbieri 103 at the EPS Conference 1989 in Madrid one finds the expected experimental errors compared in Table 1 with those actually achieved. I should remark that the error for Nv quoted as expected is from the answer which was given by the DELPHI Collaboration to the LEPC in 1982. In the end, all measurements turned out to be much more precise than expected. Despite this precision the SM continues to be in good shape.
Expected
Acknowledgments First I would like to thank the organizing committee for giving me the opportunity to present this summary talk. Preparing such a talk is not possible without numerous communications with colleagues from the LEP Collaborations. In particular I profited much from discussions and mail exchanges with P. Antilogus, E. Barberio, R. Chierici, M. Elsing, P. Gagnon, F. Glege, M. Griinewald, J. Holt, R. Jones, M. Kienzle, N. Kjaer, W. Liebig, K. Monig, S. Myers, C. Parkes, A. Stocchi, H. Voss, Ch. Weiser and D. Wicke. I am very grateful to G. Myatt for many helpful suggestions and also for carefully reading the manuscript. I thank S. Braccini for his assistance during the conference and during finalising the manuscript.
References
368
1. S. Myers, "The LEP Machine", talk presented at the LEP Fest October 2000, http://cern.web.cern.ch/CERN/ Divisions/SL/publications/talks/. 2. Design Study of a 22 to 130 GeV e + e~ Colliding Beam Machine (LEP), CERN/ISR-LEP/79-33, CERN (1979). 3. Precision Electroweak Measurements on the Z Resonance; The ALEPH, DELPHI, L3, OPAL and SLD Collaborations, the LEP Electroweak Working
J. Drees
4.
5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
21.
Review of Final LEP Results or A Tribute to LEP
Group, and the SLD Electroweak and Heavy Flavor Groups, Review article in preparation. The LEP Collaborations ALEPH, DELPHI, L3, OPAL, the LEP Electroweak Working Group and the SLD Electroweak and Heavy Flavor Groups, CERN-EP/2001-021, February 2001, updates are available on website: http://lepewwg.web.cern.ch/LEPEWWG. ALEPH Collaboration, R. Barate et al., Euro. Phys. J. C 14 (2000) 1. DELPHI Collaboration, P. Abreu et al., Euro. Phys. J. C 16 (2000) 371. L3 Collaboration, M. Acciarri et al., Euro. Phys. J. C 16 (2000) 1. OPAL Collaboration, G. Abbiendi et al., Euro. Phys. J. C19 (2001) 587. ALEPH Collaboration, A. Heister et al., Eur. Phys. J. C20 (2001) 401. DELPHI Collaboration, P. Abreu et al., Euro. Phys. J. C 14 (2000) 585. L3 Collaboration, M. Acciarri et al., Phys. Lett. B 429 (1998) 387. OPAL Collaboration, G. Abbiendi et al., Euro. Phys. J. C21 (2001) 1. SLD Collaboration, K. Abe et al., Phys. Rev. Lett. 84 (2000) 5945. SLD Collaboration, K. Abe et al., Phys. Rev. Lett. 86 (2001) 1162. ALEPH Collaboration, R. Barate et al., Phys. Lett. B 401 (1997) 163. DELPHI Collaboration, P. Abreu et al., Euro. Phys. J. C 10 (1999) 415. L3 Collaboration, M. Acciarri et al., Euro. Phys. J. C13 (2000) 47. OPAL Collaboration, G. Abbiendi et al., Euro. Phys. J. C 8 (1999) 217. M. Con very, "Recent Results from SLD", SLAC-PUB-8781 (2001). ALEPH Collaboration, H.Heister et al., "Measurement of AbFB using Inclusive b-hadron Decays", CERN-EP/2001-047, LP01 paper 217. DELPHI Collaboration, K. Munich et al., "Determination of AbFB using inclu-
369
22.
23. 24.
25.
26.
27.
28.
29.
30.
31.
sive charge reconstruction and lifetime tagging at LEP1", DELPHI 2001-048 CONF 476, LP01 paper 198. M. L. Swartz, "Precision Electroweak Physics at the Z", Proceedings of the XIX International Symposium on Lepton and Photon Interactions at High Energies, Stanford (1999) 307. L3 Annual Report to CERN (2000) LepEWWG / / S u b g r o u p , G. Geweniger et al., "Combination of the LEP II / / Results", LEP2FF/01-02 September (2001); updates can be found under http://www.cern.ch/LEPEWWG/lep2/. ALEPH Collaboration, "Fermion Pair Production in e + e~ Collisions at high energy and Limits on Physics beyond the Standard Model", ALEPH 2001-019 CONF 2001-016, LP01 paper 252. DELPHI Collaboration, A.Behrmann et al., "Results on Fermion-Pair Production at LEP running in 2000", DELPHI 2001-094 CONF 522, LP01 paper 151. L3 Collaboration, "Preliminary L3 Results on Fermion-Pair Production in 2000", L3 Note 2648, LP01 paper 673; and "Search for New Physics Phenomena in Fermion-Pair Production in e + e~ Collisions at Centre-of-Mass Energies up to 209 GeV", L3 Note 2647, LP01 paper 672. OPAL Collaboration, "Measurement of Standard Model Processes in e+e~ Collisions at v ^ ~ 203 - 209 GeV", OPAL Physics Note PN469 (2001), LP01 paper 28. OPAL Collaboration, "Limits on Low Scale Quantum Gravity in Extra Spatial Dimensions from Measurements of e+e" -+ e+e~ at LEP2", OPAL Physics Note PN471 (2001), LP01 paper 33. J. Holt, "Fermion Production above the Z resonance", Proceedings of the International Europhysics Conference on High Energy Physics, Budapest 2001. The LEP Collaborations ALEPH, DEL-
J. Drees
32. 33. 34. 35. 36.
37.
38.
39.
40.
41. 42.
Review of Final LEP Results or A Tribute to LEP
PHI, L3, OPAL, and the LEP WW Working Group, "LEP W-pair, Zpair and Single W Cross-Section Results for the Summer 2001 Conferences" , LEPEWWG/XSEC/2001-03, August 2001, updates are available on website: http://lepewwg.web.cern.ch /LEPEWWG/lepww/4f/. ALEPH Collaboration, R. Barate et al., Phys. Lett. B484 (2000) 205. DELPHI Collaboration, P. Abreu et al., Phys. Lett. B479 (2000) 89. L3 Collaboration, M. Acciarri et al., Phys. Lett. B496 (2000) 19. OPAL Collaboration, G. Abbiendi et al., Phys. Lett. B493 (2000) 249. ALEPH Collaboration, "Measurement of W-pair production and W branching ratios in e + e~ collisions up to 208 GeV", ALEPH 2001-013 CONF 2001010, LP01 paper 257. DELPHI Collaboration, P.Buschmann et al.,"Measurement of the W-Pair Production Cross Section and W Branching Ratios at y/s = 205 and 207 GeV", DELPHI 2001-104 CONF 532, LP01 paper 161. L3 Collaboration, "Preliminary Results on the Measurement of the W-Pair Cross Sections in e+e~ Interactions at yfs — 205 and 208 GeV and W-Decay Branching Fractions", L3 Note 2638 (2001), LP01 paper 664. A. Denner, S. Dittmaier, M. Roth and D. Wackeroth, Phys. Lett. B475 (2000) 127. S. Jadach, W. Placzek, M. Skrzypek, B.F.L. Ward, Z. Was, "The Monte Carlo Event Generator YFSWW3 version 1.16 for W-Pair Production and Decay at LEP2/LC Energies", CERN-TH/2001017 (2001), and refences quoted therein. A. Freitas, W. Hollik, W. Walter, and G. Weiglein, Phys. Lett. B495 (2000) 338. J.A. Aguilar-Saavedra et al., TESLA Technical Design Report, Part III,
43.
44. 45. 46. 47. 48.
49.
50.
51.
52.
370
DESY 2001-011, ECFA 2001-209, March 2001. ALEPH Collaboration, "Preliminary Measurement of the W Mass and Width in e+e~ Collisions at ^fs between 192 and 208 GeV", ALEPH 2001-020 CONF 2001-017, LP01 paper 256. ALEPH Collaboration, R. Barate et al., Eur. Phys. J. C17 (2000) 241. DELPHI Collaboration, P. Abreu et al., Phys. Lett. B511 (2001) 159. L3 Collaboration, M. Acciarri et al., Phys. Lett. B454 (1999) 386. OPAL Collaboration, G. Abbiendi et al., Phys. Lett. B507 (2001) 29. DELPHI Collaboration, D. Bloch et al., "Measurement of the mass and width of the W Boson in e + e~ collisions at V^ ~ 192 - 209 GeV", DELPHI 2001103 CONF 531, LP01 paper 160. L3 Collaboration, "Preliminary results on the Measurement of Mass and Width of the W Boson at LEP", L3 Note 2637 (2001), LP01 paper 665. The LEP Collaborations ALEPH, DELPHI, L3, OPAL and the LEP W Working Group, "Combined Preliminary Results on the Mass and Width of the W Boson Measured by the LEP Experiments", LEPEWWG/MASS/200101 (2001), updates are available under http://lepewwg.web.cern.ch/LEPEWWG /lepww/mw/. G. Gounaris et al., "Physics at LEP2", CERN 96-01, eds. G. Altarelli, T. Sjostrand, F. Zwirner, Vol. 1(1996) 525. The LEP Collaborations ALEPH, DELPHI, L3, OPAL and the LEP GC Working Group, "Combined Results for Electroweak Gauge Boson Couplings Measured on the LEP Experiments", LEPEWWG/TGC/2001-02 (2001), see also LEPEWWG/TGC/200002 (2000), updates are available under http://lepewwg.web.cern.ch/LEPEWWG /lepww/tgc/.
J. Drees
Review of Final LEP Results or A Tribute to LEP
53. ALEPH Collaboration, "Measurement of Triple Gauge-Boson Couplings in e+e~ collisions up to 208 GeV", ALEPH 2001-027 CONF 2001-021, LP01 paper 265. 54. OPAL Collaboration, G.Abbiendi et al., " Measurement of W boson polarisations and CP-violating triple gauge couplings from W+W~ production at LEP", CERN-EP/2000-113 (2000), LP01 paper 39. 55. ALEPH Collaboration, "Constraints on Anomalous Quartic Gauge boson Couplings from photon pair events at 189202 GeV", LP01 paper 262. 56. L3 Collaboration, "Measurement of the W+W~/y Cross Section and Direct Limits on Anomalous Quartic Gauge Boson Couplings at LEP", L3 Note 2675 (2001), LP01 paper 668. 57. OPAL Collaboration, G. Abbiendi et al., Phys. Lett. B471 (1999) 293. 58. ALEPH Collaboration, "Measurement of e+e~ -¥ ZZ Production Cross Section", ALEPH 2001-006 CONF 2001003, LP01 paper 258. 59. DELPHI Collaboration, G. Borisov et al., "Update of the ZZ cross-section measurement in e+e~ interactions using data collected in 2000", DELPHI 2001105 CONF 533 (2001), LP01 paper 162. 60. L3 Collaboration, "Z Boson Pair Production at LEP", L3 Note 2696 (2001), LP01 paper 675. 61. ALEPH Collaboration, "Constraints on anomalous neutral gauge boson couplings using data fromm ZZ production between 183 and 207 GeV", ALEPH 2001-014 CONF 2001-011, LP01 paper 261. 62. DELPHI Collaboration, P.Bambade et al., "Study of Trilinear Gauge Boson Couplings ZZZ, ZZ 7 and Z77", DELPHI 2001-097 CONF 525, LP01 paper 154. 63. L3 Collaboration, "Search for Anomalous ZZ7 and Z77 couplings in the pro-
371
64. 65. 66. 67. 68. 69.
70. 71. 72. 73.
74. 75. 76.
77. 78. 79.
80.
cess e+e" -> Z-y at LEP", L3 Note 2672 (2001), LP01 paper 676. OPAL Collaboration, G.Abbiendi et al., Eur. Phys. J. C I 7 (2000) 553. G. Hanson, "Searches for new particles", these proceedings. F. Zwirner, "The Higgs puzzle: experiment and theory", these proceedings. H. Burkhardt and B. Pietrzyk, Phys. Lett. B513 (2001) 46. A. D. Martin, J. Outhwaite, and M. G. Ryskin, Phys. Lett. B492 (2000) 69. ALEPH, CDF, DELPHI, L3, OPAL, SLD Collaborations, "Combined results on b-hadron production rates and decay properties", CERN-EP/2001-050 (2001), updates are available under http://lephfs.web.cern.ch/LEPHFS/. ALEPH Collaboration, R. Barate et al., Eur. Phys. J. C6 (1999) 555. DELPHI Collaboration, P. Abreu et al., Phys. Lett. B478 (2000) 14. L3 Collaboration, M. Acciarri et al., Phys. Lett. B436 (1998) 174. OPAL Collaboration, G. Abbiendi et al., "Measurement of |Vuf,| using b hadron semileptonic decay", CERN-EP/2001044 (2001), LP01 paper 53. http://battagl.home.cern.ch /battagl/vub/vub.html D. Cassel, "CLEO results: B decays, CP violation", these proceedings. Particle Data Group, Review of Particle Physics, D.E. Groom et al., Eur. Phys. J. C15 (2000) 1. C.T. Sachrajda, "Phenomenology from lattice QCD", these proceedings. Working Group on B oscillations, http://lepbosc.web.cern.ch/LEPBOSC/. DELPHI Collaboration, T. Allmendinger et al., "Search for B°s - B° oscillations in inclusive samples", DELPHI 2001-054 CONF 482, LP01 paper 204. DELPHI Collaboration, P. Kluit et al., "Search for B°s - B°s oscillations in DELPHI using high pt leptons", DELPHI
J. Drees
81.
82.
83. 84. 85. 86. 87. 88.
89.
90.
91. 92.
93.
94.
Review of Final LEP Results or A Tribute to LEP
2001-055 CONF 483, LPOl paper 205. OPAL Collaboration, G. Abbiendi et al., Eur. Phys. J. C19 (2001) 241, LPOl paper 50. All references on B° and £?° oscillation analyses can be found under http://lepbosc.web.cern.ch /LEPB OSC /references/. S. Bethke, "Determination of the QCD Coupling a s ", / . Phys. G26 (2000) R27. ALEPH Collaboration, R. Barate et al. Phys. Rep. 294 (1998) 1. DELPHI Collaboration, P. Abreu et al., Eur. Phys. J. C14 (2000) 557. L3 Collaboration, M. Acciarri et al., Phys. Lett. B489 (2000) 65. OPAL Collaboration, G. Abbiendi et al., Eur. Phys. J. C16 (2000) 185. The LEP QCD Working Group, "Preliminary Combination of as Values derived from Event Shape Variables at LEP", ALEPH 01-038 Physic 01-012, DELPHI 2001-043 PHYS 893, L3 Note 2661, OPAL TN689, 30.April 2001. D. Wicke, "Event shape studies at LEP", Proceedings of the International Europhysics Conference on High Energy Physics, Budapest 2001. A. Pich, "Tau Physics", Proceedings of the XIX International Symposium on Lepton and Photon Interactions at High Energies, Stanford (1999) 157. O. Biebel, Phys. Rep. 340 (2001) 165. S. Bethke, "QCD at LEP", talk presented at the LEP Fest October 2000, http://opal.web.cern.ch/OPAL/ talks / siggi Jepfest. D. Duchesneau, "Experimental Aspects of QCD in e+e~ Collisions", Proceedings of the 29th International Copnference on High Energy Physics, Vancouver (1998) 263. ALEPH Collaboration, "Simultaneous Measurement of the Strong Coupling Constant and the QCD Coulor Factors from 4-jet hadronic Z decays", ALEPH
2001-042 CONF 2001-026, LPOl paper 208. 95. DELPHI Collaboration, U. Flagmeyer et al., "Measurement of the strong coupling as and its energy dependence from the four jet rate of hadronic events with the DELPHI detector", DELPHI 2001-059 CONF 487, LPOl paper 164. 96. OPAL Collaboration, G. Abbiendi et al., "A Simultaneous Measurement of QCD Colour Factors and the Strong Coupling", CERN EP-2001-001, LPOl paper 24. 97. ALEPH Collaboration, "QCD Measurements in e+e~ Annihilations at 206 GeV", ALEPH 2001-007 CONF 2001004, LPOl paper 210. 98. DELPHI Collaboration, R. Reinhardt et al., "A study of the energy evolution of event shape distributions and their means with the DELPHI Detector at LEP", DELPHI 2001-062 CONF 490, LPOl paper 166. 99. DELPHI Collaboration, O. Passon et al., "QCD Results from the DELPHI Measurements of Event Shape and Inclusive Particle Distributions at the highest LEP energies", DELPHI 2001-065 CONF 493, LPOl paper 168. 100. L3 Collaboration, "QCD Results at 192 < Js < 208 GeV", LPOl paper 682. 101. OPAL Collaboration, G. Abbiendi et al., "Determination of the b Quark Mass at the Z Mass Scale", CERN-EP-2001034, LPOl paper 188. 102. G. Altarelli, "Theory of Precision Electroweak Experiments" Proceedings of the 1989 International Symposium on Lepton and Photon Interactions at High Energies, Stanford University (1989) 286. 103. R. Barbieri, "Electroweak Physics", Nucl. Phys. B (Proc. Suppl.) 16 (1990) 71.
372
J. Drees
Review of Final LEP Results or A Tribute to LEP
Discussion Alberto Sirlin, New York University: I have an observation and a question, i) With respect to the evidence for genuine electroweak corrections, I think there is a very simple argument that shows a very large signal. It consists of measuring the radiative correction Ar by using the experimental results for raw and mz, and comparing with the value Ar would have if the only contribution arose from the running of a. Last time I did this, about a year ago, I found a difference amounting to many standard deviations, ii) The question is: what is the \ 2 per degrees of freedom of the most recent electroweak global fit? J. Drees: The most recent MSM fit to all electroweak data including the direct measurements of mw and mt has X2 /ndf = 22.9/15 corresponding to the still reasonable probability of 8.6%.
373
DECLINE A N D FALL OF T H E S T A N D A R D MODEL? JOHN ELLIS Theoretical Physics Division, CERN, CH-1211 Geneva 23 E-mail: [email protected] Motivations for physics beyond the Standard Model are reviewed, with particular emphasis on supersymmetry at the TeV scale. Constraints on the minimal supersyymetric extension of the Standard Model with universal soft supersymmetry-breaking terms (CMSSM) are discussed. These are also combined with the supersymmetric interpretation of the anomalous magnetic moment of the muon. The prospects for observing supersymmetry at accelerators are reviewed using benchmark scenarios to focus the discussion. Prospects for other experiments including the detection of cold dark matter, ji —> e-y and related processes, as well as proton decay are also discussed.
1
Introduction
The empire of the Standard Model has resisted all attacks by accelerator data. Nevertheless, we theorists are driven to overcome our ignorance of the barbarian territory beyond its frontiers. In the gauge sector, the Standard Model has three independent gauge couplings and (potentially) a CP-violating phase in QCD. In the Yukawa sector, it has six random-seeming quark masses, three charged-lepton masses, three weak mixing angles and the Kobayashi-Maskawa phase. Finally, the symmetry-breaking sector has at least two free parameters. Moreover, this list of 19 parameters in the Standard Model begs the more fundamental questions of the origins of the particle quantum numbers. As if this were not enough, non-accelerator neutrino experiments x now convince us that we need three neutrino mass parameters, three neutrino mixing angles and three CP-violating phases in the neutrino sector: one observable in oscillation experiments and two that affect PPov experiments, without even talking about the mechanism of neutrino mass generation. Moreover, we should not forget about gravity, with at least two parameters to understand: Newton's constant Gjv = nip1 ~ (10 19 G e V ) - 2 and the cosmological 'constant', which recent data suggest is non-
zero 2 , and may not even be constant. Talking of cosmology, we would need at least one extra parameter to produce an inflationary potential, and at least one other to generate the baryon asymmetry, which cannot be explained within the Standard Model. Confronted by our ignorance of so much barbarian territory, we legions of theorists organize our explorations on three main fronts: unification - the quest for a single framework for all gauge interactions, flavour - the quest for explanations of the proliferation of quark and lepton types, their mixings and CP violating phases, and mass - the quest for the origin of particle masses and an explanation why they are so much smaller than the Planck mass mp ~ 10 19 GeV. Beyond all these beyonds, other scouting parties of theorists seek a Theory of Everything that includes gravity, reconciles it with quantum mechanics, explains the origin of space-time and why we live in four dimensions (if we do so). Physics beyond the Standard Model is therefore a very broad subject. However, many aspects are discussed here by other speakers: electroweak flavour physics 3 , CP violation 4 , the Higgs sector 5 , g^ — 2 6 , searches for new particles 7 , neutrinos 8 , dark matter 9 , strings and extra dimensions 10 . Therefore, in this talk I seek a complemen-
374
John Ellis
Decline and Fall of the Standard Model?
tary approach. For reasons that I describe in Section 2, many theorists believe that supersymmetry is the inescapable framework for discussing physics at the TeV scale and beyond. In the rest of this talk, I first discuss the constraints imposed on (the simplest) supersymmetric models by the available experimental and cosmological constraints, then address the prospects for understanding
2
The Electroweak Vacuum
Figure 1. Predictions for the radiative corrections £j in the Standard Model and a minimal one-generation m o d e l 1 4 are compared with the precision electroweak data 1 3 .
The generation of particle masses requires the breaking of gauge symmetry in the vacuum:
consistent with the data, but for now we focus on elementary Higgs models.
mw,z / 0 « < 0|X J i / s |0 > ^ 0
Within this framework, the data favour a relatively light Higgs boson, with m # ~ 115 GeV, just above the exclusion unit provided by direct searches at LEP, being the 'mostprobable' 15 . This is one reason why many theorists were excited by the possible sighting during the last days of LEP of a Higgs boson, with a preferred mass of 115.6 GeV 7 . If this were to be confirmed, it would suggest that the Standard Model breaks down at some relatively low energy ^ 10 3 TeV 16 . As seen in Fig. 2, above this scale the effective Higgs potential of the Standard Model becomes unstable as the quartic Higgs self-coupling is driven negative by radiative corrections due to the relatively heavy top quark 17 . This is not necessarily a disaster, and it is possible that the present electroweak vacuum might be metastable, provided that its lifetime is longer than the age of the Universe 18 . However, we would surely feel more secure if such instability could be avoided.
(1)
for some field X with isospin I and third component 73. The measured ratio P=
m
\ Q ^1 (2) mz cosz o\y tells us that X mainly has I = 1/2 n , which is also what is needed to generate fermion masses. The key question is the nature of the field X: is it elementary or composite? A fermion-antifermion condensate v = < 0|X|0 > = < 0\FF\0 > ^ 0 would be analogous to what we know from QCD, where < 0|gqjO >^= 0, and conventional superconductivity, where < 0|e~e~|0 >^= 0. However, analogous 'technicolour' models of electroweak symmetry breaking 12 fail to fit the values of the radiative corrections e^ to p and other quantities extracted from the precision electroweak data provided by LEP and other experiments, as seen in Fig. 1 13 . One cannot exclude the possibility that some calculable variant of technicolour might emerge that is 2
375
This may be done by introducing new
John Ellis
Decline and Fall of the Standard Model?
u I—,—1—.—1—^J_i—,—1—,—1—,—1—,—1—.—I 10 2 10 6 10" 10» 10'° 10' 2 10" 10'e 10'8 |i (GeV) 5
Figure 2. T h e range allowed for t h e mass of t h e Higgs boson if the Standard Model is to remain valid up to a given scale A. In the upper p a r t of the plane, the effective potential blows up, whereas in the lower part the present electroweak vacuum is unstable i r .
bosons (f> coupled to the Higgs field \22\H\2 \4>\2 : M2=X22v2
16
1—i—,—1—1—1—1—1—1—1—1—'—r—'—1—•—1-7-
102
10 4
106
10 s
10 10 10' 2 (L (GeV)
10"
10 i e
10 1S
Figure 3. (a) If the quartic coupling Mo (3) is too large, the effective potential blows up (solid line), whereas it is unstable if Mo is too small (dotted line), indicating a need for fine tuning, (b) This occurs naturally in a supersymmetric model (solid line) but not if the H are omitted (dotted line) 1 6 .
: (3)
As seen in Fig. 3a, the effective potential is very sensitive to the coupling parameter M 0 : for M0 < 70.9 GeV in this example, the potential still collapses, whereas for Mo > 71.0 GeV the potential blows up instead. Thus the bosonic coupling (3) must be finely tuned 16 . This occurs naturally in supersymmetry, in which the Higgs bosons are accompanied by fermionic partners H. As seen in Fig. 3b, again the Higgs coupling blows up in the absence of the H, whereas it is well behaved in the minimal supersymmetric extension of the Standard Model (MSSM). The avoidance of fine tuning has long been the primary motivation for supersymmetry at the TeV scale 19 . This issue is normally formulated in connection with the hierarchy problem: why/how is mw "C mp, or equivalently why is Gp ~ 1 / m ^ » GN = 1/rrip, or equivalently why does the Coulomb potential in an atom dominate over the Newton potential, e 2 ^> G^mvme ~ (m/mp)2, where the proton and electron masses? One might think naively
that it would be sufficient to set mw -C mp by hand. However, radiative corrections tend to destroy this hierarchy. For example, oneloop diagrams generate 5m2w = O ( £ ) A2 » m2w
(4)
where A is a cut-off representing the appearance of new physics, and the inequality in (4) applies if A ~ 10 3 TeV, and even more so if A ~ triGUT ~ 10 16 GeV or ~ rnp ~ 10 19 GeV. If the radiative corrections to a physical quantity are much larger than its measured values, obtaining the latter requires strong cancellations, which in general require fine tuning of the bare input parameters. However, the necessary cancellations are natural in supersymmetry, where one has equal numbers of bosons B and fermions F with equal couplings, so that (4) is replaced by 6m2w = O ( ^ )
\m% - m%\ .
(5)
The residual radiative correction is naturally
376
John Ellis
Decline and Fall of the Standard Model?
small if 2
\m% - m F\ < 1 TeV
2
(6)
Note that this argument is logically distinct from that in the previous paragraph. There supersymmetry was motivated by the control of logarithmic divergences, and here by the absence of quadratic divergences.
3
The M S S M
The MSSM has the same gauge interactions as the Standard Model, and similar Yukawa couplings. A key difference is the necessity of two Higgs doublets, in order t o give masses to all the quarks and leptons, and to cancel triangle anomalies. This duplication is important for phenomenology: it means that there are five physical Higgs bosons, two charged H± and three neutral h, H, A. Their quartic self-interactions are determined by the gauge interactions, solving the vacuum instability problem mentioned above and limiting the possible mass of the lightest neutral Higgs boson. However, the doubling of the Higgs multiplets introduces two new parameters: tan /3, the ratio of Higgs vacuum expectation values and /i, a parameter mixing the two Higgs doublets. There are two key experimental hints in favour of supersymmetry. One is provided by the LEP measurements of the gauge couplings, that are in very good agreement with supersymmetric GUTs 20 if sparticles weigh ~ 1 TeV. This agreement appears completely fortuitous in composite Higgs models 12, and is difficult (though not impossible 21 ) to reproduce accurately in models with large extra dimensions 22 . The other experimental hint is provided by the preference of the precision electroweak data for a relatively light Higgs boson 15 . In the MSSM, one predicts nih ^ 130 GeV 23,5 , right in the preferred range, whereas composite Higgs model generally predict heavier effective Higgs masses.
377
The gauge symmetries of the MSSM would permit the inclusion of interactions that violate baryon number and/or lepton number 24 : XLLEC + X'QDCL + A" UCDCDC
(7)
where the L(Q) are left-handed lepton (quark) doublets and the EC(DC, Ud) are conjugates of the right-handed lepton (quark) singlets. Their possible appearance is ignored in this talk, in which case the lightest supersymmetric particle is stable, and hence a candidate for dark matter 25 . In the following this is assumed to be a neutralino, i.e., a mixture of the 7, H and Z. The final ingredient in the MSSM is the soft supersymmetry breaking, in the form of scalar masses mo, gaugino masses m ^ and trilinear couplings A 2 6 . These are presumed to be inputs from physics at some high-energy scale, e.g., from some supergravity or superstring theory, which then evolve down to lower energy scale according to well-known renormalization-group equations. In the case of the Higgs multiplets, this renormalization can drive the effective mass-squared negative, triggering electroweak symmetry weaking 27 . In this talk, it is assumed that the TOO are universal at the input scale a , as are the rn\j2 and A parameters. In this case the free parameters are mo,mi/2,A
and
tan/3 ,
(8)
with pi being determined by the electroweak vacuum conditions, up to a sign. This constrained MSSM (CMSSM) serves as the basis for the subsequent discussion. It has the merit of being sufficiently specific that the different phenomenological constraints can be combined meaningfully. On "Universality between the squarks and sleptons of different generations is motivated by upper limits on flavour-changing neutral interactions 2 8 , but universality between the soft masses of the L, EC,QC, Dc and Uc is not so well motivated.
Decline and Fall of the Standard Model?
John Ellis the other hand, it is just one of the phenomenological possibilities offered by supersymmetry 29 .
200
4
tan p=10. u > 0
C o n s t r a i n t s on t h e C M S S M > O
Important constraints on the CMSSM parameter space are provided by direct searches at LEP and the Tevatron collider 7 , as seen in Fig. 4. One of these is the limit mx± ^ 103 GeV provided by chargino searches at LEP, where the third significant figure depends on other CMSSM parameters. LEP has also provided lower limits on slepton masses, of which the strongest is mj <; 99 GeV, again depending only sightly on the other CMSSM parameters, as long as m^ — mx ^ 10 GeV. The most important constraints on the u,d,s,c,b squarks and gluinos are provided by the Tevatron collider: for equal masses rrig = rrig ^ 300 GeV. In the case of the i, LEP provides the most stringent limit when m^ — mx is small, and the Tevatron for larger mj — mx. Their effect is almost to exclude the range of parameter space where electroweak baryogenesis is possible 30 . Another important constraint is provided by the LEP limit on the Higgs mass: mH > 114.1 GeV. This holds in the Standard Model, for the lightest Higgs boson h in the general MSSM for tan/3 ^ 5, and in the CMSSM for all tan /?, at least as long as CP is conserved b. Since rrih is sensitive to sparticle masses, particularly m^, via loop corrections: 5m\ oc
+
(9)
the Higgs limit also imposes important constraints on the CMSSM parameters, principally mi/2 as seen in Fig. 4.
III
. =
lll.(lll-
in
-H.iOeV..; il
•
•-•;«•$••>''
100
^lfl%«l 100 200 300 400 500 600 700 800
mm (GeV) 200
>
o 100
100
200
300 400
500
600
700
800
m]/2(GeV)
Figure 4. Compilations of phenomenological constraints on the CMSSM for tan j3 = 10 and (a) fj, > 0, (b) n < 0. Representative contours of the selectron, chargino and Higgs masses are indicated, as is the likely physics reach of Run II of the Tevatron Collider in (a). The dark shaded regions are excluded because the LSP is charged, whereas a neutralino LSP has acceptable relic density (10) in the lightshaded regions 3 1 . The medium-shaded region in (b) is excluded by b —* sf 3 2 .
The lower bound on the lightest MSSM Higgs boson may be relaxed significantly if C P violation feeds into the MSSM Higgs sector 3 3 .
378
John Ellis
Decline and Fall of the Standard Model?
Also shown in Fig. 4 is the constraint imposed by measurements of b —> sj 3 2 . These agree with the Standard Model, and therefore provide bounds on chargino and charged Higgs masses, for example. For moderate tan/3, the b —> S7 constraint is more important for n < 0, as seen in Fig. 4b, but it is also significant for /z > 0 when tan /? is large.
500 400
300
Fig. 4 also displays the regions where the supersymmetric relic density px = £lxpCriticai falls within the preferred range 0.1 < Q,xh2 < 0.3
(10)
The upper limit is rigorous, since astrophysics and cosmology tell us that the total matter density f2TO ^ 0.4, and the Hubble expansion rate h ~ l / v 2 to within about 10 % (in units of km/s/Mpc). On the other hand, the lower limit in (10) is optional, since there could be other important contributions to the overall matter density. As is seen in Fig. 4, there are generic regions of the CMSSM parameter space where the relic density falls within the preferred range (10). What goes into the calculation of the relic density? It is controlled by the annihilation rate 25 : Px = mxnx
• nx ~ ~
1
/
7——
"ann\XX
T
x
(11)
' • • •)
and the typical annihilation rate 1/m2 For this reason, the relic density typically increases with the relic mass, and this combined with the upper bound in (10) then leads to the common expectation that mx ^ 1 TeV. However, there are various ways in which the generic upper bound on m x can be increased along filaments in the (mi/2, mo) plane. For example, if the next-to-lightest sparticle (NLSP) is not much heavier than X- Am/mx ^j 0.1, the relic density may be suppressed by coannihilation: <j(x+NLSP—> ...) 34 . In this way, the allowed CMSSM region may acquire a 'tail' extending to large ' X ' as in the case where the NLSP is the lighter stau: fi and m^j m v as seen in 379
500
1000
1500
2000
m 1/2
Figure 5. The large-mi/2 'tail' of the x — fi coannihilation region for tan/? = 10 and fi < 0 3 5 .
Fig. 5 35 . Another mechanism for extending the allowed CMSSM region to large mx is rapid annihilation via a direct-channel pole when mx ~ \mHiggs,z 36 ' 37 . This may yield a 'funnel' extending to large m y 2 and mo at large tan/3, as seen in Fig. 6 3 7 . Another allowed region at large m y 2 and mo is the 'focus-point' region 3 8 , which is adjacent to the boundary of the region where electroweak symmetry breaking is possible, as seen in Fig. 7. However, in this region mx is not particularly large. These filaments extending the preferred CMSSM parameter space are clearly exceptional, in some sense, so it is important to understand the sensitivity of the relic density to input parameters, unknown higher-order effects, etc. One proposal is the relic-density fine-tuning measure 39 d\n(flxh2) din at
(12)
where the sum runs over the input parameters, which might include (relatively) poorlyknown Standard Model quantities such as mt
John Ellis
Decline and Fall of the Standard Model? and nib, as well as the CMSSM parameters mo,mi/2, etc. As seen in Fig. 7, the sensitivity A Q (12) is relatively small in the 'bulk' region at low mi/2, mo, and tan/3. However, it is somewhat higher in the x ~ ^i coannihilation 'tail', and at large tan/3 in general. The sensitivity measure A n (12) is particularly high in the rapid-annihilation 'funnel' and in the 'focus-point' region. This explains why published relic-density calculations may differ in these regions 40 , whereas they agree well when A n is small: differences may arise because of small differences in the treatments of the inputs.
tanp=50. »>o
It is important to note that the relicdensity fine-tuning measure (12) is distinct from the traditional measure of the finetuning of the electroweak scale 41 :
m1/2 (GeV)
Figure 6. The region where the cosmological relic density is in the preferred range (10) for tan/3 = 50 and [i > 0. Note the rapid-annihilation 'funnel' at intermediate m o / m j / 2 37 -
This electroweak fine-tuning is a completely different issue, and values of the Aj are not necessarily related to values of A n . Electroweak fine-tuning is sometimes used as a criterion for restricting the CMSSM parameters. However, the interpretation of the A, (13) is unclear. How large a value of Aj is tolerable? Different physicists may well have different pain thresholds. Moreover, correlations between input parameters may reduce its value in specific models.
5
100
200
300
400
500
600
700
800
900
1000
mi /2 (GeV) Figure 7. The ml/2,™>0 plane for tan/3 = 10 and /i > 0, including the 'focus-point' region 3 8 at large moi close to the boundary of the shaded region where electroweak symmetry breaking occurs, and exhibiting contours of the cosmological sensitivity (12) 3 9 .
Muon Anomalous Magnetic Moment
As reported at this meeting 6 , the BNL E821 experiment has recently reported a 2.6-a deviation of aM = TJ((?M — 2) from the Standard Model prediction 42 : a^P - af = (43 ± 16) x 1CT10
(14)
The largest contribution to the error in (14) is the statistical error of the experiment, which will soon be significantly reduced, as 380
Decline and Fall of the Standard Model?
John Ellis many more data have already been recorded. The next-largest error is that due to stronginteraction uncertainties in the Standard Model prediction. Recent estimates converge on an estimate of about 7x 10~ 10 for the error in the hadronic vacuum polarization contribution to (14) 43 , and the error in the hadron light-by-light scattering contribution is generally thought to be smaller 44 . Therefore, if the central value in (14) does not change substantially with the new data, this would be strong evidence for new physics at the TeV scale. 45
As many authors have pointed out , the discrepancy (14) could well be explained by supersymmetry if \i > 0 and tan (3 is not too small, as exemplified in Fig. 8. Good consistency with all the experimental and cosmological constraints on the CMSSM is found for tan/3 < 10 and m x ~ 150 to 350 GeV. Already before the measurement (14), the LHC was thought to have a good chance of discovering supersymmetry 46 . If the result (14) were to be confirmed, this would be almost guaranteed, as we now discuss.
tan p = 10, \l > 0 m
= 104 Gey
j x± 1 / 1*
700-;
= 113 GeV
1/ -
600 J
>
1) O 50B
1
4(>i
\
!
i
\\
1
3<«'
\
\
l%
'
;
^*g
\ :
:i" \ !
'•
1(JU.
. , , ^
[j;^ mm
0-=
100
200
300
400
500
700
800
000 1000
Figure 8. The medium-shaded region is that compatible with the BNL E821 measurement of g^ — 2 at the 2-CT level 6 ' 4 2 , the light-shaded region has a relic density in the preferred range (10), and the dark-shaded region does not have a neutralino LSP. Good compatibility is found between g^ — 2 and the other phenomenological constraints for tan /3 ~ 5 or more 4 5 .
• ; • » • • • • • • .
+
+
Prospects for Observing Supersymmetry at Accelerators
600
mvz (GeV)
+
6
mil
mi,, b- >s"Y
As an aid to the assessment of the prospects for detecting sparticles at different accelerators, benchmark sets of supersymmetric parameters have often been found useful 4 7 , since they provide a focus for concentrated discussion. A set of post-LEP benchmark scenarios in the CMSSM has recently been proposed 48 , and are illustrated schematically in Fig. 9. They take into account the direct searches for sparticles and Higgs bosons, b —> s"f and the preferred cosmological density range (10). About a half of the proposed benchmark points are consistent with
381
: 'i^\ j-w na Figure 9. Schematic overview of the benchmark points proposed in 4 8 . They were chosen to be compatible with the indicated experimental constraints, as well as have a relic density in the preferred range (10). The points are intended to illustrate the range of available possibilities.
John Ellis
Decline and Fall of the Standard Model?
The proposed points were chosen not to provide an 'unbiased' statistical sampling of the CMSSM parameter space, whatever that means in the absence of a plausible a priori measure, but rather are intended to illustrate the different possibilities that are still allowed by the present constraints 48 . Five of the chosen points are in the 'bulk' region at small mi/2 and mo, four are spread along the coannihilation 'tail' at larger m\/2 for various values of tan/?, two are in the 'focus-point' region at large mo, and two are in rapidannihilation 'funnels' at large m-i/2 and moThe proposed points range over the allowed values of tan/? between 5 and 50. Most of them have \i > 0, as favoured by gM - 2, but there are two points with \i < 0. Various derived quantities in these supersymmetric benchmark scenarios, including the relic density, g^ — 2,b —* S7, electroweak fine-tuning A and the relic-density sensitivity A n , are given in 4 8 . These enable the reader to see at a glance which models would be excluded by which refinement of the experimental value of g^ — 2. Likewise, if you find some amount of fine-tuning uncomfortably large, then you are free to discard the corresponding models. The LHC collaborations have analyzed their reach for sparticle detection in both generic studies and specific benchmark scenarios proposed previously 4 6 . Based on these studies, Fig. 10 displays estimates how many different sparticles may be seen at the LHC in each of the newly-proposed benchmark scenarios 4 8 . The lightest Higgs boson is always found, and squarks and gluinos are usually found, though there are some scenarios where no sparticles are found at the LHC. The LHC often misses heavier weaklyinteracting sparticles such as charginos, neutralinos, sleptons and the other Higgs bosons. The physics capabilities of linear e+e~ colliders are amply documented in various
gluino • •
squarksBH sleptons • §
x°'
" *
H
CMSSM Benchmarks
LPcr::
M
v/s=T:crreV
I
Vs=3TeV
3S
LBGCJAMHEFKD
vs=5TeV
30 • 25 20 IS 10 • s
L8SCJAMHEFKD
II II
IL8GCJAMHEFKO
Figure 10. Estimates of the numbers of different types of CMSSM particles that may be detectable 4 8 at (a) the LHC, (b) a 1-TeV linear e + e " collider 4 9 , and (c,d) a 3(5)-TeV e+e~ s 0 or n+H~ collider 5 1 . Note the complementarity between the sparticles detectable a t the LHC and at a 1-TeV linear e + e - collider.
design studies 4 9 . Not only is the lightest MSSM Higgs boson observed, but its major decay modes can be measured with high accuracy, as seen in Fig. 11. Moreover, if sparticles are light enough to be produced, their masses and other properties can be measured very precisely, enabling models of supersymmetry breaking to be tested ° 2 . As seen in Fig. 10, the sparticles visible at an e + e~ collider largely complement those visible at the LHC 4 8 . In most of benchmark scenarios proposed, a 1-TeV linear collider would be able to discover and measure precisely several weakly-interacting sparticles that are invisible or difficult to detect at the LHC. However, there are some benchmark scenarios where the linear collider (as well as the LHC) fails to discover supersymmetry. Only a linear collider with a higher centre-of-mass energy appears sure to cover all the allowed CMSSM parameter space, as
382
John Ellis
Decline and Fall of the Standard Model? fi —» e7 and related processes
M„ (GeV/c')
r:xP
"
"X
-
Figure 11. Analysis of the accuracy with which Higgs decay branching ratios may be measured with a linear e+e collider 5 3 .
— ~ »c a
o A
a ii ~ 8
H
•
6 D
seen in the lower panels of Fig. 12, which illustrate the physics reach of a higher-energy lepton collider, such as CLIC 50 or a multiTeV muon collider 51 .
K X
o X
-
-
0
(a)
I....I..
1 . . .
200
300
500
700
1000
mj. (GeV) 1 -8
B
B
'
1
1
-y^
r
--L'"X
7
-
Prospects for Other Experiments
-
— Btr
o
0
A
.»!,
10
-
A
J
a
« 5
11 " 8
D
Detection of cold dark matter
K
X
12
-
X
o
(b)
Fig. 12 shows rates for the elastic spinindependent scattering of supersymmetric relics 54 , including upper limits from the UKDMC, CDMS and Heidelberg experiments 9 , as well as the range suggested by the DAM A collaboration 55 . Also shown are the rates calculated in the proposed benchmark scenarios discussed in the previous section, which are considerably below the DAMA range, but may be within reach of future projects. Indirect searches for supersymmetric dark matter via the products of annihilations in the galactic halo or inside the Sun also have prospects in some of the benchmark scenarios
54
383
14
....I....I
100
.
.
.
.
200
300
1 . . .
500
I....I.
700
1000
mj; (GeV) Figure 12. Rates for the elastic spin-independent scattering of supersymmetric relics on (a) protons and (b) neutrons calculated in benchmark scenarios 5 4 , compared with upper limits from the UKDMC, CDMS and Heidelberg experiments 9 , as well as the range suggested by the DAMA collaboration 5 5 .
John Ellis
Decline and Fall of the Standard Model? can (B = 30 , u > 0
100
1000
2000
m„ 2 (GeV) Figure 13. Illustration in one particular leptonflavour texture model, for tan/3 = 30 and /i > 0, indicating that n —• e~t decay may occur at a rate close to the present experimental upper limit, in the CMSSM with parameters chosen 5 7 to match the measured value of o > - 2 6 ' 4 2 .
possible to measure CP violation in // —> 3e decay. This may provide another interesting interface with neutrino physics and cosmology 58 . The minimal supersymmetric seesaw model has six CP-violating phases: the MNS phase 6, two light-neutrino Majorana phases, and three phases arising from neutrino Dirac Yukawa couplings, which may be responsible for our existence via leptogenesis in the early Universe 8 . The CP-violating neutrino phases induce phases in slepton mass matrices, which may show up in fi —* 3e decay, T —• 3e//x decays and leptonic electric dipole moments. In principle, the leptogenesis phases might be obtainable by comparing CP-violating measurements in the chargedlepton and neutrino sectors 58 . Proton decay
This could be within reach, with r ( p —• e+7r°) via a dimension-six operator possibly ~ 1035y if TUGUT ~ 10 16 GeV as expected in a minimal supersymmetric GUT. Such a The BNL E821 report of a possible de- model also suggests that r ( p —» vK+) < viation from the Standard Model suggests 10322/ via dimension-five operators 59 , unless that a non-trivial \i — /j, — 7 vertex is gen- measures are taken to suppress them 6 0 . This erated at a scale ^ 1 TeV. Neutrino oscilla- provides motivation for a next-generation tions indicate that there are A£ M 7^ 0 pro- megaton experiment that could detect processes 8 , so it is natural to expect that there ton decay as well as explore new horizons in might also be a non-trivial /i — e — 7 vertex. neutrino physics 6 1 . This is indeed the case in a generic supersymmetric GUT, where neutrino mixing induces slepton mixing 5 6 . Within this framework, the measurement of #M - 2 fixes the sparticle 8 Conclusions scale, and Y[p, —» e~/) may then be calculated within any given flavour texture. Very approximately, if Qy. - 2 is within one or two a As we have seen, future colliders such as of the present central value, one may expect the LHC and a TeV-scale linear e+e~ colB(fi —» ey) with one or two orders of magni- lider have good prospects of discovering sutude of the present experimental upper limit, persymmetry and making detailed measureas illustrated in Fig. 13 57 . ments. In parallel, B and v factories have good prospects of making inroads on the The decay /z —> 3e and \i —* e conversion flavour and unification problems. Searches on nuclei are expected to occur with branch- for dark matter, stopped-muon experiments ing ratios within two or three orders of mag- and searches for proton decay also have innitude of B(n —• e-y), and it is in principle teresting prospects. 384
Decline and Fall of the Standard Model?
John Ellis Looking further beyond the Standard Model, how can one hope to test a Theory of Everything, including quantum gravity? This should be our long-term ambition, our analogue of the 'faint blue dot' towards which exoplanetary science is directed, and which motivates much of their funding. Testing a quantum theory of gravity will be relatively easy if there are large extra dimensions 10 ' 62 . Much more challenging would be the search for observable effects if the gravitational scale turns out, after all, to be of the same order as the Planck mass ~ 10 19 GeV. Perhaps the only way to reconcile relativity with quantum mechanics is to modify one or the other, or both 63 ? Testing the Theory of Everything may require thinking beyond the standard 'Beyond the Standard Model' box.
References 1. S. Aoki, J. Goodman, C. K. Jung and J. Klein, talks at this conference. 2. M. S. Turner, talk at this conference. 3. R. Barbieri, talk at this conference. 4. M. Neubert, talk at this conference. 5. F. Zwirner, talk at this conference. 6. L. Robertson, talk at this conference. 7. G. Hanson, talk at this conference. 8. H. Murayama, talk at this conference. 9. H. Klapdor-Kleingrothaus, talk at this conference. 10. L. Randall, talk at this conference. 11. D. A. Ross and M. J. Veltman, Nucl. Phys. B 95 (1975) 135. 12. For a classic review,see: E. Farhi and L. Susskind, Phys. Rept. 74 (1981) 277; for a recent review and references, see: K. D. Lane, Technicolor 2000, arXiv: hep-ph/0007304. 13. See, for example: G. Altarelli, arXiv:hep-ph/0011078, as updated in G. Altarelli, F. Caravaglios, G. F. Giudice, P. Gambino and G. Ridolfi, JHEP 0106 (2001) 018 [arXiv:hep-
385
ph/0106029]. 14. J. R. Ellis, G. L. Fogli and E. Lisi, Phys. Lett. B 343 (1995) 282. 15. J. Drees, talk at this conference. See the LEP Collaborations, ALEPH, DELPHI, L3, OPAL, the LEP Electroweak Working Group and the SLD Heavy Flavour and Electroweak Groups, LEPEWWG/2001-01, available from http://lepewwg.web.cern.ch/LEPEWWG/ Welcome.html. 16. J. R. Ellis and D. Ross, Phys. Lett. B 506 (2001) 331 [arXiv:hep-ph/0012067]. 17. For a review, see: T. Hambye and K. Riesselmann, arXiv:hep-ph/9708416. 18. G. Isidori, G. Ridolfi and A. Strumia, Nucl. Phys. B 609 (2001) 387 [arXiv:hep-ph/0104016]. 19. L. Maiani, Proceedings of the 1979 Gifsur-Yvette Summer School On Particle Physics, 1; G. 't Hooft, in Recent Developments in Gauge Theories, Proceedings of the Nato Advanced Study Institute, Cargese, 1979, eds. G. 't Hooft et al., (Plenum Press, NY, 1980); E. Witten, Phys. Lett. B 105 (1981) 267. 20. J. Ellis, S. Kelley and D. V. Nanopoulos, Phys. Lett. B 260 (1991) 131; U. Amaldi, W. de Boer and H. Furstenau, Phys. Lett. B 260 (1991) 447; C. Giunti, C. W. Kim and U. W. Lee, Mod. Phys. Lett. A 6 (1991) 1745. 21. K. R. Dienes, E. Dudas and T. Gherghetta, Phys. Lett. B 436 (1998) 55 [arXiv:hep-ph/9803466] and Nucl. Phys. B 537 (1999) 47 [arXiv:hepph/9806292]. 22. D. M. Ghilencea and G. G. Ross, Nucl. Phys. B 606 (2001) 101 [arXiv:hepph/0102306]. 23. J. Ellis, G. Ridolfi and F. Zwirner, Phys. Lett. B 257 (1991) 83; M.S. Berger, Phys. Rev. D 41 (1990) 225; Y. Okada, M. Yamaguchi and T. Yanagida, Prog. Theor. Phys. 85 (1991) 1; Phys. Lett. B 262 (1991) 54; H.E. Haber and R.
John Ellis Hempfling, Phys. Rev. Lett. 66 (1991) 1815; R. Barbieri, M. Frigeni and F. Caravaglios, Phys. Lett. B 258 (1991) 167; P.H. Chankowski, S. Pokorski and J. Rosiek, Phys. Lett. B 274 (1992) 191; J.R. Espinosa and M. Quiros, Phys. Lett. B 266 (1991) 389; J.L. Lopez and D.V. Nanopoulos, Phys. Lett. B 266 (1991) 397; M. Carena, K. Sasaki and C.E.M. Wagner, Nucl. Phys. B 381 (1992) 66; P.H. Chankowski, S. Pokorski and J. Rosiek, Phys. Lett. B 281 (1992) 100; Nucl. Phys. B 423 (1994) 437; D.M. Pierce, A. Papadopoulos and S.B. Johnson, Phys. Rev. Lett. 68 (1992) 3678; A. Brignole, Phys. Lett. B 281 (1992) 284; M. Drees and M.M. Nojiri, Phys. Rev. D 45 (1992) 2482; V. Barger, M.S. Berger and P. Ohmann, Phys. Rev. D 49 (1994) 4908; G.L. Kane, C. Kolda, L. Roszkowski and J.D. Wells, Phys. Rev. D 49 (1994) 6173; R. Hempfling and A.H. Hoang, Phys. Lett. B 331 (1994) 99; P. Langacker and N. Polonsky, Phys. Rev. D 50 (1994) 2199; H.E. Haber, R. Hempfling and A.H. Hoang, Zeit. fur Phys. C 75 (1997) 539; S. Heinemeyer, W. Hollik and G. Weiglein, Comput. Phys. Commun. 124, 76 (2000) [hepph/9812320]. 24. For a review, see: H. Dreiner, arXiv:hepph/9707435. 25. J. Ellis, J.S. Hagelin, D.V. Nanopoulos, K.A. Olive and M. Srednicki, Nucl. Phys. B 238 (1984) 453; see also H. Goldberg, Phys. Rev. Lett. 50 (1983) 1419. 26. S. Dimopoulos and H. Georgi, Nucl. Phys. B 193 (1981) 150; N. Sakai, Z. Phys. C 11 (1981) 153. 27. K. Inoue, A. Kakuto, H. Komatsu and S. Takeshita, Prog. Theor. Phys. 68 (1982) 927 [Erratum-ibid. 70 (1982) 330]; L.E. Ibafiez and G.G. Ross, Phys. Lett. B 110 (1982) 215; L.E. Ibafiez, Phys. Lett. B 118 (1982) 73; J. El-
Decline and Fall of the Standard Model?
28.
29.
30.
31.
32.
386
lis, D.V. Nanopoulos and K. Tamvakis, Phys. Lett. B 121 (1983) 123; J. Ellis, J. Hagelin, D.V. Nanopoulos and K. Tamvakis, Phys. Lett. B 125 (1983) 275; L. Alvarez-Gaume, J. Polchinski, and M. Wise, Nucl. Phys. B 221 (1983) 495. J. R. Ellis and D. V. Nanopoulos, Phys. Lett. B 110 (1982) 44; R. Barbieri and R. Gatto, Phys. Lett. B 110 (1982) 211. M. Dine and A. E. Nelson, Phys. Rev. D 48, 1277 (1993) [hep-ph/9303230]; M. Dine, A. E. Nelson and Y. Shirman, Phys. Rev. D 51, 1362 (1995) [hep-ph/9408384]; M. Dine, A. E. Nelson, Y. Nir and Y. Shirman, Phys. Rev. D 53, 2658 (1996) [hep-ph/9507378]. D. E. Kaplan, G. D. Kribs and M. Schmaltz, Phys. Rev. D 62, 035010 (2000) [hep-ph/9911293]; Z. Chacko, M. A. Luty, A. E. Nelson and E. Ponton, JHEP 0001, 003 (2000) [hep-ph/9911323]. L. Randall and R. Sundrum, Nucl. Phys. B 557, 79 (1999) [hep-th/9810155]; G. F. Giudice, M. A. Luty, H. Murayama and R. Rattazzi, JHEP 9812, 027 (1998) [hepph/9810442]. M. Fukugita and T. Yanagida, Phys. Lett. B 174 (1986) 45; for a recent review, see: W. Buchmuller, arXiv:hepph/0107153. J. R. Ellis, G. Ganis, D. V. Nanopoulos and K. A. Olive, Phys. Lett. B 502 (2001) 171 [arXiv:hep-ph/0009355]. M.S. Alam et al., [CLEO Collaboration], Phys. Rev. Lett. 74 (1995) 2885 as updated in S. Ahmed et al., CLEO CONF 99-10; BELLE Collaboration, BELLECONF-0003, contribution to the 30th International conference on High-Energy Physics, Osaka, 2000. See also K. Abe et al., [Belle Collaboration], [arXiv:hepex/0107065]; L. Lista [BaBar Collaboration], [arXiv:hep-ex/0110010]; C. Degrassi, P. Gambino and G. F. Giudice, JHEP 0012 (2000) 009; M. Carena,
John Ellis
33.
34.
35.
36.
37.
38.
39.
Decline and Fall of the Standard Model?
D. Garcia, U. Nierste and C. E. Wagner, hep-ph/0010003; G. Isidori, talk at this conference. M. Carena, J. R. Ellis, A. Pilaftsis and C. E. Wagner, Nucl. Phys. B 586 (2000) 92 [arXiv:hep-ph/O0O318O], Phys. Lett. B 495 (2000) 155 [arXiv:hepph/0009212]; and references therein. S. Mizuta and M. Yamaguchi, Phys. Lett. B 298 (1993) 120 [arXiv:hepph/9208251]; J. Edsjo and P. Gondolo, Phys. Rev. D 56 (1997) 1879 [arXiv:hepph/9704361]. C. Boehm, A. Djouadi and M. Drees, Phys. Rev. D 62 (2000) 035012 [arXiv:hep-ph/9911496]. J. Ellis, T. Falk and K. A. Olive, Phys. Lett. B 444, 367 (1998); J. Ellis, T. Falk, K. A. Olive and M. Srednicki, Astropart. Phys. 13 (2000) 181; M. E. Gomez, G. Lazarides and C. Pallis, Phys. Rev. D 61, 123512 (2000) [hep-ph/9907261] and Phys. Lett. B 487, 313 (2000) [hepph/0004028]; R. Arnowitt, B. Dutta and Y. Santoso, hep-ph/0102181. M. Drees and M. M. Nojiri, Phys. Rev. D 47, 376 (1993); H. Baer and M. Brhlik, Phys. Rev. D 53 (1996) 597 and Phys. Rev. D 57 (1998) 567; H. Baer, M. Brhlik, M. A. Diaz, J. Ferrandis, P. Mercadante, P. Quintana and X. Tata, Phys. Rev. D 63 (2001) 015007; A. B. Lahanas, D. V. Nanopoulos and V. C. Spanos, hep-ph/0009065. J. R. Ellis, T. Falk, G. Ganis, K. A. Olive and M. Srednicki, Phys. Lett. B 510 (2001) 236 [arXiv:hep-ph/0102098]. J. L. Feng, K. T. Matchev and T. Moroi, Phys. Rev. Lett. 84, 2322 (2000) [hep-ph/9908309]; J. L. Feng, K. T. Matchev and T. Moroi, Phys. Rev. D 61, 075005 (2000) [hep-ph/9909334]; J. L. Feng, K. T. Matchev and F. Wilczek, Phys. Lett. B 482, 388 (2000) [hep-ph/0004043]. J. R. Ellis and K. A. Olive, Phys. Lett. B 514 (2001) 114 [arXiv:hep-ph/0105004].
387
40. For other recent calculations, see, for example: A. B. Lahanas, D. V. Nanopoulos and V. C. Spanos, Phys. Lett. B 518 (2001) 94 [arXiv:hep-ph/0107151]; L. Roszkowski, R. Ruiz de Austri and T. Nihei, JHEP 0108, 024 (2001) [arXiv:hep-ph/0106334]. 41. J. Ellis, K. Enqvist, D. V. Nanopoulos and F. Zwirner, Mod. Phys. Lett. A 1, 57 (1986); R. Barbieri and G. F. Giudice, Nucl. Phys. B 306 (1988) 63. 42. H. N. Brown et al. [Muon g-2 Collaboration], Phys. Rev. Lett. 86, 2227 (2001) [hep-ex/0102017]. 43. R. Alemany, M. Davier and A. Hocker, Eur. Phys. J. C 2 (1998) 123 [arXivrhepph/9703220]; M. Davier and A. Hocker, Phys. Lett. B 419 (1998) 419 [arXivrhepph/9711308]; M. Davier and A. Hocker, Phys. Lett. B 435 (1998) 427 [arXivrhepph/9805470]; S. Narison, Phys. Lett. B 513 (2001) 53 [arXiv:hep-ph/0103199]; J. F. De Troconiz and F. J. Yndurain, arXiv:hep-ph/0106025. 44. J. Bijnens, talk at this conference, arXiv:hep-ph/0108111. See also J. Bijnens, E. Pallante and J. Prades, Nucl. Phys. B 474 (1996) 379 [arXiv:hepph/9511388]; M. Hayakawa and T. Kinoshita, Phys. Rev. D 57 (1998) 465 [arXiv:hep-ph/9708227]. 45. L. L. Everett, G. L. Kane, S. Rigolin and L. Wang, Phys. Rev. Lett. 86, 3484 (2001) [arXiv:hep-ph/0102145]; J. L. Feng and K. T. Matchev, Phys. Rev. Lett. 86, 3480 (2001) [arXiv:hepph/0102146]; E. A. Baltz and P. Gondolo, Phys. Rev. Lett. 86, 5004 (2001) [arXiv:hep-ph/0102147]; U. Chattopadhyay and P. Nath, Phys. Rev. Lett. 86, 5854 (2001) [arXiv:hep-ph/0102157]; S. Komine, T. Moroi and M. Yamaguchi, Phys. Lett. B 506, 93 (2001) [arXiv:hep-ph/0102204]; S. P. Martin and J. D. Wells, Phys. Rev. D 64, 035003 (2001) [arXiv:hep-ph/0103067];
John Ellis
Decline and Fall of the Standard Model?
H. Baer, C. Balazs, J. Ferrandis and X. Tata, Phys. Rev. D 64, 035004 (2001) [arXiv:hep-ph/0103280]; J. Ellis, D. V. Nanopoulos and K. A. Olive, Phys. Lett. B 508 (2001) 65 [arXiv:hepph/0102331]; R. Arnowitt, B. Dutta, B. Hu and Y. Santoso, Phys. Lett. B 505 (2001) 177 [arXiv:hep-ph/0102344]. 46. ATLAS Collaboration, ATLAS detector and physics performance Technical Design Report, CERN/LHCC 9914/15 (1999); S. Abdullin et al. [CMS Collaboration], hep-ph/9806366; S. Abdullin and F. Charles, Nucl. Phys. B 547 (1999) 60; CMS Collaboration, Technical Proposal, CERN/LHCC 94-38 (1994).
51.
52.
53. 54.
47. See, for example: I. Hinchliffe, F. E. Paige, M. D. Shapiro, J. Soderqvist and W. Yao, Phys. Rev. D 55 (1997) 5520; TESLA Technical Design Report, DESY-01-011, Part III, Physics at an e+e~ Linear Collider (March 2001). 48. M. Battaglia et al., arXiv:hep-ph/0106204. 49. S. Matsumoto et al. [JLC Group], JLC1, KEK Report 92-16 (1992); J. Bagger et al. [American Linear Collider Working Group], The Case for a 500-GeV e+e~ Linear Collider, SLAC-PUB-8495, BNL-67545, FERMILAB-PUB-00-152, LBNL-46299, UCRL-ID-139524, LBL46299,Jul 2000,hep-ex/0007022; T. Abe et al. [American Linear Collider Working Group Collaboration], Linear Collider Physics Resource Book for Snowmass 2001, SLAC-570, hep-ex/0106055, hep-ex/0106056, hep-ex/0106057 and hep-ex/0106058; TESLA Technical Design Report, DESY-01-011, Part III, Physics at an e+e~~ Linear Collider (March 2001). 50. R. W. Assmann et al. [CLIC Study Team], A 3-TeV e+e~ Linear Collider Based on CLIC Technology, ed. G. Guignard, CERN
55. 56.
2000-08; CLIC Physics Study Group, http://clicphysics.web.cern.ch/ CLICphysics/. Neutrino Factory and Muon Collider Collaboration, http://www.cap.bnl.gov/mumu/ mu_home_page. html; European Muon Working Groups, http://muonstoragerings.cern.ch/ Welcome.html. G. A. Blair, W. Porod and P. M. Zerwas, Phys. Rev. D 6 3 (2001) 017703 [hepph/0007107]. M. Battaglia and K. Desch, arXiv:hepph/0101165. J. Ellis, J. L. Feng, A. Ferstl, K. T. Matchev and K. A. Olive, arXiv:astro-ph/0110225. DAMA Collaboration, R. Bernabei et al, Phys. Lett. B 436, 379 (1998). J. Hisano, T. Moroi, K. Tobe and M. Yamaguchi, Phys. Rev. D 53 (1996) 2442; J. Hisano, D. Nomura and T. Yanagida, Phys. Lett. B 437 (1998) 351; J. Hisano and D. Nomura, Phys. Rev. D 59 (1999) 116005; W. Buchmuller, D. Delepine and F. Vissani, Phys. Lett. B 459 (1999) 171; M. E. Gomez, G. K. Leontaris, S. Lola and J. D. Vergados, Phys. Rev. D 59 (1999) 116009; J. R. Ellis, M. E. Gomez, G. K. Leontaris, S. Lola and D. V. Nanopoulos, Eur. Phys. J. C 14 (2000) 319; W. Buchmuller, D. Delepine and L. T. Handoko, Nucl. Phys. B 576 (2000) 445; J. L. Feng, Y. Nir and Y. Shadmi, Phys. Rev. D 61 (2000) 113005; J. Sato and K. Tobe, Phys. Rev. D 63 (2001) 116010; J. Hisano and K. Tobe, Phys. Lett. B 510 (2001) 197; S. Baek, T. Goto, Y. Okada and K. Okumura, hep-ph/0104146; S. Lavignac, I. Masina and C.A. Savoy, hep-ph/0106245.
57. D. F. Carvalho, J. R. Ellis, M. E. Gomez and S. Lola, Phys. Lett. B 515 (2001) 323 [arXiv:hep-ph/0103256].
388
Decline and Fall of the Standard Model?
John Ellis 58. J. R. Ellis, J. Hisano, S. Lola and M. Raidal, arXiv:hep-ph/0109125; and references therein. 59. H. Murayama and A. Pierce, arXiv:hepph/0108104. 60. J. R. Ellis, J. S. Hagelin, S. Kelley and D. V. Nanopoulos, Nucl. Phys. B 311 (1988) 1. 61. C. K. Jung, arXiv:hep-ex/0005046; Y. Suzuki et al. [TITAND Working Group Collaboration], arXiv:hepex/0110005. 62. I. Antoniadis and K. Benakli, Int. J. Mod. Phys. A 15 (2000) 4237 [arXiv:hep-ph/0007226]. 63. J. R. Ellis, N. E. Mavromatos and D. V. Nanopoulos, arXiv:gr-qc/9909085.
389
T H E HIGGS PUZZLE: E X P E R I M E N T A N D THEORY FABIO ZWIRNER Dipartimento di Fisica, Universitd di Roma 'La Sapienza', and INFN, Sezione di Roma, Piazzale Aldo Mow 2, 1-00185 Roma, ITALY E-mail: [email protected] T h e present experimental and theoretical knowledge of the physics of electroweak symmetry breaking is reviewed. Data still favor a light Higgs boson, of a kind that can be comfortably accommodated in the Standard Model or in its Minimal Supersymmetric extension, but exhibit a non-trivial structure that leaves some open questions. The available experimental information may still be reconciled with the absence of a light Higgs boson, but the price to pay looks excessive. Recent theoretical ideas, linking the weak scale with the size of possible extra spatial dimensions, are briefly mentioned. It is stressed once more that experiments at high-energy colliders, such as the Tevatron and the LHC, are the crucial tool for eventually solving the Higgs puzzle.
Rome is a city so full of religious symbols that it provides some inspiration on how to organize a talk on the physics of electroweak symmetry breaking, where firm experimental and theoretical results are mixed, so far, with a certain amount of beliefs. 1
The Standard Model (The Orthodoxy)
The obvious starting point for any discussion of the Higgs puzzle (= 'what is the physics of electroweak symmetry breaking?') is the Standard Model (SM), by now firmly established as the renormalizable quantum field theory of strong and electroweak interactions at presently accessible energies: in the spirit of the preface, it can be called 'The Orthodoxy'. The only SM ingredient still escaping experimental detection, in a theoretical construction that works incredibly well, is the Higgs 1 boson, H. Its properties are controlled by some well-known parameters of the fermion and gauge sectors (including the Fermi constant Gp, which sets the value of the weak scale) plus an independent one, the Higgs mass mn • The elementary complex spin-0 field (ft, an SU(2)L doublet of weak hypercharge Y = +1/2 (in the normalization where Q = T^L + Y), is by now considered to be an essential
part of what we call the SM. Indeed, it plays a fundamental role in the description of two symmetry-breaking phenomena. The first is the spontaneous breaking of the SU(2)L X U(l)y gauge symmetry down to the U{1)Q of QED, described by the following part of the SM Lagrangian:
CS = {D^)\D^)
-fi24>U- K4>U? • (1)
The second is the explicit breaking of the global flavor symmetry that is present if only gauge interactions are switched on. This is realized by the Yukawa part of the SM Lagrangian, CY = huq[EuR(j) + hD~qEdR(t>
+hET[eR = io'2(f>*, hU'D'E
(2)
are 3 x 3 complex
matrices and generation indices have been omitted. Over the years, our confidence in this description has been progressively reinforced by increasingly precise tests of both these symmetry-breaking phenomena (here the focus will be on gauge symmetry breaking, since the theoretical aspects of flavor symmetry breaking are discussed in another talk at this Conference 2 ). However, the ultimate, crucial test of the SM remains the direct search for the Higgs particle, by far the most important experimental enterprise in today's particle physics.
390
Fabio Zwirner 1.1
The Higgs Puzzle: Experiment and Theory
Direct searches for the SM Higgs
The experimental status of the searches for the SM Higgs particle is reviewed in detail in another talk at this Conference 3 . Here I will just summarize the present situation from the preliminary LEP-combined results 4 released for the conferences of Summer 2001: CM
• The data still show an excess, at the level of 2.1a, over the expected SM background, mainly due to ALEPH data and to the four-jet final state (to be compared with the 2.9cr excess in the preliminary data of November 2000). • The maximum likelihood occurs at rrifj = 115.6 GeV, with 3.5% probability of a fluctuation of the SM background. • The lower bound on the Higgs mass is mH > 114.1 GeV at 95% c.L, to be compared with an expected bound of 115.4 GeV. • Three out of the four LEP experiments have not yet released their final results at the time of this Conference: the final combination of the LEP results is expected for the end of 2001. 1.2
SM fits to the Higgs mass
Besides direct searches, additional information on the SM Higgs boson come from the fits to m,H based on electroweak precision data, whose experimental aspects are discussed in another talk at this Conference 5 . A popular summary of the available information 6 is the famous 'blueband' plot of the LEP Electroweak Working Group, displayed / 2 of t h e in Fig. 1: it gives the A\2 = X2 •X^4„ global SM fit to electroweak precision data as a function of m # . As evident from Fig. 1, the fit clearly favours a light Higgs. The default fit, represented by the solid curve in Fig. 1, gives mH = 88^35 GeV and mH < 196 GeV at 95% c.L. An alternative fit (see below for an explanation), represented by the dashed
391
<
Excluded
'(/
Preliminary
10 m H [GeV]
Figure 1. A x 2 as a function of m g from the global fit to the SM. The region excluded by the direct searches at LEP is also shown.
line, gives ran < 222 GeV at 95% c.L and a slightly higher central value. The band represents a (debatable) estimate of the theoretical uncertainty. Notice that the fit does not include the information coming from direct searches. Notice also that, in both fits, more than half of the x 2 curve falls in the shaded region, excluded at 95% c.L by direct searches. Given the importance of the issue, it is worth examining in more detail how the preference for a light SM Higgs arises. For given values of the remaining SM input parameters, precise electroweak data combined with updated theoretical calculations give logarithmic sensitivity to mH, mostly via two pseudo-observables: the leptonic effective electroweak mixing angle, sin 2 6*?. = (1 — vi/ai)/<±, and the mass of the W boson, mw- There are still small theoretical uncertainties in the evaluation of radiative corrections, in principle reducible by more refined calculations. A recent progress along these lines is the calculation 7 of the com-
The Higgs Puzzle: Experiment and Theory
Fabio Zwirner plete fermionic two-loop contribution to mw, but other calculations of the same order are still missing. Larger uncertainties come from non-negligible errors in other parameters entering the fit. An important one is the hadronic contribution to the running of the electromagnetic coupling constant, Aa^d, as extracted from a dispersion integral over a parametrization of the measured crosssection for e+e~ —> hadrons, including the recent data 8 from BES and CMD-2. A conservative, 'data-driven' fit 9 gives A ^ K )
= 0.02761 ± 0.00036,
Preliminary 0.23099 ± 0.00053 A,(Pt)
0.23159 ±0.00041
AiSLC
0.23098 ± 0.00026
c
T— *
A,b
0.23152 1 0 . 0 0 0 1 7 XZ/d.o.f.: 1 2 . 8 / 5
3
10 -
(3)
> O I..
£V,y
J
X i Aa,\;'a= 0.02761 ± 0.00036 sm,= 91.1875 ±0.0021 GeV am'= 174.3 ± 5 . 1 GeV
(4)
There are many other determinations, as reviewed for example in Ref. n , all consistent with the previous ones, but with a tendency to be closer to Eq. (3) than to Eq. (4). The second important uncertainty in the input parameters is the one associated with the experimental determination of the top quark mass from the CDF and DO experiments at Fermilab 12 :
0.2324 ± 0.0012
Average
0.23
= 0.02738 ± 0.00020.
0.23272 ± 0.00079
-*
corresponding to the 'default' input of the blueband plot, whereas a more aggressive, 'theory-driven' fit 10 , corresponding to the 'alternative' input of the blueband plot, gives Aa{^d(mz)
0.23226 + 0.00031
0.232
0.234
• 2jePt
sin 6 eff Figure 2. Determination of sin 2 0"Ff from the asymmetry measurements. The SM prediction is also shown, as a function of m / j , with t h e uncertainties from A a i d(mz) and mt added linearly.
quark charge asymmetry) to give larger values of sin2 #eT? (and of mjj) than the leptonic asymmetries (forward-backward asymmetry, mtop = 174.3 ± 5.1 GeV. (5) r polarization asymmetry, SLD left-right The individual experimental determina- asymmetry). The average of the hadronic l tions of sin 2 0*J*j and mw, with the corre- determinations alone gives sin e eff{had) = sponding theoretical predictions and uncer- 0.23230 ±0.00029, the average of the leptonic ones sin 2 9lef}(lep) = 0.23113 ± 0.00021, cortainties, as taken from Ref. 6 , are displayed responding to a discrepancy at the level of in Figs. 2 and 3. A careful inspection of Figs. 1-3 reveals 3.3cr. This effect was larger in Winter 2001: that the SM fit is not entirely a bed of roses, the change is mostly due to a — 0.5cr shift of the bb forward-backward asymmetry, AF'B, as stressed, for example, in Ref. 13 (on the basis of the data available in Winter 2001). after a new DELPHI analysis based on a First, the quality of the overall fit turns neural network to tag the b-charge, and an out to be acceptable but not exceptional, improvement in the jet-charge measurement of the ALEPH analysis. The most precise X2/dof = 22.9/15, corresponding to a 'probability' of 8.6%. The main reason for this hadronic determination comes indeed from can be seen from Fig. 2: there is a system- ApB, and has now a pull of 2.9<7 with reatic tendency of the hadronic asymmetries spect to the central value of the global SM fit. (bb and cc forward-backward asymmetry plus Keeping in mind the possibility of a statisti-
392
Fabio Zwirner
The Higgs Puzzle: Experiment and Theory Mass of the W Boson
Experiment
Mw
ALEPH
[GeV]
80.471 ±0.049
80.398 ±0.069 80.490 ± 0 . 0 6 5 2
X /dof = 32.5/39 80.450 ± 0.039
-mm
Aa ( 5 ) fla
had_
0.02761 ±0.00036 linearly added to
m 80.2
80.4 M w [GeV]
Mt = 174.3±5.1 GeV
80.6
Figure 3. The measurements of my/ at LEP. T h e lower plot shows the SM prediction, as a function o f m H , with the uncertainties from A a ^ ( m z ) and mt added linearly.
cal fluctuation, this can be viewed as a small problem either for the SM or for the experimental analyses. Radical modifications of the Zbb vertex appear unlikely, given the fact that Aj, from SLD and Rb are well-behaved. Also, measuring Ap^ is a very delicate experimental task, since flavor and charge of the b-quarks need to be tagged simultaneously, with more complicated systematics than in the measurement of Rb- On the other hand, all the experimental determinations of mw are in good agreement and point to a light Higgs boson: those of Fig. 3 can be combined with the ones from the UA2, CDF and DO experiments at pp colliders, giving mw = 80.454 ± 0.060, to produce a global
393
world average mw = 80.451 ± 0.033. Notice that, with the present errors, the main parametric uncertainty affecting the theoretical determination of mw is the one coming from mt, whereas mt and Aa h^d(mz) give comparable uncertainties in the theoretical determination of sin d^ff • Given the small discrepancy between hadronic and leptonic asymmetries, the exercise of looking at what happens, when dropping the hadronic asymmetries from the fit, may not be entirely academical. The result is the following: the quality of the SM fit improves, but the central value of m # is pushed down, so that the consistency of the SM fit to ran with the limits from direct searches becomes marginal [with a significant residual dependence, which should not be forgotten, on Aahld(mz) and mt]. Is there a SM crisis lurking around the corner? A prudent attitude before answering this question may be appropriate, taking into account that the final heavy-flavor analyses from LEP and improved determinations of mw and mt from the new Tevatron run will be available soon. It may well be, however, that we must wait until the discovery (or the exclusion) of a light Higgs boson to definitively settle the issue.
1.3
The SM as an effective theory
If we believed that the SM is the whole story, the talk could end here. However, we all know that the SM cannot be the ultimate theory of elementary particles, valid at arbitrarily high energy scales, since it does not contain a quantum theory of gravitational interactions and some of its couplings are not asymptotically free. Thus, the SM must be seen as an effective field theory, valid up to some physical cut-off scale A, where new physics must be introduced into the theory. On general grounds, A could be anywhere between the TeV scale and the Planck scale, MP = G~1/2/VS^ - 2.4 x 10 18 GeV, where
Fabio Zwirner
The Higgs Puzzle: Experiment and Theory tests, neutrino masses, proton decay, . . . . In this framework, an old question 14 can be addressed in the light of present experimental data: given our knowledge of the top quark mass and of the bounds on the Higgs mass, can we put some firm bounds on the cut-off scale A? The qualitative aspects of the answer can be appreciated by remembering that, in the SM, the top and Higgs masses are associated with the largest Yukawa coupling ht and with the quartic Higgs selfcoupling A, respectively, via tree-level relations of the form mt oc htv and m2H oc Au2, where v is the vacuum expectation value of the Higgs field. Also, the scale-dependence of A is controlled by the renormalization group equation
GN is Newton's constant, characterizing the observed gravitational interactions. Assuming that the SM correctly identifies the degrees of freedom at the weak scale (this may not be true, as will be discussed later, in the case of the Higgs field), we can write down the most general local Lagrangian compatible with the SM symmetries, classifying the possible operators according to their physical dimension, and scaling all dimensionful couplings by appropriate powers of A. The resulting dimensionless coefficients are then to be interpreted as parameters, which can be either fitted to experimental data or (if we are able to do so) theoretically determined from the fundamental theory replacing the SM at the scale A. Very schematically (and omitting all coefficients and indices, as well as many theoretical subtleties):
4
t-eff
2
2
2
[A + A $ ] + [(D$) + * #tf +F»VF^
+ ¥ * $ + $4
+ F^F^
*CT^*FMV
A $2F^FM + - "A2"
A
* * * *
A2 (6)
where * stands for the generic quark or lepton field, <E> for the SM Higgs field, F for the field strength of the SM gauge fields, and D for the gauge-covariant derivative. The first bracket in Eq. (6) contains two terms, a cosmological constant term and a Higgs mass term, that are proportional to positive powers of A, and are at the origin of two infamous hierarchy problems. The second bracket in Eq. (6) contains operators with no powerlike dependence on A, but only a milder, logarithmic dependence, due to infrared renormalization effects between the cut-off scale A and the weak scale. The last bracket in Eq. (6) is the starting point of an expansion in inverse powers of A, and contains operators associated with rare processes, precision
-T^TT
dlogQ
= - ^
16TT2
(A2 + A/i? - hAt;) + ... , (7) w
v
*
*
'
where Q is the renormalization scale and the dots stand for smaller one-loop contributions, controlled by the electroweak gauge couplings, and higher-order contributions. For any given values of mt and A, we can extract a 'triviality' upper bound on win observing that, if rriH is too large, A(Q) blows up at a scale Qo < A, developing a Landau pole. This leads to some well-known constraints, supported by more rigorous arguments and by lattice calculations: m # < 200 GeV if A ~ MP, mH < 600 GeV if A ~ 1 TeV. Similarly, we can extract a 'stability' lower bound on rrifj by observing that, if m,H is too small, then X(Q) becomes negative at Qo < A, and another minimum of the SM potential develops at () ~ QoSince the results of the previous subsection point to rather small values of mjj, the presently hot issue is the stability bound, recently revisited in 15 . When implementing the stability bound, three options are possible: 1) we can require absolute stability, i.e. the correct electroweak vacuum must have lower energy than the 'wrong' vacuum; 2) we can require stability with respect to high-temperature fluctuations in the cosmo-
394
The Higgs Puzzle: Experiment and Theory
Fabio Zwirner logical evolution of the early Universe; 3) we can require stability with respect to quantum fluctuations at approximately zero temperature. The latter is the most conservative option, and amounts to requiring that the lifetime of the correct electroweak vacuum should be larger than the present age of the Universe, Tu ~ 1010 yrs. The present results are illustrated 15 in Fig. 4, where
sensible value) and van = 115 GeV (close to its minimum allowed value and to the location of the slight experimental effect discussed in subsection 1.1), then option (1) leads to mt < (166 ± 2) GeV. After a new complete one-loop calculation of the tunneling probability at zero temperature, option (3) leads to mt < (175 ± 2) GeV, still in full agreement with the data. Therefore, it may be premature to claim evidence of new physics below Mp from SM vacuum stability, even if we are at the border of the allowed region, a situation for which possible theoretical reasons have been suggested 16 . 2
Figure 4. Instability, meta-stability and stability regions of the SM vacuum in the (mjf,mt) plane, for &s{mz) = 0.118 (solid curves) ±0.002 (dashed and dash-dotted curves). The shaded area indicates the experimental range fro mt, Eq. (5), at la (darker) and 2a (lighter).
as{mz) = 0.118 ± 0.002 and A = MP have been assumed, and in Fig. 5, where ran = i
" • ' i '
M S S M (The Dogma?)
In the SM effective Lagrangian of Eq. 6, the mass term for the Higgs field has a quadratic dependence on the cut-off scale A. When we try to extrapolate the SM to scales much higher than the weak scale, this gives rise to the infamous gauge hierarchy problem. The natural solution to this problem is to introduce new physics close to the weak scale. The present best candidate for such new physics is the Minimal Supersymmetric extension of the Standard Model (MSSM), extensively discussed in another talk at this Conference 17 . 2.1
Some virtues of the MSSM
Figure 5. Running of the quartic Higgs coupling A(/t) for mH = 115 GeV, mt = 165,170,175, 180, 185 GeV and as(mz) ~ 0.118. Absolute stability [X(Mweak) > 0] is still possible if mt < 166 GeV. The hatched region is excluded by the meta-stability bound.
The main virtue of the MSSM is that, if the mass splittings Amsusy that break supersymmetry (SUSY) are of the order of the weak scale Mweak ~ 1 TeV, then its cut-off scale can be naturally taken to be AMSSM = A2susy/Amsusy, where Asusy is the scale of spontaneous SUSY breaking. In hiddensector supergravity models, where Asusy ~ ^MweakMp, such cut-off scale can then be pushed very close to Mp (this is not true if SUSY breaking occurs at lower scales, as in 'gauge-mediated' models).
115 GeV and as(mz) = 0.118 have been assumed. If we set A ~ Mp (its maximum
Another virtue of the MSSM is that, in contrast with other possible solutions of the hierarchy problem, it is generically as good as
^ in GeV
395
The Higgs Puzzle: Experiment and Theory
Fabio Zwirner the SM in complying with electroweak precision data. This is due to the fact that the soft SUSY-breaking mass terms do not break the SU(2)L x U{l)y gauge symmetry. Indeed, it was recently observed 18 that, if sneutrinos and charged sleptons (and, to a lesser extent, charginos and neutralinos) have masses close to their present experimental bounds, then the MSSM may lead to an improved consistency between direct and indirect bounds on the Higgs mass, when the hadronic asymmetries are left out of the global fit. This result is illustrated in Figs. 6 and 7, drawn in planes -7
-8
o x « ID
-10
-11
1
' ' I ' ' ' ' I ' ' ' ' I + : m e a s u r e d values blue: l c r V r e d : MSSM
: SM, m H = 1 1 3 G e V - > 1 3 5 GeV I
, ,
, , I • ,
—i—i—i—I—i—i—i—i—I—i—i—i—i—I—i—i—r-
,
1 I ,
1 ,
,
6
+ : measured values blue: 1 a red: MSSM
e x x 10
Figure 7. T h e same as in Fig. 6 b u t for e\ and £3.
tonic asymmetries and the ratio Rf, have been included in the fit. The elliptic contours represent the region allowed by the data at the lcr level. The irregular contours enclose the typical MSSM predictions for a light spectrum. The fat dot with an arrow shows the : SM, m H =113 GeV-»135 GeV SM prediction for a Higgs mass varying beI , • , , I • • , , I • , , , tween 113 GeV and 135 GeV. We can see 6 that, for a light MSSM spectrum, the agreex 10 J ment between data and theoretical predictions can improve.
Figure 6. Measured values (cross) of £3 and £2, with their lcr region (solid ellipse), obtained from mw, T;, sin 2 d^fUlep) and i?(,. The area inside the irregular curve represents the MSSM prediction for m e - t between 96 and 300 GeV, m ± between 105 and 300 GeV, |/i| < 1 TeV, tan/3 = 10, miR = 1 TeV and rrij\ = 1 TeV.
characterized by two of the three flavourindependent parameters (ei,e2,£3) that are often used in non-SM fits to precision data. We remind the reader that t\, related to Veltman's parameter 5p, is mainly controlled by mt, £2 is particularly sensitive to mw and £3 is mainly controlled by sin 2 9*?Ulep). Only the W mass, the leptonic Z width, the lep-
Another important piece of indirect evidence in favour of the MSSM is the fact that, when combined with a condition on the grand unification of all gauge couplings and with the hypothesis of a 'desert' between the weak scale and the grand unification scale, it leads 19 to one successful prediction for the gauge couplings at the weak scale. To gauge the significance of this success, we can perform a simple-minded but illuminating exercise. We can consider the one-loop renormalization group equation for the running gauge coupling constants da A bA = — oc\ + . . dlogQ 2TT 396
(A = 1,2,3), (8)
The Higgs Puzzle: Experiment and Theory
Fabio Zwirner where 6.4 are the one-loop beta-function coefficients, determined by the gauge quantum numbers of the particle spectrum at the weak scale. If we are agnostic about the precise value of the unified gauge coupling and of the grand unification scale, but we assume the normalization of the U(l)y gauge coupling suggested by the simplest grand-unified models, we can perform a unification test by considering the only variable controlling the prediction for the gauge couplings at the weak scale, the ratio B = (63 — 62)/(&2 — b{). The SM value of this ratio is BSM ^ 0.53, its experimental value is Bexp ~ 0.71, and the MSSM value is BMSSM ^ 0.72. A reasonable error estimate is A B ~ 0.03, completely dominated by the the fact that we do not know the details of the MSSM spectrum and of the spectrum of the underlying theory around the grand unification scale. This is an impressive success, and it is difficult to believe that it is accidental and that we are being fooled by a malicious Nature and by theorists. Any other extension of the SM claiming to be better than the MSSM must face this important phenomenological hint.
2.2
The MSSM Higgs sector
If we take seriously the MSSM, then it is important to extract its predictions for the Higgs sector. As is well known, the MSSM Higgs sector contains two complex doublets, which after gauge symmetry breaking give rise to five physical degrees of freedom, three neutral (h, H, A) and two charged (H±). The prediction of SUSY is that the MSSM Higgs sector depends, at the tree level, only on known SM parameters and two more parameters, for example TUA and tan/3 = v^/vi. After including quantum corrections, the predictions of SUSY are not lost, but the dependences become more complicated and involve all the rest of the MSSM spectrum, in particular the parameters of the top-stop sector 20 . An intense theoretical effort has been devoted 397
over the last years to the precise computation of the MSSM Higgs properties, and we are now at the stage where the calculation of the most important two-loop corrections is being completed. When the top quark mass will be known more precisely, these calculations will be important for reliably comparing models of SUSY breaking with the available bounds on the spectrum. Of course, the relevance of all this could increase further if and when SUSY particles and SUSY Higgs bosons will be found. So far, two-loop corrections to the neutral Higgs boson masses have been computed mostly in the limit of vanishing momentum on the external lines of the Higgs and gauge boson propagators. In this limit, analytical formulae at 0(atas) a r e available, for arbitrary values of the relevant MSSM parameters 21 , and have been implemented in computer codes. As for the C ( a 2 ) corrections, which can be of comparable numerical importance, at the time of this Conference there are only partially analytic formulae 22 for m/,, valid in the limit TUA ^> mzThe general calculation of the 0 ( a 2 ) corrections (in the zero-momentum limit) has been recently completed 23 and agrees with Ref. 22 in the appropriate limit. The effects of the 0 ( Q 2 ) corrections is illustrated in Fig. 8, taken from Ref. 22 . We can see that these corrections can be sizeable, increasing rrih by several GeV in the case of large mixing in the stop mass matrix. Armed with the relevant radiative corrections [the C ( a 2 ) ones have not yet been adequately implemented in the codes, but will presumably be included in the final LEP analyses], experimentalists have searched for direct signals of the MSSM Higgs bosons, as reviewed in another talk at this Conference 3 and described in more detail in Ref. 24 . The small Higgs signal in the SM analysis has its counterpart in the MSSM analysis: some excesses at the ~ la level are reported both in the e + e~ —• hA channel, at (mh, mA) ~ (83,83), (93,93) GeV, and in the
The Higgs Puzzle: Experiment and Theory
Fabio Zwirner
140
' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' M s = l TeV
yU.= - 5 0 0 GeV
Figure 8. The mass mj, vs. the stop mixing parameter X^r , for some representative values of t h e remaining MSSM parameters. T h e two-loop corrections are included either at O(atces) (lower lines) or at 0(atas + a1) (upper lines). The fine structure corresponds to two different methods of implementing the corrections.
e+e~ —• hZ channel, for rrih ~ 97,115 GeV. The lower bounds on the MSSM Higgs masses are notoriously difficult to illustrate, due to their dependence on many parameters. Examples of exclusion plots are presented in Figs. 9 and 10, for a representative choice of MSSM parameters. In a 'benchmark' case characterized by a large mixing in the stop mass matrix, which should lead to conservative bounds on (771^,777,1) and on tan/?, the data have been interpreted 24 in terms of the following exclusion regions at 95% c.l.: (mh,mA) < (91.0,91.9) GeV and tan/? < 2.4. For small stop mixing, the limits are typically stronger. There are other recent interesting studies of the MSSM Higgs sector that would deserve to be discussed. There is just the time to briefly mention them, referring the reader to the corresponding papers. There is a new experimental analysis of the Tevatron data 25 , on the search for pp —> bb(p —• bbbb (
value, tan /3 > 40-50. Some recent theoretical studies 26 have considered the possibility of radiatively induced CP-violating effects in the Higgs sector, coming from explicit CP-violating phases in the squark-gluino sector, and have analyzed the resulting complications in the discussion of the MSSM Higgs searches. Other theoretical studies 27 have examined the implications of the experimental bounds on the MSSM Higgses for different models of SUSY-breaking 'mediation'. 2.3
Some weak points of the MSSM
It would be misleading to end this section without mentioning that, besides its virtues, the MSSM has also, in our present view, a number of weak points. To begin with, the MSSM with its soft SUSY breaking provides only an incomplete, technical solution of the hierarchy problem, since the overall mass scale of the soft terms is set 'by hand'. These soft terms also introduce a very large number of free parameters into the model: this problem is going to stay with us until when a standard model for spontaneous supersymmetry breaking will
398
Fabio Zwirner
0
20
The Higgs Puzzle: Experiment and Theory
40
60
80
100 120 140 mh0 (GeV/c2)
Figure 9. The MSSM exclusion region in t h e (mh,niA) plane, for the 'rre^-max' benchmark scenario. T h e central region is excluded by L E P searches, the lateral ones are theoretically inaccessible in such a scenario.
emerge and/or SUSY particles will be found. ' More seriously, after many years of experimental searches at increasing energy scales, which explored a large part of the theoretically most appealing region from the point of view of the hierarchy problem, no direct experimental hint for the existence of the MSSM Higgs or SUSY particles has been 3n found. Taking all this into account, we should not take the MSSM as a dogma for the new physics at the weak scale, but keep an open mind for the possible alternatives. 3
Can we do without a light Higgs? (The Heresy?)
This part of the talk will touch an issue that it often triggers heated discussions: can we do lo without a light Higgs? Some people view this is as a heresy, some others almost take it for or granted, so it is worth reviewing it, even if the le state of affairs has not changed in an imporr-
399
Figure 10. The MSSM region of the (771,4,tan/3) plane excluded by LEP searches, for the ' m / , - m a x ' benchmark scenario.
tant way during the last year. Since, to the taste of most theorists, there is no satisfactory model without a light elementary Higgs, we may take an agnostic point of view and work at the level of an effective field theory. The most drastic departure from the SM consists in getting rid of the elementary Higgs field, and having the electroweak gauge symmetry SU{2)L x U(l)y non-linearly realized (so doing, of course, we define a nonrenormalizable effective theory whose cut-off scale cannot be much above the TeV scale). This approach has a long history, from the pioneering papers of Ref. 28 to some recent phenomenological discussions 29 after the LEP and Tevatron data. In this approach, the effective Lagrangian is constructed from the Goldstone bosons wa associated with electroweak symmetry breaking, assembled into the group element £ = exp(2iwara/v), where v ~ 256 GeV. Concentrating on the terms that can affect the W and Z propagators, thus playing a major role in the discussion
The Higgs Puzzle: Experiment and Theory
Fabio Zwirner of electroweak precision tests, we can write
renormalization, must be combined with finite contributions Kg (for sin 2 O^ff) an<^ Kw £eff = jTr (D^D'tf) (for mw)- A similar phenomenon occurs in the effective theory with the Higgs field, with + X > d ; ( E , A , . . . ) , (9) the only difference that the logarithmic dei pendence is on run /rnz- At the level of both where effective theories, Kgtw depend on the cutoff and on the unknown dimensionless coefD^ = 9ME + igW^Y, - ig'ZBy (10) ficients of the higher-dimensional operators, is the covariant derivative. The first term in on which we can get reliable information only (9) describes the W and Z masses, as can be if we know about the underlying fundamental seen immediately in the unitary gauge T, — 1. theory. With the present data, it is still posThe higher-order operators Oi are scaled by sible to have A(m.tf) ^> mz without excessive appropriate powers of the cut-off A of this fine-tuning of the quantities KgtwHiggsless theory, and are characterized by diA more careful analysis, however, reveals mensionless coefficients £j. the present advantage of the light Higgs hyA less drastic approach consists in keep- pothesis. First, it must be said that, despite ing the elementary Higgs field , so that a lot of effort, so far there are no good candiSU{2)i x U(l)y can be linearly realized, dates for the underlying theory that realizes but in allowing the most general set of non- the desired situation, i.e. the phenomenorenormalizable operators compatible with logically correct magnitudes and signs of Kg the electroweak gauge symmetry and with and K\y, without disrupting the predictions Poincare invariance. Also this approach has for other observables, and avoiding 'ad hoc' a long history, from the early paper of Ref. 30 theoretical constructions. Also, it can be to other recent phenomenological discussions immediately seen, in the linear realization, 31 after the LEP and Tevatron data. In this that there is an obvious correlation: if we incase, the appropriate effective Lagrangian is crease rnu we must correspondingly decrease Ceff=CsM()
+ YlciOi(,A,...),
(11)
i
where A is the cut-off and Cj are the dimensionless coefficients of the various operators In both approaches, the theoretical expressions for the two key pseudo-observables in the fits, sin 9^?* and mw, differ from the SM ones. For given values of all the other parameters, in the SM they are just functions of TTIH, with their leading dependences proportional to log(mjj/mz)In these new frameworks, the dependences become more complicated:
^ — - { ! ° g i + ^ (*'*)• mz
l
l o
S^
\Ci,AJ
(12) In the Higgsless effective theory, the logarithmic dependence on A/mz, generated by
A, and tune the coefficients Cj, to keep agreement with the data: then mn ~ mz and A > mz looks as the most natural solution! There was a recent survey 3 2 of models that may evade the constraint of having a light Higgs with ran ~ mz- Three classes of models were identified, making reference to the (S,T) parameters, analogous to the (£3,61) parameters of Fig. 7: (i) those in which new physics can produce negative contributions to the S parameter; (ii) those with a new vector bosons Z' close to the weak scale; (iii) those in which new physics can produce positive contributions to the T parameter. Without going to the details, the important point is that all these models exhibit a rich phenomenology around the TeV scale, accessible to accelerators such as the Tevatron 3 3 , the LHC 3 4 and a possible high-energy lin-
400
The Higgs Puzzle: Experiment and Theory
Fabio Zwirner ear e+e 4
collider
.
N e w theoretical ideas (Crackpot religions?)
Electroweak symmetry breaking is a testing ground for new ideas that are restlessly explored by adventurous theorists, despite the fact that some of their most conservative colleagues may view them (in the words of Monty Python) as 'crackpot religions'. It is appropriate to comment here on some ideas that have been very actively explored in the last years, focusing on the aspects that are most strictly related with Higgs physics. The two big hierarchy problems of our present theories are the cosmological constant problem and the gauge hierarchy problem, related with the two operators of dimension d < 4 in the SM effective Lagrangian, conveniently rewritten as: ^•eff
=
h-cosm + Aweak4>
+ ... .
(13)
We must explain why the scales of the vacuum energy and of the Higgs mass satisfy the phenomenological bounds A 0(1O~ 3 eV) (as discussed in another talk at this Conference 36 ) and Aweak ~ 0 ( 1 TeV), with the intriguing numerical coincidence A cosra ~ A2weak/MP. Are the two hierarchy problems related? It turns out that they are in supergravity and superstring theories, where supersymmetry becomes a local symmetry and gravity is automatically included. In these theories, formulating an acceptable model for SUSY breaking is difficult, precisely because the problem of the weak scale (most probably linked to the scale of SUSY-breaking masses) and the problem of the cosmological constant scale must be addressed at once. Many theorists feel that models formulated in more than four space-time dimensions may offer unconventional solutions to these problems and, perhaps, some exotic phenomenology to be explored experimen-
401
tally. Since these models are the subject of another talk at this Conference 3 7 , some comments related with electroweak symmetry breaking and the gauge hierarchy problem will be sufficient here (for other recent reviews and references on extra dimensions, see e.g. Ref. 3 8 ). One of the most interesting features of models with extra dimensions is the fact that the hierarchy Mweak/Mp can be linked with some geometrical object, in the simplest case a compactification radius R characterizing the size of one or more compactified dimensions. Such a relation is strongly model-dependent. In toroidal compactifications, we can get power-like relations such as (M^eak/Mp) oc R~n, where n is an integer and the dimensionful proportionality coefficient is model-dependent. In 'warped' compactifications, we can get exponential relations of the form (Mweak/Mp) ~ exp(—MpR). The gauge hierarchy problem is then reformulated in a very interesting way: it amounts to understanding the stability and the dynamical origin of the value of the radius R that fits the phenomenological value of Mweak. There is no compelling idea so far in this direction, but some intriguing features are emerging and are at the center of an intense theoretical activity. Before describing some of the possibilities, it is worth mentioning that the problem of determining R is analogous to the problem of understanding the stability and the dynamical origin of Amsusy, the scale of SUSY-breaking mass splittings, in conventional, four-dimensional models of spontaneous SUSY breaking. The analogy becomes evident in those (higher-dimensional) superstring 39,40 and supergravity 41 models where the radius R does indeed control Amsusy. If there are symmetries of the higherdimensional theory whose breaking is nonlocal in the extra dimensions, symmetrybreaking quantities may be shielded from UV effects, and determined by the infrared dy-
Fabio Zwirner
The Higgs Puzzle: Experiment and Theory
namics. As an example, the field-dependent one-loop effective potential of some superstring 40 (and field-theory 42 ) compactifications does not contain positive powers of the string scale (cutoff scale A)
that had initially a classical scale invariance, it may be regarded as the analogous of soft breaking in the MSSM. This mechanism may be stable and related to a dimensional transmutation via the 'holographic' picture 47 . Coming back to the main subject of V1(R,4>) = R-4 + R~202 + (p4 + ... , (14) the present talk, what features may emerge for Higgs phenomenology? It may be too where all coefficients have been omitted and early to tell. One possibility is the mixthe dots stand for logarithmic corrections asing between the Higgs boson(s), charged unsociated with the infrared running of the couder the electroweak gauge symmetry, and plings. Starting from a higher-dimensional the spin-0 fields, neutral under the electheory whose symmetries forbid a Higgs mass troweak gauge symmetry, that are associated term (and ignoring the radius dynamics), ran with the compactification radius ('radion') and v = (
402
The Higgs Puzzle: Experiment and Theory
Fabio Zwirner CDF PRELIMINARY Run 1 X)
eorrbired CDF/DO thresholds
95% C.L. upper limits; II bb
X
t
•* —
' " "to bb " " ' ' qqbb vv'M.
^
% 10 "
m x
"
'"*"" -
VH combined 0 LxJ O
DC >
80
-
_J O X UJ
~
04
a.
10
90
.
.
1
100
140
Higgs mass
160
180
200
(GeV/c 2 )
Figure 12. T h e integrated luminosity required per experiment, to either exclude a SM Higgs boson at 95% c.l. or discover it at the 3cr or 5
Standard ModeT~—^^ .
120
;
UI
.
100
,
.
1
110
,
,
,
!
120
,
,
,
,
130
Higgs Mass (GeV/c2) Figure 11. Preliminary upper limits (at 95% c.l.) on production cross-sections times branching ratios, as functions of mjj, from Run I of the C D F experiment.
from CDF 49 (DO had a slightly smaller sensitivity). With the luminosity and detectors of Run I, Tevatron is still more than one order of magnitude away from the sensitivity required by the SM Higgs properties. However, as described in detail in a dedicated study 50 , and summarized in Fig. 12, things will be different, and very challenging, in the near future. For run < 135 GeV, the focus of the present attention, CDF and DO will search for the SM Higgs boson considering its associated production with a weak gauge boson, pp -> V + {H -> bb) (V = W±,Z), and looking at a number of different final states: (Zi/)(6&), {l+r)(bb), {W){bb). Serious backgrounds are Vbb, VV, it, single top, and others. For run > 135 GeV, the channel gg —> H —> W i y W becomes accessible, and the useful final states are (^l^jj) and (l+l~uV). (In this Section I will always stand for e or fi.) The results of Fig. 12 are obtained by combining the statistical power of both experiments and all the channels mentioned
above. The lower edge of the bands is the calculated threshold; the bands extend upward from these nominal thresholds by 30% as an indication of the uncertainties in btagging efficiencies, background rate, mass resolution, and other effects. The Higgs hunt will continue at the LHC (whose status is summarized by another talk at this Conference 3 4 ) . In the mass region run > 130 GeV, the job of the ATLAS and CMS experiments will be relatively easy, thanks to the gold-plated channel H -> ZZ^ -> 4 ^ , with other channels as a backup for the mass regions with less statistics: H -+ W W W —• Ivlv for mH ~ 2mw ± 30 GeV, H -> ZZ -> l+l~i>v (and possibly H —>• WW -> li/jj or H -+ ZZ -> l+l~jj) for mH > 600 GeV. In the case of a light Higgs, m # < 130 GeV, various different signals are available. Earlier studies have defined the strategies for signals such as inclusive H —> 77, it + (H —> bb, 77) and V + (H —• bb, 77). The combined discovery potential of the ATLAS and CMS experiments, for different integrated luminosities, is summarized 51 in Fig. 13. During the last year, there was progress 52 in the study of the channel qq —> (WW -> H)+jj: exploiting the two tagged forward jets, the
403
Fabio Zwirner
The Higgs Puzzle: Experiment and Theory gated
56
. The NLO QCD corrections to ~> tiH + X were computed by two different groups 57 .
PP{PP) L = 10tV' l- = 30fb'' L-lOOfb'1
ATLAS + CMS (noK-factors)
6
10
(GeV)
Figure 13. Sensitivity for the discovery of a SM Higgs boson at the LHC, as a function of mu • The overall statistical significance, integrated over different channels, is plotted for three different integrated luminosities (10, 30 and 100 f b - 1 ) , and assumes the combined statistical power of the ATLAS and CMS experiments.
background can be consistently reduced, allowing the study of decay channels such as H -^ W^W* -> lul'v, which may be a discovery mode for TO# ~ 115 GeV. There are many other recent phenomenological studies on Higgs physics at highenergy colliders that would deserve a detailed discussion. Time limitations just permit a brief mention of some SM studies, with reference to the corresponding papers, leaving aside analogous studies for the MSSM and for more exotic possibilities. Soft and virtual NNLO QCD corrections to gg —> H + X have been computed 5 3 . 'Strong' weak effects at high-energies (Bloch-Nordsieck violations) were studied 54 : in particular, (ayy/7r)log (s/m^y) corrections to a(e + chadrons) and an enhanced mu dependence in WLWL —> hadrons. The crosssection for Higgs + 2 jets via gluon-gluon fusion was computed 55 . High-px Higgs signals from WW —> H —> bb were investi-
404
Conclusions
In the presence of an experimental and theoretical puzzle, as recalled by the title assigned to this talk by the Organizers, conclusions can only be tentative. It is clear that the search for the 'Higgs boson' (or, more generally, for the dynamics underlying the spontaneous breaking of the electroweak gauge symmetry) is the main goal of high-energy physics in the present decade. Direct searches and electroweak precision tests strongly constrain the possibilities: with the presently available information, the existence of at least one light Higgs boson with SM-like (or MSSM-like) properties looks like the best bet, but there is still room for the unexpected. It is important to stress that, in all 'natural' models, the Higgs boson is not alone: the accompanying physics may be even richer in implications (as, for example, in the case of supersymmetry), and we must be prepared to fully explore the TeV scale. While not very successful so far, the theoretical search for plausible alternatives to the SM and the MSSM is worth pursuing, as confirmed by the many ongoing activities along different directions, in particular extra dimensions. The final (scientific) judgement is coming, and experiment will express it with the help of run II of the Tevatron, of the LHC, and hopefully more facilities to come . . . In our quest for the fundamental laws of Nature, there is no substitute for the highenergy frontier!
Fabio Zwirner
The Higgs Puzzle: Experiment and Theory
Questions
Pietro Slavich for useful comments. This work was supported in part by the European Union under the contracts HPRN-CT2000-00149 (Collider Physics) and HPRNCT-2000-00148 (Across the Energy Frontier).
Q. Bennie Ward, MPI and Univ. of Tennessee: There is the ultra-conservative view that we have a light Higgs, they will find it and there is nothing else. It is that way because God made it like that, unnatural or not. The theories in extra dimensions you mention are non-renormalizable so, if you would find them, you are still left with their non-renormalizable artifacts. Why would you say one of these scenarios is better than the other? A. In view of naturalness arguments, the possibility of finding a light Higgs and nothing else at the weak scale seems unlikely. Rigorously, we cannot exclude that the gauge hierarchy problem is solved by mysterious infrared-ultraviolet connections that we are unable to understand with the tools of conventional quantum field theory. However, not even the cosmological constant violates so far the naturalness criterion, since gravitational interactions have been tested only up to energy scales of the order of 10~ 3 eV, not far from the phenomenological value of Acosm in the normalization of Eq. (13). Coming to the second part of your question, extra dimensions are just one out of many possibilities for the new physics at the Fermi scale. Their phenomenology may be described by an effective field theory, but the latter must eventually find an ultraviolet completion: this may be provided, for example, by an underlying superstring theory. Acknowledgments The author would like to thank Riccardo Paramatti and Paolo Valente for precious technical help in the preparation of the talk, Giuseppe Degrassi for discussions on the content of section 1.2, Juliet Lee-Franzini and
405
References 1. P.W. Higgs, Phys. Lett. 12, 132 (1964); Phys. Rev. Lett. 13, 508 (1964); Phys. Rev. 145, 1156 (1966); F. Englert and R. Brout, Phys. Rev. Lett. 13, 321 (1964). 2. R. Barbieri, these Proceedings. 3. G.G. Hanson, these Proceedings. 4. The LEP Higgs Working Group, LHWG Note/2001-03, hep-ex/0107029. 5. J. Drees, these Proceedings. 6. A summary of the preliminary results of the LEP Electroweak Working Group is available at the LEPEWWG home page, http://lepewwg.web.cern.ch/LEPEWWG 7. A. Freitas, W. Hollik, W. Walter and G. Weiglein, Phys. Lett. B 495, 338 (2000) [hep-ph/0007091]. 8. J. Z. Bai et al. [BES Collaboration], hep-ex/0102003; R. R. Akhmetshin et al. [CMD-2 Collaboration], hepex/9904027. 9. H. Burkhardt and B. Pietrzyk, Phys. Lett. B 513, 46 (2001). 10. A.D. Martin, J. Outhwaite and M.G. Ryskin, Eur. Phys. J. C 19, 681 (2001) [hep-ph/0012231]. 11. F. Jegerlehner, hep-ph/0104304. 12. L. Demortier, R. Hall, R. Hughes, B. Klima, R. Roser and M. Strovink [The Top Averaging Group for the CDF and DO Collaborations], FERMILABTM-2084. 13. M.S. Chanowitz, hep-ph/0104024. 14. N. Cabibbo, L. Maiani, G. Parisi and R. Petronzio, Nucl. Phys. B 158, 295 (1979). 15. G. Isidori, G. Ridolfi and A. Strumia, Nucl. Phys. B 609, 387 (2001) [hep-
Fabio Zwirner ph/0104016]. 16. C D . Froggatt, H.B. Nielsen and Y. Takanishi, hep-ph/0104161. 17. J. Ellis, these Proceedings. 18. G. Altarelli, F. Caravaglios, G.F. Giudice, P. Gambino and G. Ridolfi, JHEP 0106, 018 (2001) [hep-ph/0106029]. 19. H. Georgi, H. R. Quinn and S. Weinberg, Phys. Rev. Lett. 33, 451 (1974); S. Dimopoulos, S. Raby and F. Wilczek, Phys. Rev. D 24, 1681 (1981); L. E. Ibanez and G. G. Ross, Phys. Lett. B 105, 439 (1981). 20. J. Ellis, G. Ridolfi and F. Zwirner, Phys. Lett. B 257, 83 (1991) and Phys. Lett. B 262, 477 (1991); Y. Okada, M. Yamaguchi and T. Yanagida, Prog. Theor. Phys. 85, 1 (1991) and Phys. Lett. B 262, 54 (1991); H. E. Haber and R. Hempfling, Phys. Rev. Lett. 66, 1815 (1991). 21. R. Hempfling and A.H. Hoang, Phys. Lett. B 331, 99 (1994) [hepph/9401219]; S. Heinemeyer, W. Hollik and G. Weiglein, Phys. Rev. D 58, 091701 (1998) [hep-ph/9803277]; Phys. Lett. B 440, 296 (1998) [hepph/9807423]; Eur. Phys. J. C 9, 343 (1999) [hep-ph/9812472]; Phys. Lett. B 455, 179 (1999) [hep-ph/9903404]; R. Zhang, Phys. Lett. B 447, 89 (1999) [hep-ph/9808299]; J.R. Espinosa and R. Zhang, JHEP 0003, 026 (2000) [hepph/9912236]; G. Degrassi, P. Slavich and F. Zwirner, Nucl. Phys. B 611, 403 (2001) [hep-ph/0105096]. 22. J.R. Espinosa and R. Zhang, Nucl. Phys. B 586, 3 (2000) [hepph/0003246]. 23. A. Brignole, G. Degrassi, P. Slavich and F. Zwirner, preprint DFPD-01/TH/40, RM3-TH/01-14, ROME1-1327/01, to appear. 24. The LEP Higgs Working Group, LHWG Note/2001-04 [hep-ex/0107030], LHWG Note/2001-05 [hep-ex/0107031], LHWG
The Higgs Puzzle: Experiment and Theory
25.
26.
27.
28.
29.
30. 31.
32. 33. 34. 35. 36. 37. 38.
39.
40.
406
Note/2001-06 [hep-ex/0107032] and LHWG Note/2001-08 [hep-ex/0107035]. T. AfFolder et al. [CDF Collaboration], Phys. Rev. Lett. 86, 4472 (2001) [hepex/0010052]. M. Carena, J.R. Ellis, A. Pilaftsis and C.E. Wagner, Phys. Lett. B 495, 155 (2000) [hep-ph/0009212]. S. Ambrosanio, A. Dedes, S. Heinemeyer, S. Su and G. Weiglein, hepph/0106255. T. Appelquist and C.W. Bernard, Phys. Rev. D 22, 200 (1980); A.C. Longhitano, Phys. Rev. D 22, 1166 (1980) and Nucl. Phys. B 188, 118 (1981). B.A. Kniehl and A. Sirlin, Eur. Phys. J. C 16, 635 (2000) [hep-ph/9907293]; J.A. Bagger, A.F. Falk and M. Swartz, Phys. Rev. Lett. 84, 1385 (2000) [hepph/9908327]. W. Buchmuller and D. Wyler, Nucl. Phys. B 268, 621 (1986). L.J. Hall and C. Kolda, Phys. Lett. B 459, 213 (1999) [hep-ph/9904236]; R. Barbieri and A. Strumia, Phys. Lett. B 462, 144 (1999) [hep-ph/9905281]; C. Kolda and H. Murayama, JHEP 0007, 035 (2000) [hep-ph/0003170]. M.E. Peskin and J.D. Wells, Phys. Rev. D 64, 093003 (2001) [hep-ph/0101342]. Y.Y. Kim, these Proceedings. L. Maiani, these Proceedings. R. Heuer, these Proceedings. M. Turner, these Proceedings. L. Randall, these Proceedings. I. Antoniadis and K. Benakli, Int. J. Mod. Phys. A 15, 4237 (2000) [hep-ph/0007226]; V. A. Rubakov, hepph/0104152. R. Rohm, Nucl. Phys. B 237, 553 (1984); C. Kounnas and M. Porrati, Nucl. Phys. B 310, 355 (1988); S. Ferrara, C. Kounnas, M. Porrati and F. Zwirner, Nucl. Phys. B 318, 75 (1989). I. Antoniadis, Phys. Lett. B 246, 377
The Higgs Puzzle: Experiment and Theory
Fabio Zwirner (1990). 41. J. Scherk and J. H. Schwarz, Phys. Lett. B 82, 60 (1979) and Nucl. Phys. B 153, 61 (1979); P. Fayet, Phys. Lett. B 159, 121 (1985) and Nucl. Phys. B 263, 649 (1986). 42. I. Antoniadis, S. Dimopoulos and G. R. Dvali, Nucl. Phys. B 516, 70 (1998) [hep-ph/9710204]. 43. A. Delgado, A. Pomarol and M. Quiros, Phys. Rev. D 60, 095008 (1999) [hep-ph/9812489]; I. Antoniadis, K. Benakli and M. Quiros, hep-th/0108005; R. Barbieri, L.J. Hall and Y. Nomura, Phys. Rev. D 63, 105007 (2001) [hep-ph/0011311]; N. Arkani-Hamed, L.J. Hall, Y. Nomura, D.R. Smith and N. Weiner, Nucl. Phys. B 605, 81 (2001) [hep-ph/0102090]; G.R. Dvali, S. Randjbar-Daemi and R. Tabbash, hep-ph/0102307. 44. H. Georgi, A. K. Grant and G. Hailu, Phys. Lett. B 506, 207 (2001) [hep-ph/0012379]; R. Contino, L. Pilo, R. Rattazzi and A. Strumia, JEEP 0106, 005 (2001) [hep-ph/0103104]; N. Arkani-Hamed, A. G. Cohen and H. Georgi, Phys. Lett. B 516, 395 (2001) [hep-th/0103135]. For some works appeared after the Conference, see also: D. M. Ghilencea, S. Groot Nibbelink and H. P. Nilles, hep-th/0108184; C. A. Scrucca, M. Serone, L. Silvestrini and F. Zwirner, hep-th/0110073. 45. C. Kounnas, F. Zwirner and I. Pavel, Phys. Lett. B 335, 403 (1994) [hepph/9406256]. 46. W.D. Goldberger and M.B. Wise, Phys. Rev. Lett. 83, 4922 (1999) [hepph/9907447]. 47. N. Arkani-Hamed, M. Porrati and L. Randall, JEEP 0108, 017 (2001) [hep-th/0012148]; R. Rattazzi and A. Zaffaroni, JHEP 0104, 021 (2001) [hep-th/0012248]. 48. G.F. Giudice, R. Rat-
407
49. 50. 51.
52.
53.
54.
55.
56.
57.
tazzi and J.D. Wells, Nucl. Phys. B 595, 250 (2001) [hep-ph/0002178]; E. Perazzi, G. Ridolfi and F. Zwirner, Nucl. Phys. B 574, 3 (2000) [hep-ph/0001025] and B 590, 287 (2000) [hep-ph/0005076]. L. Moneta [CDF collaboration], hepex/0106050. M. Carena et al, hep-ph/0010338. F. Gianotti, based on the ATLAS Technical Design Report, CERN/LHCC/9914. For a recent review, appeared after this Conference, see also: J. G. Branson, D. Denegri, I. Hinchliffe, F. Gianotti, F. E. Paige and P. Sphicas (for the ATLAS and CMS Collaborations), hepph/0110021. N. Kauer, T. Plehn, D. Rainwater and D. Zeppenfeld, Phys. Lett. B 503, 113 (2001) [hep-ph/0012351]. S. Catani, D. de Florian and M. Grazzini, JHEP 0105, 025 (2001) [hep-ph/0102227]; R.V. Harlander and W.B. Kilgore, Phys. Rev. D 64, 013015 (2001) [hep-ph/0102241]. M. Ciafaloni, P. Ciafaloni and D. Comelli, Phys. Rev. Lett. 87, 211802 (2001) [hep-ph/0103315] and Nucl. Phys. B 613, 382 (2001) [hepph/0103316]. V. Del Duca, W. Kilgore, C. Oleari, C. Schmidt and D. Zeppenfeld, Phys. Rev. Lett. 87, 122001 (2001) [hepph/0105129]. V.A. Khoze, A.D. Martin and M.G. Ryskin, Eur. Phys. J. C 21, 99 (2001) [hep-ph/0104230]. W. Beenakker, S. Dittmaier, M. Kramer, B. Plumper, M. Spira and P.M. Zerwas, Phys. Rev. Lett. 87, 201805 (2001) [hep-ph/0107081]; L. Reina and S. Dawson, Phys. Rev. Lett. 87, 201804 (2001) [hep-ph/0107101].
T H E M U O N ANOMALY: E X P E R I M E N T A N D T H E O R Y J. P. MILLER for the Muon (g-2) Collaboration1 Department of Physics, Boston University, 590 Commonwealth Ave., Boston, MA 02215 USA E-mail: [email protected] A summary is given of the present status of the theory and experiment of the anomalous magnetic moment of the muon. A difference between predicted and measured values is an indication of physics beyond the Standard Model. A new experimental measurement has produced a value which differs from a recent Standard Model prediction by about 1.6 standard deviations. When first announced, the discrepancy was about 2.6 standard deviations, but theorists have recently found an error in the sign of the largest term in the standard model hadronic light-by-light contribution which reduces the difference. Additional d a t a are being analyzed and elements of the theory are being scrutinized to provide, in the future, a sharper test of theory.
1
Introduction
The magnetic moment of a particle is given in terms of its spin by
U=
9&
2mc
3
and the anomaly is defined as a = \(g — 2). Historically, the measurement of particle magnetic moments has been a valuable test of existing theories. For example, magnetic moment measurements on the hyperons have provided essential information on their substructure, and the electron anomaly has been the most stringent test of QED. The Dirac theory predicts that g = 2 (a — 0) for point particles with spin \. While the hyperons have g factors very different from 2 because of their complex substructure, the leptons have j w 2 and anomalies which are nearly zero, consistent with the current evidence that they are point particles. The Standard Model predicts lepton anomalies on the order of one part in 800 due to their field interactions. In the cases of the electron and the muon, both the Standard Model predictions and the measurements are extremely precise, a relatively rare situation resulting in valuable tests of the theory. As we shall
see, however, the muon typically has a much stronger sensitivity than the electron to any physics which has not been included in the Standard Model. By far the largest contribution to the lepton anomalies comes from the lowest order electromagnetic diagram, the Schwinger term (left diagram in Fig. 1), which gives a{QED;\) = ^ , e.g. the same for muons and electrons. The next order electromagnetic diagrams, which involve virtual lepton (Fig. 1) or hadron (Fig. 2) loops, are small compared to the Schwinger term, however, they are much larger for the muon than the electron as a result of the additional available rest mass energy. The difference in the contribution between the electron and muon in diagrams involving massive virtual particles typically scales as (2^-) 2 « 40000, and it is this large factor which makes aM far more sensitive than ae to any unknown massive particles. It is instructive to compare the values of the electron and muon anomalies. From Penning trap 2 experiments, we have for the electron aeex_p = (1159652.1884 ± .0043) x 10" 9 (4 ppb) and positron aeef = (1159652.1879 ± .0043) x 10~ 9 (4 ppb), agreeing within errors as required by CPT invariance. These val-
408
J. P. Miller
The Muon Anomaly: Experiment and Theory
QED
(9.4 ppm)
H+
CERN
(10 ppm) E821 (97) n+
-•
(13 ppm)
-
(5 ppm)
E821 (98)
n+
1 o o
| o o
O O
O i—
a>
LO CD
CT> LO CO
592 ooo
Figure 1. QED contributions to the anomaly. The first diagram is the lowest-order Schwinger term. The other diagrams are representative of higher-order QED contributions.
-
(1.3 ppm) E821 (99) H+ I I 1 I II II II1M I M M I | IMI | M M | M M
CD
o o o CO
CJ> LO CO
o
o o - *
LO CD
o
o _ o ^Z LO
a)
LO CD
'
o
i— X
Figure 3. Experimental measurements of a M .
yh Figure 2. The first-order hadronic diagram.
ues are are in good agreement with the theoretical prediction of QED to fourth order in f, af = (1159652.1535 ± .0240) x 10~ 9 (21 ppb) 3 . The theoretical error is dominated by the error in the value of a taken from Quantum Hall Effect experiments. If one assumes the correctness of QED, then the best determination of a comes from the electron anomaly measurements. T h e hadronic and electroweak contributions, 1.63(3) x 1 0 - 1 2 and 0.030 x 10~ 12 , respectively, are small and the errors are negligible compared to the experimental errors. The theoretical and experimental values for the aM are not known nearly as well as for ae. In 1999, prior to t h e new published result, the world average of measured values was ae*p = (1165920.5 ± 4.6) x 10" 9 (4 ppm). This included 9.4 and 10 ppm results for the positive and negative muon, respectively, in a series of famous experiments at CERN ending in the 1970's, 6 as well as the results from the new Brookhaven muon (g-2)
409
experiment (E821) using positive muon data taken in 1997 (13 ppm) 7 and 1998 (5 ppm) 8 . All of these measurements are in agreement within their errors, as shown in Fig. 3; the negative muon data were incorporated under the assumption of CPT invariance. Using the recent published compilation of the theoretical ingredients to a^ by Czarnecki and Marciano, 3 af = (1165915.96 ±0.67) x 10" 9 (0.6 ppm), which used the hadronic evaluation by Davier and Hocker,4 we find that agreement between experiment and data was -,th good: (4.5 ± 4.7) x 10" 9 . Recently, an error in the hadronic light-by-light contribution was found, 30 which changes the theoretical prediction to off = (1165917.68 ± 0.67) x 10" 9 (0.6 ppm). This improves the agreement between theory and experiment: ,th =_ (2.8 ±4.7) x 10 - 9 ..exp The new experimental result for aM+, based on a sample of 1 billion positrons collected in 1999 by E821, 9 is ajf p (£821) = (1165920.2±1.4±0.6) x 10" 9 (1.3 ppm). The new world value changes very little, ae*p = (1165920.3 ± 1.5) x 10" 9 (1.3 ppm). Comparing with the uncorrected theoretical number gives a, 2.6 a difference between measurement and theory: Aa^ = a^p < = (4.3± 1.6) x 10" 9 (3.7 ± 1.4 ppm). Including the
The Muon Anomaly: Experiment and Theory
J. P. Miller light-by-light correction, Aa M = a^p — aft = (2.6 ± 1.6) x 10" 9 (2.2 ± 1.4 ppm), a 1.6a difference. Two additional data sets, from 2000 and 2001 runs, are currently being analyzed. The 2000 data set consists of about 4 billion events for the /x+ and 2001 consists of about 3 billion events for the /J,~ . The original stated goal of E821 was to reduce the anomaly measurement error to 0.35 ppm. Ideally, one would also equalize the errors on the /x+ and yT in order to optimally test CPT invariance and to study systematic issues in the experiment. We expect to come close to this goal, but this will require a future data run with 6 billion events. 2
Theory Status
The Standard Model contributions to a^ can be conveniently separated into QED, electroweak, and hadronic portions. Although the anomaly is dominated by the QED contribution, there are significant hadronic and EW contributions at the level of « 58.3 ppm and « 1.3 ppm, respectively. We discuss each of these contributions below, with emphasis on the hadronic contribution, whose error dominates the overall error in af/1. The QED contribution, using a from aexp a n c j c a i c u i a t e d to fifth order in — (some of the highest order diagrams were estimated), contributes 3 an error of 21 ppb to aft, a^ED = 116584705.7(2.9) x 1 0 ~ n . Kinoshita 10 reports that some of the fourth order terms are being re-calculated with 128 bit precision, which may result in a small shift in the QED contribution; this is expected to have minimal impact on the comparison between experimental and theoretical values of The lowest order electroweak diagrams, involving the exchange of a W, Z or Higgs, are shown in Fig. 4. The electroweak contribution, including the 25% reduction from higher order terms, is 3 a^w — 151(4) x
+389
-194
<1 -ll
X 10 Figure 4. Lowest-order electroweak contributions.
10 _11 (1.30 ± 0.03)ppm. The theoretical uncertainty is very small. The central value is right on the edge of the current experimental error, and it will be a significant contribution at the experimental error goal of 0.35 ppm. The lowest order hadronic diagram is shown in Fig. 2, where a hadron loop has been inserted into the Schwinger diagram. Since these contributions involve the strong interaction at low energies, they cannot be calculated from first principles. Their contribution can, however, be determined from measured e+e~ scattering cross sections over all energies through the use of the dispersion relation, Eq. 1, (also see Fig. 5(a))
a,(had;l)
= ( ^ )
2
f°° ^K(s)R(s)
(1)
K(s) is a slowly varying function and the e+e~ data enter through the ratio of cross sections, R(s) = ' ' f f . ^ ' . The low energy data are the most important as a result of their large amplitude and the \ term in the integrand, where y/s is the center-ofmass energy. High quality hadronic T decay data from LEP and Cornell can be used to augment the isovector part of the e+e~ data at energies below mTc2 (second diagram in Fig. 5) using isospin invariance (to relate for example 7r-7r° channels in r decays to ir+ir~ channels in e+e~ collisions) and the CVC hypothesis (to connect the W~ and photon intermediate states). 11 There have been a number of evaluations of a M (had;l) over the past two decades. We
410
J. P. Miller
The Muon Anomaly: Experiment and Theory
V/vwi^r (a) Figure 5. The electron scattering and tau decay diagrams relevant to the determination of the lowest order contribution to a^.
Table 1. As of 1995, contributions to a^(had;l) as a function of e+e~ energy, illustrating the (soon-to-be obsolete) sources of error. 1 2 . W i t h the incorporation of new e + e ~ and T data, the errors below 5 GeV decrease significantly from the values shown in the table.
V^GeV
a^had; 1)
Error, ppm
<1.4
87. %
1.29
1.4-• 2.0
4.6%
0.21
2 . 0 - • 3.1
4.0%
0.30
2.0 -> 2.6
2.9%
0.27
2.6-> 3.1
1.1%
0.12
J / * (6 states)
1.3%
0.08
QCD 3.1 - • oo
3.0%
0.03
Total
800
1000
900
Eon,
MeV
Figure 6. Preliminary pion form factor d a t a from Novosibirsk in the vicinity of the p resonance. 1 3 Inset: data the u> interference region.
1.37
will look in some detail at those performed since 1995. The contributions of e+e~ data to a^had; 1) in different energy ranges, from the evaluation by Brown and Worstell, 12 are indicated in Table 1. (We note that the new data which are becoming available will render this table obsolete.) The largest contribution to the value and error are for y/s < 1.4 GeV, a region dominated by the effects of the p resonance (see Figs. 6-8). The other energy ranges give a considerably smaller contribution to the error, however for an 0.35 ppm measurement they cannot be neglected, particularly in the range 1.4 —» 2.6 GeV. Since 1995, there has been a substantial improvement in the data quality, but major
411
portions of the data are either preliminary or are still in the process of being incorporated into evaluations of a^(had;l). Data on the pion form factor (which can be directly related to the e+e~ cross section) from the CMD2 and SND experiments at the VEPP-2M accelerator in Novosibirsk are nearing publication. They cover the important energy range \/s < 1.4 GeV. Preliminary data in the p resonance range (600 to 930 MeV) from CMD2 are shown in Fig. 6. 13 Their anticipated systematic error of 0.6% in this range would reduce the error contribution from the e+e~~ data in this energy region by better than a factor of two. The VEPP2000 project is an upgrade under construction at Novosibirsk which will extend quality e+e~ measurements up to 2 GeV, with an order of magnitude or more improvement in luminosity. Data relevant to the low energy region are also being taken by the KLOE 14 experiment at DA^NE. They operate at the mass, and then derive e+e~ cross sections at lower energies using the so-called radiative
The Muon Anomaly: Experiment and Theory
J. P. Miller
•
BESII 1999(Prellmlnary)
•
BESII 1998, P B L 84(2000)594
O
Gamma2
D
Mark)
•
pluto
o 60°<6r<120' • 5°<0I<21°
Figure 8. Preliminary and published e + e ~ data, 2-5 GeV, from BESII, along with older d a t a from other experiments 1 5 . 400
500
700
800 900 2 K invariant mass, MeV
been produced over the last few years as described by P. Roudeau at this conference, have significantly reduced the contribution to Figure 7. Preliminary pion form factor d a t a from rathe error of a M (had;l) for energies below the diative return measurements (KLOE). 1 4 r mass. 11 In addition, new analysis results should be available soon from ALEPH. 16 17,18 have questioned, however, return method. In this approach, the initial Some authors whether there are sufficient controls over rastate electron or positron radiates away some diative corrections and corrections to the apof its energy via a photon, providing access to scattering at energies below the <j> mass. proximations of CVC and isospin invariances. Preliminary data are shown in Fig. 7. Cur- In addition, there appears to be an overrently their systematic errors are on the or- all normalization disagreement between the 19 der of a few percent, however, they expect to T data from ALEPH and Cornell. The r improve this in the near future to the point data represent, nevertheless, valuable addi+ that they are competitive with the Novosi- tions and checks on the e e~ data at low enbirsk data. This measurement, along with ergies, and various groups plan to study these possible plans at other e+e~ machines such issues in the future. as the B-factories and Cornell which have acA list of a M (had;l) evaluations is given cess to higher energies, will contribute signif- in Table 2. Various combinations of ingrediicantly to the next wave of precision e + e~ ents were used: e+e~ data from a wide range data. of experiments, r decay data from LEP and There are new, preliminary and published data from BESII 15 in the important energy range 2 - 5 GeV (Fig. 8). Note especially that in the range 2 — 3 GeV, which has the largest contribution to ^ ( h a d j l ) , BESII data have much smaller errors and the central values are 15% lower compared to the old MARK I and Gamma2 data. High quality r decay data, which have
CESR, or theoretical input from perturbative QCD (pQCD) in the higher energy regions where it can be relied upon but where the data quality is poor. Eidelman and Jegerlehner 20 , EJ95, relied primarily on e+e~ data, using pQCD only at the highest energies (> 20 GeV). Brown and Worstell 12 , BW96, used essentially the same data set as EJ95, but took into account the
412
J. P. Miller
The Muon Anomaly: Experiment and Theory
correlation of errors among data points coming from the same experiment. BW96 and EJ95 values and errors are in excellent agreement, suggesting that the correlation issue was not so important. Adel and Yndurain 21 , AY95, used pQCD in regions where the e+e~ data were poor, and obtained a value somewhat higher than, but still in agreement with, EJ95 and BW96. The theoretical input enabled hem to quote a smaller error. In 1998, Adelman, Davier and Hocker, 11 ADH98, did a full re-evaluation using updated values of the data sets used by EJ95 and BW96, obtaining the same error and a central value which was lower but still consistent within errors. ADH98 then incorporated the high quality r data from ALEPH, producing a dramatic improvement in the error of the contribution below the T mass, and reducing the overall error in a^ by 40%; inclusion of T data increased the central value somewhat. Subsequently, Davier and Hocker 22 , DH98a, refined the e + e - and r evaluation of ADH98 by applying pQCD above 1.8 GeV in regions where the data were poor, resulting in another 20% decrease in the error and a decrease in the central value. Most of the changes can be attributed to the effect of using pQCD from 1.8 to 3 GeV, where the old data are rather poor, as seen in Fig. 8. The DH98a evaluation was done before the new BESII data (Fig. 8) were available in this energy range; their pQCD calculations are in very good agreement with the BESII data, and about 15% below the old data, pointing to the reliability of pQCD at these energies. In a subsequent work (DH98b) the same authors applied QCD sum rules at low energies, resulting in slight further reductions in error and central value. The value from DH98b was used in the Czarnecki and Marciano 3 theoretical compilation and also by E821 9 to compare with their new experimental number. Some preliminary and published theoretical evaluations of a M (had;l) have appeared since the E821 publication of Brown, et al. 9
413
Table 2. A list of a number of recent evaluations of a,n(had; 1). T h e last entry is the E821 experimental 'measurement' of a M (had;l), obtained by subtracting the QED, E W and higher order hadronic contributions from the experimental number. This is an updated version of a similar table found in Ref. 26 .
Ref.
a^,(had; 1) (xlO 1 1 )
Comment
EJ95 2 0
7024(153)
e+e-
BW96 12
7026(160)
e+e~
AY95 21
7113(103)
e+e",QCD
6950(150)
e+e-
ADH98 11
7011(94)
e+e~,T
22
6951(75)
e+e_,r pQCD
DH98b 4
6924(62)
e+e~,T pQCD, sum rules
N01 2 3
7021(76)
e+e~,T
TY01 2 4
6966(73)
e+e_,r space-like Fv
E01 2 5
6932(65)
e+e-
E821 9
7350(153)
Expt -[ QED +EW+(Had>l)]
ADH98
DH98a
11
J. P. Miller
The Muon Anomaly: Experiment and Theory
Narison 23 , N01, has used essentially, the same r and e+e~ data sets at low energies and for the resonances as ADH98, with QCD applied to the continuum at the higher energies (> 1.7 GeV), arriving at nearly the same value as ADH98, but with a slightly smaller error. Troconiz and Yndurain 24 , TY01, have applied the maximum available data (including some preliminary data) and theory, following the earlier approach of AY95, and also incorporated pion form factor data from pion scattering at low energies, to arrive at a value which compares closely to DH98b, with a slightly larger central value and error.
Fig. 10 (LOL), presents special problems because, unlike other hadronic terms, it cannot be estimated based on experimental data. The value used in Czarnecki and Marciano 3 is an average of the values and the errors of two separate determinations 28, 29 (which are in agreement within errors), a^(had;LOL)= —85(25) x 1 0 ~ n . Both calculations use models motivated by chiral perturbation theory to calculate the contributions at low energies. The largest contribution comes from the 7T° pole term, with lesser contributions from the r? and rf poles. Other contributions can be w 20 — 30% in size relative to the pole terms, but when added together they largely cancel. After the publication of the new E821 experimental result 9 and after this talk was given at LP01, Knecht and Nyffeler 30 calculated the pion pole contribution using Large-Arc and short-distance properties of QCD. They obtained virtually the same magnitude as in references 28 and 29 but with the opposite sign: a^had; LOL, 7r°) = +5.8(1.0) x 10" 1 0 . The signs of the 77 and 77' poles also change, however, Knecht and Nyffeler only estimated their magnitude using a VMD model; their total pseudo-scalar contribution is aM(had; LOL, PS) = +8.3(1.2) x 10~ 10 . The sign error has since been acknowledged by the authors of 28 and 29 (see references 3 1 and 3 2 ) . Changing the signs of the pole and axial vector terms, but keeping the other terms as calculated in 28 and 29 the same, increases the theoretical value of aM by 17.2 x 10~~10, very close to just reversing the sign on the entire LOL contribution. Knecht and Nyffeler plan in the future to calculate all of the other terms in the hadronic LOL contribution. Several other groups are also considering new ways to tackle this difficult calculation. Ultimately, the hadronic LOL term may prove to be the limiting factor in the theoretical error of a p .
Finally, Eidelman 13 , E01, has produced a preliminary number based entirely on e+e~ data (except for pQCD at the very highest energies), including the new preliminary results from Novosibirsk and BESII. He obtains a value and error which are almost the same as DH98b. What can we conclude from this series oia^had; 1) determinations? The first thing we note is that within their stated errors, all of the evaluations are in agreement. In particular, the most recent evaluations are in excellent agreement even with their smaller errors, although we note that the DH98b evaluation has the smallest value leading to the largest discrepancy between theory and experiment. Secondly, an analysis which incorporates all of the new r and e+e~ data would be helpful. Eidelman, Davier and Hocker are presently collaborating on such an evaluation. When the final analyses, which include the latest excellent data are completed, we can expect a more reliable value for a^{had; 1) with a smaller error. The higher order hadronic contributions, a^(had;> 1), can be separated into two parts. One part, involving higher order diagrams such as those in Fig. 9, have a relatively small contribution to aM and the error is negligible:27 a^{had;> 1) = -101(6) x lCT 11 . The other part, involving the hadronic light-on-light diagram in
The theoretical error on a^(had;l) has gone down dramatically over the years, and with new, more accurate calculations of the
414
J. P. Miller
The Muon Anomaly: Experiment and Theory
-101 (6) X i d " Figure 9. Some diagrams hadronic contributions.
of
the
higher-order
(Fig 4). If supersymmetry is to explain all of Aa™ew: then its contribution is large: almost two times bigger than the electroweak contribution. Of course, the presently observed AoM may also be due to a statistical variation in the experimental number or to errors in the experiment or theory. When the additional data are analyzed, and with the continued extensive studies of the theory, one can anticipate that major progress will be made in understanding any non-zero AaM.
3
Figure 10. Diagram for the hadronic light-on-light contribution.
light-by-light contribution, the error and reliability of a1^ should continue to improve significantly. New physics will be reflected by a nonzero value of Aa";ew = afj°p — a*!1. Some fj,
\L
IX
examples are muon substructure, anomalous gauge couplings, leptoquarks, or supersymmetry. In a minimal supersymmetric model with degenerate sparticle masses (see Fig. 11), the contribution to Ao™£"' would be substantial in the case where tan/3 is large: AafSY « 140 x 1 0 - 1 1 ( l ° < f ^ ) 2 t a n / 3 , where rh is the sparticle mass. For 4 < tan/3 < 40, rh « 150 - 500 GeV. Note that the supersymmetric diagrams (Fig 11) are analogous to the electroweak diagrams
Figure 11. Lowest order diagrams for SUSY contributions.
415
Experiment
The ongoing muon (g-2) experiment at Brookhaven National Laboratory, E821, had its beginnings in the early 1980's. Its original goal was to measure aM to 0.35 ppm, or about 20 times better than the CERN 33 experiment. It received Laboratory approval in 1987, and major construction on the storage ring magnet began in the early 1990's. The first data were taken in 1997, with one major run in each year 1998-2001. The experimental technique follows the general one used in the CERN experiment with a number of important improvements and innovations. The Brookhaven AGS delivers up to 7 x 10 12 protons per bunch, with energies of 24 GeV, onto a water-cooled, rotating nickel target. There are 6-12 bunches per « 2.5 second AGS cycle, each about 50 ns wide FWHM and spaced 33 milliseconds apart. Secondary pions emitted from the target with momenta of 3.1 GeV/c are sent down a 72 m straight section of alternating magnetic quadrupoles, where highly polarized muons from forward pion decays are collected. The beam is then momentum-selected for either pions or the slightly lower-momentum muons, and then is injected through a field-free inflector36 region into a circular storage ring possessing a very homogeneous magnetic field. For the case of pion injection, with the pion momentum slightly higher than that of the central storage ring momentum, a small
The Muon Anomaly: Experiment and Theory
J. P. Miller fraction (RJ 25 ppm) of the muons from pion decays will have the correct momenta and directions to be stored. The efficiency of this process is low, and the very high intensity of pions and secondary particles associated with pion interactions with surrounding materials creates severe background problems for the detector system near the injection time ('flash'). An essential improvement over prior experiments was the incorporation of direct muon injection. With muon injection, the number of stored muons is increased by a factor of 10, while the 'flash' is reduced by a factor of 50 because most of the higher-momentum pions are blocked by beam-line collimators. In the homogeneous B field of the storage ring, charged particles follow a circular path (slightly modified by the electric quadrupole field) which would cause them to strike the inflector after one revolution. Muon injection therefore requires an in-aperture magnetic pulse at 90° around the ring from the injection point (provided by the pulsed 'kicker') in order to center the muon orbits in the storage region. With either muon or pion injection, positive (negative) muons are stored in the ring with spins initially polarized anti-parallel (parallel) to their momenta. In a magnetic field (no E-field) the spins precess relative to the muon momenta according to (2a = us — ujc = — a M ^ | . Here toa and coc are the angular frequencies of spin rotation and momentum rotation (or cyclotron angular frequency), respectively. Note that all of the muons precess at the same rate in a given field, regardless of their momenta. Two quantities must be measured with precision to determine aM: (ja and B, each being time averaged over the ensemble of muons. Actually, instead of measuring B, we determine the frequency of precession of the free proton, Up, in the same average magnetic field as the muons via NMR measurements. The anomaly is given by Eq. 2,
where R = $ ^ . A = ^ = 3.183 345 39(10) is the ratio of the muon and proton magnetic moments determined from other experiments 34,5 . The analyses of < toa > and < Up > were independent, and furthermore concealed offsets were maintained in each value so that no one could calculate aM prior to the completion of the analyses. The average trajectories of muons in the storage ring can only be known moderately well. Therefore the B field needs to be as uniform as possible to minimize the dependence of a given muon's precession rate on its exact trajectory in the storage ring. This prohibits the use of a gradient magnetic field to store (focus) the beam. E821 follows the CERN approach of using a quadrupole electrostatic field to provide the focusing. In the presence of the electric field, the precession is described by Eq. 3,
wa = - — [ a „ B - (a„ - -=—-)/? x E] (3) mc 7^ — 1 The "magic" 7 « 29.3, or pM w 3.094 GeV/c, is chosen so that a^ — TTZT[ ~ 0, minimizing the effect of E on Qa. Because not all stored muons have the exact magic momentum, a small correction ( « 0.6 ppm) must be applied to the final value for wa. Equation 3 is strictly valid only for muon motion perpendicular to B; the up-down motion associated with vertical betatron oscillations leads to another small correction of « 0.2 ppm, the so-called "pitch correction". The practical limit to the strength of a ferric field with the required homogeneity is w 1.5T; E821 chose B = 1.45T, leading to a ring radius of 7.112m. The storage ring aperture radius is 4.5 cm, giving a « ±0.6% ( « ±0.4%) base-to-base range in stored momenta for pion (muon) injection. The cyclotron period is r c = 4- = — « 149.2ns, the
416
J. P. Miller
The Muon Anomaly: Experiment and Theory
1,025 miHione*(E- 2 GeV. 1999 data)
Figure 12. Spectrum of number of positrons versus time, from the 1999 d a t a sample. There are a total of 1 billion e+ above 2 GeV.
precession period is Ta « 4.365 /is and the dilated muon lifetime is r = 7-ro ~ 64.38 jus. Decays are typically measured for at least ten muon lifetimes, or about 4000 cyclotron and 150 precession periods. A log plot of the 1999 data set, folded into 100/xs periods, is shown in Fig. 12. The error on aM from the combined / i + and /i~ data sets from the CERN 6 g-2 experiment is 7 ppm with a 1.5 ppm systematic error. By comparison, E821 must keep the systematic errors in B(wp) and ua to less than a few tenths of a ppm to approach its experimental precision goal of 0.35 ppm. 3.1
The Magnet and the of u>p
Determination
The storage ring, 35 Fig. 13, is a continuous C-magnet open to the inside. A cross-section view, Fig. 14, shows its essential features. It contains more than 600 tons of magnet steel. Three superconducting coils, which provide exceptional B-field stability with time, are used to power the magnet. The entire magnet is wrapped in thermal insulation to reduce
417
Figure 13. Overhead schematic view of the storage ring magnet. T h e detectors are distributed in uniform intervals around the inside of the ring. The beam is brought in through a hole in the back of the magnet yoke, indicated by the solid line at 10 o'clock.
gap changes due to temperature change. The storage region of 4.5 cm radius is defined by a series of circular collimators inside an evacuated chamber. The pole gap is 18 cm high and 53 cm wide. Many shimming options were incorporated in order to achieve the desired field uniformity. The very high-quality steel of the pole tips is decoupled from the lower quality steel and the imperfections (including holes for cryogenic leads, etc.) of the yoke, by« means of an air gap. Iron wedges in the gap can be moved radially to locally adjust the dipole field. The thickness and position of iron pole bumps can be adjusted to minimize quadrupole and sextupole fields. Thin sheets of iron were affixed to the pole tips to improve local uniformity. Current-carrying wires attached to circuit boards and mounted on the pole faces, with one set forming a closed loop covering 12° in azimuth and another set going entirely around the ring, provide a final fine-tune of the dipole field. A continuous monitor of the B-field was
J. P. Miller
The Muon Anomaly: Experiment and Theory
ilipn!-. •MiiirciiiMi m i l
Jl'ili' 1W-|'P
l».|l-ii...
DlllCI
Ariiiy »l NMK pilllil'S I111IVI-S
Miron^li lii'.-im luhi' nil r.ililr <MI
g 2 Magnet in Cross Suction
Figure 14. Cross section view of t h e storage ring magnet. The means of shimming t h e magnet and measuring its field are indicated.
Magnetic Field Uniformity (Azimuthal Average, 1 ppm c o n t o u r s , e x c e p t '97)
/^<
-JS~~\ 11 J j
'
\
•
•
•
'
'
/
(
/
/
///l
w
V^'^
/
/"'
provided by 360 NMR probes placed in fixed positions around the ring, above and below the storage region. A subset of probes, those most highly correlated to the average B-field, provided feedback to the magnet power supply to compensate for the slight field drifting which are mainly the result of ambient temperature changes. Two separate off-line analyses of the B-field used somewhat different combinations of probes to determine the average field as a function of time, with comparable results. The B-field in the storage region was mapped in 1 cm intervals every three to four days with 17 NMR probes mounted transversely on a movable cable-driven trolley. This was accomplished inside the vacuum, with essentially no geometrical changes to the magnet or vacuum chamber configuration. The NMR probes on the trolley were calibrated against a standard spherical water NMR probe, which was normalized to the precession frequency of a free proton. The fixed probes, in the off-line analysis, tracked the trolley probes to better than 0.15 ppm over time. The steady improvement in the B-field provided by shimming is illustrated in Fig. 15. The marked improvement from 1999 to 2000 is attributable to the replacement of the inflector, which had a damaged superconducting fringe-field shield.
xlcm
1 9 9 7 run
-S^^\
1 9 9 8 run
- // \
--—"""^---''^i
' f—"^i 1 9 9 9 run
2 0 0 0 run
Figure 15. B-field contours across the storage region in E821, averaged in azimuth, for succeeding run cycles. The 1997 contours are 2 ppm, the rest 1 ppm.
418
The distribution of muons inside the storage region was determined from an analysis of the debunching of the beam as a function of time. At the time of injection, muons are localized in the ring with a full width at half maximum of about 120 degrees. As a result, the time spectrum from a given detector at early times will contain oscillations with a period equal to the cyclotron period (the so-called "fast rotation" structure). Muons with high momenta have smaller cyclotron frequencies than those with low momenta, causing the bunches to spread out around the ring and the amplitude of the oscillations to diminish with time ("debunching" lifetime w 20/xs). The analysis of the debunching ver-
J. P. Miller
The Muon Anomaly: Experiment and Theory
Table 3. Systematic errors in wp for the 1999 data set.
Source of errors
Error (ppm)
Inflector fringe field
0.20
Fixed probe calibration
0.20
Fixed probe interpolation
0.15
Trolley Bo measurements
0.10
H distribution
0.12
Absolute calibration
0.05
Others 1
0.15
Total Syst error on UJP
0.4
' higher multipoles, trolley temperature stability, kicker eddy currents.
3.2
Determination ofu>a
The decay positrons from //+ —> e+uePfl have energies in the range 0 —• 3.1 GeV. In the muon rest frame, the higher energy positrons are preferentially emitted parallel to sM+. When the muon spin is parallel to the muon momentum, there will be more high energy muons in the lab frame than when the directions are anti-parallel. The number of positrons in the lab frame above a given energy threshold Et versus time therefore oscillates at the precession frequency according to Eq. 4,
N{t) = N0e-i
x
(l + Acos(w a i + (/>a),
sus time gave the radius of curvature distribution of the muons, which in combination with simulations of the betatron motion of the muons produces the radial and vertical distribution of muons in the storage aperture. The distributions thus deduced were folded geometrically with the map of NMR frequencies to obtain < UJP >. Corroborating information on the horizontal and vertical distributions of muons, as well as information on the betatron motion, at early times, was provided by scintillating fiber hodoscopes which could be inserted into the storage region. The hodoscopes were sufficiently thin that useful beam profile data could be taken for many tens of microseconds before the beam was degraded. The final value for the average field is ^ ^ = 61 791 256 ± 25 Hz (0.4 ppm). The sources of systematic errors are given in Table 3. The improvements in the 2000 data set are the installation of a new inflector with far less fringe field, greatly reducing the first item in the Table, and better trolley calibrations.
419
(4)
where each of N0 and A depend strongly on Et, while <pa depends slightly on Et. The positrons, generally having lower momenta than the muons, are swept by the B field to the inside of the storage ring, where they are intercepted by 24 scintillating fiber/lead electromagnetic calorimeters 37 uniformly spaced around the ring. The typical energy resolution of the calorimeters is i? = E(GeV) • Since a low energy positron arrives at the detectors more quickly after muon decay than a high energy positron (the average distance traveled is less), the actual measured times at the detectors relative to the muon decay time will depend slightly on energy, therefore
J. P. Miller
The Muon Anomaly: Experiment and Theory
tomultiplier tubes, whose sum is sent to a waveform digitizer (WFD) 3 8 which samples the photomultiplier pulse height every 2.5 ns. Both the time of arrival and the energy of the positron are determined from the WFD information. For 1999, the analysis occurred in two steps. First, in the production step, WFD data were converted to positron energies and times. There were two separate productions of the data. They each developed independent algorithms to handle the WFD data, which eventually evolved to become similar. The second step involved performing \ 2 minimization fits to the data in Fig. 12 to obtain u>a. The parent distribution in Eq. 4 provided a good x2 fit to the 1998 data sample, with five variable parameters: No, r, A, uia and 4>a. It did not however provide a good fit to the 1999 data, which has 15 times more positrons. It was necessary to account for small but noticeable effects from pulses overlapping in time at high rates (pile-up), betatron motion of the stored muons, and muon losses.
At times close to injection, the bunching of the muons leads, in addition to the previously mentioned oscillations, to an enhancement of the pileup. The oscillations are eliminated from the time spectra by using a bin width = TC, adding to each arrival time a time uniformly randomized over ±-if, and summing all detectors around the ring. The pile-up enhancement is accounted for by the last term in Eq. 6; the constants are held fixed to values determined in separate pile-up studies.
There were four independent analyses of the positron time spectra for the 1999 data set, two for each production. They used different methods to handle these additional effects. The time spectrum, with these additional effects, can be described by a 14 parameter function, Eq. 5 (not all parameters are necessarily variable in a fit):
f(t) = {N0e~i [1 + A cos toat + cf>a} + p{t)} x b(t) x l(t)
(5)
The pile-up term, p(t), with parameters np, Ap, Acfip, is given by Eq. 6
p(t) =
N0e~2i
The amplitude np of pile-up was generally less than 1% even at the earliest decay times, with an asymmetry Ap small compared to the (g-2) asymmetry, A. The artificial pileup spectrum gave the expected | lifetime, and when subtracted from the main spectrum, did a very good job of eliminating events above the maximum electron energy of 3.1 GeV, which apart from energy resolution effects could only be due to pileup. The properties of the artificial pile-up spectrum matched very well with the results of multiparameter pile-up fitting. The coherent betatron oscillation (CBO) term, b(t), with parameters As, UJB, 4>B, and TB, is described by Eq. 7, b(t) = l + ABCos{ujBt + 4>B)-e~(^)'2
x (np + Ap cos (ujat + a + A0 P )) x (l + a p e " ^ ( ^ ) 2 )
Two of the analyses constructed a simulated pileup time spectrum by combining single positron pulses, from data, into pileup pulses. The pileup spectrum was then subtracted from the primary spectrum, thus eliminating the p(t) term from their fits. A third analysis varied np and Ap, with Aa and a.
(6) 420
(7)
The need for b(t) is due to the effects of betatron motion of the muons combined
The Muon Anomaly: Experiment and Theory
J. P. Miller with the restricted aperture of the inflector. Muons are injected into the ring through the inflector, whose aperture was considerably smaller than the storage ring aperture because of mechanical and geometrical limitations. This creates a muon beam with narrow horizontal (1.8 cm) and vertical (5.6 cm) waists at the inflector exit at injection time. The more important horizontal waist case will be discussed here. In a perfectly uniform B field with no electric field, the muon trajectories, projected into the plane of the magnet, are circles. After the muons are kicked, the position of the horizontal waist would ideally be in the center of the storage region. The kick, however, was generally less than 100% of its optimum value. Therefore the average radius of muons at the narrow waist, at injection time, was larger than the central storage region radius. At 180 degrees around the ring from the narrow waist, the muons are spread out to fill the ring aperture, and have an average radius more nearly equal to the central ring radius. The acceptance of the electron calorimeters depends to a slight extent on the horizontal width and especially on the average radial position of the muon beam. Thus the detector positron acceptance will be different at the narrow waist compared to the opposite side of the ring. When we add electrostatic focusing, the position of the narrow waist (or focus) will move around the ring at the so-called CBO frequency, which is the cyclotron frequency minus the horizontal betatron frequency, JCBO = /c(l - v 7 ! - n). For the field index n = -jSr^rh = 0.137, pBo or
'
ICBO ~ 475 kHz. As a result, we get a small oscillation at « 475 kHz superimposed on the time spectrum. The amplitude of the CBO is typically a few tenths of a percent of the total number of counts, and T\, is long, about 100 fj,s.
One analysis allowed all four CBO parameters to vary. Two kept UJB fixed to the value obtained from a Fourier transform of the residuals from a five-parameter fit. The
421
frequencies from the Fourier transform and the fit were in good agreement. The muon loss term, Eq. 8, has two parameters, a^L and T^L,
l{t) = 1 + a^L • e'(^]
(8)
Muon losses are thought to be caused by the slight drift of the orbits of muons whose trajectories bring them close to collimators. The drift could be caused by the small non-uniformities in the E and B- fields, although the exact mechanism is not known. Indeed, when the beam is 'scraped' for about 15 /is right after injection, by temporarily displacing the stored muon beam several millimeters vertically and horizontally in order to force the loss of muons with trajectories close to the collimators. The rate of muon loss is markedly reduced after the scraping is turned off, compared to the no scraping case. After scraping, losses are generally less than i=sl% at early decay times, with the rate of losses decreasing with a short lifetime of T nL ~ 20/j.s. In addition there was a roughly constant loss rate of « 0.1% per muon lifetime at all decay times, as determined by comparing the measured decay rate with that expected from special relativity. The two analyses which allowed both parameters to vary obtained the same loss lifetime as was observed in a third analysis which held T^L fixed to the value determined using separate muon loss detectors. It is important to realize that only two of the 14 parameters in Eq. 5 have a large correlation to uja: (j)a and A
The Muon Anomaly: Experiment and Theory
J. P. Miller
=
N++Nz-N°-N2 N+ + JV2- + ^
Table 4. Results of the four analyses for ioa. R is defined by ioa = 27r/o(l — Rx 1 0 - 6 ) . /o is the nominal precession frequency.
#Par.
2
/DOF
R(ppm)
13
1.012 ±0.023
143.24 ±1.24
10
1.005 ±0.023
143.08 ±1.24
9
1.016 ±0.015
143.30 ±1.23
3
0.986 ±0.025
143.37 ±1.28
X
Avg.
143.17 ±1.24
Table 5. Systematic errors in uia for the 1999 d a t a set.
Source of errors
Error (ppm)
Pile-up
0.13
AGS background
0.10
Lost muons
0.10
Timing shifts
0.10
E field, pitch
0.08
Binning, fit procedure
0.07
Debunching
0.04
Gain changes
0.02
Total Syst error on ua
0.3
+ N%
= Acos(cjat + <pa) + (-^)2
(9)
where N+ = N^t + ?f), JV2" = N2(t - *f), N$ = N3(t), and N% = N4(t). The muon lifetime cancels, and r(t) is sufficiently insensitive to the CBO and the muon losses that these effects can be neglected in the fit. The insensitivity to the CBO is a consequence of — being not so different from ^f. We arrive LOB
I
at a three parameter fit which has different responses to systematic errors compared to the conventional multi-parameter fits. The results of the four analyses are given in Table 4. All of the results are well within the bounds expected for correlated data sets. The final value for toa is the average of these results, f£ = 229072.8 ± 0.3 Hz(1.3 ppm), after a correction of ±0.81 ± 0.08ppm for the effects of the electric field in Eq. 3 and for vertical betatron oscillations ("pitch" correction). 8 Note that the data are not sensitive to the sign of uja; this however is welldetermined from many other experimental measurements, and is implicit in the value of A which we use to extract the value of aM. The sources of systematic error in u>a are given in Table 5. The AGS background is the result of unwanted particle injection into the ring after the initial injection. The improvements for the 2000 data run were the addition of a sweeper magnet in the beam-line to eliminate errors due to the AGS background, and an increase in the number of lost muon detectors in order to reduce the muon loss error. 4
Conclusions and Outlook
The muon anomalous magnetic moment can be both measured and calculated (within the Standard Model) to a high precision, and given its high sensitivity to new physics, its measurement affords an exceptional oppor-
422
J. P. Miller
The Muon Anomaly: Experiment and Theory
tunity to probe for new physics beyond the Standard Model. The new world average value of a+ shows a 2.2 ±1.4 ppm difference from the Czarnecki and Marciano 3 theory compilation, after the sign of the LOL pole term is corrected. Many theoretical ideas have been put forward to explain any difference, including supersymmetry, leptoquarks, muon substructure, etc. It could of course also be explained by a statistical fluctuation, an error in the experiment, or an error in the Standard Model calculation. All aspects of the theoretical calculation of aM are being heavily scrutinized. New high quality e+e" data from VEPP-2M and Beijing as well as r decay data from LEP, are being analyzed now, and should have an impact on aft in the next few months. Longer term, Novosibirsk and Beijing have upgrade plans, and DA$NE (and perhaps the B-factories and Cornell) have plans to use the radiative return process to measure e+e~ cross sections. Further calculations of the lightby-light term are being considered by several groups. One can reasonably expect a continued steady improvement in the error and reliability of aft. In E821, analysis is under way on the ss 4 billion positrons (/x+) from the 2000 run (about four times larger than the 1999 data set) and on the « 3 billion electrons (/i _ ) from the 2001 run. Systematic errors are expected to be reduced for both data sets. Once these data are analyzed, it should be possible to make a more definitive statement concerning whether the measured anomaly agrees with theory. Comparison of the [i+ and fi~ anomalies is a test of both the systematic errors in E821 and also CPT imparlance. With another data run, E821 expects to achieve 0.3 ppm statistical error on ae*p and an estimated 0.3 ppm systematic error, not far from the original goal of 0.35 ppm overall error. It is interesting to note that any new physics affecting aM may also lead to a non-
423
zero permanent electric dipole moment for the muon, through its CP violating part. Assuming that the CP violating phase for new physics 4>cp ~ 1, then dimensional arguments, along with the observed value for Aa M , give 39 dM « 10~ 22 . Even if aM experiment and theory were to agree, the muon EDM is interesting in its own right: it is the only currently accessible EDM from a second generation particle. Comparing with the electron, the current limit on the electron dipole moment is « 4 x 10"~27e — cm. If the EDM scales by the first power of the mass, then a 10~ 24 e — cm muon measurement is competitive with that of the electron. There are speculations, however, that the electron EDM could be small due to an accidental cancellation which may not apply to the muon. Or, if the scaling is with the square of the mass or higher, the muon then becomes more sensitive than the electron to new physics. 40 There are a number of models which predict dp in the range 10~22e~cm to 10 _ 2 4 e —cm.39 The presence of an EDM adds the term
~&*S
+
®
(10)
to Eq. 3, where the EDM is given by d^ = i ( ^ n c ) - ^ n t n e (§"2) experiment, the effect of the dominant (3 x B term is to tip the precession vector radially by an angle 3 = t a n - 1 TP-. This causes an increase in the precession frequency to w « uia \fl + rf. It also causes an oscillation about zero of the average vertical component of the positron momenta, which can be observed as an oscillation, with frequency u>, in the average vertical position of positrons on the face of the calorimeters. In the unlikely event that all of AaM can be attributed to a muon EDM, then d,, = (2.3 ± 0.7) x 10" 1 9 . The CERN (g2) 6 experiment has set the best limit on the muon EDM so far, d^ < 1 x 10 _ 1 8 e - cm, deduced from limits on the vertical oscillations. While this value for d^ is larger than any the-
The Muon Anomaly: Experiment and Theory
J. P. Miller ory predicts, it is nevertheless not ruled out by CERN limit. E821 expects to reduce the limit by about a factor of five from an improved measurement of the vertical oscillations. A dedicated experiment to measure the muon EDM to the 10 _ 2 4 e — cm level is currently being developed at Brookhaven National Laboratory. 41 It would use a new technique where a muon momentum and an applied electric field would be selected so that the second term cancels the first term in Eq. 3. One is only left with the motion in a vertical plane described by Eq. 10. In this technique, there is a very large enhancement of the EDM signal relative to the "noise" over the technique used in the g-2 experiments. It is planned to mount this experiment over the next few years.
5
Grosse-Perdekamp 11 , A. Grossmann 6 , M.F. Hare 1 , D.W. Hertzog 7 , V.W. Hughes 11 , M. Iwasaki 10 , K. Jungmann 6 , D. Kawall 11 , M. Kawamura 10 , B.I. Khazin 3 , J. Kindem 9 , F. Krienen 1 , I. Kronkvist 9 , R. Larsen 2 , Y.Y. Lee 2 , I. Logashenko 1 , R. McNabb 9 , W. Meng 2 , J. Mi 2 , J.P. Miller1, W.M. Morse 2 , D. Nikas 2 , C. Onderwater 7 , Y. Orlov 4 , C.S. Ozben 2 , J. Paley 1 , C. Polly 7 , J. Pretz 1 1 , R. Prigl 2 , G.zu Putlitz 6 , S.I. Redin 11 , O. Rind 1 , B.L. Roberts 1 , N.M. Ryskulov 3 , S. Sedykh 7 , Y.K. Semertzidis 2 , 3 Yu.M. Shatunov , E. Sichtermann 11 , E. Solodov3, M. Sossong7, A. Steinmetz 11 , L.R. Sulak 1 , C. Timmermans 9 , A. Trofimov1, D. Urner 7 , P. von Walter 6 , D. Warburton 2 D. Winn 5 , A. Yamamoto 8 , D. Zimmerman 9 1
Acknowledgments
The author would like to thank R. Carey, D. Hertzog, K. Jungmann and Y. Semertzidis for helpful comments on this manuscript. The author's work is supported by the U.S. National Science Foundation. Opinions regarding the theoretical content of a^ are solely those of the author and do not necessarily reflect the opinions of the E821 experiment. E821 is supported by the U.S. Department of Energy, the U.S. National Science Foundation, the German Bundesminister fiir Bildung und Forschung, the Russian Ministry of Science, and the U.S.-Japan Agreement in High Energy Physics. References 1. H.N. Brown 2 , G. Bunce 2 , R.M. Carey 1 , P. Cushman 9 , G.T. Danby 2 , P.T. Debevec7, M. Deile 11 , H. Deng 11 , W. Deninger7, S.K. Dhawan 11 , V.P. Druzhinin 3 , L. Duong 9 , E. Efstathiadis 1 , F.J.M. Farley 11 , G.V. Fedotovich3, S. Giron 9 , F. Gray 7 , D. Grigoriev3, M.
Boston University, Boston, 2 Massachusetts 02215, USA Brookhaven National Laboratory, Physics Dept., Upton, NY 11973, USA 3Budker Institute of Nuclear Physics, Novosibirsk, Russia 4 Newman Laboratory, Cornell University, Ithaca, NY 14853, USA 5 Fairfield University, Fairfield, Connecticut 06430, USA 6 University of Heidelberg, Heidelberg 69120, Germany 7 University of Illinois, Physics Dept., Urbana-Champaign, IL 61801, USA 8 KEK, High Energy Accelerator Research Organization, Tsukuba, Ibaraki 305-0801, Japan 9 University of Minnesota, Physics Dept., Minneapolis, MN 55455, USA w Tokyo Institute of Technology, Tokyo, Japan n Yale University, Physics Dept., New Haven, CT 06511,US A 2. Van Dyck, Schwinberg, Dehmelt, Phys. Rev. D34, 722(1986). 3. A. Czarnecki and W.J. Marciano, Nucl. Phys. (Proc. Suppl.J B76, 245 (1999). 4. M. Davier and A. Hocker, Phys. Lett. B435, 427 (1998).
424
The Muon Anomaly: Experiment and Theory
J. P. Miller 5. D.E. Groom, et al, the Particle Data Group, Eur. Phys. J C 1 5 , 1 (2000). 6. J. Bailey, et al, Nucl. Phys. B150, 1 (1979). 7. R.M. Carey, et a l , Phys. Rev. Lett. 82. 1632-1635(1999). 8. H.N. Brown, et al., Phys. Rev. D62, 091101 (2000). 9. H.N. Brown, et a l , Phys. Rev. Lett. 86, 2227 (2001). 10. T. Kinoshita, private communication. 11. R. Alemany, M. Davier, A. Hocker, Eur. Phys. J. C2, 123 (1998). 12. D.H. Brown and W. Worstell, Phys. Rev. D54, 3237 (1996). 13. S. Eidelman, I. Logashenko, CMD2 collaboration, private communication. 14. Data and plot from A. Denig, private communication, also see the contributed paper in this conference by the KLOE collaboration, hep-ex/0107023. 15. BESII collaboration, Phys. Rev. Lett. 84, 594 (2000); Preliminary data and plot from Z. Zhao, private communication. 16. M. Davier, private communication. 17. S.I. Eidelman and V.N. Ivanchenko, Nucl. Phys. Proc. Suppl. 40, 131 (1995). 18. K. Melnikov, SLAC-PUB-8844, hepph/0105267, May 2001. 19. Data from Cornell, Jon Urheim, private communication. 20. S. Eidelman and F. Jegerlehner, Z. Phys. C67, 585( 1995). 21. K. Adel and F.J. Yndurain, hepph/9509378,1995 and Rev. Acad. Ciencas (Esp.), 92, 736 (1985). 22. M. Davier and A. Hocker, Phys. Lett. B419, 419 (1998). 23. S. Narison, hep-ph/0103199, March 2001; Phys. Lett. B513, 53( 2001). 24. J.F. Troconiz and F.J. Yndurain, hepph/0106025, 2001. 25. S. Eidelman, preliminary result from Novosibirsk, private communication.
425
26. W. Marciano and B. Roberts, hepph/0105056, May 2001. 27. B. Krause, Phys. Lett. B390, 392 (1997). 28. M. Hayakawa and T. Kinoshita, Phys. Rev. D57, 465 (1998). 29. J. Bijnens, E. Palanta, J. Prades, Nucl. Phys. B474, 379 (1996). 30. M. Knecht and A. Nyffeler, hepph/0111058, 6 Nov 2001. Also see M. Knecht, A. Nyffeler, M. Perrottet, and E. de Rafael, hep-ph/0111059, 6 Nov 2001. 31. M. Hayakawa and T. Kinoshita, hepph/0112102, 16 Dec 2001. 32. J. Bijnens, E. Palanta, J. Prades, hepph/0112255, 19 Dec 2001. 33. J. Bailey et al., Nucl. Phys. B150, 1 (1979). 34. W. Liu, et al., Phys. Rev. Lett. 82, 711 (1999). 35. G.T. Danby, et al., Nucl. Instrum. Methods A457, 51 (2001). 36. F. Krienen, D. Loomba and W. Meng, Nucl. Instrum. Methods A283, 5 (1989). A. Yamamoto, et al., Proc. of 15th Int. Con}, on Magnetic Technology, Science Press Beijing, 246 (1998). 37. S. Sedykh, et al., Nucl. Instrum. Methods A455, 346 (2000). 38. R. Carey, et al., to be published, Nucl. Instrum. and Methods. 39. J.L.Feng, K.T. Matchev and Y. Shadmi, "Theoretical expectations for the muon's electric dipole moment," hep-ph/0107182. 40. K.S. Babu, S.M. Barr and I. Dorsner, "The scaling of lepton dipole moments with lepton mass," hep-ph/0012303. 41. R.Carey, et al., "Request for R&D Funds Toward a Proposal for the Muon Electric Dipole Moment Experiment", submitted to the Brookhaven National Laboratory Program Advisory Committee, Fall, 2001, http://www.bnl.gov/edm/.
S E A R C H E S FOR N E W PARTICLES GAIL G. HANSON Indiana University, Department of Physics, Bloomington, Indiana 4^405, USA E-mail: [email protected], [email protected] The status of searches for new particles and new physics during the past year at the Fermilab Tevatron, at HERA and at L E P is summarized. A discussion of the hints for the Standard Model Higgs boson from LEP2 data is presented. Searches for non-Standard Model Higgs bosons are also described. Many searches have been carried out for the particles predicted by supersymmetry theories, and a sampling of these is given. There have also been searches for flavor changing neutral currents in the interactions of the top quark. In addition, searches for excited leptons, leptoquarks and technicolor are summarized.
1
Introduction
much longer!
One of the most tantalizing physics topics of the past year has been the possible evidence for Standard Model Higgs boson production from the LEP experiments. Both the November 2000 combination, which led to a request for an extension of LEP running, and the combination just prepared for the summer conferences are presented here. Searches for Minimal Supersymmetric Standard Model (MSSM) Higgs bosons and searches in other extensions of the Standard Model have been performed and are summarized. A light Higgs boson with mass near 115 GeV could be the lightest SUSY Higgs h° with nearly Standard Model couplings. There have been extensive searches for the supersymmetric partners of the ordinary particles in pp, e+e~, and e^p collisions, as well as searches for excited fermions, leptoquarks, and technicolor. In the interests of fitting into the time allowed, I was able to cover only a few of the possible search topics. For example, there were many contributions in various supersymmetry scenarios which I would have liked to discuss in more detail, and I did not discuss the area of large extra dimensions. I have given references only to the experimental papers, and I refer the reader to them for the theoretical references - otherwise my list of references would have been
2
Searches for Higgs Bosons
2.1
Standard Model Higgs Search
*S
J
^
10
I'.'. 1 !'." ! "J.V.''!' 'J.L'.' ! ' " ! ' " ! ' " ! • " =
.,'- \ LJEP!
I
:
i
J
I
\
\
\
\
\
; •-^p^-p^rvSf | -j ;Expectedis+h; 4 JQ ~ _....:.j""..jExp4s.tfidJb
i
l/|
; -30-
;/ j j
;
[
j
,-! -
[ [
!—I—\—\—!—\—\—i\—!—i—!
[ _
4o
5
in ~ \. , , i , , , j , , , i , . , i , , . i , , , I , , i i , , 100 102 104 106 108 110 112 114 116 118 120 mf/GeV/c2) Figure 1. The background probability (1 — CLi) as a function of mjj f ° r the four LEP experiments combined for year 2000 data as shown at the November 3, 2000, L E P C meeting.
Preliminary results of searches for the Standard Model Higgs boson at LEP2 were presented 1 at the November 3, 2000, meeting of the LEP Experiments Committee (LEPC) using most of the data collected in 2000. The background probability as a function of the test mass mn for the combination of all four LEP experiments is shown in Fig. 1. For background only, (1 — CLb) will be 0.50 on the average. The combination of the four LEP experiments presented at the November 3 LEPC meeting showed an excess of 2.9a
426
Searches for New Particles
Gail G. Hanson significance, or (1 — CL\,) = 0.0042, at TBH ~ 115 GeV. The negative log-likelihood ratio —21nQ for the LEP combination is shown in Fig. 2. The value of m H = 115.0Jlg;| GeV is given by the point at which the observed —21nQ versus mn has its minimum value. The lower limit was raH > 113.5 GeV at 95% confidence level (C.L.) with a median limit of 115.3 GeV expected for background only.
—25
3f20 ' 75 10
5 0 -5
-10
— Observed \ -- Expected background — • Expected signal + background" — Test signal + background.
100 102 104 106 108 110 112 114 116 118 120
mJGeV/c ) --25 -' ' ' 1 ' " ! ' ' ' 1 ' ' ' 1 ' V I I O)
SSI
^20
T rp J
' 15 10
5
'-
0 : -5 -
\
^ ^ ^
Observed Expected background \ / / / " ^
~W100 102 104 106 108 110 112 114 116 118 120
m^GeV/c2) Figure 2. Negative log-likelihood ratio —21nQ as a function of mjj for the four L E P experiments combined for year 2000 d a t a as shown at the November 3, 2000, LEPC meeting.
H° Figure 3. Higgsstrahlung process for production of the Standard Model Higgs boson in e+e~ collisions at LEP.
At LEP the SM Higgs boson is expected to be produced mainly through the Higgsstrahlung process e + e~ —> HQZ°, shown in Fig. 3, with small additional contributions from i-channel W and Z boson fusion processes. Searches are performed in the channels HZ —• bbqq (four jet), HZ —> bbvv (missing energy), HZ —> bbr+T~ or T+T~qq (tau), and HZ —> bbe+e~ or bbjj,+ yT (leptonic).
427
Figure 4. Observed and expected behavior of t h e likelihood ratio - 2 1 n Q as a function of t h e test mass mH, obtained by combining t h e d a t a of all four LEP experiments. T h e solid line represents the observation; the dashed/dash-dotted lines show the median background/signal + background expectations. The dark/light shaded bands around t h e background expectation represent the ± l / ± 2 standard deviation spread of the background expectation obtained from a large number of background experiments. T h e dotted line is the result of a test where the signal from a 115 GeV Higgs boson has been added to the background and propagated through t h e likelihood ratio calculation.
Each event is assigned a probability s, of being a signal event and a probability bi of being a background event at a test Higgs mass ma. The event weight to* is given by Wi = (si + bi)/bi. The sample likelihood £ is the product of the weights. The logarithm is taken, and then the method is log-likelihood ratio:
L i k e l i h o o d r a t i o <5(TOH)
£(S + b) c(b) •
(1)
Two hypotheses are tested: background only, with compatibility measured by 1 — CLb, and signal plus background, with compatibility measured by CLs+b (CLS = CLs+b/CLb). The event weights in terms of s/b for mn = 115 GeV are given in Table 1 for the current combination 2 and for the November 3 LEPC combination 3 for the events with the ten largest weights in the current combination. Apart from the L3 missing energy event (and the OPAL event marked "*", which was
Searches for New Particles
Gail G. Hanson recorded after the deadline for the November 3 combination and was reprocessed with calibrations for that data set, giving it a higher weight), the event weights show only small changes. The L3 event weight changed because the event was unlikely either for signal or background, and higher statistics Monte Carlo simulations resulted in a lower weight. All four LEP experiments have published preliminary results for the 2000 data, 4 and L3 has published their final analysis. 5
d
10 10~ W~ 10~ 10~ io'
100 102 104 106 108 110 112 114 116 118 120 m^GeV/c2)
Figure 7. Confidence level CLS for the signal + background hypothesis. Solid line: observation; dashed line: median background expectation. T h e dark/light shaded bands around t h e median expected line correspond to the ± l / ± 2 standard deviation spreads from a large number of background experiments.
d 2a 3a Observed ; \ ,' iixjxvieti Ib£ signat+baciigrouiiii' IKxpecied foil background \>
4a -5
100 102 104 106 108 110 112 114 116 118 120 ntf/GeV/c2) Figure 5. The probability (1 — CLb) as a function of the test mass mn. Solid line: observation; dashed/dash-dotted lines: expected probability for the background/signal + background hypotheses.
£> SO. 14
tfi.12 Q
Observed Expected for background Fspecled for Msinai (iti.,-115.6 GeV
LEP
% 0.1 a 1| 0.08 Q,
0.06 0.04 0.02 0 -15
-10
10 15 -2 ln(Q)
Figure 6. Probability density functions corresponding to a test mass mjr = 115.6 GeV, for the background and signal + background hypotheses. The observed value of —2 In Q which corresponds to the data is indicated by the vertical line. T h e light shaded region is a measure of the compatibility with the background hypothesis, 1 — CLf, (3.4%), and the dark shaded region is a measure of compatibility with the signal + background hypothesis, CL3+\, (44%).
Figure 4 shows the negative loglikelihood ratio — 21nQ versus mn for the current LEP combination of the preliminary results of three LEP experiments and the final result of one experiment. The minimum is observed at WH = 115.6 GeV. The probability (1 — CLb) versusTOHis shown in Fig. 5. AtTOH= H5.6 GeV, 1 - CLb = 0.034, corresponding to a probability of background fluctuation of 2.1 standard deviations. The probability density functions corresponding to a test mass ma = 115.6 GeV, for the background and signal plus background hypotheses, are shown in Fig. 6. The area under the background curve below the observed value of — 2lnQ corresponds to 3.4% probability of compatibility with background, and the area under the signal curve above the observed value of — 21nQ corresponds to 44% probability of compatibility with signal plus background. Figure 7 shows the the distributions of CLS versusTOH,which give the lower limit of m H > 114.1 GeV at 95% C.L. with a median limit of 115.4 GeV expected for background only. Figure 8 shows the reconstructed Higgs mass distributions for special non-biasing selections with low (Fig. 8a), medium (Fig. 8b),
428
Gail G. Hanson
Searches for New Particles
Table 1. Comparison of event weights between November 3, 2000, L E P C and current combination.
mtfc (GeV)
Nov. 3 s/b
Current s/b
4-jet
114
4.7
4.7
ALEPH
4-jet
113
2.3
2.3
3
ALEPH
4-jet
110
0.9
0.9
4
L3
E-miss
115
2.1
0.7
5
OPAL*
4-jet
111
0.4
0.7
6
DELPHI
4-jet
114
0.5
0.6
7
ALEPH
Lept
118
0.6
0.6
8
ALEPH
Tau
115
0.5
0.5
9
ALEPH
4-jet
114
0.4
0.5
10
OPAL
4-jet
113
0.5
0.5
Exp.
Channel
1
ALEPH
2
-£ = 200-210 GeV
+
>
12
Vs = 200-210 GeV
+
LEP loose
H
h z
LEP light
]>&.;, j background
\^n\ background
B
Signal
h z
Signal
{nv=115 GeV)
. _jJ£~-#vttii' Reconstructed Mass mM [GeV/c ]
Reconstructed Mass mH [GeV/c2]
Reconstructed Mass m„ [GeV/c ]
ec
Figure 8. Distributions of the reconstructed Higgs mass, m ^ , from three special, non-biasing, selections with increasing purity of a signal from a 115 GeV Higgs boson.
and high (Fig. 8c) purity. 2.2
MSSM Higgs Search
In the Minimal Supersymmetric Standard Model (MSSM) there are two scalar field doublets resulting in five physical Higgs bosons: two neutral CP-even scalars, h° and H° (with m-ho < TOHO), one CP-odd scalar, A0, and two charged scalars, H^. At tree level, rriYfi < rr»z, m^o < TOH°, rnz < rn-^o, and m.}j± < mw- Loop corrections, predominantly from t and i, modify these mass relations, unfortunately for LEP2. However, in the MSSM, there must be a lowest mass neu-
429
tral Higgs boson h° with m^o < 135 GeV. At the current e+e~ center-of-mass energies accessible to LEP, the h° and H° bosons are expected to be produced predominantly via two processes: the Higgsstrahlung process e+e~ —» h°Z° (as for H$M) and the pair production process e+e~ —> h°A°. The cross sections for these two processes, <7hz and a^A, are related at tree level to the SM cross sections by the following relations:
e+e
h°Z° : a h z = sin2(/3 - Q) O%£ (2)
Searches for New Particles
Gail G. Hanson LEP8° ™n '--'" "--liminarv CQ.
c 10
0
20
40 60
80 100 120 140 mho (GeV/c2)
0
20
40
60
80
100 120 140
mho (GeV/c2)
Figure 9. The MSSM exclusion for the m h o - m a x benchmark scenario. T h e excluded (hatched) and theoretically disallowed (dark grey) regions are shown as functions of the MSSM parameters in two projections: (left) the (m h o, mAo) plane and (right) the (m h o, tan/3) plane. The dashed lines indicate the boundaries of the regions expected to be excluded at t h e 95% C.L. if only SM background processes are present.
constrained MSSM (CMSSM) the sfermion and gaugino masses are unified. In addie+e" -> h°A° : ahA = cos2(/3 - a) A a^ tion, in minimal supergravity-broken MSSM (3) (MSUGRA) the trilinear couplings are equal where a^ ls the Higgsstrahlung cross section (AQ), the scalar masses (including Higgs) are for the SM process e+e~~ —> H^MZ°, and A is unified, and the electroweak symmetry scale a factor accounting for the suppression of the determines /x. P-wave cross section near production threshThe results of the Standard Model Higgs old. searches are used for the e + e~ —> h°Z° chanThe angle /? is defined in terms of the nel, with the cross sections modified as in vacuum expectation values v\ and i>2 of the Eq. (2), and the decay branching ratios detwo Higgs field doublets: tan/3 = 1*2/^1 • The termined by the supersymmetry parameters. angle a is the mixing angle that relates the Dedicated analyses are done for the assophysical mass eigenstate h° with the field ciated production of a scalar h° and pseudoublets. doscalar (A0) Higgs. The search channels are In addition, the following parameters are hA -> bibb (Ah-4b) and hA -> T+T~bb or needed to specify the MSSM: MSUSY', /•*, the bbr+T- (Ah-tau). Higgs boson mass parameter; M\, M2, M3, The presence of an MSSM Higgs boson the gaugino masses at the electroweak scale signal is tested in a constrained MSSM in (gaugino unification gives a common gaug- which the parameter A is the common trilinino mass rrii/2 a t the GUT scale and Mi = ear Higgs-squark coupling parameter. Three (5/3) tan 2 0WM2); AT, Ab, At, the third fam- benchmark scenarios are considered: the "noily trilinear Higgs-sfermion coupling parame- mixing" scenario, in which there is no mixing ters; m,f, the scalar fermion masses (sfermion between the scalar partners of the left-handed mass unification gives a common sfermion and right-handed top quarks, MSUSY — mass mo at the GUT scale); and m^o, the 1 TeV, M 2 = 200 GeV, fi = -200 GeV, running mass of the CP-odd scalar A0. In
430
Gail G. Hanson Xt(= A - ncotp) = 0, 0.4 < tan/3 < 50, 4 GeV < m\o < 1 TeV, and the gluino mass nig = 800 GeV; the "m^o —max" scenario, which is designed to yield the maximal value of ni|,o in the model, corresponds to the most conservative range of excluded tan/3 values for fixed values of MSUSY and the top quark mass, and has the same values of the parameters as in the no-mixing scenario except for the stop mixing parameter Xt = 2MsusV, and the "large fx" scenario, which is designed to illustrate choices of MSSM parameters for which h° does not decay into pairs of b quarks and uses parameters MSUSY = 400 GeV, M2 = 400 GeV, fi = 1 TeV, m-& = 200 GeV, 4 < mAo < 400 GeV, and Xt = -300 GeV. Figure 9 shows the MSSM exclusion regions for the Who—max benchmark scenario for the combination of the preliminary results of the four LEP experiments. 6 In the m^o—max scenario, the limits obtained are mho > 91.0 GeV and mAo > 91.9 GeV at 95% C.L., and the range 0.5 < tan/? < 2.4 is excluded for a top quark mass less than or equal to 174.3 GeV.
2.3
Searches for New Particles
20
40
60
100
80
120
M„» (GeV)
Figure 10. Combined LEP experimental limits on Higgs bosons produced with Standard Model cross sections and decaying into di-photons. T h e 9 5 % C.L. upper limit on the di-photon branching fraction is shown as a function of the Higgs mass. Also shown (dotted line) is the branching fraction obtained for the benchmark fermiophobic model. The median expected limits and the ± 2 standard deviation confidence level region are denoted by the dashed curves.
LEP PRELIMINARY
Non-Standard Model Higgs Searches
Searches for Higgs bosons decaying into photons have been carried out by the four LEP collaborations, and the combination 7 of preliminary results is shown in Fig. 10. A lower bound of 108.2 GeV is set at 95% C.L. for Higgs bosons produced with the Standard Model cross section af^ and not decaying into fermion pairs. Searches for Higgs bosons produced with the Standard Model hZ cross section and decaying hadronically but not necessarily into b quarks have been combined for the four LEP experiments for the first time. 8 The combination of preliminary results in shown in Fig. 11. A lower limit of 112.9 GeV at 95% C.L was obtained for h decaying 100% hadronically. Two Higgs Doublet Models (2HDMs) are
- expected foi bjckgrouml
10
60
70
80
90
100
110
m„ (GeV)
Figure 11. Combined LEP flavor independent 95% C.L. upper limits on the production cross section as a function of Higgs mass, normalized to the expected Standard Model values and assuming BR(/i —• hadrons) = 1.0. The observed limit is shown as the solid curve and the expected median limit by the dashed curve. The bands correspond to 68.3% and 95% confidence intervals from the background-only experiments.
extensions of the Standard Model in which two scalar doublets and five physical Higgs
Searches for New Particles
Gail G. Hanson
OPAL
-Vs = 183-208 GeV
PRELIMINARY
ADLO Preliminary
^140 O 5*120 100 80
...
60 '•'•K'?,
40 >;,<,-
\' ;\ 20:^
i
°(I
2u m„ [GeV]
Figure 12. Excluded (mAo, mho) region independent of a, together with the expected exclusion limit. A particular (m A o, m h o ) point is excluded at 95% C.L. if it is excluded for 0.4 < tan/3 < 58.0 (darker grey region), 0.4 < tan/3 < 1.0 (lighter grey region) and 1.0 < tan/3 < 58.0 (hatched region) for —7r/2 < a < 7r/2. T h e cross-hatched region is excluded using constraints from Tz only. Expected exclusion limits are shown as a dashed line.
bosons occur but without the constraints of the parameters of supersymmetry. In Type II 2HDMs the first Higgs doublet couples only to down-type fermions and the second Higgs doublet couples only to up-type fermions. The Higgs sector in the MSSM is a Type II 2HDM. In Type II 2HDMs h° and A 0 are produced as in Eqs. (2) and (3), and the branching ratios are determined by a and /?, which are the only free parameters besides the Higgs masses. Limits have been obtained 9 on Type II 2HDMs with no CP violation in the Higgs sector and no additional particles besides the five Higgs bosons by interpreting the results of searches for h° and ^4°. An example is shown in Fig. 12 for which 1 < m^o < 58 GeV and 10 < mAo < 65 GeV are excluded at 95% C.L. for all values of a and tan/3 in the ranges scanned (—7r/2 < a < -n/2 and 0.4 < tan/3 < 58.0).
Figure 13. 95% C.L. lower limits for combined d a t a from the LEP experiments for the masses of righthanded scalar leptons versus the mass of the lightest neutralino. Observed limits: solid lines; expected limits: dotted lines.
3
Searches for Supersymmetry
In supersymmetric (SUSY) models each of the "normal" particles (leptons, quarks, and gauge bosons) has a supersymmetric partner (scalar leptons, scalar quarks, and gauginos) with spin differing by half a unit. Most of the searches for these supersymmetric particles are performed within the MSSM assuming .R-parity conservation. i?-parity is a multiplicative quantum number defined as Rp
= (_I)3B+L+2SJ
w h e r e
B )
Z )
a n d
g
a r e
the baryon number, lepton number, and spin of the particle, respectively. .R-parity discriminates between ordinary and supersymmetric particles: Rp = + 1 for the ordinary SM particles and —1 for their supersymmetric partners. If .R-parity is conserved, supersymmetric particles are always produced in pairs and always decay through cascade decays to ordinary particles and the lightest supersymmetric particle (LSP), which must be stable. In gravity mediated SUSY breaking, the LSP is the lightest neutralino x?> a n d the gravitino G is heavy. In gauge mediated
432
Searches for New Particles
Gail G. Hanson ALO Preliminary
ADLO Preliminary
ADLO Preliminary ,100-
"^ 100-
o
T->'bi v'
0=0°
b->bX?
6=0°
- i 9080 7060-
1
50
>
90 100 m w + „ Ms.o/GeV/c')
40J
40
9=56"/. mi l^i&T
fusion 50
60
70
80 90 1C M s,oP(GeV/c2)
mtmi
20 0
MUM
fe*^#M 50 60 70
8b 90 100 Msb«om(GeV/c2)
Figure 14. 95% C.L. lower limits for combined data from the LEP experiments for the masses of scalar top and scalar bottom quarks versus the mass of the supersymmetric decay product. Limits are shown for zero mixing angle and for the mixing angle at which the t\ (bi) decouples from the Z°.
SUSY breaking, the G is very light (LSP) and X? G7, for example. 3.1
Searches for Scalar Leptons
Scalar leptons (sleptons, £±) can be produced in pairs in e+e~~ collisions: e+e~ —> £+£~. They decay, for example, to the LSP x? and a lepton of the same flavor: £~ —> yQtr. The topology is acoplanar leptons £+£~, and observation depends on A M Mi- — M^a since the x? is undetectable. Preliminary searches for sleptons from the four LEP experiments have been combined, 10 and the excluded regions are shown in Fig. 13. The combined LEP 95% C.L. lower limits for AM > 10 GeV are: M gR > 99 GeV, M A R > 95 GeV, and M^ > 80 GeV. [Note: £^, £^ are the scalar partners of the right-handed, left-handed £~~, and + ^ R / R ) is smaller than a(e e
ilk)] 3.2
Searches for Squarks and Gluinos
The SUSY partners of top and bottom quarks (stop, t, and sbottom, b) have been searched
433
for in the LEP experiments. Stop and sbottom are mixtures of the SUSY partners of the left- and right-handed quarks, with the lowest mass squarks denoted by t\ = £L COS Of + iR,sin#j and h\ = 6 L C O S ^ + &Rsin^, where Ol and 9^ are the mixing angles. Searches are carried out for t\ —> cx°, t\ -—> b£v, and &i —> bxl- A M is the mass difference between the stop or sbottom and the SUSY decay product, Xi o r v. Exclusion regions for combinations 11 of preliminary data from the four LEP experiments are shown in Fig. 14 for no mixing and for the mixing angles for which the t\ and 61 decouple from the Z°. The 95% C.L. lower limits are shown in Table 2. Experiments at hadron colliders are sensitive to searches for scalar quarks (q) and gluinos (5). An example is shown in Fig. 15 of a recent search 12 by CDF based on the signature of large missing energy from the two LSPs and three or more hadronic jets resulting from the decays of the q and/or g. They obtain 95% C.L. limits of mg > 195 GeV independent of rriq and rrig > 300 GeV for the case rrig « mg.
Searches for New Particles
Gail G. Hanson Table 2. The excluded Mi C.L. for AM > 10 GeV.
Lower limit for t\
and M^
regions at 95%
,tanp=2 . , . , u=, -200 . , ,GeV .,, , •"-
Lower limit for b\
ADLO
Vs > 206.5 GeV
' '_ _ ~
m „ > 103.5 GeV for i t > 300 GeV
(GeV)
(GeV) *l
cx\
Uv
0
97
56
95
(°)
o
-*
—102-
(°)
bx\
97
0
100
95
68
92
ff 101-
100-
Excluded at 95 1 C.L. 99-
Mv (GeV) £p+ ^ T jets search for gluinos and squarks
ft
Figure 16. 95% C.L. lower limits for combined d a t a from the LEP experiments for the mass of the lightest chargino versus the mass of the scalar neutrino.
CDF PRELIMINARY j JLdt=84 pb ' Vs=] 8 TeV 1mSUGRA
ISAJET 7 37 + PROSPINO |
lanp=3 95% CL
L J
> o
at the kinematical limit. The exclusion region from the combination 13 of preliminary searches from the four LEP experiments is shown in Fig. 16. The 95% C.L. lower limit is M-+ > 103.5 GeV for Mv > 300 GeV. Limits on the x.i LSP mass are obtained from combined searches for charginos, sleptons, and MSSM Higgs bosons. The exclusion regions for MSUGRA constraints and for CMSSM constraints for the combined preliminary LEP searches 14 are shown in Fig. 17. The 95% C.L. lower limits are M^o > 60 GeV for MSUGRA and M*o > 45.6 GeV for CMSSM, both for 175 GeV top quark mass.
mg(GeV/c )
Figure 15. T h e 95% C.L. region in the rriq—mg plane newly excluded by CDF. Results from some previous searches by CDF, DO, LEP, UA1 and UA2 are also shown.
3.3
Searches for Charginos and Neutralinos
Charginos (x^1 can be produced in pairs in e + e~ collisions: e+e~ —> xtxi • They can then typically decay as xt ~~* XiW* —> Xi^ + i/ or XiQQ'- The signature is large missing energy and large missing transverse momentum, and detection depends on A M = M-+ — M^o. There are several topologies:
3.4
Searches for R-parity SUSY
Violating
In .R-parity violating SUSY decays, theMightest supersymmetric particle is expected to be unstable. Many searches have been performed for .R-parity violating SUSY. One example 15 from HI is shown in Fig. 18. In Xi Xj this case the squarks are assumed to be prohadronic with large multiplicity, large mulduced through an .R-parity violating Aj -fe tiplicity with isolated lepton, and low mulcoupling. They decay either through the tiplicity (acoplanar leptons). At large TOO same coupling or through an R-parity con(heavy scalar leptons) the cross section is the serving gauge decay into a x * , a x ° , or a j . largest. The lower limit for M~+ is nearly
434
Searches for New Particles
Gail G. Hanson X° Neutralino LSP Constraints in MSUGRA
X° Neutralino LSP Constraints in MSSM Unification at GUT Scale (CMSSM)
lfl
A D L O preliminary
•
With L E P Combined results
m - 0>60GeV
\
MM
-Higgsij
/ "
m0 i< 1 feV/c m,J= J7S GeV/c2
X#"--4
C'harsjinjis (large m0)
Nfegs;; | "I -jHijsgs
. v r
;
\
/iSlipiois! J •'m.j5>".54.Gey.
•.••'.'
;
V
;
r
mv, > 45.6 GeV Jinith^^dmdor
2
• . ,Mlof= i75:Ce\fA; . » M, op =;i8.0 ;
10
Excluded at 95 <mCli
. A „ ^ , rn0
Excluded al 95% C.L.
20
30
40
Figure 17. 95% C.L. lower limit for t h e combined L E P d a t a for t h e LSP mass versus tan/3 for MSUGRA (left) and CMSSM (right).
Minima? bupergravity + R Violation >. " °
J * « * J.J
3
1 V"*»v 1 \ - S V
.
K\ -- -
-
•A
- , \
*•
r\—--^
™ . «
.**-
^ .
,0
" '
fi*
-J^x" »
^ 3
**
;•-.- ,"-o aoo a s
m.
(GHV)
Figure 18. Region excluded by H I in t h e (mo, m j / a ) plane for (j, < 0, A® = 0 and tan 0 = 2 (left) or t a n j3 = 6 (right) for an .R-parity violating coupling A'131 = 0.3. The regions excluded by DO and L3 are also shown.
4
Single Top Quark Production
Searches for flavor-changing neutral currents (FCNC) have been performed at the Fermilab Tevatron in rare decays of the top quark and at LEP and HERA in single top quark
435
production. FCNC are suppressed at tree level in the Standard Model (GIM mechanism) . Small contributions occur at the oneloop levd I n e + e - t h e S M c r o g s s e c tion ^ 9 ^ Extensions to the Standard is 10 M o d e l ) s u c h a s S U S Y a n c j multiple Higgs
Gail G. Hanson
Searches for New Particles
doublet models, can allow FCNC at the tree level.
K
T
,K
7
exclusion regions in the AC7 — K% plane are shown in Fig. 20 for CDF, ZEUS, 17 HI, 1 8 and the combination 19 of preliminary LEP data. QCD and ISR corrections to the Born-level cross sections are used for the LEP combination.
. q(u,c)
LEP e+e-^tq^bWq
e"bWX
HI PRELIMINARY ]01.6pb"Vp data 94-00
Figure 19. Single top quark production at LEP and HERA.
80
b)
a) 60
-
60
40
0.8
w
X7x
by cm- k .cliidtrl b\ CD! kv l.l'P | . eluded h
x x \ \:>x\
20
4
h-
\
—^.V;-™^
HI
1 1
40
,\ \ \ : V-
+'
10
M^v(GeV)
+•' w
20
fj.
# ;
\t, P #&-4/i* M* 10
Mi v (GeV)
0.6 :
• \ x \ ,;-\v*-;
0.4 ;
Figure 21. A comparison of the HI events with isolated leptons and large missing PT with the predictions of Standard Model single W production. P* is the transverse momentum of the hadronic system and Mj," is the transverse mass of the hadronic system.
r
%
0.2
• .X^>. \
0.2
U.4
XX
0.6
" x x X :\.:x \ ' 0.8 1
Figure 20. The light grey region shows the combined LEP exclusion region at 95% C.L. in the KZ — ft-y plane for mt = 174 GeV with QCD and ISR corrections. The exclusion curves for different values of mt are also shown. The hatched area shows the CDF exclusion region, and the two vertical lines with arrows show the ZEUS and HI Ktu-y exclusion regions.
5
Events with Isolated Leptons and Large Missing pr
The HI experiment 20 at HERA observes an excess of events with isolated leptons and large missing transverse momentum (PT)The primary Standard Model process is sinCDF performed a search 16 in the top de- gle W production. The presence of these excays t —> 7c(w) and t —+ Z°c(u) in pp col- cess events increases the lower limit on K 7 in lisions at i / i = 1 . 8 TeV. Searches for single the single top quark search. Figure 21 shows a comparison of the events with the predictop production at LEP and HERA can be described by the processes shown in Fig. 19. tions for single W production for electrons and muons separately. The excess above the The FCNC transition can be described usSM expectation is mainly due to events with ing the anomalous coupling parameters K 7 transverse momentum of the hadronic sysand Kz, which represent the tree-level 7 and Z° exchange contributions. The 95% C.L. tem (P*) greater than 25 GeV where 10
436
Searches for New Particles
Gail G. Hanson events are found compared to 2.8 ± 0.7 expected. The numbers of events in the electron and muon channels for P* < 25 GeV and P* > 25 GeV are shown in Table 3. The ZEUS experiment 17 does not observe an excess; the numbers of events for P* > 25 GeV for ZEUS are also shown in Table 3. Table 3. Comparison of numbers of events with isolated leptons and large missing p x with the predictions of Standard Model single W production for HI and ZEUS. P£ is the transverse momentum of the hadronic system.
HI
electron
muon
obs.
exp.
obs.
exp.
P£ < 25 GeV
6
6.6
2
1.0
Pfi > 25 GeV
4
1.3
6
1.5
ZEUS
electron
Pfi > 25 GeV
6
Searches for Leptoquarks
Leptoquarks are resonant states carrying both baryon number and lepton number. Searches for them have been performed at the Tevatron and at LEP 2 1 and especially at HERA, 22 where they may be produced directly through e ± -quark fusion, and decay into e ± -quark or i^(i/)-quark. Leptoquarks can be scalar or vector states. F = L + 31? is preserved. Figure 22 shows ZEUS and HI limits on the Yukawa coupling constant A versus the leptoquark mass for first generation leptoquarks. 7
Searches for Excited Fermions
muon
obs.
exp.
obs.
exp.
1
1.1
1
1.3 Z.W.g
Figure 23. Excited fermion production at HERA.
Constraints on Scalar Leptoquarks Excited fermions arise naturally in models that predict a substructure in the fermion sector. Searches for pair production of excited leptons and singly produced excited leptons have been carried out at LEP. The effective electroweak Lagrangian describing chiral magnetic transitions from excited to ordinary leptons can be written -11 limit e" p. Preiun,} -II limit
C-w
f
e p, 9 4 - 97)
1 .- fAV 9f\w» t*a 2A
325 350 375 400
+9'f'^Bi
M LQ (GeV) Figure 22. Limits on the Yukawa coupling constant A versus the leptoquark mass M L Q for first generation scalar leptoquarks. Limits from L E P and Tevatron are also shown.
437
eL + h.c.
(4)
where A corresponds to the compositeness scale, the subscript L stands for left-handed, g and g' are the SM gauge coupling constants, and the factors / and / ' are weight factors associated with the two gauge groups SU(2)
Gail G. Hanson
Searches for New Particles
l
DELPHI&OPAL preliminary
f=r
f'/f e [ - 5 ; + 5 ] H1 Preliminary
Excited Electron Limits (Summer 2001)
s t
i*
J :-,/~'~~'
ep
/•
..-•'•'
/i
1
1
Vx/V\A^V—-^^^ 10 110
125
140
150
175 200 225 v* mass (GeV)
Figure 24. 95% C.L. upper limits on the ratio of the coupling to the compositeness scale for combined L E P search for excited leptons at yfs = 189 — 209 GeV (left), for ZEUS, HI and combined LEP searches for excited electrons (center), and for HI search for excited neutrinos (right).
x U(l). Typical decays of excited leptons are the following: r ± -> ^ 7 , I** -> vW±, l*± _> g±z°, v* -> iAy, v* -> i^W*, and v* -> vZ°. Excited fermion production and decay at HERA can occur through the processes shown in Fig. 23. Figure 24 shows the 95% C.L. upper limits on the ratio of the coupling to the compositeness scale for the combined LEP searches 23 for excited leptons e*, n*, and T* at y/s = 189 - 209 GeV, for ZEUS, 24 HI 2 5 and combined LEP searches for excited electrons, and for the HI search 26 for excited neutrinos. 8
Technicolor Searches
Technicolor represents an alternative to the Higgs mechanism for generating electroweak symmetry breaking. In this model the longitudinal degrees of freedom of the massive SM gauge bosons are the Goldstone bosons associated with the breaking of global chiral symmetry of a new kind of fermions, the technifermions, which besides the SM quantum numbers carry the charge of a new QCD-like interaction called Technicolor. In walking technicolor, the lightest technicolor mesons may be light enough to be observable at LEP2. DELPHI 27 and OPAL 28 have carried out searches for such technimesons: ITT
and px/ior- Figure 25 shows the 95% C.L. excluded regions for these searches. OPAL obtains mPT > 77 (62) GeV for ND = 9 (2), where N^ is the number of technifermion doublets (2 is the minimum number). DELPHI obtains mPT > 89.1 (79.8) GeV for A^D = 9 (2). 9
S u m m a r y a n d Conclusions
There is a "hint" at the two standard deviation level of a signal for a Standard Model Higgs boson from the combined search results of the four LEP experiments. The hint is weaker than it was at the November 3, 2000, LEPC presentation, but the number of events was not enough to establish a signal in any case. An extension of the LEP2 run was requested but unfortunately was not granted. Now we will have to wait until ~ 2007 to find out whether there is a light Higgs boson with TOH ~ 115 GeV. The 95% confidence level lower limit from the LEP searches is 771H > 114.1 GeV with a median limit of 115.4 GeV expected for background only. Many searches for new particles have been performed, but there have been only negative results and new limits established, so we are still left with only questions: Is there another mechanism for electroweak symmetry breaking? Supersymmetry? Tech-
438
Gail G. Hanson
Searches for New Particles DELPHI
>
j'l
01
140
o s
120
1 '
1
r
T ' 1 '
1 ' 1 '
1 '
1 \ OPAL Preliminary I \ \ i7i»y-^bbyMv=200GeV m'-J Dr4y-^bbyMv=100GeV 1 J
~
Hiij Kj —> bqbq'
100 80 60
-excluded I..i.....i i . i V . • ! ... 1 .. .1
1....I..J- 1 -
250 30u 350 400 450 500 550 bOO 2
M(pT) [GeV/c ]
m p/co [GeV]
Figure 25. The 95% C.L. excluded regions in the (MPT/U,T - M^T) plane from DELPHI (left) and OPAL (right). T h e dashed lines show the median expected exclusions for the background only hypothesis.
nicolor? Something we have not even thought of yet? In ten years' time we should have some answers. Acknowledgments I would like to acknowledge all of the contributions to the XX International Symposium on Lepton and Photon Interactions at High Energies which formed the basis for this review. I am extremely grateful to the many people who explained the details of their research to me and sent me figures. I apologize to those whose research I did not have time to cover. I would like to thank the Organizers and the Scientific Secretaries for all their help in preparing the talk. I would also like to acknowledge the support of the U.S. Department of Energy grant number DEFG0291ER40661. References 1. P. Igo-Kemenes, talk given in the LEPC open session (November 3, 2000) on behalf of the LEP Higgs Working Group. See http://lephiggs.web.cern. ch/LEPHIGGS/talks.
439
ALEPH, DELPHI, L3 and OPAL Collaborations, The LEP Working Group for Higgs Boson Searches, "Search for the Standard Model Higgs Boson at LEP," CERN-EP/2001-55 (July 11, 2001). The LEP Higgs Working Group, "Standard Model Higgs Boson at LEP: Results with the 2000 Data, Request for Running in 2001," prepared for the LEP Committee and the CERN Research Board (November 3, 2000). See http://lephiggs.web.cern.ch/LEPHIGGS /papers. ALEPH Collaboration, R. Barate et al, Phys. Lett. B495, 1 (2000); L3 Collaboration, M. Acciarri et al, Phys. Lett. B495, 18 (2000); DELPHI Collaboration, P. Abreu et al, Phys. Lett. B499, 23 (2001); OPAL Collaboration, G. Abbiendi et al., Phys. Lett. B499, 38 (2001). L3 Collaboration, P. Achard et al, Phys. Lett. B517, 319 (2001). ALEPH, DELPHI, L3 and OPAL Collaborations, The LEP Higgs Working Group, "Searches for the Neutral Higgs
Searches for New Particles
Gail G. Hanson
7.
8.
9.
10.
11.
12.
Bosons of the MSSM: Preliminary Combined Results Using LEP Data Collected at Energies up to 209 GeV," LHWG Note 2001-04 (July 9, 2001). See http://lephiggs.web.cern.ch/ LEPHIGGS/papers. ALEPH, DELPHI, L3 and OPAL Collaborations, The LEP Higgs Working Group, "Searches for Higgs Bosons Decaying into Photons: Preliminary Combined Results Using LEP Data Collected at Energies up to 209 GeV," LHWG Note 2001-08 (July 2, 2001). See http://lephiggs.web.cern.ch/ LEPHIGGS/papers. ALEPH, DELPHI, L3 and OPAL Collaborations, The LEP Working Group for Higgs Boson Searches, "Flavour Independent Search for Hadronically Decaying Neutral Higgs Bosons at LEP," LHWG Note 2001-07 (July 5, 2001). See http://lephiggs.web.cern.ch/ LEPHIGGS/papers. P. Ferrari and D. Zer-Zion, The OPAL Collaboration, "Two Higgs Doublet Model Interpretation of Neutral Higgs Boson Searches up to the Highest LEP Energies," OPAL Physics Note PN475, July 6, 2001; M. Bluj, M. Boonekamp, J. Hoffman, and P. Zalewski, The DELPHI Collaboration, "Searches for Higgs Bosons in a General Two Higgs Doublet Model," DELPHI 2001-068 CONF 496, July 5, 2001. LEP SUSY Working Group, ALEPH, DELPHI, L3 and OPAL Collaborations, LEPSUSYWG/01-01.1 (June 25, 2001). See http://lepsusy.web.cern.ch/ lepsusy/. LEP SUSY Working Group, ALEPH, DELPHI, L3 and OPAL Collaborations, LEPSUSYWG/01-02.1 (February 16, 2001). See http://lepsusy.web.cern.ch/ lepsusy/. T. Affolder et al., CDF Collaboration,
13.
14.
15. 16. 17.
18.
19.
20.
440
"Search for Gluinos and Scalar Quarks in pp Collisions at v's = 1.8 TeV using the Missing Energy plus Multijets Signature," EFI-01-22, FNAL-PUB01/084-E, June 7, 2001, submitted to Phys. Rev. Lett. , hep-ex/0106001. LEP SUSY Working Group, ALEPH, DELPHI, L3 and OPAL Collaborations, LEPSUSYWG/01-03.1 (February 22, 2001). See http://lepsusy.web.cern.ch/ lepsusy/. LEP SUSY Working Group, ALEPH, DELPHI, L3 and OPAL Collaborations, LEPSUSYWG Note. See h t t p : / / lepsusy. web.cern.ch/lepsusy/. C. Adloff et al., HI Collaboration, DESY 01-021, Eur. Phys. J. C20, 639 (2001). F. Abe et al., CDF Collaboration, Phys. Rev. Lett. 80, 2525 (1998). ZEUS Collaboration, "Search for Single Top Production in ep Collisions at HERA," submitted to the International Europhysics Conference on High Energy Physics, Budapest, Hungary, July 12-18, 2001, Abstract 650. HI Collaboration, "Search for Single Top Production in e ± p Collisions at HERA," submitted to the XX International Symposium on Lepton and Photon Interactions, Rome, Italy, July 23-28, 2001, Abstract 512. The ALEPH, DELPHI, L3 and OPAL Collaborations, and the LEP Exotica Working Group, "Search for Single Top Production Via Flavour Changing Neutral Currents: Preliminary Combined Results of the LEP Experiments," LEP Exotica WG 2001-01, Contribution to LP01, July 9, 2001. HI Collaboration, "Observation of Isolated Leptons with Missing PT and Comparison with W Production at HERA," submitted to the XX International Symposium on Lepton and Photon Interactions, Rome, Italy, July 23-28, 2001, Abstract 495.
Gail G. Hanson
Searches for New Particles
21. S. Andringa et al, DELPHI Collaboration, "Search for Single Leptoquark Production in e+e~ Collisions up to v ^ = 208 GeV with the DELPHI Detector," DELPHI 2001-080 CONF 508, June 3, 2001, submitted to the XX International Symposium on Lepton and Photon Interactions, Rome, Italy, July 23-28, 2001, Abstract 222; G. Abbiendi et al, OPAL Collaboration, "Search for Single Leptoquark and Squark Production in Electron-Photon Scattering at ^/s^ = 1 8 9 GeV at LEP," CERN-EP-2001-040, May 17, 2001, submitted to Eur. Phys. J. 22. ZEUS Collaboration, "Search for Leptoquarks in ep Collisions at HERA," submitted to the International Europhysics Conference on High Energy Physics, Budapest, Hungary, July 12-18, 2001, Abstract 600. 23. The ALEPH, DELPHI, L3 and OPAL Collaborations, and the LEP Exotica Working Group, "Search for Excited Leptons: Preliminary Combined Results of the LEP Experiments," LEP Exotica WG 2001-02, Contribution to LP01, July 9, 2001, Abstract 274. 24. ZEUS Collaboration, "Search for Excited Fermions in ep Collisions at HERA," submitted to the International Europhysics Conference on High Energy Physics, Budapest, Hungary, July 12-18, 2001, Abstract 607. 25. C. Adloff et al, HI Collaboration, DESY 00-102, Eur. Phys. J. C17, 567 (2000). 26. Hi Collaboration, "Search for Excited Neutrinos at HERA," submitted to the XX International Symposium on Lepton and Photon Interactions, Rome, Italy, July 23-28, 2001, Abstract 515. 27. W. Adam et al, DELPHI Collaboration, "Search for Technicolor with DELPHI," Contributed paper for EPS HEP 2001 (Budapest) and LP01 (Rome), DELPHI 2001-086 CONF 514, July 1, 2001.
441
28. The OPAL Collaboration, "Searches for Technicolor with the OPAL Detector in e + e~ Collisions at the Highest LEP Energies," OPAL Physics Note PN485, July 10, 2001.
This page is intentionally left blank
Neutrino Physics
Neutrino Physics Session Chair: Scientific Secretaries:
J. Goodman C.K. Jung J. Klein
H. Sugawara F. Terranova A . oatta
Results from SuperKamiokande Results from K2K Results from SNO
Session Chair: L. Maiani Scientific Secretaries: F. Spada i\. oatta S. Aoki H. Murayama
Experimental review of Neutrino Physics Theory of Neutrino masses and mixings
444
R E C E N T RESULTS F R O M S U P E R - K A M I O K A N D E J. A. GOODMAN FOR THE SUPER-KAMIOKANDE COLLABORATION Department of Physics, University of Maryland College Park, MD, USA 20742-4111 Super-Kamiokande is a 50 Kiloton water-Cherenkov t h a t detects neutrinos in the MeV energy range t h a t are produced in the Sun and neutrinos in t h e GeV energy range produced in t h e atmosphere by cosmic rays. The detector has been operational since April of 1996. In this paper results of our most recent analysis will be presented on both atmospheric and solar neutrinos.
1
Introduction
The existence of neutrinos was postulated in the 1930's. From that time the question of whether or not neutrinos have mass has yet to be conclusively answered. Direct measurements of neutrino mass have proved illusive. If neutrinos have mass then it is possible for them to change flavors as they propagate from their production point. The oscillation probability between two neutrino flavors is given by P = sin2 29 • sin2(1.27 ^ ^ Eu(GeV)
Am 2 (eV 2 ))
where 9 is the mixing angle, L is the flight length of the neutrino, Ev is the neutrino energy, Am 2 = (mf —TO2)is the mass squared difference. Super-Kamiokande is a 50 Kiloton water-Cherenkov detector designed to study neutrino oscillations. It is capable of detecting neutrinos in the MeV energy range that are produced in the Sun and neutrinos in the GeV energy range produced in the atmosphere by cosmic rays.
lected by 11,146 inward facing 50 cm PMTs mounted uniformly on the wall, providing 40% photo-cathode coverage. In the OD, 1885 outward facing 20 cm PMTs monitor the 2.5 m thick veto region. The veto tags incoming particles and is a passive shield for gamma activity from the surrounding rock. The fiducial volume for the analysis of neutrino events starts 2 m inward of the walls of the ID and contains 22.5 kton of pure water. 3
Atmospheric Neutrinos
Primary cosmic rays undergo hadronic interactions in the atmosphere producing copious amounts of pions, many of which decay before further interaction. This decay leads to the production of muons and neutrinos via:
The muons may also decay leading once again to the production of more neutrinos: ^
-> e ± + V^v^)
+ ve(Ve).
This yields an expected ratio of the flux of 2
The Super-Kamiokande Detector
( ^ + ^ ) / ( ^ e +Ve) ~ 2
Super-Kamiokande is a water-Cherenkov detector located in the Kamioka Mine in Gifu, Japan. The cylindrical detector is divided into an inner and outer detector (ID and OD, respectively) by a stainless-steel frame structure that serves as an optical barrier and a mounting point for all photo-multiplier tubes (PMTs). Cherenkov light in the ID is col-
445
In Super-Kamiokande we can observe neutrinos produced in the atmosphere anywhere around the world. Neutrinos produced overhead travel on the average 10-15 km before reaching the detector, while neutrinos produced on the opposite side of the earth travel more than 10,000 km before reaching the detector. Detailed simulations of
Recent Results from Super-Kamiokande
J. A. Goodman the atmospheric neutrino flux is discussed by Honda l.
3.1
Atmospheric neutrino data
In Super-Kamiokande we divide the observed events from atmospheric neutrino interactions into several categories depending on their energy and event topology. Events are divided into energy intervals: events with Evis > 1.33GeV are called multi-GeV events (for historical reasons) while events with E vis < 1.33GeV and P e > lOOMeV/c or P^ > 200MeV/c are called sub-GeV events. Events are characterized as fully-contained (FC) or partially-contained (PC) depending on whether their tracks extend out of the ID. FC events with only one reconstructed ring are subdivided into e-like and /t-like based on likelihood analysis of Cherenkov light pattern. Figure 1 shows a view of the differences between the pattern of light of simulated electrons and muons. Electrons shower and undergo multiple scattering while emitting Cherenkov light while muons travel in approximately straight lines. As a result muon induced rings are much sharper than electron induced rings. This is easily discerned visually and by software. The misidentification fraction is determined to less than 5%. The results of our particle identification are show in Figure 2 along with the expectation from simulation. In Table 1 we show the data from 1289 live-days along with the expectations from the atmospheric Monte Carlo calculations. By computing the double ratio of muon-like events to electron-like events divided by the expectation from Monte Carlo, the dependence on the absolute flux, where there is a ~ 20% uncertainty, is removed. The uncertainty in the Monte Carlo ratio is less than -5%
Figure 1. Shown in the top(bottom) figure is a s i m u lated muon(electron) event in the Super-K d e t e c t o r . Notice the sharpness of the muon ring compared to the electron ring).
(/x-like/e-like) DA TA (/i-like/e-like)MC 446
J. A. Goodman
Recent Results from Super-Kamiokande
Figure 2. Particle id parameter. Datais shown as dots. The lines are Monte Carlo without oscillations.
'«».
i-:^^
^—^rpi^fe^.
\
S 3
The observed values are for sub-GeV: • '
R = 0.644±g;gi? ± 0.051
-1-0.5
• <
'
Gn 0.5
1-1
and for multi-GeV:
3.2
v,M —r vT Oscillation Analysis
In Figure 3 the zenith angle distribution for e-like and mu-like data are shown. In the first of these plots the sub-GeV data with lepton momentum less than 400 MeV/c is shown. This data is insensitive to primary neutrino direction and is useful as a diagnostic. The
0n0.5
Miilti-Getf'^-like''
>
W 5200 o
R = 0.6791^32 ± ° - 0 8 0 These double ratios are consistent with our previous result 2 and shows a deficit of muon neutrinos which can potentially be explained by neutrino oscillations. We have also studied upward going muons which enter the detector from below. These muons come from neutrino interactions in the rock below the detector. Energetic muons traverse the entire detector while lower energy muons may stop in the detector. The zenith angle distribution of these events is also shown in Figure 3. Details of this analysis can be found in Ref. 2.
-0.5
COS0 M|ulti-GeV')i-lik'e'+'PC'
COS0
iZw3
-0.5
0
Q COSO
0.5
1-1
-0.6 -0.4 COS0
-0.2
0 „ 0.5 cos9
upward stopping u
upward through going u
-0.8
-0.5
0
•0.8
-0.6 -0.4 COS9
-0.2
Figure 3. Zenith angle distribution of SuperKamiokande 1289 days FC, P C and UPMU samples. Dots, green line and red line correspond to data, MC with no oscillation and MC with best fit oscillation parameters, respectively.
447
Recent Results from Super-Kamiokande
J. A. Goodman Sub-GeV (Ems < 1.33GeV) Data MC(Honda flux) 2680.4 Iring e-like 2864 4053.2 Iring //-like 2788 multi ring 2159 2680.5 Total 7811 9315.1
> E <
Sin 2 26=1.0, AM 2 =2.4x10- 3 eV 2 X2min=132.4/137 d.o.f.
Multi-GeV FC(Evis > 1.33GeV) Data MC( Honda flux) Iring e-like 626 617.8 Iring /i-like 558 834.0 multi ring 1318 1652.6 2502 Total 3104.5 Partially Contained 754 Total
0.1
1073.8
second and third set of plots shows progressively higher energy and the clear angular dependence of the muon deficit is visible. The bottom plots are for upward going and through going muons. All of this data can be used independently to test the oscillation hypothesis. Doing this yields the allowed oscillation parameters shown in Fig. 4. The minimum x2 is found to be 132.4 with 137 degrees of freedom (d.o.f.) at Am 2 = 2.4 x 10 _ 3 eV 2 , sin226> = 1.00. The deficit of upward going /i-like data is well explained by assuming v i>. —* VT oscillation. \ 2 f° r n o oscillation was found to be 299.3 for 139 d.o.f. Study ofVn Oscillations
and vn
^sterile
Since the number of light (active) neutrinos has been set at three by measurements of the Z° width. The existence of a non-interacting "sterile" neutrino (i/g) has been postulated to explain some experimental results. This neutrino would have neither a charged nor neutral current interaction. If the observed deficit of v^ is due to
02
0.3
DJ
0.3
O.S
0.7
0.8
0.9
v
ii -^ vs oscillation, then the number of interactions observed from up-going neutrinos will be reduced compared to the number that would be expected from the neutral current interactions due to vT. In addition for v^ —> vs oscillations, matter effect will suppress oscillations of high energy {Ev > 15GeV) neutrinos 6 . To observe these effects we: (1) create a neutral current enriched data sample by selecting multiring events with muons, (2) select high energy (Evis > 5GeV) muons and (3) look at upward-going though muons. These data are shown in Figure 5. Using a test of the up(cosO < -0.4)/down(cos 6 > 0.4) ratio for the first two plots and the vertical(cos G < —0.4)/horizontal(cos 0 > —0.4) ratio for the third we are able to exclude the sterile hypothesis at the 99% confidence level. The exclusion regions for these combined results are shown in Figure 6.
3.4
Atmospheric
Summary
The results from our 1289 day atmospheric sample show clear evidence for v^
448
1
sin226
Figure 4. 68,90 and 99% confidence level allowed regions for v^ —> uT oscillation obtained by SuperKamiokande 1289 days result.
Table 1. Event summary for 1289 days
3.3
Best FitforV 11 toV T
Recent Results from Super-Kamiokande
J. A. Goodman
V
^Vt
£ 'Ml
:
h .
i I » I • I •
i>
v —>v (i
S
(Am">0) <
cos0
10
70
• I i I i I • I • I
60
a 0> >
o o u
50 •
°r~i
v —»v
•
[\
s
KI0
(Am"<0) <
•*- 40
_i_
30
0>
i
i i- • i .
i
sin22G
42
s3 20 S3 10
Figure 6. Excluded regions for three oscillation modes. T h e light(dark) gray region is excluded at 90(99)% C X . by NC enriched sample and high-energy sample analysis. Thin dotted(solid) line indicates the 90(99)% C.L. allowed regions from lring-FC sample analysis.
0
COS0
dilations. Analysis of this data yields a best fit of sin2 29 = 1 and Am2 = 2.3 x 10" 3 eV 2 . In addition our data excludes v^ —> vs oscillations at the 99% confidence level. 4
Figure 5. Zenith angle distributions of: NC enriched sample (top), high-energy PC sample(middle), upthrough-going muon sample (bottom). The solid line is Vfj, —> vT while the dashed line is v^ —• vs
449
Solar N e u t r i n o s
Energy is produced in the Sun through nuclear fusion. For a star such as the Sun, this involves turning 4 protons into a helium nucleus to tap the a particle binding energy. For every a particle made, two neutrinos are produced. The neutrinos, due to their small
J. A. Goodman
Recent Results from Super-Kamiokande
interaction cross section, escape the Sun, essentially undergoing no interactions. Solar neutrinos have been previously detected in chlorine-, gallium-, and water-based detectors 7,8,9,10 . These experiments were each sensitive to different neutrino energy ranges, but all found fluxes significantly lower than those predicted by models of the solar interior, known as Standard Solar Models (SSMs; 11,12 ). This discrepancy between the predicted and measured flux of solar neutrinos is known as "the solar neutrino problem." The solution to the solar neutrino problem is generally believed to involve neutrino oscillations. In the Super-Kamiokande detector neutrinos of sufficient energy (>6 MeV) are detected in real-time by the elastic scatter of electrons. These scattered electrons are used to measure the flux of solar neutrinos, as well as searching for any possible distortions in the neutrino spectrum, and any short or long term time dependence of the flux. An observation of a measurable day/night flux difference, a distortion of the neutrino energy spectrum, or a seasonal dependence to the neutrino flux would provide strong evidence of neutrino oscillations.
4-1
Solar neutrino data
Recoil electrons from solar neutrinos seen in Super-Kamiokande have energies that range from 5 to 18 MeV. At these energies, the electron is limited to a few centimeters in range and the vertex position is found using the relative timing of hit PMTs, assuming that all Cherenkov photons came from a single point. Once a vertex is reconstructed, the direction of the electron is determined using the characteristic shape of the emitted Cherenkov radiation. The energy of the scattered electron is determined from the number of hit PMTs. In order to set the absolute energy scale of the detector, a linear accelerator of electrons (LINAC) has been installed at the
detector 14 . This system allows electrons of a known, fixed energy to be injected into the detector. The energy scale has also been cross checked with the well-known j3 decay of 16 N, which is produced in situ by an (n,p) reaction on l e O . The fast neutrons needed for this reaction are produced by a portable deuteriumtritium neutron generator (DTG; 15 ). The energy scale measured with the DTG agrees with the LINAC to within ±0.3%. The total uncertainty in the absolute energy scale is ±0.6%. We report here solar neutrino data taken between May 31, 1996 and October 6, 2000, representing 1258 live days. The raw data sample consists of 2.0 x 109 triggered events before any background reduction is performed. This sample is reduced through a series of cuts designed to remove known sources of background, including high energy cosmic ray muons and their spallation products, events generated by electrical noise and arcing in the PMTs, and external gamma-ray activity. Additional cuts are performed to remove likely background events with poorly defined vertex positions 16 . After the reduction, a sample of 236,140 events remain, with a signal-to-noise ratio of ~ 1 in the direction of the Sun. The neutrino and scattered electron have a strong angular correlation, and the solar neutrino signal is extracted from the data using the cos9sun distribution. The value of 9sun is the size of the angle between the recoil electron momentum and a vector connecting the Sun to the Earth. The distribution of cos 6sun for the reduced data sample is shown in Figure 7. A strong peak from solar neutrino events is seen. The number of solar neutrino events is extracted from the cos 0sun distribution using a likelihood function that fits the measured background shape and the expected signal shape, based on a detector simulation of solar neutrino events, to the data. This fit is also shown in Figure 7 as a solid line. After 1258 days, a total
450
J. A. Goodman
Recent Results from Super-Kamiokande 50.55
SK1258day 5.0-20MeV 22.5kt ALL (Preliminary)
(fi CO
« 0.5
(280day LE 6.5-20MeV + 978day SLE 5.0-20MeV)
Q
0.45
0.4 f
w
w
,%»..,«.....•'.»***«*.%*'"•— »*^*"i«'» " ""»
Day 0.35
Night
^ Q .«>
0
cose,,
Figure 7. The COS "sun distribution for events in the reduced sample of solar neutrino candidates. T h e best fit to signal+background is shown as the solid line, while the measured background shape is shown as a dashed line.
cose,
Figure 8. Measured solar neutrino flux as a function of zenith angle, relative to the reference SSM flux. The horizontal lin represents the flux measured for all data. The extreme right bin represents d a t a collected after passing through t h e Earth's core (cos0 z > 0.84).
The corresponding 8 B flux at 1 AU is: 2.32±0.03(stat.)+oo7(sys.) x 10 6 cm" 2s-\
of 18,464 ± 204(stat.)i^(sys.) signal events are found. 4-2
Solar Results
The number of signal events is translated into a measured flux using a full detector Monte Carlo simulation taking the input flux and spectrum from the reference SSM, and comparing this to the number of measured signal events. Our detector simulation is based on GEANT 3.21 17 . Neutrinos that produce recoil electrons with energies > 5MeV are produced almost entirely from the (3 decay of 8 B in the solar interior, with a slight admixture of neutrinos from the 3 He-p (hep) fusion reaction. The flux normalization for both fluxes and spectral shape for hep are taken from the BP2000 SSM 11 . For the 8 B neutrino spectral shape, we have taken the recent improved measurement of Ortiz et al. 18 The number of signal events obtained from the cos6sun distribution represents 45.1±0.5(stat.)t^% of the reference flux.
451
(1) The solar neutrino flux as a function of zenith angle is also measured, and the results shown in Figure 8. The zenith angle (8Z) is defined as the angle between the vertical axis at the Super-Kamiokande detector and the vector connecting the Sun to the Earth. The nighttime solar neutrino flux is measured when cos9z > 0, while the day-time flux is measured when cos# z < 0. Additionally, to search for enhancement in the flux for neutrinos passing through the core of the Earth, the night-time period is divided into 6 zenith bins. The flux measured in each of the these zenith bins, relative to the expectations from the reference SSM, are shown in Figure 8. The flux asymmetry between night- and daytime total fluxes is found to be: xf"~t\
= 0.033±0.022(stat.)+°;^ ( s y s .)
2 l * n + Wd)
(2) The flux as a function of season is shown in Figure 9, along with the expected variation from the Earth's eccentricity. The measured
J. A. Goodman
Recent Results from Super-Kamiokande
^0.55
5 CO CO
CO CO (0
13 « 0.6
«3 0.5 Q
Q ^~~~~f~~~~. •
0.45
x
I
^
J>\ ^ ^
T
'—^v-
...
\
i
J
- ^ " ^
i
1
1
'
0.5
1
J
-
0.4 -
0.35
_
m en cc >
<
LU <
z
0 . < Z>
"3 ^ 2 < s -»
O 3 <
0.4
0.3
D- H > O LU O O Ul CO O -Z. Q
Figure 9. The seasonal dependence of t h e solar neutrino flux, relative t o the reference SSM. T h e solid line shows the expected shape and shape of t h e variation in solar neutrino flux due t o the Earth's eccentricity.
data are consistent (x 2 /d.o.f. = 3.9/7) with the expected annual seasonal variation. A fit to a flat distribution has a x 2 /d.o.f. = 8.1/7. In order to search for distortions in the neutrino energy spectrum, the recoil electron spectrum is examined. This spectrum is obtained by repeating the flux measurement for small slices in recoil electron energy. The measured recoil electron spectra is steeply falling, and is therefore normalized to the expectations from the reference SSM. The normalized recoil electron spectrum is shown in Figure 10. A fit to an undistorted spectrum (flat) gives a x 2 /d.o.f. = 19.1/18. Figure 10 also presents the energy correlated systematic errors that arise from the uncertainties in the energy scale, the energy resolution and the reference 8 B spectral shape. These are the errors that could cause a systematic shift in the measured recoil electron spectral shape, and these errors are considered in the definition19 of the x 2 4-3
0.7
Solar Neutrino Oscillation Analysis
As the measured results show no significant deviation from the expectations of the reference SSM, no solar model independent ev-
j^%^|^ 10
12 14 20 Energy(MeV)
Figure 10. T h e measured recoil electron spectrum measured at Super-Kamiokande, normalized to the expectations from the reference SSM. Also shown are the energy correlated systematic errors (dotted band) that arise from the uncertainties in the energy scale, the energy resolution and the reference 8 B spectrum.
idence of neutrino oscillations is observed. Therefore, the measured results are used to generate exclusion regions in the neutrino oscillation parameter space where strong deviations are predicted. The measured flux, in combination with the measured recoil electron spectral shape and zenith angle dependence are also used to find allowed regions in this space, areas where the predicted spectral shape, zenith angle dependence and flux are consistent with the observed values. For this analysis, the data were divided into seven zenith angle bins (one day bin, 6 bins in cos# z at night). These zenith angle bins are further divided into eight recoil electron energy bins, to create a "zenith angle energy spectrum" 20 . For each set of neutrino oscillation parameters (sin2 29 and Am 2 ), the expected number of solar neutrinos and the corresponding zenith angle energy spectrum are calculated using a numerical calculation of neutrino survival probabilities, taking into account matter effects as the neutrino propagates from the center of the Sun to the SuperKamiokande detector here on the Earth. The measured and expected zenith angle energy
452
J. A. Goodman
Recent Results from Super-Kamiokande
spectrum are compared using a x2 analysis at each set of neutrino oscillation parameters. This analysis was performed under a two component neutrino oscillation hypothesis, once for active neutrinos (ue —• f^,,-) and once for sterile neutrinos (ve —* ^sterile) • The results of this analysis are shown in Figure 11 for the active case and in Figure 12 for the sterile case. The large, red shaded regions in these figures represent the 95% confidence level (C.L.) exclusion regions based on a flux-independent analysis of the zenith angle energy spectrum. Additionally, the measured zenith angle energy spectrum measured, the measured flux, and the theoretical uncertainty of 8 B neutrino flux are used to generate allowed regions at the 95% C.L. These are regions that are consistent with the observed zenith angle energy spectrum and flux measured at Super-Kamiokande and are shown in the figures as the thin, blue shaded regions near maximal mixing. Finally, the flux measured at Super-Kamiokande is combined with the total fluxes measured in the gallium and chlorine experiments to generate flux-only allowed regions at the 95% C.L. These are regions that predict the correct oscillated fluxes in all three types of solar neutrino experiments under the reference SSM inputs and are shown as green shaded regions in the figures. The flux-only allowed regions must be consistent with the flux-independent zenith angle energy spectrum measured by SuperKamiokande to be considered as a valid solution. The overlap of a particular combined flux-only allowed region (green) with the exclusion regions from the zenith angle energy spectrum analysis (red) would be strong evidence against that allowed region. Because of this, all but a portion of one flux-only allowed region (green) for active neutrinos are excluded at the > 90% C.L. The current SuperKamiokande data favor the so called "Large Mixing Angle" solution (Am 2 cz 1 0 - 4 ~ 10" 5 , sin 2 20 > 0.5) for active neutrinos. The
453
lower portion of this flux-only allowed region is excluded from a lack of a strong day-night flux asymmetry, but the upper half is consistent with the zenith angle energy s p e c t r u m and flux measured at Super-Kamiokande. All flux-only allowed regions for sterile neutrinos are disfavored at the 95% C.L.
sin*(2' ) Figure 11. Neutrino oscillation excluded/allowed regions for the case of active neutrinos. Red s h a d e d regions represent t h e 95% C.L. exclusion r e g i o n s based on a flux-independent zenith angle energy s p e c t r u m analysis. The blue shaded regions represent t h e 9 5 % C.L. allowed regions for an analysis of t h e z e n i t h angle energy spectrum with a flux constraint f o r SuperKamiokande data based on the reference S S M . T h e green shaded regions are t h e allowed r e g i o n s at t h e 95% C.L. based on a flux-only analysis b a s e d o n the results from Super-Kamiokande, in c o m b i n a t i o n with the results from the gallium and c h l o r i n e experiments.
In summary, the Super-Kamiokande detector has precisely measured the flux, recoil electron spectrum, and time variation in the flux of 8 B solar neutrinos over a 1 2 5 8 day period. These data show no strong modelindependent evidence of neutrino oscillation and are therefore used to generate exclusion
J. A. Goodman
Recent Results from Super-Kamiokande this oscillation is to active neutrino flavors (^V> vr)> then Super-Kamiokande should observe these events via their neutral current interactions. Using the SNO result we can predict the Super-Kamiokande electron scattering result ( $ e s ) : Since
~H
$ e s = ($(i/ e ) + (*(i/„, T ))/6.5))*ssAf putting in the SNO values gives: <£es (predicted) = (35% +
(65%/6.5))$SSM
* „ (predicted) = 45% ±
3%$SSM
Super-Kamiokande observes: $es(observed) = 45.1±0.5(stat.)±lJ%$ssM
sinz(2$)
This excellent agreement implies that if one assumes the SSM flux then this would be the first observational confirmation that solar electron neutrinos are oscillating t o other active flavors.
Figure 12. Neutrino oscillation excluded/allowed regions for the case of sterile neutrinos. Region definitions are the same as Figure 12.
5
regions in the neutrino oscillation parameter space. The current data favor the Large Mixing Angle solution for active neutrinos.
We gratefully acknowledge the cooperation of the Kamioka Mining and Smelting Company. This work was partly supported by the Japanese Ministry of Education, Science and Culture, the U.S. Department of Energy and the U.S. National Science Foundation.
4-4
Comparison with the SNO Results
Super-Kamiokande is sensitive to charged current (CC) interactions from electron type solar neutrinos. It is also sensitive to neutral current (NC) interactions from all neutrino flavors, but at a reduced level. The average value of the NC cross section is ~ l / 6 . 5 the CC cross section over the energy interval to which we are sensitive. Recently the SNO collaboration 21 published their measurement of charged current electron neutrinos. They report a value for the flux of
Acknowledgements
References 1. M. Honda et al, Phys. Rev. D52, 4985 (1995); V. Agrawal et al., Phys. Rev. D53, 1313 (1996). 2. Y. Fukuda et al., Phys. Lett. B433, 9 (1998); Phys. Lett. B436, 33 (1998); Phys. Rev. Lett. 8 1 , 1562 (1998); Phys. Rev. Lett. 82, 2644 (1999); Phys. Lett. B467, 185 (1999). 3. M. Ambrosio et al., Phys. Lett. B434, 451 (1998); F. Ronga, for the MACRO Collaboration, in these Proceedings. 4. W.W.M.AUison et al., Phys. Lett. B391, 491 (1997); G. Pearce, for the Soudan 2 Collaboration, in these Proceedings.
- 22 „ - l
This corresponds to ~ 35% ± 3% of the SSM. A simple way to look at this is to say that their result implies that 65% of the electron neutrinos have oscillated to other flavors. If 454
Recent Results from Super-Kamiokande
J. A. Goodman 5. M. Apollonio et al., Phys. Lett. B466, 415 (1999). 6. E. Akhmedov et al, Phys. Lett. B300, 128 (1993); P. Lipari a n d M. Lusignoli, Phys. Rev. D58, 73005 (1998); Q. Y. Liu and A. Yu. Smirnov, Nucl. Phys. B524, 505 (1998); Q. Y. Liu et al, Phys. Lett. B440, 319 (1998). 7. B.T. Cleveland et al., Astrophys. J. 496, 505 (1998). 8. J.N. Abdurashitov et al., Phys. Rev. C 60, 055801 (1999). 9. P. Anselmann et al., Phys. Lett. B 327, 377 (1994). 10. Y. Fukuda et al., Phys. Rev. Lett. 77, 1683 (1996). 11. J.N. Bahcall et al., astro-ph/0010346. 12. J.N. Bahcall et al., Phys. Lett. B 433, 1 (1998). 13. Y. Fukuda et al., Phys. Rev. Lett 8 1 , 1562 (1998). 14. M. Nakahata et al., Nucl. Instr. Meth. A 421, 113 (1999). 15. E. Blaufuss et a l , Nucl. Instr. Meth. A 458, 636 (2001). 16. S. Fukuda et a l , hep-ex/0103032 (2001). 17. GEANT Detector Description and Simulation Tool, Cern Programming Library W5013 (1994). 18. C.E. Ortiz et al., Phys. Rev. Lett. 85, 2909 (2000). 19. Y. Fukuda et al., Phys. Rev. Lett. 82, 2430 (1999). 20. S. Fukuda et a l , hep-ex/0103033 (2001). 21. SNO Collaboration, Phys.Rev.Lett. 87 (2001) 071301
455
RECENT RESULTS F R O M K2K C. K. JUNG Dept. of Physics and Astronomy, The State University of New York at Stony Brook, Stony Brook, New York 11794-3800, USA E-mail: [email protected], sunysb. edu for the K2K Collaboration K2K is a long baseline neutrino oscillation experiment using a neutrino beam produced at the KEK 12 GeV PS, a near detector complex at KEK and a far detector (Super-Kamiokande) in Kamioka, Japan. The experiment was constructed and is being operated by an international consortium of institutions from Japan, Korea, and t h e US. The experiment started taking d a t a in 1999 and has successfully taken data for about two years. K2K is the first long baseline neutrino oscillation experiment with a baseline of order hundreds of km and is the first accelerator based neutrino oscillation experiment that is sensitive to the Super-Kamiokande allowed region obtained from the atmospheric neutrino oscillation analysis. A total of 44 events have been observed in the far detector during the period of June 1999 to April 2001 corresponding to 3.85 x 10 19 protons on target. The observation is consistent with the neutrino oscillation expectations based on the oscillation parameters derived from t h e atmospheric neutrinos, and the probability that this is a statistical fluctuation of non-oscillation expectation of 63.9Jig g is less than 3%.
1
Introduction
In the late 1980's, the 1MB and Kamiokande experiments, which were both large underground water Cherenkov detectors, observed smaller atmospheric neutrino flux ratio v^/u,, than the predicted value. 1 ' 2 This was the beginning of so-called "Atmospheric Neutrino Anomaly". These results were not, however, supported by other underground experiments using iron calorimeters: Frejus 3 , NUSEX 4 and Soudan, albeit with limited statistical powers. In Table 1, the measurements of fM/^e flux ratio of these experiments are summarized along with recent measurements in a form of R = (p./e)DATA/(v/e)MC, where p. and e are the number of muon-like (/U-like) and electron-like (e-like) events observed in the detector for both data and Monte Carlo simulation. By making this double ratio one can largely cancel experimental and theoretical uncertainties, especially the uncertainty in the absolute flux. One would expect R to be unity, if the Monte Carlo simulation of an experiment accurately models the data. As can be seen from the table, the measured R
values of 1MB and Kamiokande are substantially lower than unity which could be interpreted as "disappearance" of v^ flux, possibly via neutrino oscillations. By interpreting the anomaly as neutrino oscillation phenomena the Kamiokande experiment obtained allowed region for v^ —* ue and v^ —• Vy_ oscillation scenarios, where x = s or r, utilizing zenith angle dependence of R.b While the interpretation of the anomaly as an indication of neutrino oscillation was exciting and plausible, there were many who believed that the anomaly was due to some unknown systematic errors particular to the water Cherenkov experiments, especially in particle identification. This criticism was answered by a joint (IMB-Kamiokande) beam test experiment at KEK 6 which found that the particle identification methods used by both experiments are valid well within the quoted systematic uncertainties. Furthermore, in 1997 the improved Soudan-2 experiment reported 7 that they measure the ratio closer to the water Cherenkov detector measurements than their previously reported value.
456
C. K. Jung
Recent Results from K2K
Table 1. Summary of the measurements of R. T h e first errors are the statistical uncertainties and the second errors are the systematic uncertainties. The exposure is measured in kiloton-years. The measurements of Kamiokande and Super-Kamiokande are presented separately for "sub-GeV" (EViS < 1330 MeV) and "multiGeV" (Evis > 1330 MeV) samples, where Evi3 is defined to be t h e energy of an electron that would produce the observed amount of Cherenkov light.
Experiment Kamiokande (sub-GeV) Kamiokande (multi-GeV) 1MB Soudan-2 Frejus NUSEX Super-Kamiokande (sub-GeV) Super-Kamiokande (multi-GeV)
R 0.62 0.57 0.54 0.67 1.00
± ± ± ± ±
0.06 ± 0.06 0.08 ± 0.07 0.05 ± 0.07 0.15±°$ 0.15 ± 0.08
n QQ+ 0 - 35 u . » y _ 0 25
0.64 ± 0.02 ± 0.05 0.62 ± O.03 ± 0.09
This unsettling situation, however, lasted until 1998 when the Super-Kamiokande (SuperK) experiment reported its analysis of atmospheric neutrinos with its unprecedented high statistics and high quality data resulting in a strong evidence for neutrino oscillations in the atmospheric neutrinos. 8 Figure 1 shows the updated zenith angle distributions of e-like and mu-like data sets of the SuperKamiokande atmospheric neutrino candidate sample. Zenith angle dependent deficit in the mu-like event distributions compared to the non-oscillation expectations is strikingly manifested in the data. On the other hand, the data are fit extremely well with a neutrino oscillation scenario as shown in the figure.9 This finding is a major discovery with far reaching impact in elementary particle physics, cosmology and astrophysics. The phenomenon of neutrino oscillation which requires neutrinos to have non-zero mass alters our view of the world of elementary particles. Namely, the Standard Model (SM), which is the currently prevailing theory of the elementary particles and assumes neutrinos to have zero mass, must be modified. The finding will also make the theories of the Grand Unification more viable and attractive, and thus opening a possibility of long-sought proton decays 10 ' 11 . To confirm the above finding, initially
457
Exposure 8.3 8.2 7.7 1.5 2.0 0.7 90.9 90.9
to investigate' the "atmospheric neutrino anomaly", several long baseline neutrino oscillation experiments using accelerator generated neutrino beams have been proposed. Among them, K2K (KEK to Kamioka) experiment 12 is the first one to be online. The K2K experiment consists of a neutrino beam line, a near detector complex inside the KEK laboratory near the east coast of Japan, and a far detector (the SuperKamiokande detector) 250 km away from KEK. The experiment started taking data in March 1999 and has successfully operated since then. The K2K physics program has three major components: (1) Neutrino oscillations, (2) Neutrino cross-section measurements, and (3) Study of neutrino background to proton decay searches. The neutrino oscillations program has in turn two topics: search for ve appearance signature from possible v^ —> ve oscillations, and analysis of v^ disappearance signatures for v^ —> Vy_ oscillation scenarios. The latter is the main topic of discussion of this paper. The v^ disappearance signatures can come from a comparison of the observed number of z/M events to the non-oscillation expectations at the far detector and/or from a distortion in the observed energy spectrum. If observed, the energy spectrum distortion will
Recent Results from K2K
C. K. Jung r/i
I—i—n—i—r~i—n—i—i—i i i—i—i—i—n—r
Sub-GeV e-like P<400MeV
I
1 1 1 1 I
1 1 l
1 I
1 1 1 1 I
1 1
1—r
Sub-GeV u-like P<400MeV
>
^200 o
++
S-
j¥^=¥.
s T£\
0
1
~y
• •
•
I
1 1 1 I
• r~i
'
1 , . . .
• 1 1 I
1 1 1 1 I
1 1 1
r
1
I
I
I
1 I
I
L.
Sub-GeV it-like P>40dMeV
|Sub-GeV e-like P>400MeV
¥^-
2200O
3
Vl
0
1 . . . .
1 . . . .
Mult!-^eV e-like
1
1 . . . .
• I • • • •
1 . . . .1
Muhl-lde^V-llkeiW
S200 o u =
0
_i_l
-0.5
1 1 1 i_
0 A 0.5
1-1
COS0
-0.5
0 n 0.5 cosG
Figure 1. The Super-Kamiokande atmospheric neutrino data: Zenith angle distributions oi fi-like and e-like events for sub-GeV and multi-GeV data sets. Upward-going particles have cos© < 0 and downward-going particles have cos 0 > 0. Sub-GeV data are shown separately for p < 400 M e V / c and p > 400 MeV/c. T h e solid black line shows the Monte Carlo expectation for no oscillations normalized to the data live-time. T h e grey line is the best-fit expectation for v^ <-> vT oscillations with the overall flux normalization fitted as a free parameter.
become the first direct evidence for energy dependent neutrino oscillation, which has not been firmly established so far by other exper-
iments. In Figure 2, the proposed sensitivity (circa 1998) of the K2K experiment for Vy, to vx oscillation using disappearance signatures
458
C. K. Jung
Recent Results from K2K
is shown along with other relevant contours. K2K can measure various neutrino crosssections on water, especially a relative crosssection of the neutral current n° production to the single muon production. This information is useful not only for theoretical modeling but also for a discrimination of fM —• vT oscillation scenario from v^ —> ^ s oscillation scenario in the Super-Kamiokande atmospheric neutrino analysis. A detailed description of this analysis can be found elsewhere. 13 The K2K data also provides valuable information on the proton decay search background studies. Proton decay searches performed at large water Cherenkov detectors rely on the MC simulation of the atmospheric neutrinos to estimate backgrounds to the proton decay candidates. Since the K2K neutrino beam has an energy spectrum similar to the atmospheric neutrinos, K2K can validate these atmospheric neutrino MC simulations using the lkton water Cherenkov detector data in the near detector complex, which collects high statistics neutrino events. Details of this study can be found elsewhere.11 The K2K collaboration is an international consortium of institutions from Japan, Korea, and the United States. There are about 100 collaborating members from 18 institutions.
2
K2K Neutrino Beam Line and Detectors
The K2K experiment is composed of the KEK 12 GeV Proton Synchrotron (PS), a neutrino beamline, near detector complex inside KEK and the Super-Kamiokande detector 250 km away as a far detector. The neutrino beamline is composed of an aluminum target, a pair of horn shaped magnets (so-called neutrino horns), a decay pipe, and a beam dump. The Al target (3 cm diameter, 60 cm long) is embedded inside the first horn as part of the structure and the decay pipe is 200 m long. There are several
459
V„->VV
90%C.L
10
> as
10
10
Super-Kamiokande 535 days (33i0 kiioton-years) sob-GeV + mujti-GoV, f.C.+ P C.
10
0.2
0.4
0.6
0.8
sin 2 (29) Figure 2. The K2K sensitivity for neutrino oscillation Vp to vx at 90% C.L. Also shown are the sensitivity of MINOS experiment at Fermilab along with other relevant contours. The MINOS experiment is expected to start data in 2005.
beam monitors for various purposes. A ring imaging gas Cherenkov monitor is located after the second horn to measure pion momentum and angular divergence along with a pair of rotating ionization chambers (segmented pads) which measures azimuthal symmetry and the radial profile of the secondary beam flux. Both of these monitors are remotely movable and they are in the beamline only when they are taking data. An additional ionization strip chamber is located at the entrance of the decay pipe to measure proton beam profile. At the end of the decay pipe, behind the beam dump, an ionization strip chamber and silicon pad detectors are located to measure the muon flux and profile. Since
C.K. Jung
Recent Results from K2K
the neutrino beam is generated from the pion decays in the decay pipe, the K2K neutrino source is not a point source. Thus, predicting the number of neutrino events at the far detector based on the measurements at the near detector requires a good understanding of the beam characteristics. This is the reason why many independent beam monitoring systems are employed. The near detector complex which is located 300 m from the target consists of a one kiloton water Cherenkov detector (a miniature Super-Kamiokande detector), and a finegrained detector system (a system of a scintillating fiber tracker with water targets, a scintillator veto counter, a lead glass calorimeter and a muon ranger detector). Many components of the near detector system are simply straightforward modifications of equipment available from previous projects at KEK. The layout of the experiment is shown in Figure 3. The 1 kton water Cherenkov detector (lkton) is a 10 m in height and 11 m in diameter cylindrical tank viewed by ~800 20" Hamamatsu PMTs. The scintillating fiber tracker 14 (Scifi) also uses water as target material for the neutrino beam, consisting of twenty 2.4mx2.4mx6cm segments totalling seven tons of water. The individual target segments are interleaved with scintillatingfiber tracking layers to provide good transverse vertex definition for neutrino interactions as well as track reconstruction capability. A Pb-glass wall, recycled from the TOPAZ detector at Tristan is located behind the Scifi detector to identify electrons. A Muon Ranger Detector 15 (MRD), consisting of iron plates and drift tubes that were used in VENUS, is used to measure muon energy up to 3.5 GeV with good accuracy. In order to minimize systematic uncertainties, K2K employs a common detector technology, water Cherenkov detectors at both near and far sites, and a common target, water for lkton and Scifi. The two
near-detector components are complementary. The lkton detector has its fiducial volume controlled less precisely due to coarser vertex resolution. Unlike the fine-grained detector, however, the 1-kton detector can reconstruct 7r0s, and due to the large size of the volume one can measure variations of the beam characteristics at larger distances from the beam axis. The Scifi detector has a smaller target volume, but its fiducial volume is much more precisely defined due to accurate vertex measurements. The 1-kton detector measures ve contamination and the fraction of weak neutral current (NC) events v^N —> v^N'+w0, using the same methods as in the Super-Kamiokande detector. The fine-grained detector also measures ve contamination. Quasi-Elastic (QE) reactions {y^N —* /i~N') in the fine-grained detector serve to measure beam uniformity and the neutrino energy spectrum. The far detector for K2K is the SuperKamiokande detector, which has been taking data since April, 1996. Super-Kamiokande is a water Cherenkov detector consisting of 50 kton of ultra-pure water in a 41 m (height) x 39 m (diameter) tank. Its inner volume is viewed by 11,146 20" diameter Hamamatsu photo-multiplier tubes, providing photo-cathode coverage equivalent to 40% of the detector surface area. The inner detector is surrounded by a 2 m thick outer layer of water viewed by 1885 8" diameter Hamamatsu tubes. The outer detector is used to veto particles coming from outside as well as providing shielding against backgrounds emanating from the surrounding rock. A nominal 22.5 kton fiducial volume is used to study neutrino interactions. K2K does not interfere with normal operation of Super-Kamiokande. GPS clocks at both sites are used to define appropriate DAQ time windows for KEK beam pulses arriving at Kamioka. 16 The K2K neutrino beam is generated from the fast extracted 12 GeV KEK PS pro-
460
Recent Results from K2K
C. K. Jung
SCIFI/Water target Muon Chamber
GlaSs
<[1 kt Wa"or Crete nkov i Delec-tir
f|N»utnno Beam f i *» w
11
kandt
Figure 3. K2K near detector complex.
ton beam which has l^ts spill duration with a 9 micro-bunch structure. The design intensity of the protons on target (pot) is 6 x 1012 protons per spill for every 2.2 s. The goal of the experiment is to collect data corresponding to 1020 pot for 5 years of running (4 months of run/year). The energy spectrum of the neutrino beam peaks at around 1 GeV and has a mean energy of 1.3 GeV. The beam is about 97% pure in v^ and contains 2% PM and 1% ue. The K2K near detector construction was completed in January 1999 and the neutrino beam line commissioning was started on January 27, 1999. On March 5, 1999, the K2K physics data taking run was started. By April 2001, the experiment has accumulated data corresponding to 4.6 x 10 19 pot (delivered). The average beam intensity on target was about ^5.5 x 10 12 proton per pulse and the nominal horn current was 250 kA. All detector subsystems worked very well without major problems.
461
3
Long Baseline Neutrino Oscillation Analysis
As discussed in the previous section, K2K can look for two neutrino oscillation signatures: ve appearance and v^ disappearance. However, the statistical power of the experiment so far on the ve appearance analysis is too weak to produce a competitive result. Thus, in this paper, we will focus our discussions on the Vp disappearance oscillation analysis.
3.1
Beam Aiming, Stability and Monitoring
In order to obtain an unambiguous result on neutrino oscillations, we must first ensure that the neutrino beam is aimed at the far detector correctly. The K2K neutrino beam line was constructed with an accuracy of ~0.1 mrad and was surveyed using GPS with an accuracy of ~0.01 mrad. Meanwhile the beam aiming accuracy required for the K2K experiment is about 3 ~ 5 mrad which
Recent Results from K2K
C. K. Jung
Expected v flux at 250 km, on axis and far from it On axis flux expected Flux expected lkm (4mrt away flux expected 2km (Smr) a*c
0
0.5
1
IS
2
23
3
3S
4
4£ 5 P v (GeV/c)
Figure 4. Comparison of on- a n d off-axis b e a m energy spectra at t h e far location detector
is deduced by comparing off-axis neutrino energy spectra with the on-axis energy spectrum at the far detector location. (See Figure 4.) This rather mild requirement stems from the wide-band nature of the K2K neutrino beam. The actual spill-by-spill and day-by-day variations of the neutrino beam direction with respect to the nominal beam direction are also monitored using various beam monitor systems to ensure that there are no wild fluctuations or drifts in the beam direction which would result in a reduction of the neutrino flux at the far detector. The accuracy of the monitored beam direction is found to be less than 1 mrad from all beam monitoring systems for the entire period of K2K operation. The spill-by-spill monitoring of the high energy muon direction which is highly correlated to the neutrino beam direction is done with the muon monitors (Strip ionization chambers and silicon strip pad detectors) located behind the beam dump, 200 m from the target. The measured spill-by-spill direction of high energy muons (E^ > 5 GeV) with respect to the nominal beam direction for the entire data taking time period is found to be well within the ± 1 mrad boundaries.
The day-to-day monitoring of the actual neutrino beam direction and flux stability are done with both lkton {E^ > 0.2 GeV) and MRD (£;„ > 1 GeV). Figure 5 shows the day-to-day measurements of the neutrino beam direction and neutrino flux for the entire data taking time period using MRD. The dashed lines in the left side plots indicate ± 1 mrad off-axis directions from the nominal beam directions denoted by solid lines. As can be seen almost all data points are within the ± 1 mrad boundaries. The top data point on the right side plot are CC interaction rate in unit of interactions/5 x 10 12 ppp (protons-per-pulse) and the bottom data points are the interaction rate measured in a smaller fiducial volume of the detector. Since ultimately the event rate observed at the far detector is normalized by the near detector measurement, changes in the measured event rate in the near detector are not a great concern. However, to minimize any unexpected systematic effects in the experiment, it is desirable to have the rate stable. Indeed the data points guided by the a solid line representing the average event rate show a good stability of the event rate, i.e. the neutrino flux, for the entire operation of the experiment. The apparently low event rate for the June 99 period was partly caused by a sightly different proton target area configuration, especially the neutrino horn current (200 kA instead of 250 kA).
3.2
Near Detector Data Analysis for Neutrino Oscillations
In order to observe the "fM disappearance" signatures at the far detector, we need to measure the neutrino beam flux (in practice event rates) and un-oscillated energy spectrum at the near detector. Since the lkton detector shares similar efficiencies and systematics with the SuperKamiokande detector, it provides the pri-
462
C. K. Jung
Recent Results from K2K
Neutrino beam aiming
CC-Interaction
3. ^ ~ 80 P u
60
|
40
S » 3
2C
3
;
> 0
r
20 0 -- 4
r
•wW* jtL
j^-
§0.06
m&fe
j-
'-60
-100
11/99
01/00
02/01
03/01
05/01 06/00
01/01
02/01 03/01
04/01
L T+4 +^
RPB
V !S tk
t*tt
03W1
02/01 03/01
1*i
$*,u
m
M
5*7?
D
OlApi
OOMa;
£0.04 06/99
Calendar time (1 data point / 5 days)
-100
Rate
O.0.07
100
TP i ^ r t f
IVV'HW
.W*J
ry7
*•*•**«
far# ' [A /ipfi
A"!
^ V ,yV/1 /Vw •,»«]
99Jun 99Nov OOJan OOFeb OOMarOOMay OOJun O l j s n OlFeb O l M a r O l A p r
06/99
11/99
01/00
02/01
OS/01 06AW
01/01
Calendar time (1 data point / 2 days)
04/V]
Calendar time (1 data point / 5 days)
Figure 5. Beam stability as measured by t h e muon range detector. Left is t h e centroid of reconstructed vertices each five days through all t h e d a t a analyzed. Right is the event rate measured each two days. T h e rate in June of 1999 is expected t o b e lower than later runs since the target area configuration was slightly different.
mary information on the neutrino event rate and neutrino spectrum measured at the near detector which are extrapolated to the far detector to generate the non-oscillation expectation event rate and energy spectrum. In general, for Charged Current (CC) QE neutrino interaction events (f M +n —> /j,~ +p), one can reconstruct neutrino energy spectrum by measuring muon momentum p^ and production angle cosO^ with respect to the beam direction and using simple two body kinematics: Ev
rriNE^-ml/2 mN - E^+p^ cos <9M'
where raw and mM are respective neutron and muon masses, and E^ is the muon energy. The recoil protons in the CCQE events are typically not seen in the lkton detector because most of time the protons are created with momenta below Cherenkov threshold, and thus we call these events "single muon" events. In Figure 3.2, reconstructed energy spectrum using lkton single muon event sample is shown along with MC predictions. Also
463
shown are the distributions of muon momentum and angle. As can be seen, the data and MC predictions agree fairly well. The left figures also show the background contributions to the distribution from non-QE events. It turns out that the "single muon" event sample contains substantial number of non-CCQE events that have one or more pions generated, but are not seen by the detector. In order to obtain true neutrino energy spectrum one must, then, properly subtract out these background events contributions. This can be accomplished using the data from the the scifi detector where the recoil protons can be observed and its direction can be compared with the predicted direction deduced from the muon observables. This provides us with means to estimate non-QE background contributions to the CCQE event candidate event sample. Figure 3.2 shows a cos# distribution of two track events in the Scifi detector, where cos# is defined as the angle between the reconstructed second (shorter) track and the
C. K. Jung
Recent Results from K2K
H momentum (ms 25t FC,1-ring,n)
Migrated Reconstructed Ev with acceptance correction 300O 2500 ~*~
Entris* Mean
16507 711.9
RMS
259.5
era
1500
^^^^^^^=
1000 500
: ^^^^^^fc^ 250
500
750
1000
1250
1500
1750
2000
£-
PfGeV)
0)
>
LI direction (ms 25t FC,1 ring.u) 4500 4000 3500 3000
pi-
-
16507 44.37
Entries Mean RMS
^_LS-—' -+-——
24.80
•
2500 2000
1500
ID
1=3
—
= 1
1500 1000 500
1000 80
100
120
140
160
1500
2000
180
2500
Ev (MeV)
Figure 6. Neutrino energy spectrum reconstructed from d a t a from only the l k t detector. Left are the distributions of p^ and 6n for single ring /j-like events that go into calculating a spectrum. Points are data; t h e boxes are the total prediction for simulation; the filled histogram is the non-quasi-elastic contribution to the simulated sample. Only statistical error is shown. Right is a neutrino energy spectrum after background subtraction and corrections for efficiencies and acceptance. The inclusion of information from the other detectors will raise the efficiency for high energy events.
expected proton direction calculated using two-body kinematics and assuming that the event is a CCQE event. Since at the large angle region (small cos 9 region) we expect that most of the events are coming from non-QE background events, it is possible to estimate the background contributions by fitting the MC distribution to the data at this region.
the flux are then checked by the information from the gas-Cherenkov pion monitor at the target station. From the light distribution
400 ,.QJ
350
.><"
300
3.3
.^\eu^
2S0
Near Detector to Far Detector Extrapolation
200
""""""r^s?
150
Once the event rates and energy spectra are measured at the near detector sub-systems, they are extrapolated to the far detector location to predict the event rate and energy spectrum at the Super-Kamiokande detector. The extrapolation is done by using a detailed beam MC simulation which predicts neutrino flux at the near detector and at the far detector simultaneous. Since the neutrino beam is produced by a line source rather than a single point source, it is necessary to simulate the details of the entire neutrino beam line. The energy dependent far to near ratios of
100 50
MC (all) MC (non-quasi-elastic only)
x Data
*-i -(J.« I d T
u!i ' 'oU'' old'' 'a'M ' CosA0
Figure 7. Background measurement with 2-track events from the fine-grained detector. The abscissa is the cosine of the angle between the second track reconstructed direction and the direction calculated for a proton based on the assumption that the other track was a muon and 2-body kinematics as is shown in the inset cartoon. Fitting this distribution to a pair of amplitudes times the histogram shapes from simulations can give a solid measurement of the nonquasi-elastic background in this data set.
464
Recent Results from K2K
C. K. Jung Table 2. Comparison of lkton and Super-Kamiokande Analysis Characteristics.
Fiducial Vol. E. Threshold ace. x eff. # of Events
lkton 25 tons >1000 p.e. 0.74 63,000
Table 3. Comparison of t h e Predicted Number of Events at Super-Kamiokande. The errors shown are systematic errors and the statistical errors are negligible.
SuperK 22,500 tons >300 p.e. 0.81 63.9
lkton Scifi MRD
data of the pion monitor, it is possible to extract information on the pion kinematics, i.e. pion momentum distribution and angular divergence from the proton beam direction. Since neutrinos are produced by simple two body decays of pions, the far to near neutrino flux ratio can be precisely determined by this information. For the neutrino energy regions (Ev > 1 GeV) where the pion monitor is sensitive, we find that the two predictions from the beam MC and the pion monitor match perfectly within the statistical uncertainties. A Super-Kamiokande data analysis procedure similar to that of atmospheric neutrino's is applied to the extrapolated neutrino flux to yield expected number of events and energy spectra for various sets of oscillation parameters. Table 2 shows some characteristics of lkton and Super-Kamiokande analyses as well as the observed number of events at lkton and corresponding nonoscillation predicted number of events at Super-Kamiokande. Similar procedures are applied to extrapolate measured event rates by Scifi and MRD. The resulting non-oscillation predicted number of events at Super-Kamiokande by these systems are shown in Table 3. While the three systems have quite different detector characteristics and consequently different systematic uncertainties, the predicted number of events by these systems are well within the estimated systematic uncertainties of each other. One can notice that the systematic uncertainty in the lkton predicted number of events is the smallest. This is
465
Target water water iron
SuperK Predicted 63.9±£g 64.6±|;| 69.8±j?;|
because the lkton detector and the SuperKamiokande detector have very similar detector characteristics, and consequently when far to near ratio is taken, some of the systematic uncertainties such as cross-section uncertainties cancel out. The dominant systematic uncertainties in the lkton predicted number come from the uncertainty in the far to near ratio (~7%) and the uncertainty in the lkton fiducial volume determination (~4%). 3.4
Observation of K2K Neutrino Events in Super-Kamiokande
The K2K experiment observed its first neutrino event in the Super-Kamiokande detector in June 19, 1999, which was the first long baseline (of order hundreds of km) neutrino beam event in history. The event vertex was within the 22.5 kton fiducial volume of Super-Kamiokande and it was within ±1.5/[iS time window of the expected time. The expected number of background events is estimated to be about 10~ 4 . By the end of April 2001 the experiment has observed a total of 44 fully contained (FC) events in the 22.5 kton fiducial volume (FV) of the Super-Kamiokande detector compared to the non-oscillation predicted number of 6 3 . 9 I 6 6 events. In addition, the experiment has observed 26 fully contained events in the out of fiducial volume (vertices in between the PMT planes and the surfaces 2m inward from the PMT planes). The event selection criteria for these events are very similar to that used for the atmospheric neutrino analysis. 8 The most
Recent Results from K2K
C. K. Jung powerful discrimination of the K2K events against the random atmospheric neutrino events is a timing cut utilizing the narrow time structure (l.l/us) of the K2K beam in every 2.2 s. Figure 8 shows a summary view of the data reduction for fully contained event analysis. Here AT is defined as the time difference between the SuperK event time and the beam spill time at the KEK PS subtracted by the time of flight of the neutrino from KEK to SuperK (0.83 ms). The absolute times are measured by GPS receivers located in each site and local timing clocks. The one shot timing accuracy of GPS is known to be better than 200 ns and thus the timing cut for final event sample is chosen to be —0.2 < AT < 1.3/us. As can be seen there is only one event that is out of time with the K2K beam. The event is consistent with the expected atmospheric neutrino background rate in the one millisecond time window, and the estimated background to the fully contained fiducial volume events in the signal window is of order 10~ 3 .
Fully Contained Event
Analysis
Event is not a decay electron
In order to perform an oscillation analysis using rate measurements, we must check whether the observed event rate at the far detector remains constant (within the statistical fluctuation) throughout the data taking time period. Irregular event rates beyond the statistically expected fluctuation would indicate some kind of unknown systematic problems either with the neutrino beam or with the far detector. This is checked by plotting the number of observed events (cumulative) as a function of integrated protons on target. Figure 9 shows such plots for all K2K events observed in SuperK including the Outer Detector (OD) events. The OD events are defined as the events that have appreciable OD PMT hits. While the OD events are less well understood because of the poorer resolution of the OD, they are useful in studying the time dependent event rate. The estimated background for the OD event sample is of order one. The diagonal solid lines in the figure show the average event rate slopes. While there are a few noticeable gaps in the distributions, a statistical analysis shows that they are perfectly consistent with fluctuations expected from the Poison statistics. More rigorously, a KS test of the distribution results in KS probability of 20.5%. And if only the FCFV events are used, it results in a KS probability of 32.3%.
3.5 .500 44 Contained events in fiducial 22.5ktous (70 total contained events)
10
Background ^ Oifeu; • ]0 '')
-5
A(T) |is '
Figure 8. (A simplified view of) Data reduction for fully-contained K2K events in SuperK.
Results of Oscillation Analysis
For the oscillation analysis, we use only the FCFV events since they are the most well understood events from the SuperK atmospheric neutrino analysis. In the future, however, we may be able to use the fully contained out of fiducial volume events as well for the oscillation analysis. Table 4 summarizes a break down of the FCFV events into sub-classes along with non-oscillation expectation and oscillation expectations for various Am 2 values with sin 2 26 = 1.0. While the statistical power is not over-
466
Recent Results from K2K
C. K. Jung
Table 4. Observed fully-contained K2K-SK events with vertices in the fiducial volume (FCFV). T h e indentation of entries in the first two columns indicates that some categories are subsets of one above. T h e expectation for "e-like" events changes with oscillation since some are mis-identified v^ events (not ve beam contamination). T h e oscillation expectations assume sin 2 2d = 1.0, a full mixing.
Category FC-22.5kton (all) 1-ring /tx-like e-like multi-ring
Expected for Am 2 in 10~ 3 eV 2 = 3. 5. 7. 41.5 ± 4 . 7 27.4 ± 3 . 1 23.1 ± 2.6 22.3 ± 3 . 4 14.1 ± 2 . 2 13.1 ± 2.0 19.3±3.2 11.6 ± 1 . 9 10.7 ± 1 . 8 2.9 ± 1 . 2 2.5 ± 1 . 0 2.4 ± 1 . 0 19.3 ± 3 . 4 13.3 ± 2 . 3 10.0 ± 1 . 8
Expected (non-osc.) 63.9± b 6 -j 38.4 ± 5 . 5 34.9 ± 5 . 5 3.5 ± 1 . 4 25.5 ± 4 . 3
Observed 44 26 24 2 18
FC+OD
teractions. A reconstructed neutrino energy spectrum is shown in Figure 10 along with expectations with and without oscillations. The normalizations for the expectations are done using the near detector measurements. Again while the statistics are rather poor to draw any convincing conclusions, the data appear to be consistent with the oscillated expectation given by the SuperK atmospheric neutrino best fit parameters. In order to perform an oscillation analysis using this information one must understand the associated systematic uncertainties, especially the binby-bin correlated errors. At the time of this presentation, the systematic uncertainties of the energy spectrum have not been yet well evaluated. Thus, the figure contains only the statistical error bars of the data points. Rigorous evaluation of the systematic uncertainties is being carried out at this time by the K2K collaboration along with more extensive oscillation analyses that fully utilize the information from the shape of the energy spectrum as well as the overall rate. We expect that these analyses will be completed by the spring of 2002.
aoo *
80
, / 60 A
.4
rt
40
20-
^ 0
~i S
4
/
KS, rob bility = !0.5%
i....
10
15
20
25
POT vs events
30
35
40
CT *10 1 8 POT
Figure 9. The cumulative observed events at the far detector as function of integrated protons on target. The figure includes all K2K events observed in SuperK including OD events (104 total). The solid vertical lines show the boundaries of different run periods, all but the first of which use the same beamline configuration.
whelming, the data generally indicate a deficit in the observed number of events compared to the non-oscillation expectations. A simple statistical calculation results in a probability that the observed data is a statistical fluctuation of the non-oscillation hypothesis to be less than 3%. A neutrino energy spectrum can be also constructed from these events using the method described earlier. For this purpose only the 24 1-ring /i-like events are used as they are most likely produced by CCQE in-
4
K2K Near Detector Upgrade
The original design of the K2K experiment was optimized to explore the neutrino oscillation parameter space favored by the Kamiokande experiment, i.e., Am2 10"
467
Recent Results from K2K
C. K. Jung E F.C. 22.5kt 1-ring
[i-like
5
10
0
' 0
'
' 1
'
' ' ' 2 ^(GeV)!
'
' 4
'
' 5
Figure 10. The neutrino energy spectrum reconstructed from data (points) and the predicted spectrum from MC simulations. The thick solid black line represents the expectations assuming no oscillation, and the thin dot-dashed line represents that assuming oscillations with the atmospheric neutrino best-fit parameters.
eV2 region. The well-established allowed region by the Super-Kamiokande experiment is, however, much lower in Am 2 centered around 3 x 10~ 3 eV 2 . For a fixed neutrino flight distance of 250 km, this Am 2 value corresponds to 600 MeV neutrino energy for the first peak of the oscillation probability, which in turn suggests that the distortion in the energy spectrum (ultimate signature of neutrino oscillation) may occur below 1 GeV energy region in the K2K measured energy spectrum. Thus, in order for us to reduce the systematic uncertainties in the below 1 GeV region and thereby making K2K more sensitive to the low Am 2 neutrino oscillations, it is necessary to upgrade the K2K detector. As the first step towards an upgrade, we plan to remove the lead glass calorimeter wall in the fall of 2001. This will allow us to accumulate more Scifi events below 1 GeV. In the summer of 2003, we plan to install a finely segmented liquid scintillator tracker, which will have high efficiency for short (~5 cm) tracks and will be able to detect a proton down to 400 MeV/c. It will also have a particle identification capability for p/n separation by dE/dx.
Conclusion
K2K is the first accelerator based long baseline neutrino oscillation experiment to test the allowed parameter region obtained by the Super-Kamiokande atmospheric neutrino analysis. The experiment has taken data successfully for about two years since 1999 (June 1999 - April 2001) corresponding to 4.58 (3.85) x 10 19 protons on target delivered (Super-Kamiokande live). Additional data obtained during the period May - July 2001 are being analyzed and the goal of the experiment is to accumulate data corresponding to 10 20 protons on target. With 44 observed events at the far detector compared to the non-oscillation expectation of 63.9l 6 ' 6 , the non-oscillation hypothesis is disfavored more than 2a (97%) level. Both the observed event rates and the energy spectrum are consistent with those expected from neutrino oscillations with the best fit parameters from the Super-Kamiokande atmospheric neutrino analysis. K2K provides a proof of principle for long baseline experimentation with baseline over 100 km. Its success leads us to rich neutrino physics programs of future long baseline experiments: MINOS, CNGS, JHF to Super-Kamiokande, super-beams and ultimately neutrino factories. Neutrino oscillation is the only evidence we have for physics beyond the Standard Model. During the past two decades, we have devoted much of our effort to understand the nature of the quark mixing and associated CP violation. With neutrino oscillation, we have now started a new era of particle physics in which we must dedicate our effort to understand the lepton mixing and possible CP violation in the lepton sector. With some good cooperation of nature, we may ultimately be able to explain the nature of matter-antimatter asymmetry in the Universe through lepto-genesis.
468
Recent Results from K2K
C. K. Jung Acknowledgments The K2K experiment is co-hosted by KEK, and ICRR, University of Tokyo, and it is supported by the Ministry of Education, Culture, Sports, Science and Technology, Government of Japan, the U.S. Department of Energy, the Korea Research Foundation, and the Korea Science and Engineering Foundation. The collaboration gratefully acknowledges the cooperation of the Kamioka Mining and Smelting Company. The author wishes to acknowledge the support of The Research Foundation of the State University of New York at Stony Brook, and the U.S. Department of Energy under contract^ DEFG0292ER40697.
10. 11.
12.
13.
References 1. K.S. Hirata et al, Phys. Lett. B 205, 416 (1988); K.S. Hirata et al, Phys. Lett. B 280, 146 (1992). 2. D. Casper, et al, Phys. Rev. Lett. 66, 2561 (1991); R. Becker-Szendy et al., Phys. Rev. D 46, 3720 (1992). 3. K. Daum, et al., Z. Phys. C 66, 417 (1995). 4. M. Aglietta, et al, Europhys. Lett. 8, 611 (1989). 5. Y. Fukuda et al, Phys. Lett. B 335, 237 (1994). 6. S. Kasuga et al, Phys. Lett. B 374, 238 (1996). 7. W.W.M. Allison et. al, Phys. Lett. B 391, 491 (1997). 8. Y.Fukuda et al, Phys. Lett. B 433, 9 (1998); Y.Fukuda et al, Phys. Lett. B 436, 33 (1998); Y.Fukuda et al, Phys. Rev. Lett. 81, 1562 (1998). 9. For most recent update of the SuperKamioknde results, see J. Goodman's presentation and paper presented in this conference: XX International Symposium on Lepton and Photon Interac-
469
14.
15. 16.
tions at High Energies, Rome, Italy, July 2001, http://www.lp01.infn.it. K.S. Babu, J. C. Pati and F. Wilczek, [arXiv:hep-ph/9812538]. " Physics Potential and Feasibility of UNO," UNO "white paper" edited by D. Casper, C. K. Jung, C. McGrew and C. Yanagisawa, June 2001. Preprint #:SBHEP01-3. The full report is available on the web at: http://nngroup.physics.sunysb.edu/uno/ S. H. Ahn et al [K2K Collaboration], Phys. Lett. B 511, 178 (2001) [arXiv:hep-ex/0103001]. For more details on this subject, see C. Mauger's presentation in NuIntOl workshop: The First International Workshop on Neutrino-Nucleus Interactions in the Few GeV Region, in Tsukuba, Japan, December 2001, http://neutrino.kek.jp/nuint01/. A. Suzuki et al. [K2K Collaboration], Nucl. Instrum. Meth. A 453, 165 (2000), [arXiv:hep-ex/0004024]. T. Ishii et al. [K2K Collaboration], [arXiv:hep-ex/0107041]. H. G. Berns and R. J. Wilkes, IEEE Nucl. Sci. 47, 340 (2000).
SOLAR N E U T R I N O RESULTS FROM T H E S U D B U R Y N E U T R I N O OBSERVATORY JOSHUA R. KLEIN, FOR THE SNO COLLABORATION Department of Physics and Astronomy, University of Pennsylvania, Philadelphia, PA 19104-6396 We describe here the measurement of the flux of neutrinos created by the decay of solar 8 B by the Sudbury Neutrino Observatory (SNO). The neutrinos were detected via t h e charged current (CC) reaction on deuterium and by the elastic scattering (ES) of electrons. T h e CC reaction is sensitive exclusively to ue's, while t h e ES reaction also has a small sensitivity to i/M's and uT's. T h e flux of ve's from 8 B decay measured by the CC reaction rate is e) = 1.75 ± 0.07 ( s t a t . ) l ° ' J j (sys.) ± 0.05 (theor.) x 10 6 c m ~ 2 s - 1 . Assuming no flavor transformation, the flux inferred from the ES reaction rate is ES(ux) = 2.39 ± 0.34 ( s t a t . ^ o J f (sys.) x 10 6 c m _ 2 s _ 1 . Comparison of <j>cc(ve) t o the Super-Kamiokande Collaboration's precision value of (j>ES{vx) yields a 3.3CT difference, assuming the systematic uncertainties are normally distributed, providing evidence that there is a non-electron flavor active neutrino component in t h e solar flux. The total flux of active 8 B neutrinos is thus determined to be 5.44±0.99 x 10 6 c m _ 2 s _ 1 , in close agreement with the predictions of solar models.
1
Introduction
Over thirty years of solar neutrino experiments i'2'3'4*5'6 have demonstrated that the flux of neutrinos from all sources within the Sun is significantly smaller than predicted by models of the Sun's energy generating mechanisms 7 ' 8 . The deficit is not only universally observed but has an energy dependence which makes it hard to attribute to astrophysical sources: the data are consistent with a negligible flux of neutrinos from solar 7 Be 9 ' 10 , though neutrinos from 8 B (a product of solar 7 Be reactions) are observed. A natural explanation for the observations is that neutrinos born as j/ e 's change flavor on their way to the Earth, thus producing an apparent deficit in experiments detecting primarily j/ e 's. Neutrino oscillations—either in vacuum or matter—provide a mechanism both for the flavor change and the observed energy variations. While these deficit measurements argue strongly for neutrino flavor change through oscillation, a far more compelling demonstration would not resort to model predictions at all but look for non-^e flavors coming from the Sun. The Sudbury Neutrino Observatory
(SNO) was designed to do just that: provide direct evidence of solar neutrino flavor change through the inclusive appearance of non-electron neutrino flavors from the Sun. We present here the first solar neutrino results from SNO, which have also been described in an earlier publication n .
2
SNO D e t e c t o r
SNO is an imaging water Cerenkov detector, which uses heavy water (D2O) as both the interaction and detection medium 12 . Figure 1 shows a diagram of the detector. SNO is located ~ 2 km (6020 k.w.e.) underground in INCO Ltd.'s Creighton Mine, deep enough that the rate of cosmic ray muons passing through the entire active volume is just 3/hour. The 1000 tons of heavy water is contained in a 12 m diameter transparent acrylic vessel, and is surrounded by 2 ktons of light water shielding. The Cerenkov light produced by neutrinos and radioactive backgrounds is detected by an array of 9500 8 inch photomultiplier tubes (PMTs), supported by a stainless steel geodesic sphere. Each PMT is surrounded by a light concen-
470
Joshua R. Klein
Solar Neutrino Results from the Sudbury Neutrino Observatory cesses: vx + e~ —> vx + e~ ve+d^p + p + e~ ux+d^p + n + ux
The first reaction, the elastic scattering (ES) of electrons, has been used to detect solar neutrinos in other water Cerenkov experiments. It has the great advantage that the recoil electron direction is strongly correlated with the direction of the incident neutrino (and hence the direction of the Sun). In addition, this reaction has sensitivity to all neutrino flavors. For ve's, the elastic scattering reaction has both charged and neutral current components, making the cross section for ^ e 's 6.5 times larger than that for u^s or
Figure 1. Diagram of SNO Detector.
trator, which increases the photocathode coverage to nearly ~ 55%. The front-end discriminator thresholds are set to fire on 1/4 of a photoelectron of charge. Outside the PMT support sphere is another 7 ktons of light water shielding. The detector is also equipped with a flexible calibration system, capable of placing sources almost everywhere in either the x — z or y — z plane. The sources that can be deployed include a diffuse multi-wavelength laser for measurements of optical parameters and PMT timing, a 16 N source which provides a triggered sample of 6.13 MeV 7's, and a 8 Li source delivering 0's with an endpoint near 14 MeV. In addition, high energy energy (19.8 MeV) 7's are provided by a 3 H(p,7) 4 He ('pT') source 13 and neutrons by a Cf source. Some of the sources can also be deployed on vertical axes within the light water volume between the acrylic vessel and PMT support sphere. 3
(ES) (CC) (NC)
SNO R e a c t i o n s
SNO can provide direct evidence of solar neutrino flavor change through comparisons of the interaction rates of three different pro471
The deuterium in the heavy water makes the second process possible: an exclusively charged current (CC) reaction which (at solar energies) occurs only for i/e's. In addition to providing exclusive sensitivity to ve 's, this reaction has the advantage that the recoil electron energy is strongly correlated with the incident neutrino energy, and thus can provide a good measurement of the 8 B energy spectrum. The CC reaction also has an angular correlation with the Sun which falls as (1 — 0.340cos(# Q )) 14 , and has a much larger cross section (~ 10 times larger) than the ES reaction. The third reaction—also unique to heavy water—is a purely neutral current process. This has the obvious advantage that it is equally sensitive to all neutrino flavors, and thus provides a direct model-independent measurement of the total flux of neutrinos from the Sun. For both the ES and CC reactions, the recoil electrons are directly detected through their production of Cerenkov light. For the NC reaction, the neutrons are not seen directly, but are detected in a multi-step process. When a neutrino liberates a neutron from a deuteron, the neutron wanders within
Joshua R. Klein
Solar Neutrino Results from the Sudbury Neutrino Observatory
the D2O and is eventually captured by another deuteron, releasing a 6.25 MeV 7 ray. The 7 Compton scatters an electron and it is this secondary particle which is detected. Although the data we present here were acquired with the acrylic vessel filled with pure D2O, the detector is now running with NaCl added to the heavy water. The addition of the salt provides chlorine which has a larger capture cross section (and hence a higher detection efficiency) for the neutrons. The capture on chlorine also yields multiple 7's instead of the single 7 from the pure D2O phase, which aids in the identification of neutron events. Eventually, discrete He 3 proportional counters will be added which will count neutrons exclusively. To determine whether neutrinos which start out as ue's in the solar core convert to another flavor before detection on Earth, we have two choices: comparison of the CC reaction rate to the NC reaction rate, or comparison of the CC rate to the ES rate. The former has the advantage of high sensitivity— we compare the total flux to the ve flux and therefore expect to see a large difference if the true neutrino flux agrees with standard solar models (which predict a total flux two to three times larger than previous measurements). In addition, uncertainties in the cross sections for the two processes will largely cancel. The second comparison has the advantage that both the CC and ES recoil electrons provide neutrino spectral information. The spectral information can ultimately be used to show that any excess in the ES reaction over the CC reaction is not caused by a difference in the energy thresholds used to analyze the two reactions. The CC-ES comparison also has the advantage that the strong angular correlation with the Sun demonstrates that any excess seen is not due to some unexpected non-solar background. Lastly, the CC-ES comparison can be made with fairly high precision despite the small ES reaction
cross section, because the Super-Kamiokande collaboration has already made a precision, high statistics measurement of the ES rate 5 . For the results presented here, only the CCES comparison will be described. 4
Data Analysis
The goal of the data analysis is the determination of the relative sizes of the three signals (CC, ES, and neutrons) and ultimately the comparison of the rates. In the pure D2O detector configuration—the configuration with which these data were taken—we cannot separate the signals on an event-by-event basis. Instead, we 'extract' the signals statistically by using the fact that they are distributed distinctly in the following three derived quantities: the kinetic energy of the recoil electron or capture 7 ray (T), the reconstructed radial position of the interaction (R3), and the reconstructed direction of the event relative to the Sun
(COS#Q).
Figure 2 shows these distributions for each of the signals. The top row of Figure 2 plots the different energy distributions for the three signals. We see in the figure that the strong correlation between the electron energy and the incident neutrino energy for the CC interaction produces a spectrum which resembles the initial 8 B neutrino spectrum, while the recoil spectrum for the ES reaction is much softer. The NC reaction is—within the resolution smearing of the detector— essentially a 5-function, because the 7 produced by the neutron capture on deuterium always has the same 6.25 MeV. The bottom row of Figure 2 shows the reconstructed direction distribution of the events. In the middle of that row we see the familiar peaking for the ES reaction, pointing toward the Sun. The ~ 1 — 1/3 cos 6Q distribution of the CC reaction is also clear in the left hand side of the bottom row. Not surprisingly, the NC reaction shows no correlation with the solar direction—the 7 ray from
472
Joshua R. Klein
Solar Neutrino Results from the Sudbury Neutrino Observatory
CC
ES
this is that the capture cross section for neutrons on deuterium is very small—the neutron wanders around long enough inside the D2O that it may leak outside and be captured by hydrogen in either the acrylic vessel or the light water. Such hydrogen captures produce a much lower energy 7 ray (~ 2.2 MeV), below the analysis threshold. Therefore the acceptance for events which are produced near the edge of the volume is reduced, because the probability of leakage there is correspondingly higher than for events produced near the center.
NC
R' (AV rodii)
-1
0
1
u
- , cosOsun
Figure 2. The energy (top), radial (middle), and directional (bottom) distributions used to build pdfs to fit the SNO signal data.
the captured neutron knows nothing about the incident neutrino. The distributions of reconstructed event positions is shown in the middle row of Figure 2. These distributions are plotted as a function of R3, with R3 = 1 occurring at the radius of the acrylic vessel (the edge of heavy water volume). We see here that the CC reactions—which occur only on deuterons—produce events distributed uniformly within the heavy water, while the ES reaction (which occurs on a n y electron) produces events distributed uniformly well beyond the heavy water volume. The small leakage of events just outside the heavy water volume for the CC reaction is due to the resolution tail of the reconstruction algorithm. The NC reaction, however, does not have a uniform distribution inside the heavy water like the CC reaction, b u t instead monotonically decreases from the central region to the edge of the acrylic vessel. The reason for 473
One last point needs to be made regarding the distributions labelled ' N C in Figure 2: they represent equally well the detector response to all neutrons, not just those produced by neutral current interactions, as long as the neutrons are produced uniformly in the detector. For example, neutrons produced through photodisintegration by 7 rays emitted by U or Th chain daughters inside the D2O will have the same distributions of energy, radial position, and solar direction as those produced by solar neutrinos. In the analysis described here, no separation is done between these neutrons and those from the NC reaction. To determine the size of the three signals, then, we use these nine distributions to create probability density functions (pdfs) and perform a generalized maximum likelihood fit to the same distributions in the data. There are, however, two principal prerequisites that must be satisfied before we can even begin this 'signal extraction' process. First, we need to process the data so that it is in a form we can use to do the fits. For example, we need to reconstruct the events to give us positions and directions that can be used to produce distributions, and we need to calibrate the energy of each event. Even more importantly, we must be sure that the only signals present in the data are the three for which we are doing the fits—we have implicitly assumed that the backgrounds are neg-
Joshua R. Klein
Solar Neutrino Results from the Sudbury Neutrino Observatory
ligible. To accomplish this, we apply cuts to the data to eliminate backgrounds and we must ultimately demonstrate that any residual backgrounds are negligible. The second signal extraction prerequisite is that the distributions used in the fitting process must be good representations of the detector's true response. In other words, we must build a model of the detector's response to the three signals which can be used to generate the pdfs used in the fit. The model needs to reproduce the response at all places in the detector, for all neutrino directions, for all neutrino energies, and for all times. The last requirement is necessary because the detector's response changes over time due to things like failed PMT's or electronics channels. The analysis we describe here therefore has three major components before the final fitting stage: the processing of data to remove backgrounds, the building of a model to fit the data, and the demonstration that the residual backgrounds are small enough to use the signal-only model in the fits.
4-1
Data Processing
We recorded the data set used in this analysis between November 2, 1999, and January 15, 2000. Roughly 40% of the time during this period was taken up either by calibration source runs or downtime caused by mine power outages. Of the remaining good data, we selected runs to analyze based on criteria which were 'blind' to the data itself—whether enough channels were live, whether calibration sources were present, whether water assays were being run, etc. After passing this run selection stage, no further run removal was allowed from the data set, and the final total livetime amounted to 241 days. Approximately 30% of the data was put aside to serve as a blind test of statistical bias. As no significant differences were found between this sample and the other 70%, all subsequent
discussion here refers to the full data set. During this time, the primary trigger threshold was set to fire on a ~ 100 ns coincidence of 18 PMT's each exceeding a channel threshold of ~ 1/4 photoelectron. This trigger threshold corresponds to an energy of roughly 2 MeV. The trigger reaches 100% efficiency at 23 hit PMTs. In addition to the 100 ns coincidence, we ran simultaneously with other triggers, such as a pre-scaled (1:1000) lower threshold (11 PMTs) trigger, a trigger on PMT pulse height sums, and a pulsed (random) trigger. The raw data set is far from the clean distributions shown in Figure 2. In particular, the data is contaminated by instrumental backgrounds arising primarily from PMT light emission ('flasher PMTs'), static discharges in the neck of the acrylic vessel, or electronic pickup. Although these instrumental backgrounds are very distinct from the neutrino signal, they occur at far higher rates: flasher events, for example, occur roughly once each minute compared to the five to ten neutrino events we expect each day. We therefore developed a suite of low level cuts designed to remove instrumental backgrounds while losing a minimum of neutrino events. These cuts were applied before any reconstruction of the data was done, and used only primitive information such as the PMT charge distributions, the raw and calibrated time distributions, hits in veto tubes, and event-to-event time correlations. Figure 3 shows the effects of the progressive application of these instrumental background cuts to the raw data set, illustrating the multiple orders of magnitude reduction in the overall number of events. In any case in which such a large reduction is obtained, the obvious question is what is the consequent reduction in good events— how much acceptance loss have we incurred by applying cuts which remove more than three orders of magnitude of the instrumental backgrounds? To measure this loss, we
474
Joshua R. Klein
Solar Neutrino Results from the Sudbury Neutrino Observatory
Progression of Instrumentol Cuts pr
Figure 3. Effects of progressive application of instrumental background cuts.
i _
:^+
;++
Figure 4. Acceptance loss from low level instrumental background cuts, measured with calibration sources.
used triggered calibration sources which provided both samples of Cerenkov events and isotropic light events, and applied the same cuts to the source data as to the neutrino data. Figure 4 shows the acceptance loss as a function of the number of hit PMTs, for 16 N, 8 Li, and laser data. Although there is evidence of a bias at high energies (high number of hit PMTs), the overall scale of the loss is very small, ~ 0.5%. For events passing this first stage, we reconstructed the vertex position and direction of the particle using the calibrated times and positions of the hit PMTs. The reconstruction algorithm begins with maximum likelihood fits using only PMT times, seeded by positions fixed to a grid throughout the detector volume. The best fit vertex from this 475
o.t*ta:--o'± crc2£t~;:
Figure 5. Resolution in x for 8 Li source data.
grid-seeded procedure is then used as a seed for a second level of fitting which uses both the PMT times and their angular distribution to simultaneously fit both the position and the direction of the event. The fitting process includes cuts on angular figures-ofmerit which test both the quality of the fit and the hypothesis that the event is a single Cerenkov electron. Figure 5 shows the vertex resolution for electrons produced by the 8 Li source, which provides a localized (~ 5 cm) set of electrons with a broad spectrum of energies. At 16 N energies, the vertex resolution is 16 cm and the angular resolution is 26.7°. For each event surviving the reconstruction stage, we assigned an energy based on the hypothesis that the event was a single Cerenkov electron. While the number of hit PMTs by itself is directly related to the event energy, it must be corrected for the number of live channels online when the event was recorded, any change in the overall detector gain with time, and the optical effects of the intervening media between the Cerenkov production point (the event position) and the photon detection points (the hit PMTs). The optical corrections were calculated using insitu measurements of the detector's optical properties (attenuation lengths, PMT angular responses, etc.) and account for both the vertex position and the event direction. To minimize uncertainties associated with late
Joshua R. Klein
Solar Neutrino Results from the Sudbury Neutrino Observatory
hits (reflections, scattering, noise), the optical corrections use only prompt (in-time) photons by requiring the fitted time residuals of the PMT hits to be within a narrow window around At — 0 ns. Figure 6 shows the calibrated response of the detector to 16 N data. In Figure 6a we see the energy distributions for data taken with the source at the center and at R = 465 cm. The only corrections made here are for the number of live channels online, and therefore the shift in the mean of the two distributions is due to the different optical response at the two positions. Figure 6b demonstrates how the two distributions coincide once the optical corrections are applied.
High level p a r a m e t e r s performance
l}.4 -
- Instrumental background
0.2 i-
<• '*N i-alibrsiiimi source
• Candidate neutrino event:
0
i).l
0.2
0.3
(1.4
0i 0.6 11.7 0-S 11.9 ] Fraction "Chits in prompt lime window-
Figure 7. Cerenkov box cuts defined by average angle between hit P M T s and ratio of prompt t o total light.
With the event positions and directions fit and the energy calibrated, we passed the data through a final stage of cuts aimed at ensuring that the remaining events were consistent with Cerenkov light. We defined Cerenkov light with two orthogonal cuts: one which tests the narrowness of the timing distribution, and one which tests the angular distribution of PMT hits. The former is done by cutting on the ratio of prompt (in-time) hits to the total number of hits in the event, and the latter by using the average angular distance between hit PMTs in the event. Figure 7 shows three data sets distributed in these two variables: data tagged by the low level instrumental background cuts (triangles), Cerenkov data from the 16 N source (open circles), and neutrino data (closed circles). The box used to define Cerenkov light is also shown, illustrating how both the source data and the neutrino data lie inside, while the instrumental backgrounds stay well outside. We required all data in the final signal sample to lie within the box shown in Figure 7.
Table 1. Data processing steps.
Analysis step Total event triggers Neutrino data triggers ^hit > 3 0
Inst, bkgrnd cuts Muon followers Cerenkov box cuts Fiducial volume cut Threshold cut Total events
No. of events 355 320 964 143 756 178 6 372 899 1 842 491 1 809 979 923 717 17 884 1 169 1 169
ing Cerenkov and laser calibration sources. The systematic uncertainties on these losses are associated with the calibration sources themselves (source reflectivity and shadowing), the low level electronic calibrations (for example, ADC pedestals), and changes in the detector over time. Using the calibration source data, we find that the total loss for all cuts is 1.4+°;£%. For the final signal sample, we further restricted events to be within a fiducial volume of 550 cm and have a kinetic energy T > 6.75 MeV. The fiducial volume restriction minimizes backgrounds associated with The reconstruction quality cuts and the the acrylic vessel, light water, and PMTs, 'Cerenkov box' cuts contribute to the overall while the energy threshold reduces radioacacceptance loss, and we measured the scale tive backgrounds and neutron events in the of this loss along with the losses by the low final signal sample. Table 1 summarizes the level instrumental background cuts by us- data processing, from the total number of 476
Solar Neutrino Results from the Sudbury Neutrino Observatory
Joshua R. Klein
Souit»«M85em
X 1-
{—
_ ~
_ •r —
>-i * eBr-r
&-X.
JT _
T . . . .T —
t....r..
«al««l
(b)
(a) Figure 6. Comparison of
16
N energy distribution before (a) and after (b) optical corrections.
events in the raw data set to the final sample of 1169 events. The table also includes the effects of cuts which remove the spallation products of cosmic ray muons. These cuts remove all events within 20 s of a parent muon. At this stage of the analysis, we have a data set which has been reconstructed and calibrated, and has had the majority of backgrounds removed. However, before fitting the resulting distributions, we still need to build a model of the detector's response, and demonstrate that the background removal has been successful enough that we can perform the fits using a signal-only model.
4-2
Model Building
The model of detector response we have used in this analysis takes as its inputs the physics of electron and 7 interactions in matter, the geometry of the detector, the behavior of the front-end data acquisition electronics and trigger, and—most importantly—the same measured optical parameters used in the energy calibration described in the previous section. The model is a Monte Carlo simulation, which combines these inputs as well as the state of the detector as a function of time (the number of channels online, the overall energy scale determined by the 16 N source, etc.) to produce a predicted response function for all event positions, directions, and energies. This response function is what we 477
Figure 8. Deployment positions for D2O scan and an H2O scan.
16
N source for a
use to create the pdfs which are ultimately used to fit the data. To ensure that the model is correct, we tested it against Cerenkov data representative of the neutrinos we are trying to detect, for as many positions, directions, and energies as possible. The degree to which the model does not correctly reproduce the various measurements sets the scale for the systematic uncertainties on the predicted response function. Figure 8 depicts the positions inside the D2O and H2O for some of the 16 N scans. For the dependence on energy of the energy response, we compared 16 N data to pT data (6.13 MeV 7's to 19.8 MeV 7's). For the dependence on position and direction we compared different source positions and different sources—the Cf neutron source, for example,
Joshua R. Klein
Solar Neutrino Results from the Sudbury Neutrino Observatory
a. 0.05
-1 fo.04 - 1 ~0.03 - 1
— • — H^hrate
,6
N
Low rate Ct 11850 -•
— H£h rate Ct 12511
— — Ptdata
0.02 -
j
0.01
0 —\ -0.01
T
f
•
T
!
*
i
V
M
1 \
|k|
-1- 1
\
I A 1
-
-0.02
-
\
-
\
^).03 -0.04 -0.05
, j . .
i
. . j , , i , , j , , i , ,
. .
Nov 99 Jan 00 Mar 00 May 00 Jul 00 Aug 00 Oct 00 Dec 00 Mar 01 May 01 Figure 9. Differences between predicted energy response and measured response for different sources as a function of time. In addition t o the 1 6 N source, are a 19.8 MeV 7 from the p T source and a 6.25 MeV 7 from the neutrons produced by the Cf source.
provides a very different event position distribution than the 16 N source does, and samples many more positions within the volume. We also tested the dependence on data rate by varying the rates for some of the calibration sources. Figure 9 summarizes the differences between the predicted energy response and the measured response for various sources as a function of time. The overall systematic uncertainty on the energy scale determined through these measurements is 1.4%. We performed the same kinds of tests for the prediction of the reconstruction accuracy, and Figure 10 compares the vertex resolution measured with the 16 N source at various positions to the model prediction. There is a small systematic shift (~ 1 cm) between the two, but otherwise the model tracks the data well. The model prediction of the angular resolution agrees very well with the measurements made with the 16 N source, and has a negligible contribution to the overall systematic uncertainty on the measured fluxes.
Vertex Resolution with N16 Source •. DIO i - « u rat (X=T=0>: Nhir>4S
Figure 10. Differences between predicted and measured vertex resolution for the 1 6 N source as a function of position in the detector.
4-3
Backgrounds
We are not quite ready yet to fit for the signal amplitudes, because we must still demonstrate that the data is free enough from backgrounds to justify the use of a model which contains only signal distributions. There are three classes of background: the instrumental backgrounds discussed in Section 4.1, high energy 7 rays from the phototube support sphere and cavity walls, and low energy backgrounds from radioactivity both within and
478
Joshua R. Klein
Solar Neutrino Results from the Sudbury Neutrino Observatory
without the D 2 0 volume. To measure the residual instrumental backgrounds, we used the 'Cerenkov box' cuts described earlier. The low level cuts aimed at reducing the instrumental backgrounds and the higher level Cerenkov box cuts are independent and orthogonal—and so the fraction of the instrumental backgrounds which lie inside the Cerenkov box is the same whether the instrumental background cuts identify the events or not. We therefore used the fraction of identified instrumental backgrounds which lie inside the box and multiplied it by the number of events in the 'clean' data sample which lie outside. From this, we found the fraction of the clean data sample inside the Cerenkov box which may be due to instrumental backgrounds missed by the low level cuts. As a fraction of the final CC data sample, this is < 0.2%, small enough to ignore in the fit for signals. The determination of the high energy backgrounds was similar in principle, but here we had at our disposal calibration sources which provide triggered samples of high energy 7 rays. The 16 N source is nearly ideal for this measurement, as it acts as a triggered 'point source' of events which— with the exception of energy spectrum—look exactly like the background we are trying to measure. To use this source to measure the backgrounds, we deployed it near where the backgrounds originate—out (and beyond) the detector's active volume. We then measured the ratio of the number of inward-going 7 events reconstructing just inside the source position (the 'monitoring' box) to the number of events reconstructing inside the 550 cm fiducial volume. With the number of events in the final data sample which reconstruct inside the same (but now spherically symmetric) monitoring box, we determined the number of background events which lie insde the 550 cm volume by multiplying by the source-measured ratio. We explored systematic uncertainties by varying
479
the monitoring box size, the deployment position, and by using Monte Carlo simulation to explore the variation of the leakage with energy. The final limit on this source of background measured in this way is < 0.8%. Low energy backgrounds originate from several sources: radioactivity in the heavy water, the acrylic, the light water, and the PMTs. Their typical energy is ~ 2 MeV, and our energy threshold of 6.75 MeV is high enough that the leakage can only come from the tail of the background energy spectrum. The small fiducial volume of 550 cm also greatly restricts the number of events from the PMTs, light water, and acrylic. To estimate the number of events from low energy backgrounds which leak above the signal energy threshold or inside the fiducial volume, we used a combination of radioassays, encapsulated U and Th calibration sources, and Monte Carlo simulation. Figure 11 shows that the radioactivity in the heavy water— as determined by radioassays—is well below the original target values. At these levels, simulation shows that the tail of the backgrounds above our energy threshold is negligible. In the light water, assays also show that the the backgrounds are near or below target levels, but because these levels are still relatively high, we deployed calibration sources to measure the fraction that reconstruct within the 550 cm fiducial volume. Of the external sources of background, by far the largest is the radioactivity in the PMTs themselves. With calibration sources placed near the PMT sphere, we measured an upper limit on the leakage of the PMT radioactivity of < 0.2% of the final CC rate.
5
Results and Implications
We now have satisfied all the pre-requisites for doing a signal extraction: we have a clean data set in which the backgrounds are low enough to justify a signal-only fit, and a model which correctly predicts the response
Solar Neutrino Results from the Sudbury Neutrino Observatory
Joshua R. Klein
Radioactivity in D.O from Water Assays SNO Preliminary Rrsufcl
Table 2. Systematic uncertainties on fluxes.
^Itn [U]
..^,»iS-"«*
' ' •
"T
J,
J.
.!*
»
Error source
?
~*fU(Th]
' * > '
•
-! ...
Energy scale Energy resolution En. non-linearity Vertex accuracy Vertex resolution Angular resolution High energy 7's Low energy bkgrnd Inst, bkgrnd Trigger efficiency Live time Cut acceptance Earth orbit ecc.
&M»'_ I T : * Kiap<[.
|
'iff
'
1 Oeai. 4,5J- iO'"jrVs
f •
T 199*5
3)00
2000.5 Titae (yen)
3001
3B1.5
Figure 11. Water assay measurements of radioactivity in D2O.
function of the detector as measured by calibration sources. For the neutrino spectrum input to the model, we use an undistorted 8 B shape 15 ' 18 . The maximum likelihood fit to the 1169 events in our sample gives us 975.4±39.7 CC events, 106.1±15.2 ES events, and 87.5±24.7 neutron events, where the uncertainties given are statistical only. Figure 12 shows the best fit to the distribution of event directions with respect to the Sun. The elastic scattering peak can clearly be seen, but with the available statistics, only a hint of the slope of the CC electrons. To convert the CC and ES event numbers into fluxes, we need to correct for the acceptance of the cuts, the energy threshold, and the fiducial volume restriction. We then need to normalize by the interaction cross sections and the number of deuterons and electrons inside the fiducial volume. For the CC cross section, we use the calculation of Butler et al16, and do not include any radiative corrections. The radiative corrections may serve to increase the cross section by up to a few percent 17 , and therefore decrease the measured value of the flux (and ultimately increase the significance of any difference between the CC and ES fluxes). We also include small corrections due to the isotopic abundances of 1 7 0 and 1 8 0 , upon which CC reactions can also
17
o, 18o
Exp. uncertainty Cross section Solar Model
CC error (percent) -5.2, +6.1 ±0.5 ±0.5 ±3.1 ±0.7 ±0.5 -0.8, ±0.0 -0.2, ±0.0 -0.2, ±0.0 0.0 ±0.1 -0.6, ±0.7 ±0.1 0.0 -6.2, ±7.0 3.0 -16 , ±20
ES error (per cent) -3.5 ,±5.4 ±0.3 ±0.4 ±3.3 ±0.4 ±2.2 -1.9, ±0.0 -0.2, ±0.0 -0.6, ±0.0 0.0 ±0.1 -0.6, ±0.7 ±0.1 0.0 -5.7, ±6.8 0.5 -16 , ±20
occur. Finally, we normalize by the overall livetime. Table 2 lists the systematic uncertainties on the flux measurements. The dominant uncertainties arise from our lack of knowledge of the true response function of the detector. As described above, we characterize the scale of the uncertainties on the model by comparing the model predictions to measurements made with calibration sources, for example the 1.4% on the energy scale. To derive the uncertainties on the fluxes shown in Table 2, we varied the model predictions over the range of the uncertainties and repeated the analysis. In some cases this resulted in a larger uncertainty on the flux measurement— the 1.4% uncertainty on the energy scale becomes a ~ 6% uncertainty on the flux derived from the CC rate, for example. In addition to the measurement of the systematic uncertainties, we have explored the systematic behavior of our results under many different analysis approaches: for ex-
480
Solar Neutrino Results from the Sudbury Neutrino Observatory
Joshua R. Klein
~i—i—i—i—i—i—r-
-r-n-r-
i ' ' ' i
(a)
jioo > 75 ^
J
50
25 0
J
1_1
l_l
I
l_l
I
L_l
l_l
l_J
I
I
I
1 -0.8 -0.6 -0.4 -0.2
'
I
I
0
I
I
I
1_J
I
I
I I I
0.2 0.4 0.6 0.8 a cos 0.0
Figure 12. Reconstructed direction distribution of electrons with respect to the Sun.
ample, comparing different suites of low level cuts, reconstruction algorithms, Cerenkov box cuts, and choices of fiducial volume. We have also compared the results we get using the total number of hit PMTs as the measure of energy scale (thus changing the sensitivity to the knowledge of the late light distribution) to the results from the energy calibration described above. Lastly, we have performed fits using an analytical (as opposed to Monte Carlo) model of the detector response. In all cases, the results from these alternative approaches agree with the fluxes presented here to well within the systematic uncertainties quantified in Table 2. Converting the fit numbers to fluxes and including the systematic uncertainties listed in Table 2, we find that the flux of neutrinos inferred from the ES reaction (assuming no flavor transformation) is ^SNol^)
2.39 ± 0.34(stat.) - a i 4 (sys.) x 10 6 cm- 2 s
and the flux of 8 B i/e's measured by the CC reaction is i>$go(ve) = 1-75 ±0.07 (stat.)±0;J 2 (sys.) ±0.05 (theor.) x 106 c m _ V
481
where the theoretical uncertainty comes from the uncertainty in the CC cross section 16 . The difference between these two numbers is 1.6(7, assuming that the systematic errors are distributed normally. The low significance of this result is driven mainly by the large statistical errors on the ES measurement. However, the Super-Kamiokande collaboration has measured the flux with the ES reaction to high precision 5 , and finds JES
§£(I/X)
= 2.32 ±0.03 (stat.)!™? ( s y s j l ) x!0b c m _ V
(2)
The difference between SNO's measurement using the CC reaction (sensitive only to Ve&) and Super-Kamiokande's measurement using the ES reaction (sensitive to i/^s and uT's as well as pe's in the ratio of 1./6.5) is 3.3o\ This difference is therefore evidence of an active, non-fe component to the solar 8 B neutrino flux. Figure 13 summarizes the situation for all published solar neutrino experiments, including SNO. The points are plotted as ratios of the measured fluxes to the Standard Solar Model predictions of Bahcall, Pinnsoneault and Basu (BPB01 7 ), for the energy threshold used in each experiment. Here we can see the 3.3a difference between the SNO and
Joshua R. Klein
Solar Neutrino Results from the Sudbury Neutrino Observatory
ENERGY (MeV)
Figure 13. Summary of solar neutrino rate measurements from various experiments relative to BPB01 standard solar model, including SNO and the derived 8 B flux from the SNO and Super-Kamiokande rates.
Super-Kamiokande measurements as well as the poor statistical accuracy of the SNO ES. Also plotted in Figure 13 is the total flux of all 8 B neutrinos using the SNO CC measurement and the Super-Kamiokande ES measurement: (vx) = 5.44 ± 0.99 x 106 c n r V 1 . We see in the figure that the agreement between the measurement and the model prediction is very good. Figure 13 also shows that the differences in the thresholds for the SNO and Super-Kamiokande measurements allows for the possibility that there is some spectral distortion which could be causing the difference. Such a spectral distortion could occur if, for example, the oscillation were into a sterile neutrino. To look for such an effect, we can first examine the spectrum of recoil electrons created by the CC interactions relative to the prediction for an undistorted 8 B spectrum. We derive such a spectrum by re-fitting the data energy bin-by-energy bin, without using the pdf for the CC energy spectrum. Figure 14 shows the ratio of the spectrum derived this way to the standard solar model prediction. The dominant systematic uncertainties are indicated by the horizontal lines on the plot. Figure 14 shows that there is no large distortion in the expected spectrum. We can also eliminate the possibility of
a spectral distortion leading to the difference in the SNO CC and Super-Kamiokande ES measurements by comparing the two measurements for the same neutrino energy. As described by Fogli et al 19 , this can be done by using different recoil energy thresholds for the SNO and Super-Kamiokande measurement. For these 'matched' thresholds (~ 8.5 MeV for the Super-Kamiokande measurement compared to SNO's 6.75 MeV) we still get a difference of 3.1
6
Future and Conclusions
SNO's current and future data sets will provide many more interesting measurements. We are now analyzing the pure D2O data in order to make a the measurement of the NC rate. The NC measurement should give us a confirmation of the CC-ES result, a higher precision measurement of the total 8 B flux, and a higher significance for the excess of non-z/e flavors (because we will be comparing a ue flux of 1.75 x 10 6 cm~ 2 s _ 1 to ~ 5 x 10 6 cm~ 2 s^ 1 rather than to ~ 2.3 x 10 6 cm~ 2 s - 1 ). We are also analyzing the data in day and night bins, to determine whether any asymmetry is present (which would indicate that matter oscillations are the cause of the flavor change as well as better restrict the allowed regions in the (tan 2 9,Am2) plane. In addition, we are working on an analysis which includes the hep neutrinos. Beyond the pure D2O data, we will also have the salt data set, which should provide us with an even better NC measurement, as well as new measurements of the other fluxes as well. Non-solar neutrino physics analyses are underway as well—looking at atmospheric neutrinos, anti-neutrinos (for which SNO has an exclusive coincidence tag), and
482
Solar Neutrino Results from the Sudbury Neutrino Observatory
Joshua R. Klein
^0.6 >-.
~^T
(c)
"3 o
PQ oo 0.4
im^EE^3:
o OH
Super-K ES flux
0.2
0
->
_L_L.
I
7
I
10 v„ . 11 1 2 / A . .,20 Kinetic energy (MeV)
Figure 14. Spectrum of electrons from ve interactions compared to prediction of the BPBOl standard solar model.
supernova searches. SNO's first results, in combination with the Super-Kamiokande collaboration's measurements, provide direct evidence that solar neutrinos undergo flavor change on their way from the Sun to the Earth. They also show that the Standard Solar Model prediction of the 8 B flux is correct within the uncertainties of both the prediction and the measurement. We expect many more interesting measurements to come out of SNO for a long time to come.
Acknowledgments This research was supported by the Natural Sciences and Engineering Research Council of Canada, Industry Canada, National Research Council of Canada, Northern Ontario Heritage Fund Corporation and the Province of Ontario, the United States Department of Energy, and in the United Kingdom by the Science and Engineering Research Council and the Particle Physics and Astronomy Research Council. Further support was provided by INCO, Ltd., Atomic Energy of Canada Limited (AECL), Agra-Monenco, Canatom, Canadian Microelectronics Corporation, AT&T Microelectronics, Northern
483
Telecom and British Nuclear Fuels, Ltd. The heavy water was loaned by AECL with the cooperation of Ontario Power Generation.
References 1. B.T. Cleveland et al, Astrophys. J. 496, 505 (1998). 2. K.S. Hirata et al, Phys. Rev. Lett. 65, 1297 (1990); K.S. Hirata et al, Phys. Rev. D 44, 2241 (1991), 45 2170E (1992); Y. Fukuda et al, Phys. Rev. Lett. 77, 1683 (1996). 3. J.N. Abdurashitov et al, Phys. Rev. C 60, 055801, (1999). 4. W. Hampel et al, Phys. Lett. B 447, 127 (1999). 5. S. Fukuda et al, Phys. Rev. Lett. 86, 5651 (2001). 6. M. Altmann et al, Phys. Lett. B 490, 16 (2000). 7. J.N. Bahcall, M. H. Pinsonneault, and S. Basu, astro-ph/0010346 v2. The reference 8 B neutrino flux is 5.05x 106 —9
—1
cm s . 8. A.S. Brun, S. Turck-Chieze, and J.P. Zahn, Astrophys. J. 525, 1032 (1999); S. Turck-Chieze et al, Ap. J. Lett., v. 555 July 1, 2001.
Joshua R. Klein
Solar Neutrino Results from the Sudbury Neutrino Observatory
9. N.Hata, S. Bludman, and P. Langacker, Phys. Rev. D 49, 3622 (1994) 10. K.M. Heeger and R.G.H. Robertson, Phys. Rev. Lett. 77, 3720 (1996) 11. Q.R. Ahmad et al., Phys. Rev. Lett. 87, 071301 (2001) 12. The SNO Collaboration, Nucl. Instr. and Meth. A449, 172 (2000). 13. A.W.P. Poon et al, Nucl. Instr. and Meth. A452, 115, (2000). 14. J.F. Beacom and P. Vogel, hepph/9903554, Phys. Rev. Lett. 83, 5222 (1999). 15. C.E. Ortiz et al, Phys. Rev. Lett. 85, 2909 (2000). 16. S. Nakamura, T. Sato, V. Gudkov, and K. Kubodera, Phys. Rev. C 63, 034617 (2001); M. Butler, J.-W. Chen, and X. Kong, Phys. Rev. C 6 3 , 035501 (2001); G. 't Hooft, Phys. Lett. 37B 195 (1971). The Butler et al. cross section with LjtA = 5.6 fm3 is used. 17. I. S. Towner, J. Beacom, and S. Parke, private communication; I. S. Towner, Phys. Rev. C 58 1288 (1998), J. Beacom and S. Parke, hep-ph/0106128; J.N. Bahcall, M. Kamionkowski, and A. Sirlin, Phys. Rev. D 51 6146 (1995). 18. Given the limit set for the hep flux by Ref. 5 , the effects of the hep contribution may increase this difference by a few percent. 19. G. L. Fogli, E. Lisi, A. Palazzo, and F.L. Villante Phys. Rev. D 63, 113016 (2001); F.L. Villante, G. Fiorentini and E. Lisi Phys. Rev. D 5 9 013006 (1999).
484
E X P E R I M E N T A L R E V I E W OF N E U T R I N O PHYSICS SHIGEKI AOKI for CHORUS and DONUT collaboration, Tsurukabuto, Nada, Kobe, 657-8501, JAPAN E-mail: [email protected] Dedicated to Prof. Masami Nakagawa who passed away in March 2001.
served 7r+ decays at rest. They confirmed constant range of fi+ indicating a two body 1.1 ve decay. This is also the "first observation" of the production of atmospheric neutrinos. At The first "detection" of the neutrino was the that time, it was not known whether they observation of a continuous spectrum of /?were same neutrinos as those from /3-decay 1 ray measured by Chadwick in 1914. Here, or not. "detection" means a discovery of experimenIn 1962, interactions of neutrinos from tal "indirect" evidence. the decays of TT produced by an accelerator In 1956 (42 years later!), Cowan were observed 7 and confirmed that v^ is difand Reines 2 observed interactions of antielectron-neutrinos from a nuclear power reac- ferent particle from ve8. Soon after, Maki, tor by the reaction V2 + p —• n + e + in a wa- Nakagawa and Sakata discussed the possibility of Vfj, - ve oscillation when two neutrinos ter target containing dissolved CdCh which was sitting between liquid scintillator detec- have different masses. tors. About the same time, Davis 3 was try1.3 vT ing to detect the (not anti-)electron-neutrino by the radiochemical method using the reac- After the r lepton was discovered in 1975 by tion ve + 37Cl —> 37Ar + e~. Although this Perl et al.9, it was thought that the vT must method is not suitable for the detection of VI exist as the third generation neutrino. Howfrom a nuclear reactor, this experiment mo- ever, the charged current interactions of v T tivated Pontecorvo 4 to discuss the possibility had not been observed until DONUT collabof v - V oscillation. oration's report 10 . 1
Introduction
In 1964, Davis started the Homestake Experiment to observe ue from the Sun by the same detection technique described above, which is good to detect (not-anti) ve. Four years later, Davis 5 reported a deficit of the solar neutrino. This was the beginning of the "Solar Neutrino Problem". 1.2 i/„ The muon neutrino was "detected" in a experiment to observe Yukawa meson (i.e. n) •decaying into muon in 1947. In a balloonborne emulsion experiment, Powell et al.6 ob-
485
2
Solar Neutrinos
For solar neutrino detection, there have been three types of experiments, shown in Table 1. Two of them are radiochemical experiments using chlorine or gallium. These radiochemical experiments can measure the total flux above its energy threshold. Timing information is limited by the period of chemical extraction. The other type of experiment is water Cerenkov experiment, which can measure timing and energy of each event. Kamiokande, Super-Kamiokande and SNO
Shigeki Aoki
Experimental Review of Neutrino Physics Table 1. Solar Neutrino Experiments
Type
Experiments
Process
Chlorine
Homestake 11
ve+
Gallium
SAGE 12 Gallex+GNO 1 3
ve+71Ga^
H20
Kamiokande 14 Super-K 15
vx + e" —> vx + e~
Ee > 6.5 - 7 MeV Ee > 5.5 MeV
SNO 16
vx + e~ —> ux + e~ ve + d —> p + p + e~ vx + d —> p + n + vx
Ee > 6.75 MeV Ee > 6.75 MeV Ev > 2.2 MeV
D20 (heavy water)
Gallium
37
Cl -»
37
Ar + e~
71
Ge + e~
Ev > 817 keV Ev > 235 keV
Total Hates: Standard Mode! vs. Experiment
SuperK, SNO
iChlorine.
Energy Threshold
BohcaH-pinsonneeult 2000
10". 10" i JO"
j3
10'
5
Ipt'f-
i
I
- -J
10" • 10"
•
'""{
^ W I
S,
-°"rK'jJ
Theory
N e u t r i n o E n e r g y (MeV)
Figure 1. The SSM flux prediction as a function of energy. The sensitivity of each experiment is also shown. From http://www.sns.ias.edu/~jnb/
®K
"'Be a P-P> P e P "B • CNO
Experiments
Figure 2. The SSM flux prediction with four energy thresholds corresponding to the type of experiment, From http://www.sns.ias.edu/~jnb
are this type. They can see the electron from the elastic scattering (ES) vx-\-e~ —> vx + e~.
trino flavors, can also be detected by 7 from neutron capture. By measuring CC/ES flux ratio (or CC/NC flux ratio), evidence of flaHowever, SNO is a qualitatively new type vor transformation can be extracted indepenof experiment, since D 2 0 (heavy water) is dent of solar model flux calculations. used as detector media. It can also see elecComparing CC flux measurement from trons from the charged current reaction (CC) SNO to ES flux measurement from Super-K, ve+d —> p + p + e~. The CC reaction is sen- a 3.3 p+n+vx, ing angle solutions are favored17. which is equally sensitive to all active neu-
486
Experimental Review of Neutrino Physics
Shigeki Aoki KamLAND and Borexino are experiments designed to detect low energy neutrinos using liquid scintillator. They are "realtime" detectors like water Cerenkov experiment, but the energy threshold can be lower. KamLAND 18 is a long baseline reactor neutrino experiment, planning to detect Ve from nuclear power reactors located 175 ± 35km away. Their sensitivity for neutrino oscillations completely covers the LMA solution to the solar neutrino problem. Borexino 19 is aiming to provide direct measurement of the 7Be neutrino flux, which is very crucial for the flux independent analysis of the solar neutrino oscillation. 3
Atmospheric Neutrinos
The sources of atmospheric neutrinos are charged 7r's produced in cosmic ray reactions with nuclei of atmospheric molecules. Since TT^ decays to / ^ + ^M(X^) and ^ decays to e* + ve{T7l) + T^iVfj,), the flux ratio of Vn/ve is expected to be about 2. If the primary cosmic ray arrives isotropically, upward and downward symmetry is expected. But in case of neutrino oscillations, different results are expected, depending on the energy and the zenith angle (i.e. the flight length of the neutrino). Super-K has reported their 79.3 ktonyear (1289 days with 22.5kton fiducial mass) data analysis. From their v^ - vT oscillation analysis, they got 90% C.L. allowed region at 1.6 x 1CT3 eV 2 < Am2 < 4 x lO" 3 eV 2 and sin2 20 > 0.89. Detailed discussions can be found elsewhere15. Soudan-2 is a different type of experiment, with 963 tons of fine tracking calorimeter made of 1.5cm diameter drift tubes installed in honeycomb lattice steel cell. From preliminary results based on 5.1 kton-years data, their v^ - vT oscillation-allowed region is consistent with Super-K. MACRO is another type of experiment with large area streamer tubes for tracking
and liquid scintillator counters for time-offlight. From their zenith angle distribution of upward-going muons produced by atmospheric neutrinos in the rock below their detector, they observed deficit of upgoing muon flux20. It agrees well with v^-vT oscillations with maximal mixing around Am2 ~ 2.4 x 10~ 3 eV 2 . Some models predict that the atmospheric v^ deficit is due to its oscillation into "sterile" neutrinos (us) that do not interact, even via neutral current(NC). If this is the case, the number of NC like events for upward going neutrino should also be reduced. Moreover, matter effect in the Earth suppress the oscillation at high energy region (Ev > 15GeV). Both Super-K 15 and MACRO 20 have reported that pure Vp,-va oscillation is disfavored with 99% C.L. 4
Long Baseline Experiments (1st generation)
In order to investigate the atmospheric neutrino deficit, the K2K experiment was proposed as the first long baseline experiment using a v,j, beam from an accelerator. 3.85 x 10 19 POT were accumulated from June 1999 to April 2001. 44 events were observed while 64lg'g events were expected. The probability of null oscillation is less than 3%. Energy spectrum analysis is underway. Detailed discussions can be found elsewhere21. 5
"Medium Baseline" Experiments
The LSND experiment and the KARMEN experiment use neutrinos from 7r+ —> //+ + i/M decays at rest followed by fi+ —> e + + ue + ~v^ decays at rest in the low energy (~800 MeV) proton beam dump. ir~ decay is suppressed by nuclear capture. There is a minor fraction of 7r~ decay in flight followed by /j~ decay, which is also supressed by capture process \i~ +p —> n + i v
487
Experimental Review of Neutrino Physics
Shigeki Aoki
Table 2. LSND and KARMEN
LSND
KARMEN
proton beam intensity
1000 nA
200 //A
reptition rate
120 Hz
50 Hz
spill structure
600 us
100 ns double pulse (225 ns apart)
baseline length
30 m
17 m
detector mass
167 t
56 t
liquid scintillator
mineral oil based
organic scintillator
detector structure
5.7mc/> x 8.3m cylinder
512 segmented rectangular modules
(water Cerenkov like)
with Gd^Oo, coated paper within the acrylic module wall
neutron detection
+ Gd + 7 (8MeV)
n+p^d
to
9 Hi
17.5
*
UJ
15
• '_
^ ^
CO UQ
Beam Excess
12.5
P(v„-ve,e>
ESI
P(v„,e>
mm
other
10 7.5 5 2.5 0 ^^^P^W^^^^^^WX •
0.4
.
.
,
i
,
0.6
,
.
i
0.8
.
.
PsqpSfc:* ,
i
,
,
,
i
I
I
I
I
1 1.2 1.4 UEV (meters/MeV)
Figure 3. The LvjEv distribution for events observed by LSND. The expected distribution from backgrounds and V^. —> z/J oscillations at low A m 2 are also shown. From Phys. Rev. D 64, 112007.
488
sin 26
Figure 4. T h e allowed region of neutrino oscillation parameter by the LSND experiment is shown shaded. Also shown are the excluded regions (90% C.L.) by the KARMEN, NOMAD, CCFR and Bugey experiment. From Phys. Rev. D 64, 112007.
Shigeki Aoki
Experimental Review of Neutrino Physics
This gives a very small VI contamination flux below the 1 0 - 3 level, which is futher reduced by software cuts. Both experiments seached for V^ —> V^ oscillation in appearance mode by detecting the positron from VI-f p —-> n + e+, and the delayed 7 ray from neutron capture. Figure 3 shows the L/E distribution by LSND. The LSND result gives evidence for an excess of V^ events, while KARMEN result 23 does not support it. Comparison of both experiments is shown in Table 2. Figure 4 shows the allowed reagion of oscillation parameters describing LSND result, with excluded regions by other experiments. A narrow region still remains as a possible region for V^ —> 17^ oscillation. The primary goal of the Mini-BooNE 24 experiment is to confirm or disprove the neutrino oscillation signal suggested by LSND. The neutrino beam is produced by 8GeV protons from the Fermilab Booster with horn focusing and variable (25 to 50m) decay path. The detector is a 769 ton (445 ton fiducial) pure mineral oil Cerenkov detector installed 500m away from the production target. A Vp, —> ve oscillation to explain LSND results would produce 1000 ve charged current (CC) events, while 500,000 v^ CC events and 1700 ve CC events due to the intrinsic ve component are expected.
6
Short Baseline Experiments
The CHORUS and NOMAD experiments searched for v^ —> vT oscillation in the Am2 > leV 2 region, which corresponds to the vT mass as the hot component of the Dark Matter of the Universe. They were installed one behind the other in the wide-band neutrino beam from the CERN 450 GeV proton synchrotorn. The average neutrino energy was about 27 GeV and the distance between the neutrino source and detector was ~600m on average. The intrinsic vT compo-
489
nent in Ufi beam was as low as 3.3 x 10~ 6 vT charged current (CC) interaction per v^ CC interaction. The vr is identified by detecting r~ produced by vT CC interaction. CHORUS and NOMAD employ two different approaches to identify r ^ decay. NOMAD uses a purely kinematical technique to identify vT CC interaction. Precise measurements of the secondary particle momenta are required. The core of the detector consisted of 2.7 tons of fiducial mass made of a series of drift chambers in a 0.4 Tesla magnetic field, acting both as target and as spectrometer. The vT CC interaction is differentiated essentially by selecting events using three vectors: the missing transverse momentum, the transverse momentum of hadron jet, and the transeverse momentum of the r decay track(s). Figure 5 shows the final NOMAD results 25 on v^ - vT and ve - vT oscillations together with the results by other experiments. The two flavor mixing limit is excluded down to sin2 W^r < 3.3 x 10" 4 at large Am2 and sin2 29eT < 1.5 x 10~ 2 at large Am2. CHORUS used nuclear emulsion acting as the target and as detector of the interaction vertex and the decay of r~ lepton. The nuclear emulsion can provide a three dimensional topology of charged tracks with suborn spatial resolution. 0.77 tons of nuclear emulsion was exposed to the neutrino beam with an integrated intensity corresponding to 5.06 x 10 19 proton on target. The search for vT interaction has been performed for the muonic ("lyu") and the hadronic ("0/z") one prong decay. In Phase-I analysis, ~1.7 xlO 5 neutrino interactions have been located in the emuslion and no r candidate event was observed, while 0.1 and 1.1 background events were expected for 1/z and 0/j, samples respectively. The 90% C.L. excluded region from CHORUS Phase-I result 26 is also shown in Figure 5. The mixing limit is excluded
Experimental Review of Neutrino Physics
Shigeki Aoki
"i
-^10 >
3
'
>
( trlORUS'
•
CHORUS NOMAD /
e
10
CDHS-.: ..i
10
-
-:
v — v 90% C.L. 10
10
10 2
'_ N O M A L K A ; - . .
10
•
3
-
10 =-
:
1
• •'""
E531 :,"CCFR
r
0)
S,A2
•
;
T
F 90% C.L.
CH002>^\q
•'
10
1
10
sin2 20
10
1
sin2 20
Figure 5. Contours outlining a 90% C.L. region in the Am2 - sin 2 28 plane for fM - vr oscillation (left) and ve - vT oscillation (right). From Nucl. Phys. B 6 1 1 , 3 (2001).
down to sin2 20^T < 6.8 x 10~ 4 at large Am2. This number can not be directly compared to NOMAD result, since the statistical treatment 27 of the data is different. Using the same statistical treatment 2 8 as the NOMAD experiment for comparison, sin 2 28'MT < 4.4 x 10~ 4 at large Am 2 is obtained. To improve sensitivity, CHORUS continues Phase-II analysis with a new scanning method, called "Netscan", developed in the DONUT experiment described below. Improvement is expected by higher kink finding efficiency and electron identification to search electronic decay mode of the T. As an example of the power of "Netscan" method in Phase-II analysis, from ~2.5 x 104 u^ CC interactions, 283 D° meson decay is observed with an estimated background of 9.2 K° and A decays 29 . The DONUT experiment has observed the charged current interactions of vT by identifying the r lepton as the only lepton created at the interaction vertex. The neutrino beam was created using 800 GeV protons from the Fermilab Tevatron interacting in a one meter long tungsten
beam dump. The nuclear emulsion target was placed 36m downstream from the dump. ve, v^ and vT that interacted in the emulsion target mostly originated from the decays of charmed mesons in the beam dump. But Pfj, also had large (about half) component from 7r and K decays. The primary source of vT is the leptonic decay of a Dg meson into T^ + vT(p^), and the subsequent decay of the T^ to T^{vT) + anything. The other products from the beam dump, mostly muons, were absorbed or swept away from the emulsion region using magnets with concrete, iron and lead shilding. The length of exposure for each target was set by the accumulated track density of muon, with a limit of 105 c m - 2 for the analysis. ECC ("Emulsion Cloud Chamber") module had a repeated structure of emulsion plates interleaved with 1 mm thick stainless steel sheets. Scintillating fiber trackers, distributed between the emulsion modules, were used to reconstruct vertex and provide the information to locate vertex in the emulsion module. The spectrometer downstream provided the information about lepton identification and energy measurement.
490
Shigeki Aoki
i
— ...
Experimental Review of Neutrino Physics
---•• * -r
dain«itiii«Miiilh irmllil
Jj^^^a^s^M^SSSm^.'^mti'Tsmfu
f
uA^aaas&ffiite&fi^Q
\ "'\
ah^ggKj^^gg^^^g^^^^m^KC^IjB^wgjijria
WV^ p\
Figure 6. Example of NETSCAN in the DONUT experiment.
(top left) after alignment (top right) after rejection of penetrating track (bottom left) after rejection of non-connected track (bottom right) after requiring a small impact parameter
491
Experimental Review of Neutrino Physics
Shigeki Aoki A total of 4.0 x 106 triggers were recorded from 3.54x 1017 protons incident on the tungsten beam dump. Prom the spectrometer information, 898 events were selected as neutrino interaction candidates, and 727 of them had a predicted vertex with in a fiducial emulsion volume. Among them, 566 events were scanned to locate a vertex using "Netscan" method 30 as follows. All track segments in entire predicted volume are read out by the fully automated system, called the Ultra Track Selector31. A typical scanning volume is 5mm x 5mm in the transverse direction and ~15mm in the longitudinal direction, typically coresponding to 12 emulsion plates. There are approximately 104 track segments found in each emulsion plate. Tracks that are recognized as passing through the volume are used for plate-to-plate alignment and are eliminated as candidate tracks from a neutrino interaction. The distance of closest approach between any two tracks that started in the same or neighboring emulsion plate is calculated and those within 4 /xm are retained as candidates to form a two-track vertex. From the cluster of two-track vertices, the interaction vertex is identified. An example is shown in Figure 6
Table 3. r candidate in D O N U T T
charm BG
hadronic int. BG
short flight
1
0.13
0.22
long flight kink trident
4 2
0.30 0.61
0.27 0.08
total
7
1.04
0.57
They are planning to perform high precision measurements of spectrum distortion in fMdisapperance, high precision measurements of mixing angle, or ^-appearance detection in order to establish the clear evidence of oscillation at Am 2 region by atmospheric neutrino experiments. The MINOS 32 detector consists of 2.54 cm thick magnetized iron interleaved with 4cm wide, 1 cm thick and 8 m long scintillator strips with wavelength shifting fiber read-out, which provides both calorimetric and tracking information. NC/CC ratio can discriminate Ufj, —> vT and v^ —> vs.
Out of 566 events, a valid vertex was located for 344 events. To 337 out of the 344 events, systematical decay search was applied and 7 T candidates were identified as shown in Table 3. The main sources of backgrounds are the decays of charmed particles produced by fM or ve CC interaction with inefficient lepton identification at the interaction vertex, and hadronic interactions in the steel plate. The total background is estimated to be 1.61 events (prelimary result).
The JHF-Kamioka neutrino project 33 plans the high precision measurement of neutrino mass and mixing. The first phase using the Super-Kamiokande detector plans (1) an order of magnitude better precision in the v^ —> vT oscillation measurement. (2) a factor of 20 more sensitive search in the v n —> ve appearance. (3) a confirmation of the v^ —• vT oscillation or discovery of vs by detecting the neutral current events.
7
Both ICARUS and OPERA are vT appearance search experiments. The ICARUS 34 detector is a liquid Argon time projection chamber with ~ 1 mm spatial resolution in three dimensional reconstruc-
Long Baseline Experiments (2nd generation)
In addition to KamLAND, there are three long baseline projects shown in Table 4.
492
Experimental Review of Neutrino Physics
Shigeki Aoki
Table 4. Future Long Baseline Project
NuMI
JHF-Kamioka
CNGS
accelerator
Main Injector (Fermilab)
JHF (JAERI)
SPS (CERN)
experiment
MINOS
Super-K
ICARUS/OPERA
proton beam energy
120 GeV
50 GeV
400 GeV
proton beam intensity
4.0 x 10 13 ppp
3.3 x 10 14 ppp
4.8 x 10 13 ppp
machine cycle
2.0 sec
3.4 sec
6 - 2 7 sec
20
/year
1 x 10
3.7 x 10
baseline length
732 km
295 km
730 km
mean v energy
6.8(5.2,12) GeV
0.7 GeV
17 GeV
number of v^ CC int.
1270(470,2740)/kt/year
100/kt/year
3200/kt/year
fiducial mass
3.3 kt
22.5 kt
1.2 kt / 1.8 kt
Summary
SNO has begun a new era of measurements of Solar neutrinos, and the large mixing angle solutions seem to be favored. KamLAND has good chance to get a positive result. ' B e neutrino detection by Borexino is also crucial. LSND's large Am 2 solutions are ex-
493
/year
4.5 x 1019 /year
proton on target
tion capabilities. For T~ identification, the kinematical approach is used like NOMAD. The OPERA 3 5 experiment uses same detection technique as the DONUT experiment. The building unit of the detector is an ECC ("Emulsion Cloud Chamber") brick, consisting of machine-coated emulsion films interleaved with 1mm thick lead plates. 0.23 millions of bricks will be used to build a 1.8 kton target. Using information from an electronic detector, the bricks containing the neutrino interaction will be extracted on a daily basis. The emulsion analysis will progress simultaneous with the exposure. For Am 2 = 2.5 x 10~ 3 eV 2 , 2.8 r event/year is expected while 0.11 background event/year is expected.
8
21
cluded. But its small Am 2 solutions are still possible, which will be covered by MiniBooNE. From the atmospheric neutrino experiis favored over vu ments, vu And DONUT confirmed that vT does interact in the same manner as ve and v^. The next step is getting clear evidence of oscillations by vT appearance detection and high precision measurements of the oscillation parameter. Acknowledgments I thank Byron Lundberg, Naoki Nonaka and Masahiro Komatsu for their help for this article. I am also grateful to Juliet Lee Franzini for her encouragement to complete this article. References 1. J. Chadwick, Verhandl. Dtsh. phys. Ges. 16, 383 (1914). 2. C.L. Cowan et al., Sience 124, 103 (1956).
Experimental Review of Neutrino Physics
Shigeki Aoki 3. R. Davis Jr., Phys. Rev. 97, 766 (1955). 4. B. Pontecorvo, JETP(USSR) 34, 247 (1958). 5. R. Davis Jr. et al., Phys. Rev. Lett. 20, 1205 (1968). 6. G.P.S. Occhialini and C.F. Powell, Nature 159, 186 (1947). 7. G.T. Danby et al., Phys. Rev. Lett. 9, 36 (1962). 8. Z. Maki, M. Nakagawa and S. Sakata, Prog. Theo. Physics 28, 247 (1962). 9. M.L. Perl et al., Phys. Rev. Lett. 35, 1489 (1975). 10. M. Nakamura, Nucl. Phys. B (Proc. Suppl.) 77, 259 (1999). K. Kodama et al., Phys. Lett. B 504, 218 (2001). 11. B.T. Cleveland et al, Astrophys. J. B 496, 505 (1998). 12. J.N. Abdurashitov et al, Phys. Rev. C 60, 055801 (1999). V. Gavrin et al, Nucl. Phys. B (Proc. Suppl) 91, 36 (2001). 13. M. Altman et al, Phys. Lett. B 490, 16 (2000). E. Bellotti et al, Nucl. Phys. B (Proc. Suppl) 91, 44 (2001). 14. Y. Fukuda et al, Phys. Rev. Lett. 77, 1683 (1996). 15. J. Goodman, these proceedings. S. Fukuda et al, Phys. Rev. Lett. 86, 5651 (2001). S. Fukuda et al, Phys. Rev. Lett. 85, 3999 (2000). 16. J. Klein, these proceedings. Q.R. Ahmad et al., Phys. Rev. Lett. 87, 071301 (2001). 17. Fogli, Lisi, Montanino and Palazzo, Phys. Rev. D 64, 093007 (2001). Bahcall, Gonzalez-Garcia and PenaGaray, JHEP 0108, 014 (2001). 18. A. Piepke et al, Nucl. Phys. B (Proc. Suppl) 91, 99 (2001). 19. G. Alimonti et al, Astropart. Phys. 16, 205 (2002). 20. M. Ambrosio et al., Phys. Lett. B 517,
59 (2001). 21. C.K. Jung, these proceedings. 22. A. Aquilar et al., Phys. Rev. D 64, 112007 (2001). 23. K. Eitel et al, Nucl. Phys. B (Proc. Suppl) 91, 191 (2001). 24. A.O. Bazarko et al, Nucl. Phys. B (Proc. Suppl) 91, 210 (2001). 25. P. Astier et al., Nucl. Phys. B 6 1 1 , 3 (2001). 26. E. Eskut et al., Phys. Lett. B 497, 8 (2001). 27. T. Junk, Nucl. Lnstr. and Meth. A 434, 435 (1999). 28. G.J. Feldman and R.D. Cousins, Phys. Rev. D 57, 3873 (1998). 29. A. Kayis-Topaksu et al., CERN-EP 2002-005 (2002). to be published in Phys. Lett. B. 30. K. Kodama et al., to be published in Nucl. Instr. and Meth. 31. S. Aoki et al., Nucl. Instr. and Meth. B 51 466 (1990). T. Nakano, Ph.D Thesis, Nagoya Univ., (1997). 32. S.G. Wojcicki, Nucl. Phys. B (Proc. Suppl) 91, 216 (2001). 33. Y. Itow et al., [hep-ex/0106019] (2001) 34. A. Rubbia, Nucl. Phys. B (Proc. Suppl) 91, 223 (2001). F. Arneodo et al., [hep-ex/0103008] (2001)) 35. M. Gular et al., CERN/SPSC 2000-028, SPSC/P318, LNGS P25/2000, (2000).
494
THEORY OF N E U T R I N O M A S S E S A N D M I X I N G S HITOSHI MURAYAMA Department of Physics, University of California Berkeley, CA 94720, USA Theoretical Physics Group, Lawrence Berkeley National Laboratory University of California, Berkeley, CA 94720, USA E-mail: [email protected] Neutrino physics is going through a revolutionary progress. In this talk I review what we have learned and why neutrino mass is so important. Neutrino masses and mixings are already shedding new insight into the origin of flavor. Given t h e evidences for neutrino mass, leptogenesis is gaining momentum as the origin of cosmic baryon asymmetry. Best of all, we will learn a lot more in t h e coming years.
1
Introduction
There is no question that we are in a truly revolutionary moment in neutrino physics. Prom the previous speakers, Jordan Goodman, 1 Josh Klein, 2 Chang-Kee Jung, 3 and Shigeki Aoki,4 we have seen a wealth of new experimental data in neutrino physics since the previous Lepton Photon conference. We are going through a revolution in our understanding of neutrinos, even in broader context of flavor physics, unification, and cosmology. My job is to go through some of the exciting aspects of these topics. I found that not only us but also general public is excited about neutrino physics; I've found fortune cookies from SuperK brand on my way from Snowmass meeting to Denver airport at a Chinese restaurant. We have learned the following important points since the previous Lepton Photon conference: • Evidence for v^ deficit in atmospheric neutrinos is stronger than ever, with the up/down asymmetry established at more than 10 a level in i^-induced events. 1 We are more than 99% certain v^ are converted mostly to vT. Current K2K data support this evidence. 3 • Putting SuperK amiokande and SNO 495
Figure 1. Fortune cookies from SuperK brand I found in Colorado.
data together, we are certain at 3 a level that the solar ve must have converted to i/M or vT? • Such neutrino conversions are most likely due to neutrino oscillations. Other possibilities, such as neutrino decay, violation of equivalence principle, spinresonant rotation, exotic flavor-changing interactions, are still possible, but are either squeezed phenomenologically or rely on models that are theoretically not motivated or esthetically not pleasing. • Tiny neutrinos masses required in neutrino oscillation for atmospheric and so-
Theory of Neutrino Masses and Mixings
Hitoshi Murayama lar neutrinos are the first evidence for the incompleteness of the Minimal Standard Model.] Even apart from oscillations, any explanation to these phenomena require physics beyond the Standard Model. Given this dramatic progress in experimental results, it is interesting to see how good insight theorists had had on neutrino physics. Here is the list of typical views among theorists on neutrino physics back in 1990: • Solar neutrino problem must be solved by the small angle MSW solution because it is so cute. • The natural scale for fM-^T oscillation is Am = 10-100 eV 2 because it is cosmologically interesting. • The mixing angle for v^-Vj. oscillation is of the order of V^,. • Atmospheric neutrino anomaly must go away because it requires an ugly large angle, not suggested by simple grandunified theories. Looking back at this list, the first one is most likely wrong as we will see later, and (2-4) are all wrong. You can see that theorists had had great insight into the nature of neutrinos. The rest of the talk is organized in the following manner. I first review the situation with the global fit to solar neutrino data after SNO results came out. Then models of neutrino masses and mixings, in the broader context of models of flavor, are briefly reviewed. After that, the idea of leptogenesis is explained, which is one of the main ideas now to explain the cosmological matter anti-matter asymmetry. Finally I will discuss what we can look forward to in the near future.
2
Global Fits
I review the global fits to the solar neutrino data including SNO. But before doing so, I spend some time discussing the parameter space of two-flavor neutrino oscillation. Traditionally, neutrino oscillation data had been shown on the (sin2 29, Am2) plane. This parameterization, however, covers only a half of the parameter space. We instead need to use, for example, (tan 2 9, Am 2 ) to present data. This point had been recognized for a long time in the context of three-flavor mixing, 5 but two-flavor analyses had always been presented on the (sin2 29, Am2) except as a limit of three-flavor analyses. Because you will see plots that cover the full parameter space soon and also in the future, I will briefly explain this point. The oscillation occurs because the flavor eigenstate (i.e. SU(2)L partner of charged leptons) and the mass eigenstates are not the same. Let us talk about ve and v^ for the sake of discussion. The mass eigenstates are then given by two orthogonal linear combination of them:
v\ = ve cos 9 + v^ sin 0,
(1)
V2 = —ye sin 6 + VJJ, cos 9. As a convention, we can always choose vi to be heavier than v\ without a loss of generality. Now the question is how much we should vary 9. If you use sin 2 29 e [0,1] as your parameter, 9 can go from 0 to 45° to exhaust all possibilities of sin 2 29. However, for 9 < 45°, V\ always contains more ve than v^. To represent the possibility of v\ dominated by v^ rather than ve, we need to allow 9 to go beyond 45° up to 90°. Then sin2 29 folds over at 1 and comes back down to 0. Clearly, sin2 29 is not the right parameter for global fits to the data. When the oscillation is purely that in the vacuum (no matter effect), the oscillation
(2)
Hitoshi Murayama
Theory of Neutrino Masses and Mixings
probability depends only on sin 2 29 P{ye -> Vp) 2
2,
„
92
10'
Am 2 _
sin 26> sin 1 . 2 7 ^ — L , (3)
£
10
2
with Am in eV , E in GeV and 1/ in km. Therefore, it is the same for 9 and 90° — 6, explaining why sin2 29 had been used for fits in the past. In the presence of the matter effect, however, the oscillation probability is different for 6 and 90° - 6. To cover all physically distinct possibilities, one possible parameter choice is sin 2 9, that ranges from 0 to 1 for 9 e [0°, 90°], and shows the symmetry under 9 <-> 90° — 9 in the absence of the matter effect as a reflection with respect to the axis 9 = 45° on a linear scale. This is a perfectly adequate choice for representing atmospheric neutrino data, for instance when matter effect is included in v ii <-> vs oscillation. On a log-scale, however, tan 2 9 shows the symmetry manifestly as 9 —> 90° - 9 takes tan 2 9 -> cot 2 9 = 1/ tan 2 9. Especially for fits to the solar neutrino data, we need to cover a wide range of mixing angles, and hence a log-scale; that makes tan 2 9 essentially the unique choice for graphically presenting the fits. The range of mixing angle 0° < 9 < 45° is what had been covered traditionally with the parameter sin2 29, and we call it "the light side," while the remaining range 45° < 9 < 90° "the dark side" because it had been usually neglected in the fits.6 Now that SNO charged-current data is available, we would like to see its impact on the global fit to the solar neutrino data. Of course, the most important lesson from the SNO data is that there is an additional active neutrino component va (i.e., a linear combination of v^ and vT) coming from the Sun. The next quantitative lesson is that BP00 flux calculation of 8 B neutrino flux is verified within its error, as discussed by Josh Klein in his talk. 2 I show results from the global fits.7 The first one is a two-flavor fits for ve to va oscillation, using BP2000 solar neutrino flux calculations in Fig. 2. Four regions of the param-
497
10
-5 •SMA
Iff"
^ion « iff*
2 d.o.f.
10*% 10 10
Active i- Includes SK< •12 •11
n.-4
10
10'
10'2
10'1
tan2^ Figure 2. Two-flavor fit to solar neutrino d a t a for i/e to va oscillation 7 . T h e confidence levels are 90%, 95%, 99%, and 99.73% (3a).
eter space, LMA (Large Mixing Angle MSW solution), SMA (Small Mixing Angle MSW solution), LOW (MSW solution with LOW Am 2 ), and VAC (VACuum oscillation solultion), that fit the data can be seen. LMA solution is the best fit to the current data. One of the concerns in solar neutrino fits has been that been that solar 8 B neutrino flux is very sensitive to solar parameters." Thanks to the SNO data, we can now drop the predicted flux entirely from the fit and obtain equally good fit, as shown in Fig. 3. To account for both solar and atmospheric neutrino oscillations, we need threeflavor analyses. The standard parameterization of the mixing among neutrinos, MNS
"This concern has been greatly ameliorated by the agreement of helioseismology data and the standard solar model, however. 8
Hitoshi Murayama
Theory of Neutrino Masses and Mixings
10'
Fr'ee'^B
10' -5
10
Figure 4. T h e mass spectrum of three neutrino mass eigenstates.
io-6
«
E
ios
<10'9 10 10 10
11
10
•12
Figure 5. The 3 + 1 mass spectra of four neutrino mass eigenstates.
2 d.o.f. Active Includes SKs
10'4
10'3
10'2
10'1 Figure 6. The 2 + 2 mass spectra of four neutrino mass eigenstates.
tan¥ Figure 3. Two-flavor fit taking t h e solar 8 B neutrino flux as a free parameter. 7
(Maki-Nakagawa-Sakata) matrix, is given by
/ uel ue2 ue3 UMNS
=
U^i U^
\uTl
U^3
UT2UT3, C13
0
0 si3eiS
1 0
s13e
-i«>
0 ci 3 (4)
The angle 8\3 is currently undetermined, except that there is an upper bound sin #13 < 0.16 from reactor neutrino experiments 10 for Am 2 = 3 x 10" 3 eV2 preferred by the atmospheric neutrino data. Fortunately, the smallness of #13 essentially decouples solar and atmospheric neutrino oscillation, allowing us to interpret data with separate twoflavor fits. The convention here is that the mass eigenstates V\ and v2 have small mass splitting for the solar neutrino oscillation Am 2 2 < 2 x !CT4eV2, while v2 and v3 have
the splitting ATO23 — 3 x 10~ 3 eV 2 , as shown in Fig. 4. I've put v2 above v\ assuming we are on the "light side" of solar neutrino oscillation. The "dark side" of solar neutrino oscillation would correspond to the opposite order. Future improvements in the solar neutrino data would allows us to discriminate between the two. In addition, we still do not know if v3 should be above or below the solar doublet. The remaining issues are (1) to determine the solar neutrino oscillation parameters, (2) the ordering of mass eigenstates, and (3) #13. When LSND oscillation signal (see talk by Aoki 4 ), that prefers Am 2 ~ eV , is also considered, we have to accommodate three different orders of magnitude of Am 2 values and hence four mass eigenstates. Because of three Am 2 , the mass spectrum now has 3! = 6 possibilities shown in Figs. 5 and 6. Now that both solar and atmospheric neutrino data disfavor oscillations into pure sterile state, the scenario with sterile neutrino is getting squeezed. For example, the comparison of sterile and active case can be
498
Hitoshi Murayama
Theory of Neutrino Masses and Mixings
•s
10
•4
10
I I I HUM
rTTTTTTTT—
TTTTT—i i LLimi—i i 11 MI E ^ — ii i i i I I I I I — i ri 11rmi i mill 1 —r
i IIIHI—i
i~r11111r—TTTTrre
1 F?ee%
Free°5
10' -6 -
10
« 10*
E
io10 J1
Active
Sterile
10 L Includes SKsp(D/N) •12 i i mill i i i 11 i i mill i i i nml 10 10'4 10'3 10'2 10'1 1
Includes SKsp(D/N) i i 11 mil
I i nml
10.
i i i mill
i i i mill
10'3 W'2
LI, 11 lin
10'1
1
10.
2
tan i? Figure 7. A global fit interpolating the pure active and pure sterile cases. 7 See text for the precautions on the definition of confidence levels that differ from Figs. 2 and 3.
done by interpolating the two cases with an additional parameter r\. The state that solar ve oscillates to is a mixture of an active and sterile neutrino, va cos 77 + vs sin 77. Then one can fit the data with three degrees of freedom, Am 2 , tan 2 8, and 77, and finds that the fit is better for 77 = 0 (pure active case). A word of caution here is that the definition of the confidence levels is now looser, making the allowed regions bigger. The confidence levels 90%, 95%, 99%, and 99.73% in Fig. 7 would correspond to 96%, 98%, 99.7%, and 99.92% for pure two-flavor case. Nonetheless, it is clear that the pure sterile case provides a much less good fit. However, the phenomenological motivation for having a sterile state comes from the combination of LSND, atmospheric, and solar neutrino data, and it requires full fourflavor analysis. In the case of 3 + 1 spec-
499
tra, the mixing angle for LSND oscillation is related to the MNS matrix elements as sin220LSND = 4|[/ e 4 | 2 |C/ M | 2 > 0.01. On the other hand, 1^41 cannot be too large because of CDHS data and atmospheric neutrino data, while |C e 4| 2 either because of Bugey reactor neutrino data. 1 1 This type of spectra is only marginally allowed12, and fits prefer 2+2 spectra. 13 However, 2+2 spectra are also getting squeezed now that both solar and atmospheric neutrino data disfavor oscillations into pure sterile state. If solar neutrino oscillation is into a pure active state, the atmospheric neutrino oscillation must be into a pure sterile state, and vice versa. The way this type of spectra can fit the data is to find a compromise between the requirements of sufficient active component in both oscillations. Detailed numerical analysis showed that such a compromise is still possible and
Theory of Neutrino Masses and Mixings
Hitoshi Murayama
fermion
masses
Vji
1= n> <
c •
WH*
(large angle MSW)
t*
•H@V2»V3
B
a> <
CD
FT
<
CD
g
a
<
H
a> <
G>
<
n> <:
Figure 8. Mass spectrum of quarks and leptons. LMA solution to the solar neutrino problem is assumed, while the range given for vi mass is basically just a guess.
the sterile neutrino is still allowed.9 The final verdict will be given by Mini-BooNE as we will see later. 6 3
Models
Now that neutrinos appear massive, despite what Standard Model has assumed for decades, we need to somehow incorporate the neutrino masses into our theory. The most striking facts about the neutrino sector are (1) the masses are very small, and (2) mixing angles appear large. Looking at the mass spectrum of quarks and leptons, it is especially bizarre that even the third-generation mass is so low compared to quarks and charged leptons. An explanation is clearly called for. The minute you talk about masses of spin 1/2 particles, you need both spin up and down states because you can stop any massive particle. When the particle is at a relativistic speed, a more useful label is leftor right-handed states. For strictly mass"There is an intriguing possibility that all three oscillation data can be explained without a sterile neutrino, if there is C P T violation that allows different mass spectra between neutrinos and anti-neutrinos. 1 4 This possibility can be tested by having anti-neutrino run at Mini-BooNE beyond t h e planned neutrino run.
less particles, left- and right-handed states are completely independent from each other and you do not need both of them; this is how neutrinos are described in the Standard Model. Once they are massive, though, we need both, so that we can write a mass term using both of them: C„
mD{vLvR
+
uRvL).
(5)
But then the mass term is exactly the same as the other quarks and leptons, and why are neutrinos so much lighter? The so-called seesaw mechanism 15 is probably the most motivated explanation to the smallness of neutrino masses. 0 The first step is to rewrite the mass term Eq. (5) in a matrix form C„
1
VL VR)
0 ran
mo 0
VR
+ c.c.
(6) _ Here, I had to put UL and UR (CP conjugate of VR) together so that both of them are left-handed and are allowed to be in the same multiplet. The problem was that we (at least naively) expect the "Dirac mass" mo to be of the same order of magnitudes as other quarks and lepton masses in the same generc Recently, alternative explanation using extra dimensions had appeared. 1 6
500
Theory of Neutrino Masses and Mixings
Hitoshi Murayama ation which would be way too large (Fig. 9). The point is that the right-handed neutrino is completely neutral under the standard-model gauge groups and is not tied to the electroweak symmetry breaking (v = 246 GeV) to acquire a mass. Therefore, it can have a mass much larger than the electroweak scale without violating gauge invariance, and the mass term is (Fig. 10) r
1,
_ , / 0
C^s = -(vLVR)^mD
mflU"i\
±
MJ^_J+c.C..
(7) Because one of the mass eigenvalues is clearly dominated by M > mo, while the determinant is —mD, the other eigenvalue must be suppressed, -m2D/M <§; m a (Fig. 11). This way, physics at high-energy scale M suppresses the neutrino mass in a natural way. In order to obtain the mass scale for the atmospheric neutrino oscillation ( A t n ^ ) 1 ' 2 ~ 0.05 eV, and taking the third generation mass mo ~ mt ~ 170 GeV, we find M = mD/m„ ~ 0.6 x 10 15 GeV. It is almost the grand-unification scale 2 x 10 16 GeV where all gauge coupling constants appear to unify in th minimal supersymmetric standard model. Indeed, the seesaw mechanism was motivated by SO(10) grand-unified models which include right-handed neutrinos automatically together with all other quarks and leptons in irreducible 16-dimensional multiplets. SO(10) GUT also has an esthetic appeal, being the smallest anomaly-free gauge group with chiral fermions, while not requiring additional fermions beyond the right-handed neutrinos. But it has a slight problem: it is too predictive. The simplest version of the model predicts mc = ms = mM at the GUTscale and no CKM (Cabibbo-KobayashiMaskawa) mixing. The art of unified model building is how to break the naive prediction to a realistic one. There are many many models, with or without grand unification, of neutrino masses and mixings. There are papers submitted by C.S. Lam, Alexandre Khodin, Joe Sato,
501
Figure 9. Too large Dirac mass of neutrinos.
Figure 10. We can put a large mass to the righthanded neutrino without violating gauge invariance.
Figure 11. Then the mass of the neutrino becomes light.
Koichi Matsuda, Bruce McKellar, and Carl H. Albright to this conference. The bottomline is that one can construct nice unified models of quark and lepton masses, especially within SU(5) or S"O(10) unification. I do not intend to go into an exhaustive review of proposed models because I can't. d Instead, I'd like to discuss how we might eventually arrive at understanding masses and mixings in a bottom-up approach. Looking at quarks, the masses are very hierarchical, and mixings are small. The masses of charged leptons are also very hierarchical. Even though we are used to hierarchical mass spectrum, it is actually quite bizarre: why do particles with exactly the same quantum numbers have such different masses, and mix little? In graduate quantum mechanics, didn't we learn that states with same quantum numbers have typically similar energy levels and mix substantially? A very naive answer to the puzzle is that different generations of quarks presumably have d I hope no list angers less people than an incomplete list.
Hitoshi Murayama different hidden quantum numbers we have not identified yet. We call them "flavor quantum numbers." Flavor quantum numbers distinguish different generations, allowing them to have very different energy levels (masses) and forbidding them to mix substantially. First question we should ask then is if there is need for fundamental distinction among three neutrinos. If you look at the currently-favored LMA solution to the solar neutrino problem, two mass-squared differences are not that different, A m ^ ~ 1-7 x l(T 3 eV 2 , AmlUA ~ 0.2-6 x 10~ 4 eV 2 (both at 99% CL). The mixing angles 612, #23 are both large. Even though #i 3 is usually said to be small, \Ue3\ = sin#i 3 < 0.16 at Am 2 t m = 3 x 10" 3 eV 2 , it is smaller than It^sl - \UT3\ ^ 1/V2 = 0.71 only by a factor of 2.3. Furthermore at the low end of Am 2 t m , sin #i3 may be still sizable (0.2 (0.4) at Am 2 t m = 2(1) x 10" 3 eV 2 ). It is not clear yet if #13 is so small. If there is no hierarchy and mixing is large, we apparently do not need new quantum numbers to distinguish three generation of neutrinos. But isn't near-maximal mixing suggested by atmospheric neutrino data special? Isn't there a special reason for it? If there is no distinction among three generations of neutrinos, isn't even a small hierarchy between atmospheric and solar mass-squared differences puzzling? To address these questions, we ran Monte Carlo over seesaw mass matrices. 17 It turns out that the distribution is peaked at sin 2# = 1 for atmospheric neutrino mixing, and the ratio of two masses-squared differences Am 2 o l a r /Am^ t m ~ 0.1 is the most likely value. The lesson here is that, if there is no fundamental distinction among three generations of neutrinos, the apparent pattern of masses and mixings comes out quite naturally. Of course, "randomness" behind Monte Carlo is just a measure of our ignorance. But complicated unknown dynamics of flavor at some high-energy scale may well appear to produce random numbers in the low-energy
Theory of Neutrino Masses and Mixings
complex
DiracJ- -
-
seesaw
-
P --'" -.',"
'--.-J
Majorana j —
,--'"' 10- 4
10- 3
lO- 2 = m S0laT/Am2Mm
R
A
10- 1
1
2
Figure 12. T h e ratio of two mass-squared differences in randomly generated 3 x 3 neutrino mass matrices. 1 8 For the seesaw case, the peak is around ATO
solar/Amatm~0-1-
0.2
0.4
0.6
0.8
1
2
sin 20 Figure 13. sin 2 2023 in randomly generated 3 x 3 neutrino seesaw mass matrices. 1 8
theory. We call such a situation "anarchy." And the peak in the mixing angle distribution can be understood in terms of simple group theory. 18 In fact, the anarchy predicts that all three mixings angles are peaked in sin2 2# distributions at maximum. This does not sound quite right for sin2 2#i 3 , but we find that three out of four distributions, #12, #23, Am 2 o l a r /Am 2 t m , prefer what data suggest. I find it quite reasonable. If you take this idea seriously, then sin2 2#i 3 must be basically just below the current limit and we hope to see it sometime soon. What about quarks and charged leptons? 502
Theory of Neutrino Masses and Mixings
Hitoshi Murayama Here we clearly see a need for fundamental distinction among three generations. 6 They are definitely not anarchical; they needed ordered hierarchical structure. Let us suppose the difference among three generations is just a new charge, namely a flavor U{\) quantum number. As a simple exercise, we can assign the following flavor charges consistent with SC/(5)-type unification: 10(Q,L/,£)(+2,+l,0), 5*(L,D)(+1,+1,+1).
(8)
All three generations of L have the same charge because of anarchy: no fundamental distinction among them. As we saw, neutrino masses and mixings come out reasonably well from this charge assignment. With SU(5)like unification, right-handed down quarks, that belong to the same 5* multiplets with left-handed lepton doublets, also have the same charge for all three generations. It is intriguing that large mixing among righthanded quarks is consistent with what we know about quarks, because the CKM matrix is sensitive only to particles that participate in charged-current weak interaction, namely left-handed quarks. Even if right-handed quarks are maximally mixed, we wouldn't know. On the other hand, 10 multiplets, that contain left-handed quark doublets, righthanded up quarks and right-handed leptons, need differentiation among three generations. Here we assigned the charges +2 for the first generation, +1 for the second, and no charge for the third. This way, top quark Yukawa coupling is allowed by the flavor charges, but all other Yukawa couplings are forbidden. Now suppose the flavor charge is broken by a small breaking parameter e ~ 0.04 that carries charge — 1. Then all other entries of Yukawa matrices are now allowed but regulated by powers of the small parameter e. Then we find that the ratio of quarks and e
See also Riccardo Barbieri at this conference. 1
lepton masses are mu : mc : mt ~ md : ms : mb ~ m 2 :ml : m2. ~ e4 : e2 : 1.
(9)
Namely, the up quarks are doubly hierarchical than down quarks and charged leptons, consistent with what we see. I'd like to emphasize that the "anarchy" is actually a peaceful ideology, nothing radical. According to Merriam-Webster dictionary, anarchy is defined as "A Utopian society of individuals who enjoy complete freedom without government." I'm just saying that neutrinos work peacefully together to freely mix and abolish hierarchy, without being forced by any particular structure. It predicts LMA solution to the solar neutrino problem, sin 2 2#i3 must be just below the current limit, and 0(1) CP violating phase. This is an ideal scenario for very long-baseline neutrino oscillation experiments, requiring many countries to be involved. It is therefore proglobalization! We have seen how new flavor quantum numbers can determine the structure of masses and mixings among quarks and leptons. Theorists of course argue about what the correct charge assignment is. How we will know if any of such flavor quantum numbers are actually right? It will be a long-shot program, needing many new data such as sin2 2^13, solar neutrinos, possible CP violation in the neutrino sector as well as more details in the quark sector and even charged leptons (EDM), B-physics, Lepton Flavor Violation, even proton decay. Because the difference in flavor quantum numbers suppress flavor mixing, the pattern of flavor violation must be consistent with assigned flavor quantum numbers. Details of these flavor-violating phenomena could eventually tell us what new quantum number assignment is correct. It is not clear if the origin of flavor is at the energy scale accessible by any accelerator experiment; it may be up at the unification scale. However, we may still
503
Theory of Neutrino Masses and Mixings
Hitoshi Murayama learn enough information to be able to reconstruct a plausible theory of the origin of flavor, masses, and mixings. I call such a program "archaeology" in the best sense of the word. For example, we can never recreate the conditions of ancient world. But by studying fossils, relics, ruins, and employing multiple techniques, we can come up with satisfactory plausible theories of what had happened. Cosmology is by nature an archaeology. You can't recreate Big Bang. But cosmic microwave background is a wonderfully colorful beautiful dinosaur that tells us great deal about the history of Universe. In the same sense, a wealth of flavor data can point us to the correct theory. For this purpose, Lepton Flavor Violation (LFV) is a crucial subject. Now that neutrinos appear to oscillate, i.e., convert from one flavor to another, there must be corresponding process among charged leptons as well. Unfortunately neutrino masses themselves do not lead to sizable rates of LFV processes. However, many extensions of physics beyond the Standard Model, most notably supersymmetry, 20 tend to give LFV processes at the interesting levels, such as n —> ej, fi —> e conversion, T —> /j,-y, etc. The violation of overall lepton numbers, such as neutrinoless double beta decay, would be also extremely important/
4
Leptogenesis
One of the primary interest in flavor physics is the origin of the cosmic baryon asymmetry. From the Big-Bang Nucleosynthesis, we know that there is only a tiny asymmetry in the baryon number, n s / n 7 « 5 x 10~ 10 . In other words, there was only one excess quark out of ten billion that survived the annihilation with anti-quarks. Leptogenesis is a possible origin of such a small asymmetry using •^See also John Ellis at this conference, 21 and papers submitted by Funchal Renata Zukanovich, and Bueno et al to this conference.
neutrino physics, a possibility that is gaining popularity now that we seem to see strong evidence for neutrino mass. The original baryogenesis theories used grand-unified theories, 22 because grand unification necessarily breaks the baryon number. Suppose a GUT-scale particle X decays out of equilibrium with direct CP violation B(X -> q) ^ B(X -> q). Then it can create net baryon number 5 ^ 0 in the final state from the initial state of no baryon number B = 0. It is interesting that such a direct CP violation indeed had been established in neutral kaon system in this conference (see R. Kessler 23 and L. IconomidouFayard 24 in this proceedings). However, the original models preserved B — L and hence did not create net B — L, that turned out be a problem. The Standard Model actually violates B.25 In the Early Universe when the temperature was above 250 GeV, there was no Higgs boson condensate and W and Z bosons were massless (so where all quarks and leptons). Therefore W and Z fields were just like electromagnetic field in the hot plasma and were fluctuating thermally. The quarks and leptons move around under the fluctuating Wfield background. To see what they do, we solve the Dirac equation for fermions coupled to W. There are positive energy states that are left vacant, and negative energy states that are filled in the "vacuum." As the Wfield fluctuates, the energy levels fluctuate up and down accordingly. Once in a while, however, the fluctuation becomes so large that all energy levels are shifted by one unit. Then you see that one of the positive energy states is now occupied. There is now a particle! This process occurs in the exactly the same manner for every particle species that couple to W, namely for all left-handed lepton and quark doublets. This effect is called the electroweak anomaly. Therefore the electroweak anomaly changes (per generation) AL = 1, and Ag = 1 for all three colors, and hence
504
Theory of Neutrino Masses and Mixings
Hitoshi Murayama
Figure 14. The energy levels of the Dirac equation in the presence of fluctuating W-field move up and down. All negative energy states are occupied while the positive energy states vacant in the "vacuum" configuration.
1001—c 100
< 150
L
• 200
' 250
M2 Figure 16. Constraint on the MSSM chargino parameter space in electroweak baryogenesis. 31 To generate r) = 5 x 1 0 - 1 0 , the parameters must lie inside the contour labeled "5." It implies light charginos. Shaded region is excluded by LEP.
not get washed out further by the electroweak anomaly. The other is the leptogenesis, 28 where you try to generate L ^ 0 but no B from neutrino physics well before the electroweak phase transition, and L gets parFigure 15. Once in a while, the fluctuation in the Wfield becomes so large that the energy levels of the tially converted o B due to the electroweak Dirac equation in the presence of fluctuating W-field anomaly. shift all the way by one unit. Then a positive energy The electroweak baryogenesis is not posstate is occupied and a particle is created. sible in the Standard Model, 29 but is still a possibility in the Minimal Supersymmetric Standard Model. However, the model is getting cornered; the available parameter AB = 1. Note that A(B - L) = 0; the elecspace is becoming increasingly limited due troweak anomaly preserves B — L. Because of this process, the pre-existing to the LEP constraints on chargino, scalar top quark and Higgs boson. 30,31 We are supB and L are converted to each other to find the chemical equilibrium at B ~ 0.35(5 — L), posed to find a right-handed scalar top quark, L ~ — 0.65(5 — L). 2 7 In particular, even if charginos "soon" with a large CP violation in the chargino sector. There is possibly a dethere was both B and L, both of them get tectable consequence in 5-physics as well. 32 washed out if B — L was zero. Given this problem, there are now two major directions in the baryogenesis. One is the electroweak baryogenesis, 26 where you try to generate B = L at the time of the electroweak phase transition so that they do
505
In leptogenesis, you generate L ^ 0 first. Then L gets partially converted to B by the electroweak anomaly. The question then is how you generate L ^ 0. In the original proposal, 28 it was done by the decay of a
Theory of Neutrino Masses and Mixings
Hitoshi Murayama
^H
K
y
H
hlk Figure 17. The tree-level and one-loop diagrams of right-handed neutrino decay into leptons and Higgs. The absorptive part in the one-loop diagram together with CP-violating phases in the Yukawa couplings leads to the direct CP violation T(N\ —• IH) ^ r(7Vi - • IH).
right-handed neutrino (say TVi), present in the seesaw mechanism, with a direct CP violation. At the tree-level, a right-handed neutrino decays equally into l + H and l + H*. At the one-loop level, however, the interference between diagrams shown in Fig. 17 cause a difference in the decay rates of a right-handed neutrino into leptons and anti-leptons proportional to ^(hijhikhlf.hjj). Much more details had been worked out in the light of recent neutrino oscillation data and it had been shown that a right-handed neutrino of about 10 10 GeV can well account for the cosmic baryon asymmetry from its out-ofequilibrium decay.33 There is some tension in the supersymmetric version because of cosmological problems caused by the gravitino. But it can be circumvented for instance using the superpartner of right-handed neutrino that can have a coherent oscillation after the inflation.34 Leptogenesis can work.
generation seesaw mechanism is enough to have CP violation that can potentially produce lepton asymmetry, unlike the minimum of three-generations for CP violation in neutrino oscillation. However, we decided that if we will see (1) electroweak baryogenesis ruled out, (2) lepton-number violation e.g. in neutrinoless double beta decay, and (3) CP violation in the neutrino sector e.g., in very longbaseline neutrino oscillation experiment, we will probably believe it based on these "archaeological" evidences. 5
Future
Can we prove leptogenesis experimentally? Lay Nam Chang, John Ellis, Belen Gavela, Boris Kayser, and myself got together at Snowmass and discussed this question. The short answer is unfortunately no. There are additional CP violating phases in the heavy right-handed neutrino sector that cannot be seen by studying the light lefthanded neutrinos. 9 For example, even two-
Even though dramatic progress had been made, many questions remain. The important aspect of neutrino physics is that we don't stop here. It is quite a healthy field with many studies done jointly by theorists and experimentalists looking forward. Here I will discuss what is coming in the near future.' 1 First of all, the oscillation signal from LSND will be verified or refuted at high confidence levels by Mini-BooNE experiment. It will start taking data in 2002. The result will mark a major branch point in the development of neutrino physics: do we need a sterile neutrino? There is a series of experiments aimed at the oscillation signal in the atmospheric neutrinos after K2K. MINOS, a long-baseline oscillation experiment from Fermilab to Sudan, Minnesota, will start in 2004, and will determine Am.23 precisely. This will provide crucial input to design very long-baseline neutrino oscillation experiments of later generations. OPERA and ICARUS in Gran Sasso, using the CNGS beam from CERN, will look for r appearance in the v^ beam for the same A m ^ . All of these experiments extend
9 If you believe in a certain scenario of supersymmetry breaking, low-energy lepton-flavor violation can carry information about CP violation in the right-handed neutrino sector. 35 However, such connection depends
on the assumptions in the origin of supersymmetry breaking. ''Plots are shown only for experiments that start d a t a taking by the next Lepton Photon conference.
506
Hitoshi Murayama
Theory of Neutrino Masses and Mixings KamLAND exclusion. Rate analysis. 90% C L .
J 10 u
10- l "dashed red
Sensitivity shown in proposalO^
dotted blue
E898 single-hom design ideal cut efficiencies E898 singe-hom design conservative cut efficiency estimation
10r2 ^solid blue
sin^e
v^vu 90% CL 10r3 E Sensitivity 10,-4
10
-3
Figure 19. KamLAND expected 90% confidence level sensitivity limits. 3 7 KamLAND 3 years
io-
2
10
-l
1
sin226 Figure 18. MiniBooNE expected 90% level sensitivity limits. 3 6
confidence
% E
<
reach in sin 2 20 13 as well. MONOLITH, if approved, will study atmospheric neutrinos using iron calorimetry and verify the oscillation dip. On the solar neutrino oscillation signal, KamLAND will be the first teerrestial experiment to attack this problem. Construction has completed in 2001. Using reactor neutrinos over the baseline of about 175 km, it will verity or exclude currently-favored LMA solution to the solar neutrino problem. If signal will be found, it will determine oscillation parameters quite well. 38,39,40 If this turns out to be the case, it will be a dream scenario for neutrino oscillation physics. This is because Am 2 olar is within the sensitivity of terrestrialscale very long baseline experiments, making CP violation in neutrino oscillation a possible target. For example, the difference in neutrino and anti-neutrino oscillation rates is given by P(ve
~ P(*e - P„) 507
"IO"5 10'r i
10u
10 1
tan2 6 Figure 20. Measurement of oscillation p a r a m e t e r s at KamLAND. 3 9
=
16Si2Cl2Si3e?13 3 S23C23Sin<5 *
2 Aro? 2 T . A m 2 3 . . A m 23 L Sin L Sin ~~&t ~~4E^ ~AELL ( 1 0 ) using the notation of the MNS m a t r i x in Eq. (4). 5 is the CP-violating phase. Clearly, Am 2 2 has to be sizable in order for t h e difference not to vanish. At the same time, large S13 is preferred. On the other hand, if LMA will be excluded, study of low-energy solar neutrinos will be crucial. KamLAND will be another major branch point. There are already proposals t o extend reach in sin 2f?i3 as well as possibly detect CP violation, often called neutrino super beam experiments. One possibility is to Sitl
Theory of Neutrino Masses and Mixings
Hitoshi Murayama
Table 1. Physics Sensitivity for Current Superbeam Proposals.' 11
Name
Years of Running
kton
JHF to SuperK SJHF to HyperK CERN to UNO
5 years v 2 years v, 6 years v 2 years v, 10 years v
50 1000 400
build a neutrino beam line using 50 GeV protons from JHF (Japan Hadron Facility) under construction, and aim the beam at SuperKamiokande. A possible upgrade of SuperKamiokande to an even bigger HyperKamiokande (?) together with more intense neutrino beam is also being discussed. On the other hand, CERN is discussing the Super Proton Linac (SPL), a GeV proton accelerator using LEP superconducting cavities. It can be aimed at a water Cherenkov detector at a modest distance to study possible CP violation thanks to its low energy. See Table 1 for their sensitivity. At Snowmass, a case was made that a higher energy superbeam with a longer baseline will be beneficial.41 For example, the matter effect can discriminate between two possible mass spectra in Fig. 4, and a longer baseline is needed for this purpose. If KamLAND excludes the LMA solution, the study of low-energy solar neutrinos will be crucial. The predicted spectrum of solar neutrinos is shown in Fig. 21. Only realtime experiments had been done so far using 8 B neutrinos. Borexino will be the first real-time experiment to detect lower-energy solar neutrinos from the 7 Be line. They expect data taking starting in 2002. What is so crucial about the 7 Be neutrinos is that it is mono-energetic (with some broadening due to thermal collisions in the Sun). The VAC solution to the solar neutrino problem can be studied by looking for anomalous seasonal variation. The distance between the Sun and the Earth is not constant; the Sun is closer in the winter
sin 2#i3 sensitivity (3a) 0.016 0.0025 0.0025
CP Phase S sensitivity (3 15° > 40°
Eu (GeV) 0.7 0.7 0.3
Neutrino Energy (MeV) Figure 21. T h e spectrum of solar neutrinos.
while farther in the summer in the northern hemisphere. This seasonal change in the distance causes a change in the oscillation probability for this solution and the event rate would be modulated beyond the trivial factor of 1/r2; see Fig. 22. Borexino can look for such an anomalous seasonal variation. The sensitivity region is shown in Fig. 23. At the low Am 2 region, the region is reflection symmetric. However for larger values of Am 2 , the region is asymmetric showing the impact of matter effect. The fact that matter effect is important even for the VAC solution had not been realized until recently.44 This parameter region was named "quasi-vacuum". 45 The LOW solution would lead to a large Earth matter effect for 7 Be neutrinos. 46 ' 47 Again once found, the zenith angle dependence of the event rate depends sensitively on the oscillation parameters and can be determined quite well. If the SMA solution turns out to be true,
508
Theory of Neutrino Masses and Mixings
Hitoshi Murayama
10"
7000
•SJi^^l
6000
* »f|*d[| 10"
4000
E
<
10" Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 22. Illustration of the effect of vacuum oscillations on the shape of the seasonal variation of the solar neutrino data. The points with statistical error bars represent the number of events/month expected at Borexino after 3 years of running for A m 2 = 3 x 1CT 10 eV 2 , sin 2 20 = l . 4 3
10" 0.01
0.1
1 tan2 8
10
100
Figure 23. The parameter space Borexino is sensitive to by looking for anomalous seasonal variation in the event rate.
SNO may eventually see a distortion in the charged-current spectrum. However, the determination of parameters would require the study of pp neutrinos. The point is that there is a sharp fallofF in the survival parameter in the pp spectrum. This is because the levelcrossing does not occur for low energy for which the matter effect does not overcome the mass splitting 4g£ > y/2GFne(0). The location of the fallofF determines Am 2 . On the other hand, the slow rise at higher energy depends both on the mixing angle and the mass-squared difference. GALLEX, SAGE, and now GNO experiments had detected pp neutrinos, but they cannot measure neutrino energy spectrum and hence cannot separate pp neutrinos from other solar neutrinos. There had been many proposals to study pp neutrinos. Some of them use neutrino-electron elastic scattering, sensitive to both charged- and neutral-current interactions (gaseous Helium TPC, HERON with superfluid He, liquid Xe, GENIUS using solid Ge), while others use charged-current reactions on nuclei (LENS, both Yb and In versions, and Moon on Mo). I hope that some of them will work in the end and provide us crucial information.
509
6
Conclusions
Neutrino physics is going through a revolution right now. We had learned a lot already, and will learn a lot more, especially on solar neutrinos. It provides an unambiguous evidence for physics beyond the minimal Standard Model. Given strong evidences for neutrino mass, leptogenesis has gained momentum as the possible origin of cosmic baryon asymmetry. Establishing lepton-number violation would be crucial, while seeing CP violation in neutrino oscillation would boost the credibility. Neutrino superbeams and eventually neutrino factory could play essential role in this respect if the currently-favored LMA solution turns out to be correct. We may not be lucky enough to test directly leptogenesis and models of neutrino masses, mixings, and flavor in general. But we can collect more "fossils" to gain insight into flavor physics at high energies, such as lepton flavor violation, combination of quark flavor physics, and even proton decay. Data may eventually point to new flavor quantum numbers that control the masses and mixings.
Theory of Neutrino Masses and Mixings
Hitoshi Murayama
Solar neutrino spectrum
10"
1 1 1 1 1 |
o 10" CD
5.10"
1
•r ° E
7
8 7
.
Ba;
;
Be
—
—
o 104 ="
"J 10;
10"
"r
S 5. 10" 7
10:
lO"7
1 •? 0.8 E T 0.6
i
'
i.-»-i^r+'
<
">
104 Energy (keV)
Survival probability
5. 10 " 8
Borexino sensitivity 10"
'
1
I
_LMA ( A m ^ . v l O ' W , sin226=0.85)
•£• 0 . 4 ; D.
0.2^ 0 102
o.ooi
T
~~>SMA (Am2=5x10-c,eV2, sin22B=8x10'3)
0.5
1
Figure 24. 95% (darkest), 3a (dark), and 5
103
10" Energy (keV)
Figure 25. The survival probability as a function of neutrino energy. For SMA, there is a sharp falloff in the survival probability.
Acknowledgments This work was supported in part by the DOE Contract DE-AC03-76SF00098 and in part by the NSF grant PHY-0098840.
8.
References 9. 1. 2. 3. 4. 5.
Jordan Goodman, in this proceedings. Josh Klein, in this proceedings. Chang-Kee Jung, in this proceedings. Shigeki Aoki, in this proceedings. G. L. Fogli, E. Lisi and D. Montanino, Phys. Rev. D 54, 2048 (1996) [arXiv:hep-ph/9605273]. 6. A. de Gouvea, A. Friedland and H. Murayama, Phys. Lett. B 490, 125 (2000) [arXiv:hep-ph/0002064]. 7. J. N. Bahcall, M. C. Gonzalez-Garcia and C. Pena-Garay, JHEP 0108, 014 (2001) [arXiv:hep-ph/0106258]. See also G. L. Fogli, E. Lisi, D. Montanino and A. Palazzo, Phys. Rev. D 64, 093007 (2001) [arXiv:hep-ph/0106247]; A. Bandyopadhyay, S. Choubey, S. Goswami and K. Kar, Phys. Lett. B 519, 83 (2001)
10.
11.
12.
13.
14. 510
[arXiv:hep-ph/0106264]; P. I. Krastev and A. Y. Smirnov, arXiv:hepph/0108177. P. Aliani, V. Antonelli, M. Picariello and E. Torrente-Lujan, arXiv:hep-ph/0111418. J. N. Bahcall, M. H. Pinsonneault and S. Basu, Astrophys. J. 555, 990 (2001) [arXiv:astro-ph/0010346]. For the most recent analysis with fourflavors including a sterile neutrino, see M. C. Gonzalez-Garcia, M. Maltoni and C. Pena-Garay, arXiv:hep-ph/0108073. M. Apollonio et al. [CHOOZ Collaboration], Phys. Lett. B 466, 415 (1999) [arXiv:hep-ex/9907037]; F. Boehm et al., Phys. Rev. D 64, 112001 (2001) [arXiv:hep-ex/0107009]. S. M. Bilenkii, C. Giunti, W. Grimus and T. Schwetz, Phys. Rev. D 60, 073007 (1999) [arXiv:hep-ph/9903454]; V. D. Barger, B. Kayser, J. Learned, T. Weiler and K. Whisnant, Phys. Lett. B 489, 345 (2000) [arXiv:hepph/0008019]. M. Maltoni, T. Schwetz and J. W. Valle, Phys. Lett. B 518, 252 (2001) [arXiv:hep-ph/0107150]. H. Murayama and T. Yanagida, Phys.
Hitoshi Murayama
Theory of Neutrino Masses and Mixings
Lett. B 520, 263 (2001) [arXiv:hepph/0010178]. 15. T. Yanagida, "Horizontal Symmetry And Masses Of Neutrinos", Prog. Theor. Phys. 64 (1980) 1103, and in Proceedings of the "Workshop on the Unified Theory and the Baryon Number in the Universe", Tsukuba, Japan, Feb 13-14, 1979, Eds. O. Sawada and A. Sugamoto, KEK report KEK-79-18, p. 95; M. Gell-Mann, P. Ramond and R. Slansky, in "Supergravity" (North-Holland, Amsterdam, 1979) eds. D.Z. Freedman and P. van Nieuwenhuizen, Print80-0576 (CERN). 16. K. R. Dienes, E. Dudas and T. Gherghetta, Nucl. Phys. B 557, 25 (1999) [arXiv:hepph/9811428]; N. Arkani-Hamed, S. Dimopoulos, G. R. Dvali and J. MarchRussell, talk presented at SUSY 98 Conference, Oxford, England, 11-17 Jul 1998. arXiv:hep-ph/9811448; Y. Grossman and M. Neubert, Phys. Lett. B 474, 361 (2000) [arXiv:hep-ph/9912408]. 17. L. J. Hall, H. Murayama and N. Weiner, Phys. Rev. Lett. 84, 2572 (2000) [arXiv:hep-ph/9911341]. 18. N. Haba and H. Murayama, Phys. Rev. D 63, 053010 (2001) [arXiv:hepph/0009174]. 19. Riccardo Barbieri, in this proceedings. 20. For a review, see H. Murayama, lectures at TASI 2000, h t t p : / / h i t o s h i . berkeley.edu/TASI/ 21. John Ellis, in this proceedings. 22. M. Yoshimura, Phys. Rev. Lett. 41, 281 (1978) [Erratum-ibid. 42, 746 (1978)]; A. Y. Ignatev, N. V. Krasnikov, V. A. Kuzmin and A. N. Tavkhelidze, Phys. Lett. B 76, 436 (1978). 23. R. Kessler, in this proceedings. 24. L. Iconomidou-Fayard, in this proceedings. 25. G. 't Hooft, Phys. Rev. Lett. 37, 8 (1976).
511
26. V. A. Kuzmin, V. A. Rubakov and M. E. Shaposhnikov, Phys. Lett. B 155, 36 (1985). 27. S. Y. Khlebnikov and M. E. Shaposhnikov, Nucl. Phys. B308 (1988) 885; J. A. Harvey and M. S. Turner, Number Violation," Phys. Rev. D 42 (1990) 3344. 28. M. Fukugita and T. Yanagida, Phys. Lett. B 174, 45 (1986). 29. See a review and references therein: A. G. Cohen, D. B. Kaplan and A. E. Nelson, Ann. Rev. Nucl. Part. Sci. 43, 27 (1993) [arXiv:hep-ph/9302210]. 30. M. Carena, J. M. Moreno, M. Quiros, M. Seco and C. E. Wagner, Nucl. Phys. B 599, 158 (2001) [arXiv:hepph/0011055]. 31. J. M. Cline, M. Joyce and K. Kainulainen, JHEP 0007, 018 (2000) [arXiv:hep-ph/0006119]; Erratum, arXiv:hep-ph/0110031. 32. H. Murayama and A. Pierce, in preparation. 33. See, e.g., W. Buchmuller, Presented at 8th International Symposium on Particle Strings and Cosmology (PASCOS 2001), Chapel Hill, North Carolina, 10-15 Apr 2001, arXiv:hep-ph/0107153, and references therein. 34. K. Hamaguchi, H. Murayama and T. Yanagida, arXiv:hep-ph/0109030. 35. J. R. Ellis, J. Hisano, S. Lola and M. Raidal, arXiv:hep-ph/0109125. 36. A. Bazarko [MiniBooNE Collaboration], talk presented at 19th International Conference on Neutrino Physics and Astrophysics - Neutrino 2000, Sudbury, Ontario, Canada, 16-21 Jun 2000. Published in proceedings, Nucl. Phys. Proc. Suppl. 9 1 , 210 (2000) [arXivrhepex/0009056]. 37. J. Busenitz et al., "Proposal for US Participation in KamLAND," March 1999, http://kamland.lbl.gov/. 38. V. D. Barger, D. Marfatia
Theory of Neutrino Masses and Mixings
Hitoshi Murayama
39.
40.
41.
42. 43.
44. 45.
46.
47.
and B. P. Wood, Phys. Lett. B 498, 53 (2001) [arXiv:hep-ph/O011251]. H. Murayama and A. Pierce, Phys. Rev. D 65, 013012 (2002) [arXiv:hepph/0012075]. A. de Gouvea and C. Pena-Garay, Phys. Rev. D 64, 113011 (2001) [arXiv:hepph/0107186]. "El working group summary: Neutrino factories and muon colliders," in Proc. of the APS/DPF/DPB Summer Study on the Future of Particle Physics (Snowmass 2001) ed. R. Davidson and C. Quigg, arXiv:hep-ph/0111030. Taken from h t t p : //www. sns. i a s . edu/ "jnb/SNviewgraphs/. A. de Gouvea, A. Friedland and H. Murayama, Phys. Rev. D 60, 093011 (1999) [arXiv:hep-ph/9904399]. A. Friedland, Phys. Rev. Lett. 85, 936 (2000) [arXiv:hep-ph/0002063]. G. L. Fogli, E. Lisi, D. Montanino and A. Palazzo, Phys. Rev. D 62, 113004 (2000) [arXiv:hep-ph/0005261]. A. de Gouvea, A. Friedland and H. Murayama, JHEP 0103, 009 (2001) [arXiv:hep-ph/9910286]. G. L. Fogli, E. Lisi, D. Montanino and A. Palazzo, Phys. Rev. D 61, 073009 (2000) [arXiv:hep-ph/9910387].
512
Cosmology and Astrophysics
Cosmology and Astrophysics Session Chair: Scientific Secretaries:
H. Klapdor Kleingrothaus F. Halzen M. Turner
M. Danilov M. Castellani F. Spada Dark Matter searches The highest energy cosmic rays, gamma rays and neutrinos Highlights in cosmology
514
DARK MATTER SEARCH H.V. KLAPDOR-KLEINGROTHAUS Max-Planck-Institut fur Kernphysik, P.O. Box 10 39 80, D-69029 Heidelberg, Germany Spokesman of HEIDELBERG-MOSCOW and GENIUS Collaborations E-mail: [email protected], Home-page: http://www.mpi-hd.mpg.de.non.ace/ Dark matter is at present one of the most exciting field of particle physics and cosmology. We review the status of undergound experiments looking for cold and hot dark matter.
1
Introduction MATTER
The physics motivations to search for dark matter are manyfold. Recent investigation of the cosmic microwave background radiation (MAXIMA, BOOMERANG, DASI) together with large scale structure results fix £7A + Qm ~ 1; where Q\ (= PA/PC) stands the for dark energy and Qm for matter. With the early nucleosynthesis constraint of ftbar ^0.04 the need for non-baryonic dark matter is evident. This is true even for our galaxy, since MACHOs represent only a small fraction of galactic dark matter 1 . Natural candidates for cold non-baryonic dark matter exist in the lightest SUSY particles, usually assumed to be the neutralinos. Although there exist other candidates such as axions 2 , neutralinos seem to be the favored candidates at present. Hot dark matter, according to CMB and LSS (Redshift-Survey) results still contribute up to 38% of the dark matter 3 7 . This corresponds to a sum of neutrino masses <5.5 eV. Neutrino oscillation experiments proved that the neutrino mass is not vanishing, can, however, not give absolute mass scales. Our present picture of the mass/energy distribution in the Universe is as given in Fig.l. In this presentation we shall concentrate on the terrestrial direct search for cold and hot dark matter. In section 2 we discuss the
515
COMPOSITION
1 COLD DARK CDM MATTER T0.35+/-O.1
•AO 2 T
D A R K
0.7+/-0.2
I'
0.1BARYONS
J0.05+/-0.0O5
ENERGY
T
I 0.3+/-0.1
IIE1DEI.BI".R<:MOSCOW |3|i best value
NEUTRINOS 0.01-
i.
g-2
STARS I 0.005+/-0.002
0.001-
a.
Figure 1. Our present understanding of the mass/energy composition of the Universe.
expectations for cold dark matter from SUSY models. In section 3 we discuss the experimental situation in direct search for cold dark matter in Underground Laboratories, and the future possibilities. In section 4 we outline hot dark matter search. The most sensitive way to look for an absolute neutrino mass is at present double beta decay. Section 5 gives a conclusion.
Dark Matter Search
H. V. Klapdor-Kleingrothaus 2
Expectations
liniLBERG •§, I ) \ \ U (l(«) kB N il)-r>idrn/ for WIMP-Signal (2000)^v ^—^•"*~ i p ^ i 2 *> kj; enreh. Ge)
1
(b)
,
„
„
- „
^ 2ffH(l 40Skg Ge BLIP) "~-HI)MS«1.2kg ,3 Ge> ;l 4 fcMIS IT V<-40kj.'nit.Ge)
^ r 'UiKN IL S • . . ; - \ k » y ? j * J \ - • t. v -
200 300 m j . (GeV)
500
700
the WIMP signal resulting from the seasonal variation of the earth's velocity against the WIMP 'wind'. The expectation for neutralino elastic scattering cross sections and masses have been extensively analysed in many variants of SUSY models. Figs.2,3 represent the present situation. The SUSY predictions Fig.2a, are from the MSSM with relaxed unification conditions 8 and the MSUGRA model 9 . Fig.2b shows the result of a study 'at Post-LEP Benchmark points' based again on the MSUGRA 10 . While Figs.2a,b are for spin-independent interaction, Fig.3 shows the case for spindependent interaction. Present experiments only just touch the border of the area predicted by the MSSM. The experimental DAMA evidence for dark matter lies in an area, in which MSUGRA models do not expect dark matter. They would require beyond GUT physics in this frame 11 .
1000
Figure 2. (a): WIMP-nucleon cross section limits in pb for scalar interactions (a)as function of the WIMP mass in GeV. Shown are contour lines of present experimental limits (solid lines) and of projected experiments (dashed lines). Also shown is the region of evidence published by DAMA. T h e theoretical expectations from the MSSM are shown by two scatter plots, - for accelerating and for non-accelerating Universe (from 8 ) and from the SUGRA by the grey region (from 9 ). Only GENIUS will be able to probe the shown range also by the signature from seasonal modulations, (b): W I M P - proton elastic scattering cross sections according to various MSUGRA models (see text). From 1 0 .
Direct search for WIMPs can be done (a) by looking for the recoil nuclei in WIMP- nucleus elastic scattering. The signal could be ionisation, phonons or light produced by the recoiling nucleus. The typical recoil energy is a few lOOeV/GeV WIMP mass. (b) by looking for the modulation of
100 m
1000 (GeV)
Figure 3. Expectations (grey area) from the MSUGRA model 9 some as in (Fig.2a), b u t for spindependent interaction. Solid lines correspond to present experimental limits, dashed lines to future projects.
It is clear from extensive theoretical work that high-sensitivity dark matter experiments can yield an important contribution to SUSY search. Fig.4 (from12) shows, in
516
Dark Matter Search
H. V. Klapdor-Kleingrothaus the MSUGRA model, the SUSY reach contours for different accelerators (LEP2, Tevatron, LHC, NLC) together with direct detection rates in a 73Ge detector. It is visible that a detector of GENIUS-sensitivity operates in SUSY search on the level of LHC and NLC. Fig.5 shows another study in the MSSM with relaxed unification conditions. Non-observation of Dark Matter with GENIUS would exclude a 'light' SUSY spectrum (all sfermion masses lighter than 300 400 GeV) and any possibility for a light Higgs sector in the MSSM.
0
100
200
300
400
500
Figure 4. Direct detection fate G rt/m events/kg day in a 7 3 Ge detector. Added are SUSY reach contours for LEP2, Tevatron (MI), LHC slepton signal and NLC. TH - excluded by theoretical considerations. EX - excluded by collider searches for SUSY particles (from 12 ).
If classifying the SUGRA models into more g^-2-friendly (I,L,B,G,C,J) and less gM2-friendly models, according to 1 0 , the former ones have good prospects to be detectable by LHC and/or a 1 TeV collider. GENIUS could check not only the larger part of these ones, but in addition two of the less gM-2-friendly models (E and F), which will be difficult to be probed by future colliders (see Fig.2b). This demonstrates nicely the complementarity of
517
Lighl spmnim
100
1000
Figure 5. Different lower bounds forH?nelaf(Mivevent rate in 7 3 Ge (events/day kg) versus mass of the LSP (GeV). Here Msq^cH,Hl denote masses of the squark, the charged Higgs boson, and the light neutral CPeven Higgs boson, respectively. The heavy chargino mass is denoted as M C ( , _ J J . "Full" corresponds to the lower bound obtained from main (unconstrained) scan, and "light spectrum" denotes the lower bound for R, which is obtained with all sfermion masses lighter than about 3O0GeV. T h e horizontal dotted line represents the expected sensitivity for the direct dark matter detection with GENIUS (from 8 ).
collider and underground research. It might be mentioned that in case of gM2 - unfriendly models, i.e. those with very low cross sections in Figs.2b, it might be required to turn from spin-zero targets and looking for spin-independent interaction, which usually for not too light nuclei gives the largest cross sections, to spin- non-zero target nuclei and spin-dependent interaction 8 . It has been shown recently 8 that if spin-zero experiments with sensitivities of 1 0 _ 5 - 1 0 - 6 events/kg day will fail to detect a dark matter signal, an experiment with nonzero spin target and higher sensitivity will be able to detect dark matter only due to the spin neutralino-quark interaction (see Fig.6). 3
Cold Dark M a t t e r - P r e s e n t and Future
Summarizing the present experimental status, present and also future projects can be categorized in two classes: 1. Sensitivity (or sensitivity goal) 'just
Dark Matter Search
H. V. Klapdor-Kleingrothaus
rejection procedure. This is also the reason, why simple cryogenic calorimeters such as the CRESST I 27 - 16 and the Milano 43 ' 42 bolometers seem not fully competitive.
Direct 0
200
TO
6Q-3
SOS
LSP Mass, G«V
'"
o g
0 7
0
.g
o a
,
detection;
- Ionization detectors - Ge HEIDELBERG-MOSCOW HDMS (Heidelberg Dark Malter Sean* exp.
Gniii'mC traction
-Scintillation detectors - Nal . i ; <- : . a « w"*i
Rots,rveni/Kg/doi-
^*M»,«?NM5, P^
Figure 6. Ratio of spin-dependent event rate to spinindependent event rate in 73Ge as function of LSP mass (upper left), of total (spin-dependent plus spinindependent) event rate (lower left part), and of scalar cross section of neutralino-proton interaction (lower right) obtained with O . K Qxho2 < 0 . 3 . The vertical line gives the expected sensitivity of GENIUS. In the region above the horizontal line the spin contribution dominates. T h e total event rate versus gaugino fraction of LSP is also given (upper right). (Prom 8 ).
i •:
=
• Drift chamber (diurnal modulation) DK11T • Nt* Iwiiz«ilif:i ni.'1'.vfur .Xp^roac'sM-
Indirect for' confirmation of DAMA. 2. Sensitivity to enter deeply into the range of SUSY predictions. Only very few experiments may become candidates for category 2 in a foreseeable future (see Figs.2,3), and as far as at present visible, of those only GENIUS will have the chance to search for modulation, i.e. to check, like DAMA, positive evidence for a dark matter signal. Figs.8,7 give an overview of present and future experiments. Present sensitivity limits are given in Figs.2,3. The at present most sensitive experiments DAMA 45 ' 46 , CDMS 47 (and Edelweiss17) are claimed 13 not to be fully consistent, although CDMS can at present not exclude the full DAMA evidence region 15>13. Some problems in the data analysis of CDMS have been revised recently 14 . One of the main problem of the cryodetectors is to obtain good numbers of background in the raw data, i.e. of the starting values for the
&r • » ; / • • - • - : . ' • ' . . - " "
- Superheated drtjplt'f
detection:
'•Viii-
The at present most sensitive experiments using raw data are the HEIDELBERG-MOSCOW experiment 24 and the HEIDELBERG Dark Matter Search Experiment (HDMS) 28 - 26 - 30 . HDMS uses a special configuration of Ge detectors, with an enriched 73 Ge detector embedded into a natural Ge-detector (Fig.9a). Results of the pilote project are given in 28 ' 26 . The final configuration with the enriched 73 Ge inner detector is in operation since February 2001. The experimental future of the field is also illustrated in Figs.2,3. A useful overview is given in 3 , and for Ge detectors in 21 . For an earlier review see 38 . The cryogenic projects are CDMS II, CREST II, Edelweiss II. In contrast to CDMS and Edelweiss which do its active back-
518
H. V. Klapdor-Kleingrothaus
Dark Matter Search
The Potencial of World Dark Matter Underground Experiments
"Rittt 96 llv« to» in **, Kpt *m HM A« p * wj, •«* w* «*
• MP- M
.H&tftr2(MMf CRESST-1
CaWO.
( KKSI:T !f
t'!|\H).
NO l 7 t t5Y k |? d Run iu t'WHsoim."
1.048
month*
at I? keV NO
Figure 8. Status of some running and future Dark Matter experiments.
DRIFT
HDMS x W
N
39.4 mm
a u
w
WIMP wind X^4 recoil ^ ' / V nuclei
H
!&>"
OS S-
12:00
o to s o «
a
H H
s££
s = Q
diurnal directional modulation
T3
O
u
O
(BOULBY mine) p.. a
o
a
o
•
aJ • cathode
as
Xe+CS-, gtu
^anodes
"3 c ^
Q
CSj'uDfis dnft
s a
(a)
Ge detector
(h)
a
a
a
q..a|
TPC with Xe f Art
Figure 9. HDMS and DRIFT experiments.
519
a
Dark Matter Search
H. V. Klapdor-Kleingrothaus ground rejection by looking for ionisation and phonons, CRESST II plans to use simultaneous detection of light and phonons. CDMS II plans to use 42 detectors with a total mass of 6.8 kg of Ge by 2006 in the Soudan mine, CRESST II plans to have - 1 0 kg of C d W 0 4 in the Gran Sasso in some future. The cryogenic experiments are, however, operating at present only 600 g of detectors or less, after a decade of development. CDMS has collected only 10.6 kg d of data over this time (in 1999, since then no measurement) 13 , Edelweiss only 4.53 kg d 17 . Therefore, they may have severe problems to expand their small detector masses to several tens of kg or better 100 kg, as required for modulation search. This means that although e.g. CDMS II may reach a future sensitivity in an exclusion plot as shown in Fig.2b, it will not be able to look for the modulation signal. A general problem in the present stage still seems to be the reproducibility of the highly complicated cryo detectors. In spite of this, phantasy is large enough, to dream already about 1 ton cryo detectors systems 13 . Other far future projects are the superheavy droplet detectors PICASSO/SIMPLE 3 . They are working at present on a scale of 15 and 50 g detectors. Their idea is to use 10-100/mi diameter droplets of volatile C4F10, C 3 F 8 , ... in metastable superheated condition and to choose critical energy and radius such that only nuclear recoils can trigger a phase transition, but not 7 and /? particles. The acoustic signal of the explosive bubble formation will be observed. The expected sensitivity of a 1 ton module for spin-dependent WIMPnucleon interaction is shown in Fig. 3 - as SDD 1 ton (for an assumed U / T h contamination of 10~ 15 g/g - U/Th a-emitters can cause recoil events!). A drawback is t h a t these detectors cannot measure energy spectra of WIMPs. A very promising project which would yield a nice signal identification, is DRIFT in the Boulby mine. It is aiming at looking for
the diurnal directional modulation (Fig.9b). The idea is to detect tracks of nuclear recoils in a TPC with Xe(Ar) by a multiwire readout. A 1 m 3 prototype (1.5 kg Xe) is under construction. It is seen as first component of the full 10m 3 DRIFT experiment 3 ' 4 . ( 1
1" f
Towards Data Acquisit "l
r
«jl
t
*
Hi**
. •j
f
t
<&' *'<•&
' >
, • " . * • * •
•"
St Boron Polyethelene
Figure 10. (a): Conceptual design of the Genius T F . Up to 14 detectors will be housed in the inner detector chamber, filled with liquid nitrogen. As a first shield 5 cm of zone refined Germanium, or extremely low-level copper will be used. Behind the 20 cm of polystyrene isolation another 35 cm of low level lead and a 15 cm borated polyethylene shield will complete the setup, (b): GENIUS: 100 kg of naked Ge detectors are suspended in a huge tank of liquid nitrogen.
The ZEPLIN project uses scintillation and electro-luminescence in two-phase xenon. Plans for ZEPLIN II are 0.01 - 0.1 520
Dark Matter Search
H. V. Klapdor-Kleingrothaus
counts/kg d and 20 kg of Xe. While still waiting for results of ZEPLIN I, plans are already discussed for a ZEPLIN IV 18 . To return to the more 'earth-bound' projects: DAMA will extend their mass to 250 kg, and plans to start operation in summer 200215. Also the NAIAD project (Boulby mine) plans to use Nal - 40-100 kg of Nal in a liquid scintillator Compton veto 19 . The projected NAIAD limits for lOOkgy exposure are 0.1c/kg d. Because of the large mass it will be possible to look for modulation. The HDMS experiment and the GENIUS-TF experiment 40 ' 41 - 28 aim at probing the DAMA evidence (see Fig.ll). GENIUS-TF consisting of 40 kg of Ge detectors in liquid nitrogen (Fig.10a) could also measure the modulation signal 41 ' 20 . Up to summer 2001, already 6 detectors of 2.5 kg each, with an extreme low-energy threshold of ~500eV have been produced. A similar potential is aimed at by the GEDEON project 21 , which plans to use 28 Ge diodes in one single cryostat. GENIUS-TF is already under installation in the Gran-Sasso laboratory and should start operation by end of 2002 41 . The probably most far reaching project is GENIUS 7 ' 31 (Fig.lOb). Since it is based on conventional techniques, using Ge detectors in liquid nitrogen, is may be realized in the most straightforward way. GENIUS would already in a first step, with 100 kg of natural Ge detectors in three years of measurement, cover a significant part of the SUSY parameter space for prediction of neutralinos as cold dark matter (Fig. 2). For this purpose the background in the energy range < 100 keV has to be reduced to 10~ 2 (events/kgykeV). At this level solar neutrinos as source of background are still negligible. Of particular importance is to shield the detectors during production (and transport) to keep the background from spallation by cosmic rays sufficiently low (for de521
•.
„
"lu
-
DAMA
., ' .- .
cnMS Edelweiss
!
10
Figure 11. The potential of GENIUS-T^wfoT [ S a r P matter search. WIMP-nucleon cross section limits as a function of the W I M P mass for spin-independent interactions. The solid lines are current limits of the HEIDELBERG-MOSCOW experiment, the HDMS, the DAMA and the CDMS experiments. T h e dashed curves are t h e expectation for HDMS, and for GeniusT F with an energy threshold of l l k e V and 2keV (no tritium contamination) respectively, and a background index of 2 events/kg y keV below 50 keV. T h e filled contour represents the evidence region of the DAMA experiment.
tails see 35 - 36 - 39 ). The sensitivity of GENIUS for Dark Matter corresponds to that obtainable with a 1 km 3 AMANDA detector for indirect detection (neutrinos from annihilation of neutralinos captured at the Sun) (see 44 ). Interestingly both experiments would probe different neutralino compositions: GENIUS mainly gaugino-dominated neutralinos, AMANDA mainly neutralinos with comparable gaugino and Higgsino components (see Fig. 38 in 44 ). 4
Hot Dark Matter Search
According to the recent indication for the neutrinoless mode of double beta decay 22 , neutrinos should still play an important role as hot dark matter in the Universe. The effective mass has been determined to be 22 (m)= (0.05 - 0.84) eV at a 95% c.l. (best value 0.39 eV) including an uncertainty of ±50% of the nuclear matrix elements. With the limit deduced for the effective neutrino mass, the HEIDELBERGMOSCOW experiment excludes several of the neutrino mass scenarios allowed from present neutrino oscillation experiments (see
Dark Matter Search
H. V. Klapdor-Kleingrothaus
BEST VALUE
2001 mmmmimmm0mmv^wmEm!ms&m;Kmmgmmmm"Pin'FL.BERQ-
ORE 600kg project
MlUSIOt project
Ilierurchy
I'arUiillv = •-. s.-« i.---r-.ic> |> t .«.i. n crar»
i'ii>;.i'Ikrur^r.
~ .-: r. >.
Figure 12. The impact of the evidence obtained for neutrinoless double b e t a decay 2 2 (best value of the effective neutrino mass (m) = 0.39 eV, 95% confidence range (0.05 - 0.84) eV - allowing already for an uncertainty of the nuclear matrix element of dr 50%, on possible neutrino mass schemes. T h e bars denote allowed ranges of (m) in different neutrino mass scenarios, still allowed by neutrino oscillation experiments. Hierarchical models are excluded by the new Ov/3/3 decay result. Also shown are the expected sensitivities for t h e future potential double beta experiments CUORE, MOON, EXO, and the 1 ton and 10 ton project of G E N I U S 3 1 ' 3 9 ' 3 6 (from 23 ).
Fig.12) - allowing mainly only for degener- dication for as anomalous magnetic moment ate and partially degenerate mass scenarios of the muon 50 . It lies in a range of interest and an inverse hierarchy 3v - scenario (the also for Z-burst models recently discussed as latter being, however, strongly disfavored by explanation for super-high energy cosmic ray a recent analysis of SN1987A). In particular events beyond the GKZ-cutoff49. The sensitivity of the present result is already in the hierarchical mass schemes are excluded. range to be probed by the satellite experiAssuming the degenerate scenarios to be ments MAP and PLANCK (Fig.13). realized in nature we fix - according to the 5 6 formulae derived in ' - the common mass The neutrino mass deduced allows neueigenvalue of the degenerate neutrinos to m trinos to still play an important role as hot = (0.05 - 3.4) eV. Part of the upper range dark matter in the Universe. is already excluded by tritium experiments, New approaches and considerably enlarged experiments (as discussed, e.g. which give a limit of m < 2.2 eV (95% c.l.) 51 . m7,3i,39,35,36,32,48-j w jjj ^e required in future The full range can only partly (down to ~ 0.5 eV) be checked by future tritium de- to fix the neutrino mass and the contribution cay experiments, but could be checked by of neutrinos to hot dark matter with higher some future (3(3 experiments (see, e.g. 7 ' 39,36 ). accuracy. The deduced best value for the mass is conAgain GENIUS is the most promising of sistent with expectations from experimental them. With a mass of 1 ton of enriched 76 Ge /i —* ej branching limits in models assuming it would cover the sensitivity range of the efthe generating mechanism for the neutrino fective mass (m) down to 0.02 eV. Already mass to be also responsible for the recent in100 kg of enriched Ge would be sufficient to
522
Dark Matter Search
H. V. Klapdor-Kleingrothaus References m,(eV) HEIDELBERG-MOSCOW Positive EVIDENCE
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1 tan29
Figure 13. Double beta decay observable (m) and oscillation parameters: The case for degenerate neutrinos. Plotted on the axes are the overall scale of neutrino masses mo and mixing t a n 2 912. Also shown is a cosmological bound deduced from a fit of CMB and large scale structure 3 3 and the expected sensitivity of the satellite experiments MAP and PLANCK. T h e present limit from tritium /3 decay of 2.2 e V 5 1 would lie near the top of the figure. T h e range of (m) fixed by the HEIDELBERG-MOSCOW experiment is, in the case of small solar neutrino mixing, already in the range to be explored by MAP and PLANCK.
reach a sensitivity down to 0.04eV 7 .
5
Conclusion
Dark matter search is presently one of the most exciting fields of particle physics and cosmology. Underground experiments at present only marginally touch in their sensitivity the range of present SUSY predictions for cold dark matter. Of future experiments the GENIUS project has the best prospects to cover a large part of the predicted range. GENIUS will provide information complementary to future collider search. This information is indispensable, even if LHC would find supersymmetry, since in any case it still has to be shown that SUSY particles indeed form the cold dark matter in the Universe. GENIUS will simultaneously be the most straightforward way to fix the neutrino mass and the contributions of neutrinos to hot dark matter with higher accuracy.
523
1. K. Freese et al. in Proc. of DARK'98, Heidelberg, July 1998, IOP, Bristol (1999) eds. H.V. Klapdor-Kleingrothaus et al. 352; D. Graff, in Proc. of DARK 2000, Heidelberg, July 2000, Springer, Heidelberg (2001) ed. H.V. KlapdorKleingrothaus, 352 and K. Freese et al. in Proc. of DM'2000, WS (2001) 213 223 and astro-ph/'0007'444. 2. S. Asztalos et a l , Phys. Rev. D 64 (2001) 092003; P. Sikivie in Proc. of BEYOND'99, Castle Ringberg, Germany, June 1999, edited by H.V. KlapdorKleingrothaus and I. V. Krivosheina, IOP, Bristol 816-828 (2000) 3. V. Zacek NPB, in Proc. of NEUTRINO2000, Sudbury, Canada, June 2000, ed. J. Law et al. (2001), Nucl. Phys. Proc. Suppl. 91 (2001) 368. 4. M.J. Lehner et al., in Proc.of DARK'2000, Heidelberg, July 2000, Springer, Heidelberg (2001), ed. H.V. Klapdor-Kleingrothaus, 590597, C.J. Martoff, DRIFT Collab. in Proc. York 2000, ed. J.C. Spooner et al. 463-474. 5. H.V. Klapdor-Kleingrothaus, H. Pas and A.Yu. Smirnov, Preprint: hepp / i / 0 0 0 3 2 1 9 . (2000) and in Phys. Rev. D (2001). 6. H.V. Klapdor-Kleingrothaus, H. Pas and A.Yu. Smirnov, in Proc. of DARK2000, Heidelberg, 10-15 July, 2000, Germany, ed. H.V. Klapdor-Kleingrothaus, Springer, Heidelberg (2001) 420-434. 7. H.V. Klapdor-Kleingrothaus, "60 Years of Double Beta Decay", World Scientific, Singapore (2001) 1253 p. 8. V.A. Bednyakov and H.V. KlapdorKleingrothaus, Phys. Rev. D 62 (2000) 043524/1, and Phys. Rev. D 63 (2001) 095005. 9. J. Ellis et al. Phys. Lett. B 481 (2000) 304 and Phys. Rev. D 63 (2000) 065016.
Dark Matter Search
H. V. Klapdor-Kleingrothaus 10. J. Ellis et al., hep-ph/0111294. 11. R. Arnowitt, private communication, 2001 12. M. Brhlik in Proc. of DARK'98, Heidelberg, July 1998, IOP, Bristol (1999) eds. H. V. Klapdor-Kleingrothaus et al. 499-515. 13. R.J. Gaitskell, NANP2000, Dubna, Russia, June 2000, to be publ. in Nuclei, Letters (2002). 14. B. Sadoulet in Proc. of TAUP'2001, September 2001, Gran Sasso, Italy, ed. A. Bettini. 15. R. Bernabei et al. in Proc. of TAUP'2001, September 2001, Gran Sasso, Italy, ed. A. Bettini and of Erice'2001, September 2001, Erice, Italy, ed. A. Faessler. 16. F. Probst in Proc. of TAUP'2001, September 2001, Gran Sasso, Italy, ed. A. Bettini. 17. J. Gascon et al. in Proc.of DARK'2000, Heidelberg, July 2000, Springer, Heidelberg (2001), ed. H.V. KlapdorKleingrothaus, 575-580 and A. de Lesquen et al. in Proc. of NANP2000, Dubna, Russia, June 2000, ed. V. Bednjakov, to be publ. in Nuclei, Letters (2002). 18. D.B. Cline, astro-ph/0111098. 19. J.C. Spooner in Proc. of York'2000, ed. J.C. Spooner et al., World Scientific, Singapore (2001). 20. H.V. Klapdor-Kleingrothaus et al., subm for publ.. 21. A. Morales. , hep-ex/0111089. 22. H.V. Klapdor-Kleingrothaus, A. Dietz, H.L. Harney and I.V. Krivosheina; Modern Physics Letters A 16, N o . 37 (2001) 2409-2420. 23. H.V. Klapdor-Kleingrothaus and U. Sarkar, Modern Physics Letters A 16, N o . 38 (2001) 2469-2482. 24. HEIDELBERG-MOSCOW Collaboration, Phys. Rev. D 59, 022001 (1998).
25. A. Staudt, K. Muto and H.V. KlapdorKleingrothaus, .EWop/i.Lett.l3(1990)31. 26. L. Baudis, A. Dietz, B. Majorovits, F. Schwamm, H. Strecker and H.V. Klapdor-Kleingrothaus, Phys. Rev. D 63 , 022001 (2000). 27. Y. Ramachers for the CRESST Collaboration in Proc. of Xlth Rencontres de Blois, Frontiers of Matter, France, June 27-July 3, 1999. 28. H.V. Klapdor-Kleingrothaus et al. in Proc. of DARK2000, Heidelberg, Germany, July 10-15, 2000, Springer, Heidelberg (2001), ed. H.V. KlapdorKleingrothaus 553- 568. 29. H.V. Klapdor-Kleingrothaus in Proc. of BEYOND'97, Castle Ringberg, Germany, 8-14 June 1997, edited by H.V. Klapdor-Kleingrothaus and H. Pas, IOP Bristol 485-531 (1998) 30. H.V. Klapdor-Kleingrothaus and B. Majorovits, in Proc. of York'2000, ed. J.C. Spooner et al., World Scientific, Singapore (2001). 31. H.V. Klapdor-Kleingrothaus et al. M P I - R e p o r t MPI-H-V26-1999 and Preprint: hep-ph/9910205 and in Proc. of BEYOND'99, Castle Ringberg, Germany, 6-12 June 1999, edited by H. V. Klapdor-Kleingrothaus and I. V. Krivosheina, IOP Bristol, 915 - 1014 (2000). 32. H.V. Klapdor-Kleingrothaus, in Proc. of NEUTRINO 98, Takayama, Japan, 4-9 Jun 1998, (eds) Y. Suzuki et al. Nucl. Phys. Proc. Suppl. 77 (1999) 357. 33. R.E. Lopez, astro-ph/99094:14; J.R. Primack and M.A.K. Gross, astro-ph/0007165; J.R. Primack, astro-ph/0007187; J. Einasto, in Proc. of DARK2000, Heidelberg, Germany, July 10-15, 2000, Ed. H.V. Klapdor-Kleingrothaus, Springer, Heidelberg, (2001) 3-11. 34. HEIDELBERG-MOSCOW Coll., Phys. Rev. Lett. 8 3 (1999) 41-44.
524
H. V. Klapdor-Kleingrothaus
Dark Matter Search
35. H.V. Klapdor-Kleingrothaus, in Proc. of Int. Conference NOW2000, Otranto, Italy, Sept. 2000, Nucl. Phys. B (2001) ed. G. Fogli. 36. H.V. Klapdor-Kleingrothaus, in Proc. of NOON2000, Tokyo, Dec. 2000, World Scientific, Singapore (2001). 37. M. Tegmark, M. Zaldarriaga and A.J.S. Hamilton, Preprint: /iep-p/i/0008145. 38. H.V. Klapdor-Kleingrothaus and Y. Ramachers, Eur. Phys. J. A 3 (1998) 85. 39. H.V. Klapdor-Kleingrothaus, in Proc. of LowNu2, Dec. 4-5. ed: Y. Suzuki et al. World Scientific, Singapore (2001). 40. H.V. Klapdor-Kleingrothaus, L.Baudis, A.Dietz, G.Heusser, I.Krivosheina, B.Majorovits, H. Strecker, et al. Internal Report MPI-H-V32-2000. 41. H.V. Klapdor-Kleingrothaus et al., hepea:/0103082, sub. for publication. 42. E. Fiorini et al., Phys. Rep. 307, 309 (1998). 43. A. Alessandrello et al., Phys. Lett. B 486 (2000) 13-21. 44. J. Edsjo, home page: http://www.physto.se/edsjo/ 45. R. Bernabei et al., Nucl. Phys. B 70 (Proc. Suppl) (1998) 79. 46. R. Bernabei et al., Phys. Lett. B 424 (1998) 195, Phys. Lett. B 450 (1999) 448, Phys. Lett. B 480 (2000) 23. 47. R. Abusaidi et al. (CDMS Collaboration), Nucl. Instrum. Meth. A 444 (2000) 345, Phys. Rev. Lett. 84 (2000) 5699-5703. 48. H.V. Klapdor-Kleingrothaus, in Proc. of NANPino-2000, Dubna, July 19-22, 2000, Part, and Nucl. Lett. iss. 1/2 (2001) 20-39 and hep-ph/ 0102319. 49. T.J. Weiler, in Proc. BEYOND'99 Tegernsee, Germany, 6-12 Juni 1999, eds. H.V. Klapdor-Kleingrothaus and I.V. Krivosheina, IOP Bristol, (2000) 1085-1106; H. Pas and T.J. Weiler, Phys. Rev. D 63 (2001) 113015. 50. E. Ma and M. Raidal, Phys. Rev. Lett.
525
87 (2001) 011802; Erratum-ibid. 87 (2001) 159901 and hep-ph/0102255. 51. C. Weinheimer, in Proc. of DARK'2000, Heidelberg, Germany, 10-16 July, 2000, ed. H.V. Klapdor-Kleingrothaus, Springer, Heidelberg (2001) 513 - 519.
T H E HIGHEST E N E R G Y COSMIC RAYS, G A M M A - R A Y S A N D N E U T R I N O S : FACTS, F A N C Y A N D R E S O L U T I O N FRANCIS HALZEN Department of Physics, University of Wisconsin, Madison, WI53706,
USA
Although cosmic rays were discovered 90 years ago, we do not know how and where they are accelerated. There is compelling evidence t h a t the highest energy cosmic rays are extra-galactic — they cannot be contained by our galaxy's magnetic field anyway because their gyroradius exceeds its dimensions. Elementary elementary-particle physics dictates a universal upper limit on their energy of 5 x 10 1 9 eV, the so-called Greisen-Kuzmin-Zatsepin cutoff; however, particles in excess of this energy have been observed, adding one more puzzle to the cosmic ray mystery. Mystery is nonetheless fertile ground for progress: we will review the facts and mention some very speculative interpretations. There is indeed a realistic hope t h a t the oldest problem in astronomy will be resolved soon by ambitious experimentation: air shower arrays of 10 4 km 2 area, arrays of air Cerenkov detectors and kilometer-scale neutrino observatories.
1
The New A s t r o n o m y
Conventional astronomy spans 60 octaves in photon frequency, from 10 4 cm radio-waves to 10 _ 1 4 cm photons of GeV energy; see Fig. 1. This is an amazing expansion of the power of our eyes which scan the sky over less than a single octave just above 1 0 - 5 cm wavelength. The new astronomy, discussed in this talk, probes the Universe with new wavelengths, smaller than 10~ 1 4 cm, or photon energies larger than 10 GeV. Besides gamma rays, gravitational waves and neutrinos as well as very high energy protons that are only weakly deflected by the magnetic field of our galaxy, become astronomical messengers from the Universe. As exemplified time and again, the development of novel ways of looking into space invariably results in the discovery of unanticipated phenomena. As is the case with new accelerators, observing the predicted will be slightly disappointing. Why do high energy astronomy with neutrinos or protons despite the considerable instrumental challenges which we will discuss further on? A mundane reason is that the Universe is not transparent to photons of TeV energy and above (units are: GeV/TeV/PeV/EeV/ZeV in ascending factors of 10 3 ). For instance, a PeV energy photon 7 cannot reach us from a source at the
edge of our own galaxy because it will annihilate into an electron pair in an encounter with a 2.7 degree Kelvin microwave photon 7CMB before reaching our telescope. Energetic photons are absorbed on background light by pair production 7+7bk g nd —• e+ + e~~ of electrons above a threshold E given by 4Ee ~ (2m e ) 2 ,
(1)
where E and e are the energy of the highenergy and background photon, respectively. Eq. (1) implies that TeV-photons are absorbed on infrared light, PeV photons on the cosmic microwave background and EeV photons on radio-waves. Only neutrinos can reach us without attenuation from the edge of the Universe. At EeV energies proton astronomy may be possible. Near 50 EeV and above, the arrival directions of electrically charged cosmic rays are no longer scrambled by the ambient magnetic field of our own galaxy. They point back to their sources with an accuracy determined by their gyroradius in the intergalactic magnetic field B: d
dB
-"-gyro
-^
where d is the distance to the source. Scaled
526
Francis Halzen
Facts, Fancy and Resolution lo,[Flux/(«,enr«,-»-)] -8 1 2Zl' -
-4 1 ' 1 '
^
0 4 8 12 1 < 1 ' ! ' 1 ' I i 1 ' 1 ' ) '"1
A*»
r^
^ 16 20 ' 1 i 1 I 1™
/
-; =
~
/
T
™
-
yTeV sources!.:
r
V \% cosmic J / \ \ rays z
i
/ / -20 " i " t ....! ..!,.
i i l . i l l . i l
r. *-
-
/ % I 1 • - 112 . 1 , 1 . I .-20 ! . 1 .-24 1 1
log{X/cm) Figure 1. The diffuse flux of photons in the Universe, from radio waves to GeV-photons. Above tens of GeV only limits are reported although individual sources emitting TeV gamma-rays have been identified. Above GeV energy cosmic rays dominate t h e spectrum.
to units relevant to the problem, 6 ^ \l MpcJ VlO-9G/ ( E \ 0.1° \3xl0
20
(3)
eV J
Speculations on the strength of the intergalactic magnetic field range from 1 0 - 7 to 10~ 12 Gauss. For a distance of 100 Mpc, the resolution may therefore be anywhere from sub-degree to nonexistent. It is still reasonable to expect that the arrival directions of the highest energy cosmic rays provide information on the location of their sources. Proton astronomy should be possible; it may also provide indirect information on intergalactic magnetic fields. Determining their strength by conventional astronomical means has turned out to be challenging. 2
The Highest Energy Cosmic Rays: Facts
In October 1991, the Fly's Eye cosmic ray detector recorded an event of energy 3.0 ±554 xl0 2 0 eV. 1 This event, together with an
527
event recorded by the Yakutsk air shower array in May 1989,2 of estimated energy ~ 2 x 10 20 eV, constituted at the time the two highest energy cosmic rays ever seen. Their energy corresponds to a center of mass energy of the order of 700 TeV or ~ 50 Joules, almost 50 times LHC energy. In fact, all experiments 3 have detected cosmic rays in the vicinity of 100 EeV since their discovery by the Haverah Park air shower array. 4 The AG AS A air shower array in Japan 5 has by now accumulated an impressive 10 events with energy in excess of IO20 eV.6 How well experiments can determine the energy of these events is a critical issue. With a particle flux of order 1 event per km 2 per century, these events can only be studied by using the earth's atmosphere as a particle detector. The experimental signatures of a shower initiated by a cosmic particle are illustrated in the cartoon shown in Fig. 2. The primary particle creates an electromagnetic and hadronic cascade. The electromagnetic shower grows to a shower maximum, and is
Francis Halzen
Facts, Fancy and Resolution
X max •max
N L^max
Ne
Figure 2. Particles interacting near the top of the atmosphere initiate an electromagnetic and hadronic particle cascade. Its profile is shown on the right. The different detection methods are illustrated. Mirrors collect the Cerenkov and nitrogen fluorescent light, arrays of detectors sample the shower reaching the ground, and underground detectors identify the muon component of the shower.
subsequently absorbed by the atmosphere. This leads to the characteristic shower profile shown on the right hand side of the figure. The shower can be observed by: i) sampling the electromagnetic and hadronic components when they reach the ground with an array of particle detectors such as scintillators, ii) detecting the fluorescent light emitted by atmospheric nitrogen excited by the passage of the shower particles, iii) detecting the Cerenkov light emitted by the large number of particles at shower maximum, and iv) detecting muons and neutrinos under-
ground. Fluorescent and Cerenkov light is collected by large mirrors and recorded by arrays of photomultipliers in their focus. The bottom line on energy measurement is that, at this time, several experiments using the first two techniques agree on the energy of EeV-showers within a typical resolution of 25%. Additionally, there is a systematic error of order 10% associated with the modeling of the showers. All techniques are indeed subject to the ambiguity of particle simulations that involve physics beyond LHC. If the final outcome turns out to be erroneous inference
528
Facts, Fancy and Resolution
Francis Halzen of the energy of the shower because of new physics associated with particle interactions, we will be happy to contemplate this discovery instead. Whether the error in the energy measurement could be significantly larger is a key question to which the answer is almost certainly negative. A variety of techniques have been developed to overcome the fact that conventional air shower arrays do calorimetry by sampling at a single depth. They give results within the range already mentioned. So do the fluorescence experiments that embody continuous sampling calorimetry. The latter are subject to understanding the transmission of fluorescent light in the dark night atmosphere — a challenging problem given its variation with weather. Stereo fluorescence detectors will eliminate this last hurdle by doing two redundant measurements of the same shower from different locations. The HiRes collaborators have one year of data on tape which should allow them to settle any doubts as to energy calibration once and for all.
i i i i |
-|—i
1
1—i
i i i i i |
r
^
5
—*-^jk r LU - I D * 111
Uniform sources
23|_J
i i i i i I
l
l
l
i i i i i I
1D
**.
i
10
Energy [eV]
HiRes-1 m o n o E > I D 1 " eV HiBes-2 m o n o 10"-° eV < E < 10 ,fll eV o r i g i n a l FE s t e r e o s p e c t r u m
S 10'"
1 1
^ m & * <J *
10"
The premier experiments, HiRes and AG AS A, agree that cosmic rays with energy in excess of 10 EeV are not a feature of our galaxy and that their spectrum extends beyond 100 EeV. They disagree on almost everything else. The AGASA experiment claims evidence that they come from point sources, and that they are mostly heavy nuclei. The HiRes data do not support this. Because of statistics, interpreting the measured fluxes as a function of energy is like reading tea leaves; one cannot help however reading different messages in the spectra (see Fig. 3). More about that later. 3
3.1
10"
10"
10" Energy (eV)
10"
10"
Figure 3. The cosmic ray spectrum peaks in the vicinity of 1 GeV and has features near 10 1 5 and 10 1 9 eV. They are referred to as the "knee" and "ankle" in the spectrum. Shown is the flux of the highest energy cosmic rays near and beyond the ankle measured by t h e AGASA and HiRes experiments.
larger than the gyroradius of the particle:
R > Rgyro = I •
(4)
I.e. the accelerating magnetic field must contain the particle orbit. This condition yields a maximum energy
T h e Highest E n e r g y Cosmic R a y s : Fancy
E = TBR
Acceleration to > 100 EeV?
It is sensible to assume that, in order to accelerate a proton to energy E in a magnetic field -B, the size R of the accelerator must be 529
(5)
by dimensional analysis and nothing more. The T-factor has been included to allow for the possibility that we may not be at rest
Francis Halzen
Facts, Fancy and Resolution
Conditions with E ~ 10 EeV • quasars flil B S 103 G M S 10 9 M sun 3 • blasars T > 10 5 ^ 10 G M ^ 109Afsun • neutron stars T ^ 1 B ^ 10 12 G M ^ M s u n black holes • grb
r > 102
5 =* 10 12 G
in the frame of the cosmic accelerator resulting in the observation of boosted particle energies. Theorists' imagination regarding the accelerators is limited to dense regions where exceptional gravitational forces create relativistic particle flows: the dense cores of exploding stars, inflows on supermassive black holes at the centers of active galaxies, annihilating black holes or neutron stars? All speculations involve collapsed objects and we can therefore replace R by the Schwartzschild radius R ~ GM/c2
(6)
E ~ TBM .
(7)
to obtain
Given the microgauss magnetic field of our galaxy, no structures are large or massive enough to reach the energies of the highest energy cosmic rays. Dimensional analysis therefore limits their sources to extragalactic objects; a few common speculations are listed in Table 1. Nearby active galactic nuclei distant by ~ 100 Mpc and powered by a billion solar mass black holes are candidates. With kilo-Gauss fields we reach 100 EeV. The jets (blazars) emitted by the central black hole could reach similar energies in accelerating substructures boosted in our direction by a T-factor of 10, possibly higher. The neutron star or black hole remnant of a collapsing supermassive star could support magnetic fields of 1012 Gauss, possibly larger. Shocks with T > 102 emanating from the collapsed black hole could be the origin of gamma ray bursts and, possibly, the source of the highest energy cosmic rays.
M S Msun
The above speculations are reinforced by the fact that the sources listed happen to also be the sources of the highest energy gamma rays observed. At this point however a reality check is in order. Let me first point out that the above dimensional analysis applies to the Fermilab accelerator: lOkGauss fields over several kilometers yield 1 TeV. The argument holds because, with optimized design and perfect alignment of magnets, the accelerator reaches efficiencies matching the dimensional limit. It is highly questionable that Nature can achieve this feat. Theorists can imagine acceleration in shocks with efficiency of perhaps 10%. The astrophysics problem is so daunting that many believe that cosmic rays are not the beam of cosmic accelerators but the decay products of remnants from the early Universe, for instance topological defects associated with a grand unified GUT phase transition. A topological defect will suffer a chain decay into GUT particles X,Y, that subsequently decay to familiar weak bosons, leptons and quark- or gluon jets. Cosmic rays are the fragmentation products of these jets. We know from accelerator studies that, among the fragmentation products of jets, neutral pions (decaying into photons) dominate protons by two orders of magnitude. Therefore, if the decay of topological defects is the source of the highest energy cosmic rays, they must be photons. This is a problem because the highest energy event observed by the Fly's Eye is not likely to be a photon. 7 A photon of 300 EeV will interact with the magnetic field of the earth far above
530
Francis Halzen
Facts, Fancy and Resolution
the atmosphere and disintegrate into lower energy cascades — roughly ten at this particular energy. The measured shower profile of the event does not support this assumption; see Fig. 4. One can live and die by a single event!
The major source of proton energy loss is photoproduction of pions on a target of cosmic microwave photons of energy e. The Universe is therefore also opaque to the highest energy cosmic rays, with an absorption length: A7P =
O c M B 0>+7CME
^ lOMpc,
Figure 4. The composite atmospheric shower profile of a 3 x 10 2 0 eV 7-ray shower calculated with Landau-Pomeranchuk-Migdal (solid) and BetheHeitler (dashed) electromagnetic cross sections. T h e central line shows the average shower profile and the upper and lower lines show 1 a deviations — not visible for the BH case, where lines overlap. The experimental shower profile is shown along with the d a t a points. It does not fit the profile of a photon shower.
3.2
Are Cosmic Rays Really Protons: the GZK Cutoff?
(9) (10)
or only tens of megaparsecs when their energy exceeds 50 EeV. This so-called GZK cutoff establishes a universal upper limit on the energy of the cosmic rays. The cutoff is robust, depending only on two known numbers: ncMB = 400 c m - 3 and o>+ 7cMB = 10- 2 8 cm 2 . Protons with energy in excess of 100 EeV, emitted in distant quasars and gamma ray bursts, will have lost their energy to pions before reaching our detectors. They have, nevertheless, been observed, as we have previously discussed. They do not point to any sources within the GZK-horizon however, i.e. to sources in our local cluster of galaxies. There are three possible resolutions: i) the protons are accelerated in nearby sources, ii) they do reach us from distant sources which accelerate them to much higher energies than we observe, thus exacerbating the acceleration problem, or iii) the highest energy cosmic rays are not protons.
The first possibility raises the challenge of finding an appropriate accelerator by conAll experimental signatures agree on the parfining these already unimaginable sources to ticle nature of the cosmic rays — they look our local galaxy cluster. It is not impossible like protons, or, possibly, nuclei. We menthat all cosmic rays are produced by the actioned at the beginning of this article that tive galaxy M87, or by a nearby gamma ray the Universe is opaque to photons with enburst which exploded a few hundred years ergy in excess of tens of TeV because they ago. The sources identified by the AGASA annihilate into electron pairs in interactions array do not correlate however with any such with background light. Also protons intercandidates. act with background light, predominantly Stecker8 has speculated that the highby photoproduction of the A-resonance, i.e. est energy cosmic rays are Fe nuclei with a V + ICMB —> A —> TX + p above a threshold delayed GZK cutoff. The details are compienergy Ep of about 50 EeV given by: cated but the relevant quantity in the problem is 7 = E/AM, where A is the atomic 2Epe > (m\ (8) 531
Francis Halzen
Facts, Fancy and Resolution
number and M the nucleon mass. For a fixed observed energy, the smallest boost above GZK threshold is associated with the largest atomic mass, i.e. Fe.
3.3
magnitude short of the strong cross sections required to make a neutrino interact in the upper atmosphere to create an air shower. Could EeV neutrinos be strongly interacting because of new physics? In theories with TeV-scale gravity one can imagine that graviton exchange dominates all interactions and thus erases the difference between quarks and neutrinos at the energies under consideration. Notice however that the actual models performing this feat require a fast turn-on of the cross section with energy that violates Swave unitarity. 9 We thus exhausted the possibilities: neutrons, muons and other candidate primaries one may think of are unstable. EeV neutrons barely live long enough to reach us from sources at the edge of our galaxy.
Could Cosmic Rays be Photons or Neutrinos?
When discussing topological defects, I already challenged the possibility that the original Fly's Eye event is a photon. The detector collects light produced by the fluorescence of atmospheric nitrogen along the path of the high-energy shower traversing the atmosphere. The anticipated shower profile of a 300 EeV photon is shown in Fig. 4. It disagrees with the data. The observed shower profile roughly fits that of a primary proton, or, possibly, that of a nucleus. The shower profile information is however sufficient to conclude that the event is unlikely to be of photon origin. The same conclusion is reached for the Yakutsk event that is characterized by a huge number of secondary muons, inconsistent with an electromagnetic cascade initiated by a gamma-ray. Finally, the AGASA collaboration claims evidence for "point" sources above 10 EeV. The arrival directions are however smeared out in a way consistent with primaries deflected by the galactic magnetic field. Again, this indicates charged primaries and excludes photons. Neutrino primaries are definitely ruled out. Standard model neutrino physics is understood, even for EeV energy. The average x of the parton mediating the neutrino interaction is of order x ~ ^jM^js ~ 1 0 - 6 so that the perturbative result for the neutrinonucleus cross section is calculable from measured HERA structure functions. Even at 100 EeV a reliable value of the cross section can be obtained based on QCD-inspired extrapolations of the structure function. The neutrino cross section is known to better than an order of magnitude. It falls 5 orders of
4
A Three Prong Assault on the Cosmic Ray Puzzle
We conclude that, where the highest energy cosmic rays are concerned, both the accelerator mechanism and the particle physics are totally enigmatic. The mystery has inspired a worldwide effort to tackle the problem with novel experimentation in three complementary areas of research: air shower detection, atmospheric Cerenkov astronomy and underground neutrino physics. While some of the future instruments have other missions, all are likely to have a major impact on cosmic ray physics. 4-1
Giant Cosmic Ray Detectors
With super-GZK fluxes of the order of a single event per kilometer-squared per century, the outstanding problem is the lack of statistics; see Fig. 3. In the next five years, a qualitative improvement can be expected from the operation of the HiRes fluorescence detector in Utah. With improved instrumentation yielding high quality data from 2 detectors operated in coincidence, the interplay between sky transparency and energy measure-
532
Francis Halzen
Facts, Fancy and Resolution
ment can be studied in detail. We can safely anticipate that the existence of super-Greisen energies will be conclusively demonstrated by using the instrument's calorimetric measurements. A mostly Japanese collaboration has proposed a next-generation fluorescence detector, the Telescope Array. The Auger air shower array is tackling the low rate problem with a huge collection area covering 3000 square kilometers on an elevated plain in Western Argentina. The instrumentation consists of 1600 water Cerenkov detectors spaced by 1.5 km. For calibration, about 15 percent of the showers occurring at night will be viewed by 3 HiResstyle fluorescence detectors. The detector will observe several thousand events per year above lOEeV and tens above 100 EeV, with the exact numbers depending on the detailed shape of the observed spectrum which is at present a matter of speculation; see Fig. 3.
4-2
Gamma-Rays from Cosmic Accelerators
An alternative way to identify the sources of the cosmic rays is illustrated in Fig. 5. The cartoon draws our attention to the fact that cosmic accelerators are also cosmic beam dumps producing secondary photon and neutrino beams. Accelerating particles to TeV energy and above requires high-speed, massive bulk flows. These are likely to have their origin in exceptional gravitational forces associated with dense cores of exploding stars, inflows onto supermassive black holes at the centers of active galaxies, annihilating black holes or neutron stars. In such situations, accelerated particles are likely to pass through intense radiation fields or dense clouds of gas leading to production of secondary photons and neutrinos that accompany the primary cosmic-ray beam. An example of an electromagnetic beam dump is the X-ray radiation fields surrounding the central black holes of active galaxies. The target material, whether
533
a gas or particles or of photons, is likely to be sufficiently tenuous so that the primary beam and the photon beam are only partially attenuated. However, it is also a real possibility that one could have a shrouded source from which only the neutrinos can emerge, as in terrestrial beam dumps at CERN and Fermilab. The astronomy event of the 21st century could be the simultaneous observation of TeV-gamma rays, neutrinos and gravitational waves from cataclysmic events associated with the source of the cosmic rays. We first concentrate on the possibility of detecting high-energy photon beams. After two decades, ground-based gamma ray astronomy has become a mature science. 10 A large mirror, viewed by an array of photomultipliers, collects the Cerenkov light emitted by air showers and images the showers in order to determine the arrival direction as well as the nature of the primary particle; see Fig. 2. These experiments have opened a new window in astronomy by extending the photon spectrum to 20 TeV, possibly beyond. Observations have revealed spectacular TeVemission from galactic supernova remnants and nearby quasars, some of which emit most of their energy in very short burst of TeVphotons. But there is the dog that didn't bark. No evidence has emerged for n° origin of the TeV radiation and, therefore, no cosmic ray sources have yet been identified. Dedicated searches for photon beams from suspected cosmic ray sources, such as the supernova remnants IC433 and 7-Cygni, came up empty handed. While not relevant to the topic covered by this talk, supernova remnants are theorized to be the sources of the bulk of the cosmic rays that are of galactic origin. The evidence is still circumstantial. The field of gamma ray astronomy is buzzing with activity to construct secondgeneration instruments. Space-based detectors are extending their reach from GeV
Francis Halzen
Facts, Fancy a n d Resolution
NEUTRINO BEAMS: HEAVEN & EARTH
o accelerator
•> black hole
P
© target
•> radiation enveloping black hole directional beam
magnetic fields Figure 5.
t o TeV energy with A M S and, especially, GLAST, while the ground-based Cerenkov collaborations are designing instruments with lower thresholds. In t h e not so far future b o t h techniques should generate overlapping measurements in the 1 0 ~ 1 0 2 G e V energy range. All ground-based air Cerenkov experiments aim at lower threshold, b e t t e r angular- and energy-resolution, and a longer d u t y cycle. One can however identify t h r e e pathways t o reach these goals: 1. larger mirror area, exploiting t h e parasitic use of solar collectors during nighttime ( C E L E S T E , STACEY and SOLARII),11
is a n a r r a y of 9 upgraded Whipple telescopes, each with a field of view of 6 degrees. These can be operated in coincidence for improved angular resolution, or be pointed at 9 different 6 degree bins in t h e night sky, t h u s achieving a large field of view. T h e H E G R A collaboration 1 4 is already operating four telescopes in coincidence a n d is building an upgraded facility with excellent viewing a n d optimal location near t h e equator in Namibia. T h e r e is a dark horse in this race: Milagro. 1 5 T h e Milagro idea is t o lower t h e threshold of conventional air shower arrays to 100 GeV by instrumenting a pond of five million gallons of ultra-pure water with photomultipliers. For time-varying signals, such as bursts, t h e threshold may be lower.
2. better, or rather, u l t i m a t e imaging w i t h t h e 17 m M A G I C mirror, 1 2 4-3 3. larger field of view using multiple telescopes (VERITAS, H E G R A and HESS). T h e Whipple telescope pioneered t h e atmospheric Cerenkov technique. V E R I T A S 1 3
High Energy Neutrino
Telescopes
Although neutrino telescopes have multiple interdisciplinary science missions, t h e search for t h e sources of t h e highest-energy cosmic rays stands o u t because it clearly iden-
534
Francis Halzen
Facts, Fancy and Resolution
tifies the size of the detector required to do the science.16 For guidance in estimating expected signals, one makes use of data covering the highest-energy cosmic rays in Fig. 3 as well as known sources of non-thermal, highenergy gamma rays. Accelerating particles to TeV energy and above involves neutron stars or black holes. As already explained in the context of Fig. 5, some fraction of them will interact in the radiation fields surrounding the source, whatever it may be, to produce pions. These interactions may also be hadronic collisions with ambient gas. In either case, the neutral pions decay to photons while charged pions include neutrinos among their decay products with spectra related to the observed gamma-ray spectra. Estimates based on this relationship show that a kilometer-scale detector is needed to see neutrino signals. The same conclusion is reached in specific models. Assuming, for instance, that gamma ray bursts are the cosmic accelerators of the highest-energy cosmic rays, one can calculate from textbook particle physics how many neutrinos are produced when the particle beam coexists with the observed MeV energy photons in the original fireball. We thus predict the observation of 10-100 neutrinos of PeV energy per year in a detector with a kilometer-square effective area. In general, the potential scientific payoff of doing neutrino astronomy arises from the great penetrating power of neutrinos, which allows them to emerge from dense inner regions of energetic sources. Whereas the science is compelling, the real challenge has been to develop a reliable, expandable and affordable detector technology. Suggestions to use a large volume of deep ocean water for high-energy neutrino astronomy were made as early as the 1960s. In the case of the muon neutrino, for instance, the neutrino (i/^) interacts with a hydrogen or oxygen nucleus in the water and produces a muon travelling in nearly the same direction as the neutrino. The blue Cerenkov
535
light emitted along the muon's ~kilometerlong trajectory is detected by strings of photomultiplier tubes deployed deep below the surface. With the first observation of neutrinos in the Lake Baikal and the (under-ice) South Pole neutrino telescopes, there is optimism that the technological challenges to build neutrino telescopes have been met. The first generation of neutrino telescopes, launched by the bold decision of the DUMAND collaboration to construct such an instrument, are designed to reach a large telescope area and detection volume for a neutrino threshold of order 10 GeV. The optical requirements of the detector medium are severe. A large absorption length is required because it determines the spacings of the optical sensors and, to a significant extent, the cost of the detector. A long scattering length is needed to preserve the geometry of the Cerenkov pattern. Nature has been kind and offered ice and water as adequate natural Cerenkov media. Their optical properties are, in fact, complementary. Water and ice have similar attenuation length, with the role of scattering and absorption reversed. Optics seems, at present, to drive the evolution of ice and water detectors in predictable directions: towards very large telescope area in ice exploiting the long absorption length, and towards lower threshold and good muon track reconstruction in water exploiting the long scattering length. DUMAND, the pioneering project located off the coast of Hawaii, demonstrated that muons could be detected by this technique, but the planned detector was never realized. A detector composed of 96 photomultiplier tubes located deep in Lake Baikal was the first to demonstrate the detection of neutrino-induced muons in natural water. 17 In the following years, NT-200 will be operated as a neutrino telescope with an effective area between 10 3 ~5 x 10 3 m 2 , depending on energy. Presumably too small to detect neutrinos from extraterrestrial sources, NT-
Facts, Fancy and Resolution
Francis Halzen 200 will serve as the prototype for a larger telescope. For instance, with 2000 OMs, a threshold of 10^20 GeV and an effective area of 5 x 10 4 ~10 5 m 2 , an expanded Baikal telescope would fill the gap between present detectors and planned high-threshold detectors of cubic kilometer size. Its key advantage would be low threshold. The Baikal experiment represents a proof of concept for deep ocean projects. These do however have the advantage of larger depth and optically superior water. Their challenge is to find reliable and affordable solutions to a variety of technological challenges for deploying a deep underwater detector. The European collaborations ANTARES 18 and NESTOR 19 plan to deploy large-area detectors in the Mediterranean Sea within the next year. The NEMO Collaboration is conducting a site study for a future kilometer-scale detector in the Mediterranean. 20 The AMANDA collaboration, situated at the U.S. Amundsen-Scott South Pole Station, has demonstrated the merits of natural ice as a Cerenkov detector medium. 21 In 1996, AMANDA was able to observe atmospheric neutrino candidates using only 80 eight-inch photomultiplier tubes. 21 With 302 optical modules instrumenting approximately 6000 tons of ice, AMANDA extracted several hundred atmospheric neutrino events from its first 130 days of data. AMANDA was thus the first first-generation neutrino telescope with an effective area in excess of 10,000 square meters for TeV muons. 22 In rate and all characteristics the events are consistent with atmospheric neutrino origin. Their energies are in the 0.11 TeV range. The shape of the zenith angle distribution is compared to a simulation of the atmospheric neutrino signal in Fig. 6. The variation of the measured rate with zenith angle is reproduced by the simulation to within the statistical uncertainty. Note that the tall geometry of the detector strongly influences the dependence on zenith
Figure 6. Reconstructed zenith angle distribution. The points mark the d a t a and the shaded boxes a simulation of atmospheric neutrino events, the widths of the boxes indicating the error bars.
-90°
Figure 7. Distribution in declination and right ascension of the up-going events on the sky.
angle in favor of more vertical muons. The arrival directions of the neutrinos are shown in Fig. 7. A statistical analysis indicates no evidence for point sources in this sample. An estimate of the energies of the up-going muons (based on simulations of the number of reporting optical modules) indicates that all events have energies consistent with an atmospheric neutrino origin. This enables AMANDA to reach a level of sensitivity to a diffuse flux of high energy extraterrestrial neutrinos of order 22 dN/dEu = 10~ 6 £'~ 2 cm~ 2 s~ 1 sr _ 1 GeV" 1 , assuming an E~2 spectrum. At this level they exclude a variety of theoretical models which assume
536
Facts, Fancy and Resolution
Francis Halzen the hadronic origin of TeV photons from active galaxies and blazars. 23 Searches for neutrinos from gamma-ray bursts, for magnetic monopoles, and for a cold dark matter signal from the center of the Earth are also in progress and, with only 138 days of data, yield limits comparable to or better than those from smaller underground neutrino detectors that have operated for a much longer period. In January 2000, AMANDA-II was completed. It consists of 19 strings with a total of 677 OMs arranged in concentric circles, with the ten strings from AMANDA forming the central core of the new detector. First data with the expanded detector indicate an atmospheric neutrino rate increased by a factor of three, to 4-5 events per day. AMANDA-II has met the key challenge of neutrino astronomy: it has developed a reliable, expandable, and affordable technology for deploying a kilometer-scale neutrino detector named IceCube.
Neutrino flavor
Log(energy/eV) Figure 8. Although IceCube detects neutrinos of any flavor above a threshold of ~ 0.1 TeV, it can identify their flavor and measure their energy in the ranges shown. Filled areas: particle identification, energy, and angle. Shaded areas: energy and angle.
IceCube is an instrument optimised to detect and characterize sub-TeV to multiPeV neutrinos of all flavors (see Fig. 8) from extraterrestrial sources. It will consist of 80 strings, each with 60 10-inch photomultipli537
Figure 9. Simulation of a ultra-high energy taulepton by the interaction of a 10 million GeV tauneutrino, followed by the decay of the secondary taulepton. The color represents the time sequence of the hits (red-orange-yellow-green-blue). The size of the dots corresponds to the number of photons detected by the individual photomultipliers.
ers spaced 17 m apart. The deepest module is 2.4 km below the surface. The strings are arranged at the apexes of equilateral triangles 125 m on a side. The effective detector volume is about a cubic kilometer, its precise value depending on the characteristics of the signal. IceCube will offer great advantages over AMANDA II beyond its larger size: it will have a much higher efficiency to reconstruct tracks, map showers from electron- and tau-neutrinos (events where both the production and decay of a r produced by a vT can be identified; see Fig. 9) and, most importantly, measure neutrino energy. Simulations indicate that the direction of muons can be determined with sub-degree accuracy and their energy measured to better than 30% in the logarithm of the energy. Even the di-
Facts, Fancy and Resolution
Francis Halzen
4. M. Ave et al, Phys. Rev. Lett. 85, 2244 (2000). 5. h t t p : / / www-akeno.icrr.u-tokyo.ac.jp/AGASA/ 6. Proceedings of the International Cosmic Ray Conference, Hamburg, Germany, August 2001. Some of the results described here can be found in the rapporteur's talks of this meeting which was held two weeks after this conference. 7. R. A. Vazquez et al, Astroparticle Physics 3, 151 (1995). 8. F. W. Stecker and M. H. Salamon, astroph/9808110 and references therein. 9. J. Alvarez-Muniz et al, hep-ph/0107057; R. Emparan et al, hep-ph/0109287 and references therein. 10. T. C. Weekes, Status of VHE Astronomy c.2000, Proceedings of the International Symposium on High Energy GammaRay Astronomy, Heidelberg, June 2000, astro-ph/0010431; R. A. Ong, XIX International Symposium on Lepton and Photon Interactions at High Energies, Stanford, August 1999, hep-ex/0003014. 11. E. Pare et al, astro-ph/0107301. 12. J. Cortina for the MAGIC collaboration, Proceedings of the Very High Energy Phenomena in the Universe, Les Arcs, France, January 20-27, 2001, astroph/0103393. 13. http://veritas.sao.arizona.edu/ 14. http://hegral.mppmu.mpg.de 15. h t t p : / / www.igpp.lanl.gov/ASTmilagro.html 16. For reviews, see T.K. Gaisser, F. Halzen and T. Stanev, Phys. Rep. 258(3), 173 (1995); J.G. Learned and K. Mannheim, Ann. Rev. Nucl Part. Science 50, 679 (2000); R. Ghandi, E. Waxman and T. Weiler, review talks at Neutrino 2000, Sudbury, Canada (2000). 17. I. A. Belolaptikov et al, Astroparticle Physics 7, 263 (1997). 18. E. Aslanides et al, astro-ph/9907432 (1999).
rection of showers can be reconstructed to better than 10° in both 9, 0 above lOTeV. Simulations predict a linear response in energy of better than 20%. This has to be contrasted with the logarithmic energy resolution of first-generation detectors. Energy resolution is critical because, once one establishes that the energy exceeds 100 TeV, there is no atmospheric neutrino background in a kilometer-square detector. At this point in time, several of the new instruments, such as the partially deployed Auger array and HiRes to Magic to Milagro and AMANDA II, are less than one year from delivering results. With rapidly growing observational capabilities, one can express the realistic hope that the cosmic ray puzzle will be solved soon. The solution will almost certainly reveal unexpected astrophysics, if not particle physics. Acknowledgements I thank Concha Gonzalez-Garcia and Vernon Barger for comments on the manuscript. This research was supported in part by the U.S. Department of Energy under Grant No. DE-FG02-95ER40896 and in part by the University of Wisconsin Research Committee with funds granted by the Wisconsin Alumni Research Foundation. References 1. D. J. Bird et al, Phys. Rev. Lett. 7 1 , 3401 (1993). 2. N. N. Efimov et al, ICRR Symposium on Astrophysical Aspects of the Most Energetic Cosmic Rays, ed. M. Nagano and F. Takahara (World Scientific, 1991). 3. h t t p : / / www.hep.net/experiments/all_sites.html, provides information on experiments discussed in this review. For a few exceptions, I will give separate references to articles or websites.
538
Facts, Fancy and Resolution
Francis Halzen 19. L. Trascatti, in Procs. of the 5th International Workshop on "Topics in Astroparticle and Underground Physics (TAUP97), Gran Sasso, Italy, 1997, ed. by A. Bottino, A. diCredico, and P. Monacelli, Nucl. Phys. B 7 0 (Proc. Suppl.), p-442 (1998). 20. Talk given at the International Workshop on Next Generation Nucleon Decay and Neutrino Detector (NNN99), Stony Brook, 1999, Proceedings to be published by AIP. 21. The AMANDA collaboration, Astroparticle Physics, 13, 1 (2000). 22. E. Andres et al., Nature 410, 441 (2001). 23. F. Stecker, C. Done, M. Salamon, and P. Sommers, Phys. Rev. Lett. 66, 2697 (1991); erratum Phys. Rev. Lett. 69, 2738 (1992).
539
T H E N E W COSMOLOGY MICHAEL S. TURNER Center for Cosmological Physics Departments of Astronomy & Astrophysics and of Physics Enrico Fermi Institute, The University of Chicago Chicago, IL 60637-1433, USA NASA/Fermilab Astrophysics Center Fermi National Accelerator Laboratory Batavia, IL 60510-0500, USA E-mail: [email protected] Over the past three years we have determined the basic features of our Universe. It is spatially flat; accelerating; comprised of 1/3 a new form of matter, 2/3 a new form of energy, with some ordinary matter and a dash of massive neutrinos; and it apparently began from a great burst of expansion (inflation) during which quantum noise was stretched to astrophysical size seeding cosmic structure. This "New Cosmology" greatly extends the highly successful hot big-bang model. Now we have to make sense of it. What is t h e dark matter particle? What is the nature of the dark energy? Why this mixture? How did the matter - antimatter asymmetry arise? W h a t is the underlying cause of inflation (if it indeed occurred)?
1
The N e w Cosmology
^o
Cosmology is enjoying the most exciting period of discovery yet. Over the past three years a New Cosmology has been emerging. It incorporates the highly successful standard hot big-bang cosmology1 and may extend our understanding of the Universe to times as early as 10~ 32 sec, when the largest structures in the Universe were still subatomic quantum fluctuations. This New Cosmology is characterized by • Flat, critical density accelerating Universe • Early period of rapid expansion (inflation) • Density inhomogeneities produced from quantum fluctuations during inflation • Composition: 2/3 dark energy; 1/3 dark matter; 1/200 bright stars • Matter content: (29±4)% cold dark matter; ( 4 ± 1)% baryons; > 0.3% neutrinos
2.725 ± 0.001 K
• t0 = 14 ± 1 Gyr • H0 = 7 2 ± 7 k m s - 1 M p c " 1 The New Cosmology is not as well established as the standard hot big-bang cosmology. However, the evidence is growing. 1.1
Mounting Evidence: Recent Results
The position of the first acoustic peak in the multipole power spectrum of the anisotropy of the cosmic microwave background (CMB) radiation provides a powerful means of determining the global curvature of the Universe. With the recent DASI observations of CMB anisotropy on scales of one degree and smaller, the evidence that the Universe is at most very slightly curved is quite firm.2 The curvature radius of the Universe (= i?Curv) and the total energy density parameter flo = PTOT/Pcrit, are related: -Rcurv — HQ
540
/\Q,O — 1
,1/2
The New Cosmology
Michael S. Turner The spatial flatness is expressed as fio = l-0± 0.04, or said in words, the curvature radius is at least 50 times greater than the Hubble radius. I will discuss the evidence for accelerated expansion and dark energy later. The series of acoustic peaks in the CMB multipole power spectrum and their heights indicate a nearly scale-invariant spectrum of adiabatic density perturbations with n = 1 ± 0.07. Nearly scale-invariant density perturbations and a flat Universe are two of the three hallmarks of inflation. Thus, we are beginning to see the first significant experimental evidence for inflation, the driving idea in cosmology for the past two decades. The striking agreement of the BBN determination of the baryon density from measurements of the primeval deuterium abundance, 3 - 4 Q.Bh2 = 0.020 ± 0.001, with those from from recent CMB anisotropy measurements, 2 Q,Bh2 = 0.022 ±0.004, make a strong case for a small baryon density, as well as the consistency of the standard cosmology (h = i?o/100kmsec - 1 M p c - 1 ) . There can now be little doubt that baryons account for but a few percent of the critical density. Our knowledge of the total matter density is improving, and becoming less linked to the distribution of light. This makes determinations of the matter less sensitive to the uncertain relationship between the clustering of mass and of light (what astronomers call the bias factor b).7 Both the CMB and clusters of galaxies allow a determination of the ratio of the total matter density (anything that clusters - baryons, neutrinos, cold dark matter) to that in baryons alone: QM/^B = 7.2±2.1 (CMB), 5 9±1.5 (clusters). 6 Not only are these numbers consistent, they make a very strong case for something beyond quarkbased matter. When combined with our knowledge of the baryon density, one infers a total matter density of 0 « = 0.33 ± 0.04.7 The many successes of the cold dark mat-
541
ter (CDM) scenario - from the sequence of structure formation (galaxies first, clusters of galaxies and larger objects later) and the structure of the intergalactic medium, to its ability to reproduce the power spectrum of inhomogeneity measured today - makes it clear that CDM holds much, if not all, of the truth in describing the formation of structure in the Universe. The two largest redshift surveys, the Sloan Digital Sky Survey (SDSS) and the 2degree Field project (2dF), have each recently measured the power spectrum using samples of more than 100,000 galaxies and found that it is consistent with that predicted in a flat accelerating Universe comprised of cold dark matter. 8 The SDSS will eventually use a sample of almost one million galaxies to probe the power spectrum. [Interestingly enough, according to the 2dF Collaboration, bias appears to be a small effect, b = 1.00 ± 0.09 9
]
All of this implies that whatever the dark matter particle is, it moves slowly (i.e., the bulk of the matter cannot be in the form of hot dark matter such as neutrinos) and interacts only weakly (e.g., with strength much less than electromagnetic) with ordinary matter. The evidence from SuperKamiokande 10 for neutrino oscillations makes a strong case that neutrinos have mass (J^ i m„ > 0.1 eV) and therefore contribute to the mass budget of the Universe at a level comparable to, or greater than, that of bright stars. Particle dark matter has moved from the realm of a hypothesis to a quantitative question - how much of each type of particle dark matter is there in the Universe? Structure formation in the Universe (especially the existence of small scale structure) suggests that neutrinos contribute at most 5% or 10% of the critical density, corresponding to Y^i rrinU =
J2imu/90h2eV<5eVn Even the age of the Universe and the pesky Hubble constant have been reined in.
Michael S. Turner
The New Cosmology
The uncertainties in the ages of the oldest globular clusters have been better identified and quantified, leading to a more precise age, t0 = 13.5 ± 1.5 Gyr. 12 The CMB can be used to constrain the expansion age, independent of direct measurements of Ho or the composition of the Universe, i e x p = 14 ± 0.5 Gyr. 13 A host of different techniques are consistent with the Hubble constant determined by the HST key project, H0 = 72 ± 7 k m s _ 1 M p c ~ 1 . Further, the error budget is now well understood and well quantified.14 [The bulk of the ± 7 uncertainty is systematic, dominated by the uncertainty in the distance to the LMC and the Cepheid period - luminosity relation.] Moreover, the expansion age derived from this consensus Hubble constant, which depends upon the composition of the Universe, is consistent with the previous two age determinations. The poster child for precision cosmology continues to be the present temperature of the CMB. It was determined by the FIRAS instrument on COBE to be: T 0 = 2.725 ± 0.001 K. 15 Further, any deviations from a black body spectrum are smaller than 50 parts per million. Such a perfect Planckian spectrum has made any noncosmological explanation untenable.
1.2
Successes and Consistency
Tests
To sum up, we have determined the basic features of the Universe: the cosmic matter/energy budget; a self consistent set of cosmological parameters with realistic errors; and the global curvature. Two of the three key predictions of inflation - flatness and nearly scale-invariant, adiabatic density perturbations - have passed their first significant tests. Last but not least, the growing quantity of precision data are now testing the consistency of the Friedmann-Robertson-Walker framework and General Relativity itself. In particular, the equality of the baryon densities determined from BBN and CMB
anisotropy is remarkable. The first involves nuclear physics when the Universe was seconds old, while the latter involves gravitational and classical electrodynamics when the Universe was 400,000 years old. The entire framework has been tested by the existence of the aforementioned acoustic peaks in the CMB angular power spectrum. They reveal large-scale motions that have remained coherent over hundreds of thousands of years, through a delicate interplay of gravitational and electromagnetic interactions. Another test of the basic framework is the accounting of the density of matter and energy in the Universe. The CMB measurement of spatial flatness implies that the matter and energy densities must sum to the critical density. Measurements of the matter density indicate QM — 0.33 ± 0.04; and measurements of the acceleration of the Universe from supernovae indicate the existence of a smooth dark energy component that accounts for Qx ~ 0.67. [The amount of dark energy inferred from the supernova measurements depends its equation of state; for a cosmological constant, fU = 0.8 ± 0.16.] Finally, while cosmology has in the past been plagued by "age crises" - time back to the big bang (expansion age) apparently less than the ages of the oldest objects within the Universe - today the ages determined by very different and completely independent techniques point to a consistent age of 14 Gyr.
2
Mysteries
Cosmological observations over the next decade will test - and probably refine - the New Cosmology.16 If we are fortunate, they will also help us to make better sense of it. At the moment, the New Cosmology has presented us with a number of cosmic mysteries opportunities for surprises and new insights. Here I will quickly go through my list, and save the most intriguing to me - dark energy - for its own section.
542
Michael S. Turner 2.1
The New Cosmology
Dark Matter
By now, the conservative hypothesis is that the dark matter consists of a new form of matter, with the axion and neutralino as the leading candidates. That most of the matter in the Universe exists in a new form of matter - yet to be detected in the laboratory - is a bold and untested assertion. Experiments to directly detect the neutralinos or axions holding our own galaxy together have now reached sufficient sensitivity to probe the regions of parameter space preferred by theory. In addition, the neutralino can be created by upcoming collider experiments (at the Tevatron or the LHC), or detected by its annihilation signatures — high-energy neutrinos from the sun, narrow positron lines in the cosmic rays, and gammaray line radiation. 17 While the CDM scenario is very successful there are some nagging problems. They may point to a fundamental difficulty or may be explained by messy astrophysics. 18 The most well known of these problems are the prediction of cuspy dark-matter halos (density profile PDM —* l/rn as r —> 0, with n ~ 1 — 1.5) and the apparent prediction of too much substructure. While there are plausible astrophysical explanations for both problems, 19 they could indicate an unexpected property of the dark-matter particle (e.g., large self-interaction cross section 20 , large annihilation cross section 21 , or mass of around 1 keV). While I believe it is unlikely, these problems could indicate a failure of the particle dark-matter paradigm and have their explanation in a radical modification of gravity theory. 18 I leave for the "astrophysics to do list" an accounting of the dark baryons. Since £lB ~ 0.04 and 0* ~ 0.005, the bulk of the baryons are optically dark. In clusters, the dark baryons have been identified: they exists as hot, x-ray emitting gas. Elsewhere, the dark baryons have not yet been identi-
543
fied. According to CDM, the bulk of the dark baryons are likely to exist as hot/warm gas associated with galaxies, but this gas has not been detected. [Since clusters account for only about 5 percent of the total mass, the bulk of the dark baryons are still not accounted for.]
2.2
Baryogenesis
The origin of quark-based matter is not yet fully understood. We do know that the origin of ordinary matter requires a small excess of quarks over antiquarks (about a part in 109) at a time at least as early as 10~ 6 sec, to avoid the annihilation catastrophe associated with a baryon symmetric Universe. 1 If the Universe underwent inflation, the baryon asymmetry cannot be primeval, it must be produced dynamically ("baryogenesis") after inflation since any pre-infiation baryon asymmetry is diluted away by the enormous entropy production associated with reheating. Because we also now know that electroweak processes violate B + L at a very rapid rate at temperatures above 100 GeV or so, baryogenesis is more constrained than when the idea was introduced more than twenty years ago. Today there are three possibilities: 1) produce the baryon asymmetry by GUT-scale physics with B — L^= 0 (to prevent it being subsequently washed away by B + L violation); 2) produce a lepton asymmetry (L ^ 0), which is then transmuted into the baryon asymmetry by electroweak B + L violation; 22 or 3) produce the baryon asymmetry during the electroweak phase transition using electroweak B violation. 23 While none of the three possibilities can be ruled out, the second possibility looks most promising, and it adds a new twist to the origin of quark-based matter: We are here because neutrinos have mass. [In the lepton asymmetry first scenario, Majorana neutrino mass provides the requisite lepton number violation.] The drawback of the first possibility
The New Cosmology
Michael S. Turner is the necessity of a high reheat temperature after inflation, T R H S> 10 5 GeV, which is difficult to achieve in most models of inflation. The last possibility, while very attractive because all the input physics might be measurable at accelerator, requires new sources of CP violation at TeV energies as well as a strongly first-order electroweak phase transition (which is currently disfavored by the high mass of the Higgs). 23
2.3
veal much about the underlying scalar potential driving inflation. Measuring their spectral index - a most difficult task - provides a consistency test of the single scalar-field model of inflation. 25 2.4
of Space-time
Are there additional spatial dimensions beyond the three for which we have very firm evidence? I cannot think of a deeper question in physics today. If there are new dimensions, they are likely to be relevant for cosmology, or at least raise new questions in cosmology (e.g., why are only three dimensions large? what is going on in the bulk? and so on). Further, cosmology may well be the best means for establishing the existence of extra dimensions.
Inflation
There are still many questions to be answered about inflation, including the most fundamental: did inflation (or something similar) actually take place! A powerful program is in place to test the inflationary framework. Testing framework involves testing its three robust predictions: spatially flat Universe; nearly-scale invariant, nearly power-law spectrum of Gaussian adiabatic, density perturbations; and a spectrum of nearly scale-invariant gravitational waves. The first two predictions are being probed today and will be probed much more sharply over the next decade. The value of 00 should be determined to much better than 1 percent. The spectral index n that characterizes the density perturbations should be measured to percent accuracy. Generically, inflation predicts \n — 1| ~ O(0.1), where n = 1 corresponds to exact scale invariance. Likewise, the deviations from an exact power-law predicted by inflation,24 \dn/d\nk\ ~ 10" 4 - 10" 2 will be tested. The CMB and the abundance of rare objects such as clusters of galaxies will allow Gaussianity to be tested. Inflationary theory has given little guidance as to the amplitude of the gravitational waves produced during inflation. If detected, they are a smokin' gun prediction. Their amplitude is directly related to the scale of inflation, hGW a i/i„flation/"ipi- Together measurements of n — 1 and dnj lnfc, they can re-
The Dimensionality
2.5
Before Inflation, Other Big-bang Debris, and Surprises
Only knowing everything there is to know about the Universe would be worse than knowing all the questions to ask about it. Without doubt, as our understanding deepens, new questions and new surprises will spring forth. The cosmological attraction of inflation is its ability to make the present state of the Universe insensitive to its initial state. However, should we establish inflation as part of cosmic history, I am certain that cosmologists will begin asking what happened before inflation. Progress in cosmology depends upon studying relics. We have made much of the handful we have - the light elements, the baryon asymmetry, dark matter, and the CMB. The significance of a new relic cannot be overstated. For example, detection of the cosmic sea of neutrinos would reveal the Universe at 1 second. Identifying the neutralino as the dark matter particle and determining its properties at an accelerator laboratory would open a
544
The New Cosmology
Michael S. Turner window on the Universe at 1 0 - 8 sec. By comparing its relic abundance as derived from its mass and cross section with its actual abundance measured in the Universe, one could test cosmology at the time the neutralino abundance was determined. And then there may be the unexpected. Recently, a group reported evidence for a part in 105 difference in the fine-structure constant at redshifts of order a few from its value today. 26 I remain skeptical, given possible astrophysical explanations, other much tighter constraints to the variation of a (albeit at more recent times), and the absence of a reasonable theoretical model. For reference, I was also skeptical about the atmospheric neutrino problem because of the need for largemixing angles. 3
Dark Energy: Seven Things We Know
The dark energy accounts for 2/3 of the stuff in the Universe and determines its destiny. That puts it high on the list of outstanding problems in cosmology. Its deep connections to fundamental physics - a new form of energy with repulsive gravity and possible implications for the divergences of quantum theory and supersymmetry breaking - put it very high on the list of outstanding problems in particle physics. 27,28 What then is dark energy? Dark energy is my term for the causative agent for the current epoch of accelerated expansion. According to the second Friedmann equation,
this stuff must have negative pressure, with magnitude comparable to its energy density, in order to produce accelerated expansion [recall q = — (R/R)/H2; R is the cosmic scale factor]. Further, since this mysterious stuff does not show its presence in galaxies and clusters of galaxies, it must be relatively smoothly distributed.
545
That being said, dark energy has the following defining properties: (1) it emits/absorbs no light; (2) it has large, negative pressure, px ~ —px\ (3) it is approximately homogeneous (more precisely, does not cluster significantly with matter on scales at least as large as clusters of galaxies); and (4) it is very mysterious. Because its pressure is comparable in magnitude to its energy density, it is more "energy-like" than "matter-like" (matter being characterized by p < p ) . Dark energy is qualitatively very different from dark matter, and is certainly not a replacement for it.
3.1
Two Lines of Evidence for an Accelerating Universe
Two independent lines of reasoning point to an accelerating Universe. The first is the direct evidence based upon measurements of type la supernovae carried out by two groups, the Supernova Cosmology Project 29 and the High-z Supernova Team. 30 These two teams used different analysis techniques and different samples of high-z supernovae and came to the same conclusion: the expansion of the Universe is speeding up, not slowing down. The recent serendipitous discovery of a supernovae at z = 1.76 bolsters the case significantly31 and provides the first evidence for an early epoch of decelerated expansion. 32 SN 1997ff falls right on the accelerating Universe curve on the magnitude - redshift diagram, and is a magnitude brighter than expected in a dusty open Universe or an open Universe in which type l a supernovae are systematically fainter at high-2. The second, independent line of reasoning for accelerated expansion comes from measurements of the composition of the Universe, which point to a missing energy component with negative pressure. The argument goes like this: CMB anisotropy measurements indicate that the Universe is nearly flat, with density parameter, CIQ = 1.0 ±0.04.
The New Cosmology
Michael S. Turner In a fiat Universe, the matter density and energy density must sum to the critical density. However, matter only contributes about 1/3 of the critical density, 9,M = 0.33 ± 0.04. (This is based upon measurements of CMB anisotropy, of bulk flows, and of the baryonic fraction in clusters.) Thus, two thirds of the critical density is missing! Doing the bookkeeping more precisely, £tx = 0.67 ± 0.06.7 In order to have escaped detection, this missing energy must be smoothly distributed. In order not to interfere with the formation of structure (by inhibiting the growth of density perturbations), the energy density in this component must change more slowly than matter (so that it was subdominant in the past). For example, if the missing 2/3 of critical density were smoothly distributed matter (p = 0), then linear density perturbations would grow as R1'2 rather than as R. The shortfall in growth since last scattering (z ~ 1100) would be a factor of 30, far too little growth to produce the structure seen today.
In Newton's theory, mass is the source of the gravitational field and gravity is always attractive. In General Relativity, both energy and pressure source the gravitational field: R/R oc -(p + Zp), cf., Eq. 1. Sufficiently large negative pressure leads to repulsive gravity. While accelerated expansion can be accommodated within Einstein's theory, that does not preclude that the ultimate explanation lies in a fundamental modification of Einstein's theory. Lacking any good ideas for such a modification, I will discuss how accelerated expansion fits in the context of General Relativity. If the explanation for the accelerating Universe ultimately fits within the Einsteinian framework, it will be a stunning new triumph for General Relativity.
The pressure associated with the missing energy component determines how it evolves:
3.3
px oc
Gravity Can Be Repulsive in Einstein's Theory, But ...
The Biggest Embarrassment in all of Theoretical Physics
Einstein introduced the cosmological constant to balance the attractive gravity of matter. He quickly discarded the cosmological constant after the discovery of the expansion of the Universe. The advent of quantum field theory made consideration of the cosmological constant obligatory, not optional: The only possible covariant form for the energy of the (quantum) vacuum,
(2)
where w is the ratio of the pressure of the missing energy component to its energy density (here assumed to be constant). Note, the more negative w, the faster the ratio of missing energy to matter decreases to zero in the past. In order to grow the structure observed today from the density perturbations indicated by CMB anisotropy measurements, w must be more negative than about — | . 3 3 For a fiat Universe the deceleration parameter today is 1
3.2
R~^+w)
c< (1 + zfw
=> PX/PM
energy makes the supernova case all the more compelling.
T
vlc
3
r^ 1 10 = ~ + -jWtlx ~ - + W Therefore, knowing w < — | implies Qo < 0 and accelerated expansion. This independent argument for accelerated expansion and dark
=
PVAC9^V,
is mathematically equivalent to the cosmological constant. It takes the form for a perfect fluid with energy density PVAC and isotropic pressure PVAC = ~PVAC (i-e-> w = —1) and is precisely spatially uniform. Vacuum energy is almost the perfect candidate for dark energy.
546
The New Cosmology
Michael S. Turner Here is the rub: the quantum zero-point contributions arising from well-understood physics (the known particles, integrating up to 100 GeV) sum to 10 55 times the present critical density. (Put another way, if this were so, the Hubble time would be 10~ 10 sec, and the associated event horizon would be 3cm!) This is the well known cosmological-constant problem. 27 ' 28 While string theory currently offers the best hope for marrying gravity to quantum mechanics, it has shed precious little light on the cosmological constant problem, other than to speak to its importance. Thomas has suggested that using the holographic principle to count the available number of states in our Hubble volume leads to an upper bound on the vacuum energy that is comparable to the energy density in matter + radiation. 34 While this reduces the magnitude of the cosmological-constant problem very significantly, it does not solve the dark energy problem: a vacuum energy that is always comparable to the matter + radiation energy density would strongly suppress the growth of structure. The deSitter space associated with the accelerating Universe may pose serious problems for the formulation of string theory. 35 Banks and Dine argue that all explanations for dark energy suggested thus far are incompatible with perturbative string theory. 36 At the very least there is high tension between accelerated expansion and string theory. The cosmological constant problem leads to a fork in the dark-energy road: one path is to wait for theorists to get the "right answer" (i.e., fix = 2/3); the other path is to assume that even quantum nothingness weighs nothing and something else with negative pressure must be causing the Universe to speed up. Of course, theorists follow the advice of Yogi Berra: "When you see a fork in the road, take it."
3.4
Parameterizing Dark Energy: For Now, It's w
Theorists have been very busy suggesting all kinds of interesting possibilities for the dark energy: networks of topological defects, rolling or spinning scalar fields (quintessence and spintessence), influence of "the bulk", and the breakdown of the Friedmann equations. 28 ' 38 An intriguing recent paper suggests dark matter and dark energy are connected through axion physics. 37 In the absence of compelling theoretical guidance, there is a simple way to parameterize dark energy, by its equation-of-state w.33 The uniformity of the CMB testifies to the near isotropy and homogeneity of the Universe. This implies that the stress-energy tensor for the Universe must take the perfect fluid form.1 Since dark energy dominates the energy budget, its stress-energy tensor must, to a good approximation, take the form Txt « diag[px, ~Px,~Px,
~Px]
(3)
where px is the isotropic pressure and the desired dark energy density is px=2.7x
10~ 47 GeV 4
(for h = 0.72 and fix = 0.66). This corresponds to a tiny energy scale, p-% = 2.3 x 10" 3 eV. The pressure can be characterized by its ratio to the energy density (or equation-ofstate): w =
Px/px
Note, w need not be constant; e.g., it could be a function of px or an explicit function of time or redshift. (w can always be rewritten as an implicit function of redshift.) For vacuum energy w = — 1; for a network of topological defects w = —N/3 where N is the dimensionality of the defects (1 for strings, 2 for walls, etc.). For a minimally coupled, rolling scalar field,
•u,= j f - V W 547
(4)
Michael S. Turner
The New Cosmology
which is time dependent and can vary between — 1 (when potential energy dominates) and +1 (when kinetic energy dominates). Here V(cf>) is the potential for the scalar field. 3.5
• The CMB has limited power to probe w (e.g., the projected precision for Planck is aw = 0.25) and no power to probe its time variation. 41
The Universe: The Lab for Studying Dark Energy
Dark energy by its very nature is diffuse and a low-energy phenomenon. It probably cannot be produced at accelerators; it isn't found in galaxies or even clusters of galaxies. The Universe itself is the natural lab - perhaps the only lab - in which to study it. The primary effect of dark energy on the Universe is determining the expansion rate. In turn, the expansion rate affects the distance to an object at a given redshift z [= r(z)\ and the growth of linear density perturbations. The governing equations are: H2(z) = H%{1 + zf
pM
+nx(l
r(z) = f du/H{u) Jo 0 = 5k + 2H5k - 4xGpM6k
of the Universe, the most sensitive redshift interval for probing dark energy is z = 0.2-2.41
+ z)3w]
(5)
where for simplicity w is assumed to be constant and 8k is the Fourier component of comoving wavenumber k and overdot indicates d/dt. The various cosmological approaches to ferreting out the nature of the dark energy all of which depend upon how the dark energy affects the expansion rate - have been studied. 40 Based largely upon my work with Dragan Huterer, 41 I summarize what we now know about the efficacy of the cosmological probes of dark energy:
• A high-quality sample of 2000 SNe distributed from z = 0.2 to z = 1.7 could measure w to a precision aw = 0 . 0 5 (assuming an irreducible systematic error of 0.14 mag). If &M is known independently to better than O~QM = 0.03, aw improves by a factor of three and the rate of change of w' = dw/dz can be measured to precision aw> = 0.16. 41 • Counts of galaxies and of clusters of galaxies may have the same potential to probe w as SNe la. The critical issue is systematics (including the evolution of the intrinsic comoving number density, and the ability to identify galaxies or clusters of a fixed mass). 39 • Measuring weak gravitational lensing by large-scale structure over a field of 1000 square degrees (or more) could have comparable sensitivity to w as type la supernovae. However, weak gravitational lensing does not appear to be a good method to probe the time variation of- 42 The systematics associated with weak gravitational lensing have not yet been studied carefully and could limit its potential.
With the exception of vacuum energy, all the other possibilities for the dark energy cluster to some small extent on the largest scales.44 Measuring this clustering, while extremely challenging, could rule out vacuum • Because dark energy was less important or help to elucidate the nature of the dark in the past, px/pM oc (1 + z)3w —> 0 energy. Hu and Okamoto have recently sugas z —> oo, and the Hubble flow at low gested how the CMB might be used to get at this clustering. 45 redshift is insensitive to the composition • Present cosmological observations prefer w = — 1, with a 95% confidence limit w < -0.6. 4 3
548
Michael S. Turner
The New Cosmology
While the Universe is likely the lab where dark energy can best be attacked, one should not rule other approaches. For example, if the dark energy involves a ultra-light scalar field, then there should be a new long-range force46. 3.6
The Nancy Kerrigan Problem
A critical constraint on dark energy is that it not interfere with the formation of structure in the Universe. This means that dark energy must have been relatively unimportant in the past (at least back to the time of last scattering, z ~ 1100). If dark energy is characterized by constant w, not interfering with structure formation can be quantified as: w < - | . 3 3 This means that the dark-energy density evolves more slowly than i?~ 3 / 2 (compared to R~3 for matter) and implies PX/PM
^0
PX/PM
—> oo
for t —> 0 for t —> oo
That is, in the past dark energy was unimportant and in the future it will be dominant! We just happen to live at the time when dark matter and dark energy have comparable densities. In the words of Olympic skater Nancy Kerrigan, "Why me? Why now?" Perhaps this fact is an important clue to unraveling the nature of the dark energy. Perhaps not. I shudder t o say this, but it could be at the root of an anthropic explanation for the size of the cosmological constant: The cosmological constant is as large as it can be and still allow the formation of structures that can support life.48 3.7
Dark Energy and Destiny
Almost everyone is aware of the connection between the shape of the Universe and its destiny: positively curved recollapses, flat; negatively curved expand forever. The link
549
between geometry and destiny depends upon a critical assumption: that matter dominates the energy budget (more precisely, that all components of matter/energy have equation of state w > — g)- Dark energy does not satisfy this condition. In a Universe with dark energy the connection between geometry and destiny is severed. 47 A flat Universe (like ours) can continue expanding exponentially forever with the number of visible galaxies diminishing to a few hundred (e.g., if the dark energy is a true cosmological constant); the expansion can slow to that of a matter-dominated model (e.g., if the dark energy dissipates and becomes sub-dominant); or, it is even possible for the Universe to recollapse (e.g., if the dark energy decays revealing a negative cosmological constant). Because string theory prefers anti-deSitter space, the third possibility should not be forgotten. Dark energy is the key to understanding our destiny. 4
Closing Remarks
As a New Cosmology emerges, a new set of questions arises. Assuming the Universe inflated, what is the physics underlying inflation? What is the dark-matter particle? How was the baryon asymmetry produced? Why is the recipe for our Universe so complicated? What is the nature of the Dark Energy? Answering these questions will help us make sense of the New Cosmology as well as revealing deep connections between fundamental physics and cosmology. There may even be some big surprises - time variation of the constants of Nature, or a new theory of gravity that eliminates the need for dark matter and dark energy (though I for one am not betting on either!). There is an impressive program in place, with telescopes, accelerators, and laboratory experiments, both in space and on the ground: the Sloan Digital Sky Survey; the
Michael S. Turner
The New Cosmology
Hubble Space Telescope and the Chandra Xray Observatory; a growing number of large ground-based telescopes; the Tevatron and B-factories in the US and Japan; specialized dark-matter detectors; gravity-wave detectors; a multitude of ground-based and balloon-borne CMB anisotropy experiments; the MAP satellite (which is already taking data) and the Planck satellite (to be launched in 2007). Still to come are: the LHC; a host of accelerator and nonaccelerator neutrino-oscillation and neutrino-mass experiments; the Next Generation Space Telescope; gravity-wave detectors in space; cluster surveys using x-rays and the Sunyaev - Zel'dovich effect. And in the planning: dedicated ground and space based wide-field telescopes to study dark energy, the next linear collider and on and on. Any one, or more likely several, of these experiments will produce major advances in our understanding of the Universe and the fundamental laws that govern it.
ph/0104490 3. D. Tytler et al, Physica Scripta T 8 5 , 12 (2000); J.M. O'Meara et al, Astrophys. J. 552, 718 (2001) 4. S. Buries et al, Phys. Rev. D 63, 063512 (2001); Astrophys. J. 552, LI (2001) 5. C.B. Netterfield et al, astro-ph/0104460; C. Pryke et al, astro-ph/0104490 6. See e.g., J. Mohr et al, Astrophys. J. 517, 627 (1998) 7. M.S. Turner, astro-ph/0106035 8. W.J. Percival et al (2dF), astroph/0105252; S. Dodelson et al (SDSS), astro-ph/0107421 9. O. Lahav et al (2dF), astro-ph/0112162 10. Y. Fukuda et al, Phys. Rev. Lett. 8 1 , 1562 (1998) 11. R. Croft, W. Hu and R. Dave, Phys. Rev. Lett. 83, 1092 (1999) 12. L. Krauss and B. Chaboyer, astroph/0111597; I. Ferreres, A. Melchiorri, and J. Silk, MNRAS 327, L47 (2001) 13. L. Knox, N. Christensen, and C. Skordis, astro-ph/0109232 14. W.L. Freedmanet al, Astrophys. J. 553, 47 (2001) 15. J. Mather et al, Astrophys. J. 512, 511 (1999) 16. M.S. Turner, PASP 113, 653 (2001) (astro-ph/0102057) 17. See e.g., B. Sadoulet, Rev. Mod. Phys. 71, S197 (1999) or K. Griest and M. Kamionkowski, Phys. Rep. 333-4, 167 (2000) 18. See e.g., J. Sellwood and A. Kosowsky, astro-ph/0009074 19. See e.g., J. Bullock, A. Kravtsov and D. Weinberg, Astrophys. J. 539, 517 (2000); R.S. Somerville, astroph/0107507; A.J. Benson et al, astroph/0108218; M.D. Weinberg and N. Katz, astro-ph/0110632; A. Klypin, H.S. Zhao, and R.S. Somerville, astroph/0110390; D. Merritt and F. Cruz, Astrophys. J. 551, L41 (2001); M. Milosavljevic and D. Merritt, Astrophys.
The progress we make over the two decades will determine how golden our age of cosmology is. Acknowledgments This work was supported by the DoE (at Chicago and Fermilab) and by the NASA (at Fermilab by grant NAG 5-7092).
References 1. See e.g., S. Weinberg, Gravitation and Cosmology (Wiley & Sons, NY, 1972); or E.W. Kolb and M.S. Turner, The Early Universe (Addison-Wesley, Redwood City, CA, 1990) 2. P. de Bernardis et al, Nature 404, 955 (2000); S. Hanany et al, Astrophys. J. 545, L5 (2000); C.B. Netterfield et al, astro-ph/0104460; C. Pryke et al, astro550
Michael S. Turner
20. 21.
22. 23.
24. 25.
26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39.
40. 41.
The New Cosmology
J. 563, 34 (2001); M. Milosavljevic et al, astro-ph/0110185 D.N. Spergel and P.J. Steinhardt, Phys. Rev. Lett. 84, 3760 (2000) M. Kaplinghat, L. Knox and M.S. Turner, Phys. Rev. Lett. 85 3335 (2000) W. Buchmuller, hep-ph/0107153 J.M. Cline et al, JHEP 0007 (2000) 018; A. Cohen, D. Kaplan, and A. Nelson, Ann. Rev. Nucl. Part. Sci. 43, 27 (1994) A. Kosowsky and M.S. Turner, Phys. Rev. D 52, R1739 (1995) M.S. Turner, Phys. Rev. D 48, 5539 (1993); J.E. Lidsey et al, Rev. Mod. Phys. 69, 373 (1997) J.K. Webb et al, Phys. Rev. Lett. 87 091301 (2001) S. Weinberg, Rev. Mod. Phys. 61, 1 (1989) http://www.livingreviews.org/Articles/ Volume4/2001-lcarroll S. Perlmutter et al, Astrophys. J. 517, 565 (1999) A. Riess et al, Astron. J. 116, 1009 (1998) A. Riess et al, Astrophys. J. 560, 49 (2001) M.S. Turner and A. Riess, astroph/0106051 (Astrophys. J., in press) M.S. Turner and M. White, Phys. Rev. 56, R4439 (1997) S. Thomas, hep-th/0010145 E. Witten, hep-th/0106109 T. Banks and M. Dine, hep-th/0106276 S. Barr and D. Seckel, astro-ph/0106239 M.S. Turner, Physica Scripta T 8 5 , 210 (2000) See e.g., J. Newman and M. Davis, Astrophys. J. 534, L l l (2000); G.P. Holder et al, Astrophys. J. 553, 545 (2001); S. Podariu and B. Ratra, astro-ph/0106549 http://supernova.lbl.gov/~evlinder/ sci.html D. Huterer and M.S. Turner, Phys. Rev.
551
D 64, 123527 (2001) 42. D. Huterer, astro-ph/0106399 (submitted to Phys. Rev. D) 43. S. Perlmutter, M.S. Turner, and M. White, Phys. Rev. Lett. 83, 670 (1999) 44. K. Coble, S. Dodelson, and J. Prieman, Phys. Rev. D 55, 1851 (1997) 45. W. Hu and T. Okamoto, astro-ph/0111606 46. S. Carroll, Phys. Rev. Lett. 8 1 , 3067 (1998) 47. L. Krauss and M.S. Turner, Gen. Rel Grav. 31, 1453 (1999) 48. H. Martel, P. Shapiro, and S. Weinberg, Astrophys. J. 492, 29 (1998)
This page is intentionally left blank
Future High Energy Facilities
Future High Energy Facilities Session Chair: Scientific Secretaries:
L. Maiani R. Heuer S. Geer
J. Lee-Franzini M. Boscolo A. Denig
LHC machine program''' e + -e~ colliders Muon storage rings and Neutrino factories
Session Chair: S. Yamada, P.I.P. Kalmus Scientific Secretaries: M. Moulson D. Meloni S. Miscetti L. Randall N. Cabibbo
Superstrings, duality, large extra dimensions Concluding remarks
f Prepared by P. Valente
554
LHC MACHINE P R O G R A M LUCIANO MAIANI CERN, Geneva, Switzerland Because Prof. Luciano Maiani could not submit his manuscript in time for t h e proceedings our Scientific Secretary, Paolo Valente, selected some pictures from t h e presentation for inclusion here.
Point 8
*K*vlT***r
lv*C
Figure 1. A view of the LHC accelerator complex at CERN, between France Switzerland: a schematic view of the main buildings and the underground tunnel with the main experimental areas is shown, with t h e G e n e v a lake and t h e J u r a mountains on t h e background.
555
LHC Machine Program
Luciano Maiani
Internal ional Col] aboration for LHC construction **
Gioss NMS contributions l-S: 200 M$ Russia: 100 MCHF ".• MCHF Japan. 170 Canada: 30 MCHF India: 25 MS Cost slianna t'oi LHC (BCilF). MS. Material: 2.1 MS. Personnel. i i (appnn 0.2 Hns! States: NMS (net r
0.6 ( 15%) " W
¥ &
V-^fV
'*
' !• WV,!> K V R M W v
' ,,! m
(til lW."k".
'rtjji t Jtyf' i-
DM.L'pum IR!lMt-V.,'VW,-lT
JV, *
HilTPn-twh.
/
HM'SiiM-'.lhfs'. '
'
HNIvfutlu ll\Ri UtlMbJi klkUikU-j
Figure 2. CERN member and non-member States cost contribution for LHC: gross (money and support) contribution (upper table) and approximate cost sharing (lower table).
.-
'evolution du demant&ement LEP Machine: >50% dismanik •.; lend of works March 2( \; ?.
i^BEsaasiiHiB Figure 3. Dismantling of the LEP: picture of the empty tunnel.
556
L H C Machine P r o g r a m
Luciano Maiani
n
)
l3j
-3><
Figure 4. Overall picture of civil engineering works for LHC.
Civil engineering at Point 1
m
W^Cs^-jJ'^
,v
s
" • »
,.
. -,. . - - 1
«*fcr. .4"
IK
«•'
-!••
»
•_*h*>5 Concreting vault end in August 01 Bench Excavation starts until April 02
Figure 5. Excavation and concreting of the ATLAS cavern (Point 1).
557
Luciano Maiani
LHC Machine Program
Pillar concreting ends in August 01
Figure 6. Excavation and concreting of the CMS cavern (Point 5).
m Dipole n. 360 in Novosibirsk
ifPiiliiiilJiiiiilliiiili! Figure 7. Russia contribution: dipole and quadrupoles for the transfer line between SPS and LHC, built in Novosibirsk (the last one is shown in the upper picture).
558
Luciano Maiani
LHC Machine Program
Non-Member States: FNAL, United States
Low-|3 prototype coil (interaction regions)
*^Jk
Figure 8. United States contribution: low-/3 super-conducting quadrupole (a prototype is shown).
Ci>lljhi>i\ilii>n v.ill] J . I | M I I •
•>
*
•
»&!B .-
iX^i*
->.•
l|ll.llllll|lii|i>
i lllk-ljv.1 it'll iv.'lull i ' .;* Figure 9. Japan contribution: super-conducting quadrupoles for t h e interaction region.
559
LHC Machine Program
Luciano Maiani
BO in SACLAY.
...and in the ATLAS test Stand at CERN 25
BO Test Current 17-20 July 01 1
20 <15 §10 tl 3
U5 7/17/2001
7/18/2001
7/19/2001
7/20/2001
Figure 10. Picture of BO prototype of the ATLAS coil, being assembled in Saclay.
560
Luciano Maiani
LHC Machine Program
« • J.
L = 10fl>"L L = 30fb" i i . = 100 ft"1
ATLAS + C M S (no K-foctors)
— io'
10
LEP2
i j
Most difficult region: H-* ttH-^ttbb
io' mH(GeV)
10
Figure 11. Significance for Higgs boson discovery t h a t can be obtained by combining t h e two LHC experiments, at different luminosities.
1EO
1 30
Figure 12. Signal for H —• 77 at LHC for 100 fb
561
l
of integrated luminosity.
Luciano Maiani
LHC Machine Program
O H. WH. t±H7y) A WH. ttH (H~>bb) D H->ZZ->4i * H->WW-*lvlv 0 WH(H-»WW-»Mv) A all channels
10
-2
10
10
ATLAS + CMS j L d t = 300fb" 1 "10 10' m H (GeV)
10
— 0.5 CD X 0,
A H - J W W - 4 Ivlv •
H-^ZZ'-»4I
g0.4
0.3 ATLAS + CMS ,[Ldt = 300fb" 1
0.2
0.1 -
SC >
*
)»1
10 m H (GeV)
10
Figure 13. At LHC a precision at the level of 1 0 - 3 for Higgs mass (top) and 10% for the coupling (bottom) can be obtained all over t h e Higgs mass range, using t h e different channels (with 300 f b _ 1 by combining ATLAS and CMS).
562
Luciano Maiani
LHC Machine Program
ri
r
^•-Jf
G
ZrM•;vi
c«
^ / *
™ ™ . ™
sr ! » 6 !
7 •
v.?i ..dice,
Raftazzi, Weils
e""e" to
avitons
-W-
Figure 14. Extra dimensions: electroweak particles are confined t o a surface in a multi-dimensional space while only gravity can get into "real" D dimensions; this would give a sizeable difference for the gravity constant and this can be explored with high energy collisions at LHC.
563
Luciano Maiani
LHC Machine Program
Early Universe LH<" F - P b ( 2 0 0 5 ) /A i i vTeV"
Quark-Gluon. P l a s m a , ,-. deconfinem ent R J11C: A xi - A u (n o w!) V , 200 AGeV
1
J
SPS
PL
E, -
I D ^ \ ;• v
!
fio-A i
rip
i'.i ' > e \
V"*".. i
/1
"V \
\ G , J Au-An, (past') F - 11 AGeV
H a d i o n Gab i mfii< cm ent neutron stars?
M13 ;
Figure 15. High energy nuclear physics: temperature vs. baryon density. A very high energy density can be reached at the LHC in nuclei collisions, extending t h e SPS results, close t o t h e conditions of the early Universe (very low chemical potential).
,r-
I
Figure 16. C P violation: t h e unitarity triangle can be closed by combining the measurement of sin 2/3 and Vub: the allowed regions for p and r) are compared with t h e uncertainty bands for | V^5|/[ V^j,[, Amd and A m a / A m j .
564
E L E C T R O N - P O S I T R O N - COLLIDERS R.-D. HEUER Institut fur Experimentalphysik, Universitdt Hamburg, Luruper Chaussee 149, 22761 Hamburg Germany E-mail: rolf-dieter.heuer@desy. de An electron-positron linear collider in the energy range between 500 and 1000 GeV is of crucial importance to precisely test the Standard Model and to explore t h e physics beyond it. The physics program is complementary to t h a t of the Large Hadron Collider. Some of the main physics goals and the expected accuracies of the anticipated measurements at such a linear collider are discussed. A short review of the different collider designs presently under study is given including possible upgrade paths to the multi-TeV region. Finally a framework is presented within which the realisation of such a project could be achieved as a global international project.
1
Introduction
A coherent picture of matter and forces has emerged in the past decades through intensive theoretical and experimental studies. It is adequately described by the Standard Model of particle physics. In the last few years many aspects of the model have been stringently tested, some to the per-mille level, with e+e", ep and pp machines making complementary contributions, especially to the determination of the electroweak parameters. Combining the results with neutrino scattering data and low energy measurements, the experimental analysis is in excellent concordance with the electroweak part of the Standard Model. Also the predictions of QCD have been thoroughly tested, examples being precise measurements of the strong coupling as and probing the proton structure to the shortest possible distances. Despite these great successes there are many gaps in our understanding. The clearest one is the present lack of any direct evidence for the dynamics of electroweak symmetry breaking and the generation of the masses of gauge bosons and fermions. The Higgs mechanism which generates the masses of the fundamental particles in the Standard Model, has not been experimentally established though the indirect evidence from precision measurements is very strong. Even if
565
successfully completed, the Standard Model does not provide a comprehensive theory of matter. There is no explanation for the wide range of masses of the fermions, the grand unification between the two gauge theories, electroweak and QCD, is not realised and gravity is not incorporated at the quantum level. Several alternative scenarios have been developed for the physics which may emerge beyond the Standard Model as energies are increased. The Supersymmetric extension of the Standard Model provides a stable bridge from the presently explored energy scales up to the grand unification scale. Alternatively, new strong interactions give rise to strong forces between W bosons at high energies. Quite general arguments suggest that such new phenomena must appear below a scale of approximately 3 TeV. Extra space dimensions which alter the high energy behaviour in such a way that the energy scale of gravity is in the same order as the electroweak scale are another proposed alternative. There are two ways of exploring the new scales, through attaining the highest possible energy in a hadron collider and through high precision measurements at the energy frontier of lepton colliders. This article is based on the results of
R.-D. Heuer
Electron-Positron-Colliders
many workshops on physics and detector studies for linear colliders. Much more can be found in the respective publications 1,2'3>4 and on the different Web sites 5>6>7>8. Many people have contributed to these studies and the references to their work can be found in the documents quoted above.
2
Complementarity of Lepton and Hadron Machines
It is easier to accelerate protons to very high energies than leptons, but the detailed collision process cannot be well controlled or selected. Electron-positron colliders offer a well defined initial state. The collision energy ,/s is known and it is tuneable thereby allowing the choice of the best suited centre-of-mass energy, e.g. for scanning thresholds of particle production. Furthermore, polarisation of electrons and positrons is possible. In proton collisions the rate of unwanted collision processes is very high, whereas the pointlike nature of leptons results in low backgrounds. In addition, a linear collider offers besides e + e~ collisions the options of e~e~, ej and 77 collisions which could provide important additional insight. Hadron and Lepton Colliders are complementary and the present state of knowledge in particle physics would not have been achieved without both types of colliders running concurrently. Telling examples from the past are internal consistency tests of the electroweak part of the Standard Model. In 1994, the precision electroweak measurements of the Z° boson predicted a mass of the top quark from quantum corrections of M t o p = 178±11 t.\g GeV. The direct measurement at the Tevatron in the following years yielded M t o p = 174.3 ±5.1 GeV. The indirect measurement of M w = 80.363 ± 0.032 GeV agrees well with the direct mass measurements from Tevatron and LEP of M w = 80.450 ± 0.039 GeV. The Standard Model has been tested and so far
confirmed at the quantum level. Much progress about the possible mass range of the Higgs boson, if it exists, has been achieved in the past around five years. Lower bounds on the mass have been derived through direct searches at LEP running with ever increasing centre-of-mass energies until the year 2000. The Standard Model Higgs contribution to electroweak observables through loop corrections provides further indirect information. Although these corrections vary only logarithmically, oc log(M#/Mvi/), the accuracy of the electroweak data obtained at LEP, SLC and the Tevatron, provides sensitivity to MH and in turn an upper bound for the allowed mass range. The development of these bounds is shown in figure 1. The 95% upper limit for MH of 196 GeV is well within the reach of a linear collider with a centre-of-mass energy of 500 GeV.
| 95% limit LEPEWIH
> o
year
Figure 1. Development over the past years of limits for the mass of the Higgs boson from direct and indirect measurements,
In general, the physics target of the next generation of electron-positron linear colliders will be a comprehensive and high precision coverage of the energy range from Mz up to around 1 TeV. Energies up to around 3 to 5 TeV could be achieved with the follow-
566
R.-D. Heuer
Electron-Positron-Colliders
ing generation of colliders. The physics case for such a machine will depend on the results from the LHC and the linear collider in the sub-TeV range. 3
Selected Physics Topics
In this chapter, some of the main physics topics to be studied at a linear collider will be discussed. Emphasis is given to the study of the Higgs mechanism in the Standard Model, the measurements of properties of supersymrnetric particles, and precision tests of the electroweak theory. More details about these topics as well as information about the numerous topics not presented here can be found in the physics books published in the studies of the physics potential of future linear colliders 1-2>3'4. 3.1
Standard Model Higgs Boson
The main task of a linear electron-positron collider will be to establish experimentally the Higgs mechanism as the mechanism for generating the masses of fundamental particles: • The Higgs boson must be discovered. • The couplings of the Higgs boson to gauge bosons and to fermions must be proven to increase with their masses. • The Higgs potential which generates the non-zero field in the vacuum must be reconstructed by determining the Higgs self-coupling. • The quantum numbers (Jpc must be confirmed.
= 0++)
The main production mechanisms for Higgs bosons in e + e~ collisions are Higgsstrahlung e+e~ —> HZ and WW-fusion e + e~ —> veveH, and the corresponding crosssections as a function of MH are depicted in figure 2 for three different centre of mass energies. With an integrated luminosity of 500
567
100
200
300
400
500
600
700
Figure 2. T h e Higgs-strahlung and W W fusion production cross-sections as a function of M g for different %/s.
/ 6 _ 1 , corresponding to about two years of operation, some 10 5 events will be produced and can be selected with high efficiency and very low background. The Higgs-strahlung process e + e~ —> ZH, with Z —-> £+£~, offers a very distinctive signature ensuring the observation of the Standard Model Higgs boson up to the kinematical limit independently of its decay as illustrated in figure 3. The Higgs-strahlung process allows to measure the decay branching ratios of the Higgs boson and to test their dependence on the mass of the fundamental particles. The detectors proposed for linear colliders have excellent flavour tagging capability in order to distinguish the different hadronic decay modes (see for example 9 ). Therefore, the branching ratios can be determined with accuracies of a few percent, as shown in figure 4. The determination of the Yukawa coupling of the Higgs boson to the top quark is provided by the process e+e~ —> ttH at y/s of about 800 GeV; for 1000 fb~l an accuracy of 6% can be expected. The Higgs boson quantum numbers can be determined through the rise of the cross section close to the production threshold and through the angular distributions of the H and Z bosons in the continuum.
Electron-Positron-Colliders
R.-D. Heuer
•
Data ZH->p.nX
O 200-
ui £1
B
120
140
Figure 4. The predicted Standard Model Higgs boson branching ratios (bands) and the expected experimental accuracies (points with error bars).
160
Recoil Mass [GeV]
Figure 3. The M+/x recoil mass distribution in the process e + e ~ —» HZ —> p+n~ for M H = 1 2 0 GeV, 500 fb'1 at ^ = 3 5 0 GeV
The Higgs boson production and decay rates discussed above, can be used to determine the Higgs couplings to gauge bosons and fermions. A global fit to the measured observables optimises the available information, accounts properly for the experimental correlations between the different measurements and allows to extract the Higgs couplings in a model independent way. As an example for the accuracies reachable with the newly developed program HFITTER 1, figure 5 shows Id and 95% confidence level contours for the fitted values of the couplings gc and gb to the charm and bottom quark with comparison to the sizes of changes expected from the minimal supersymmetric extension to the Standard Model (MSSM).
V = Xv2H2 + XvH3 + \XH\ The trilinear Higgs coupling XJJHH = 6At> can be measured directly in the double Higgs-strahlung process e+e~ —> HHZ —• qqbbbb. The final state contains six partons resulting in a rather complicated experimental signature with six jets, a challenging task calling for excellent granularity of the tracking device and the calorimeter 9 . Despite the low cross section of the order of 0.2 fb for MH = 120 GeV at y/s = 500 GeV, the coupling can be measured with an accuracy of better than 20% for Higgs masses below 140 GeV at ^ = 5 0 0 GeV with an integrated luminosity of lab~l as shown in figure 6. Measurements of Higgs boson properties and their anticipated accuracies are summarised in table 1. In summary, the Higgs mechanism can be established in an unambiguous way at a high luminosity electron-positron collider with a centre-of-mass energy up to around one TeV as the mechanism responsible for the spontaneous symmetry breaking of the electroweak interactions.
To generate a non-zero value of the Higgs field in the vacuum, the minimum o = v/\/2 of the self potential of the Higgs field V = X(4>2 — 2y2) must be shifted away from the 3.2 Supersymmetric Particles origin. This potential can be reconstructed by measuring the self couplings of the physi- Supersymmetry (SUSY) is considered the cal Higgs boson as predicted by the potential most attractive extension of the Standard
568
R.-D. Heuer
Electron-Positron- Colliders
SM Double Higgs-strahlung: e+ e~ H> ZHH a [ft] If)
. MSSM prediction: _
_
2O0 GeV < m A < 400 GoV
400 GeV <mA<
600 GeV
6O0GeV<mA
800 GeV
S
-
-- ,
-
^-_-v"<j
8 0 0 G e V < m A - lOQflfGeV
"
(b) '_
—
'"^StEBSm
v
•
\ - LC 95% CL -LC1o
MHfGeV]
Figure 6. T h e cross-section for double Higgsstrahlung in the Standard Model at y/s of 500 and 800 GeV together with the experimental accuracies expected for 1 afr - 1 (points with error bars).
mH = 120 GeV _ 1
.
1
,
1
,
1
9c/gc(SM)
Figure 5. Higgs coupling determination: The contours for
Model, which cannot be the ultimate theory for many reasons. The most important feature of SUSY is that it can explain the hierarchy between the electroweak scale of «100 GeV, responsible for the W and Z masses, and the Planck scale Mpi ~ 10 19 GeV. When embedded in a grand-unified theory, it makes a very precise prediction of the electroweak mixing angle sin 2 9\y in excellent concordance with the precision electroweak measurement. In the following, only the minimal supersymmetric extension to the Standard Model (MSSM) will be considered and measurements of the properties of the supersymmetric particles will be discussed. Studies of the supersymmetric Higgs sector can be found elsewhere 1 ' 2,3 ' 4 . In addition to the particles of the Standard Model, the MSSM contains their supersymmetric partners: sleptons l^^i (I = e, ju, r ) , squarks q, and gauginos g, X ^ X°- In the MSSM the multiplicative quantum number R-parity is conserved, Rp = +1 for par-
569
Table 1. Precision of the possible measurements of the Higgs boson properties.
mH mass spin CP total width 9HZZ 9HWW 9Hbb 9Hcc 9HTT
9HU
^HHH
120 GeV 0.06% yes yes 6% 1% 1% 2% 3% 3% 3% 20%
140 GeV 0.05% yes yes 5% 1% 2% 2% 10% 5% 6% ~30%
tides and Rp = — 1 for sparticles. Sparticles are therefore produced in pairs and they eventually decay into the lightest sparticle which has to be stable. As an example, smuons are produced and decay through the process e+e~" —» /t + /x~ —> n+/J.~X?Xi with X? as the lightest sparticle being stable and, therefore, escaping detection. The mass scale of sparticles is only vaguely known. In most scenarios some sparticles, in particular charginos and neutralinos, are expected to lie in the energy region accessible by the next generation of e+e~
Electron-Positron-Colliders
R.-D. Heuer mSUGRA
GMSB
t
800
=t 5
600
> 400
200
-
1 1 1 I
s
i
'•
V
i-
™
%=
}£-=-;;
tL
fe
(« =\x?
cant part of the spectrum should be measureable. In general, at an e + e~ collider produc" « » • ' = - = ft \%,, •— ^ " tion cross sections are large and backgrounds are rather small. Masses of sparticles can be ; determined from the decay kinematics, measured in the continuum. An example for such i! mesurements is given in figure 8. Typical ac- curacies are of the order 100 to 300 MeV. Excellent mass resolutions of the order of 50 MeV with an integrated luminosity of 100 fb~l can be obtained for the light charginos and neutralinos through the measurement of in mSUGRA, the excitation curves at production threshold, as also shown in figure 9. AMSB
x*
^=
— if i-<
zzzz In
Figure 7. Examples of mass spectra GMSB and AMSB models.
_
^
colliders also supported by the recent measurement of (g — 2)^ 10 . Examples of mass spectra for three SUSY breaking mechanisms (mSUGRA, GMSB, AMSB) are given in figure 7. The most fundamental problem of supersymmetric theories is how SUSY is broken and in which way this breaking is communicated to the particles. Several scenarios have been proposed in which the mass spectra are generally quite different as illustrated in figure 7. High precision measurements of the particle properties are therefore expected to distinguish between some of these scenarios. The study and exploration of Supersymmetry will proceed in the following steps: • Reconstruction of the kinematically accessible spectrum of sparticles and the measurement of their properties, masses and quantum numbers • Extraction of the basic low-energy parameters such as mass parameters, couplings, and mixings • Analysis of the breaking mechanism and reconstruction of the underlying theory. While it is unlikely that the complete spectrum of sparticles will be accessible at a collider with y/s up to around 1 TeV, a signifi-
40 80 120 di—jel energy EJJ [GeV]
20 40 d i - j e t m a s s mw
60 [GeV]
Figure 8. Di-jet energy and mass spectra at ^/I=320 GeV and 160 / 6 _ 1 for the reaction e+e~ —» X\X\ ~*
The reconstruction of the mechanism which breaks supersymmetry will give significant insight into the laws of Nature at energy scales where gravity becomes important as quantum effect. Various models like minimal supergravity mSURGA), gauge mediated SUSY breaking (GMSB), or anomaly mediated SUSY breaking (AMSB) have been proposed. These mechanisms lead to different spectra of sparticle masses as was shown already in figure 7. The supersymmetric renormalisation group equations (RGE's) are largely independent of the assumed properties of the specific SUSY model at high energies. This can be used to interpret the measured SUSY spectra. In a 'bottom up'
570
R.-D. Heuer
Electron-Positron-Colliders 3.3
^
200
Precision
Measurements
X1X1
256
260
264
Ecm.
268
[GeV]
Figure 9. Cross section near threshold for the process e + e ~ —> xtx-i > 10 fb~l per point.
approach, the measured electroweak scale SUSY parameters are extrapolated to high energies using these RGE's. Due to the high precision of the measured input variables, only possible at the linear collider, an accurate test can be performed at which energy scale certain parameters become equal. Most interesting, the assumption of grand unification of forces requires the gaugino mass parameters Mi,M2,M% to meet at the GUT scale (figure 10 (left)). Different SUSY breaking mechanisms predict different unification patterns of the sfermion mass parameters at high energy. With the high accuracy of the linear collider measurements these models can be distinguished as shown in figure 10 for the case of mSUGRA (middle) and GMSB (right). In summary, the high precision studies of supersymmetric particles and their properties can open a window to energy scales far above the scales reachable with future accelerators, possibly towards the Planck scale where gravity becomes important. 571
The primary goal of precision measurements of gauge boson properties is to establish the non-abelian nature of electroweak interactions. The gauge symmetries of the Standard Model determine the form and the strength of the self-interactions of the electroweak bosons, the triple couplings WWj and WWZ and the quartic couplings. Deviations from the Standard Model expectations for these couplings could be expected in several scenarios, for example in models where there exists no light Higgs boson and where the W and Z bosons are generated dynamically and interact strongly at high scales. Also for the extrapolation of couplings to high scales to test theories of grand unification such high precision measurements are mandatory. For the study of the couplings between gauge bosons the best precision is reached at the highest possible centre of mass energies. These couplings are especially sensitive to models of strong electroweak symmetry breaking. W bosons are produced either in pairs, e+e" —• W+W~ or singly, e + e~ —> Weu with both processes being sensitive to the triple gauge couplings. In general the total errors estimated on the anomalous couplings are in the range of few x 10~ 4 . Figure 11 compares the precision obtainable for A K 7 and AA7 at different machines. The measurements at a linear collider are sensitive to strong symmetry breaking beyond A of the order of 5 TeV, to be compared with the electroweak symmetry breaking scale h-BWSB = 4irv « 3 TeV. One of the most sensitive quantities to loop corrections from the Higgs boson is the effective weak mixing angle in Z boson decays. By operating the collider at energies close to the Z-pole with high luminosity (GigaZ) to collect at least 109 Z bosons in particular the accuracy of the measure-
Electron-Positron-Colliders
R.-D. Heuer „500 5 450 2 400 350 300
250 200 150 100
:
_—
50 0
,l ,l
2
1o
105
I ,l
108
10"
I ,l , l , l
I
101410 0 (GeV)
Q (GeV)
Figure 10. Extrapolation of SUSY parameters measured at the electroweak scale t o high energies.
u~
^V*'S;.'--
<1
' -2
10
III!
|§|
§§i&-
?Mk?
" W8&
-
W$-. KjpjH'-
IP mi'4
-3
10
:
mm T4&
-4
H
LEP
i-%&$%•&?
IH TEV
1 •
w$$ LHC
TESLA 500
« TESLA 800
LEP
TEV
LHC
TESLA 500
TESLA 800
Figure 11. Comparison of constraints on the anomalous couplings Are-y and AA 7 at different machines
ment of sin 2 0\ *, can be improved by one order of magnitude wrt. the precision obtained today n . With both electron and positron beams longitudinally polarised, sin 2 9leff can be determined most accurately by measuring the left-right asymmetry ALR = Ae = 2veae/(v1 + a 2 ) with ve (a e ) being the vector (axialvector) couplings of the Z boson to the electron and ve/ae = 1 - 4sin 2 0g,, for pure Z exchange. Particularly demanding is the precision of 2 x 1 0 - 4 with which the polarisation needs to be known to match the statistical accuracy. An error in the weak mixing angle of A s i n 2 ^ / / = 0.000013 can be expected. Together with an improved determination of the mass of the W boson to
a precision of some 6 MeV through a scan of the WW production threshold and with the measurements obtained at high energy running of the collider this will allow many high precision tests of the Standard Model at the loop level. As an example, figure 12 shows the variation of the fit x 2 to the electroweak measurements as a function of MH for the present data and for the data expected at a linear collider. The mass of the Higgs boson can indirectly be constraint at a level of 5%. Comparing this prediction with the direct measurement of MH consistency tests of the Standard Model can be performed at the quantum level or to measure free parameters in extensions of the Standard Model. This is
572
Electron-Positron-Colliders
R.-D. Heuer
Figure 12. A x 2 as a function of t h e Higgs boson mass for the electroweak precision d a t a today (2000) and after GigaZ running (LC).
of particular importance if M # > 200 GeV in contradiction to the current electroweak measurements. In summary, there is strong evidence for new phenomena at the TeV energy scale. Only the precision exploration at the linear collider will allow, together with the results obtained at the Large Hadron Collider, the understanding of the underlying physics and will open a new window beyond the centreof-mass energies reachable. Whatever scenario is realized in nature, the linear collider will add crucial information beyond the LHC. There is global consensus in the high energy physics community that the next accelerator based project needs to be an electronpositron linear collider with a centre-of-mass energy of at least 500 GeV. 4
Electron-Positron Linear Colliders
The feasibility of a linear collider has been successfully demonstrated by the operation 573
of the SLAC Linear Collider, SLC. However, aiming at centre-of-mass energies at the TeV scale with luminosities of the order of 1 0 3 4 c m _ 2 s _ 1 requires at least two orders of magnitude higher beam power and two orders of magnitude smaller beam sizes at the interaction point. Over the past decade, several groups worldwide have been pursuing different linear collider designs for the centre-ofmass energy range up to around one TeV as well as for the multi-TeV range. Excellent progress has been achieved at various test facilities worldwide in international collaborations on crucial aspects of the collider designs. At the Accelerator Test Facility at KEK 12 , emittances within a factor two of the damping ring design have been achieved. At the Final Focus Test Beam at SLAC 13 demagnification of the beams has been proven; the measured spot sizes are well in agreement with the theoretically expected values. The commissioning and operation of the TESLA Test Facility at DESY 14 has demonstrated the feasibility of the TESLA technology. In the following, a short review of the different approaches is given.
4-1
TeV range
Three design studies are presently pursued: JLC 15 , NLC 16 and TESLA 17 , centred around KEK, SLAC and DESY, respectively. Details about the design, the status of development and the individual test facilities can be found in the above quoted references as well as in the status reports presented at LCWS2000 18 ' 19 . 20 . A comprehensive summary of the present status can be found in the Snowmass Accelerator R&D Report 21 , here only a short discussion of the main features and differences of the three approaches will be given with emphasis on luminosity and energy reach. One key parameter for performing the physics program at a collider is the centreof-mass energy achievable. The energy reach
Electron-Positron-Colliders
R.-D. Heuer
the accelerating structures. The design of NLC is based on normalconducting cavities using fRF of 11.4 GHz (X-band), for JLC two options, X-band or C-band (5.7 GHz) are pursued. The TESLA concept, developed by the TESLA collaboration, is using superconducting cavities (1.3 GHz). As an example for a linear collider facility, figure 13 shows the schematic layout of TESLA.
of a collider with a given linac length and a certain cavity filling factor is determined by the gradient achievable with the cavity technology chosen. For normalconducting cavities the maximum achievable gradient scales roughly proportional to the RF frequency used, for superconducting Niobium cavities, the fundamental limit today is around 55 MV/m. The second key parameter for the physics program is the luminosity £, given by
nbNJf, e J rep 4ira*
HD
(1)
where n^ is the number of bunches per pulse containing Ne particles, frep is the pulse repetition frequency, a*x the horizontal (vertical) beamsize at the interaction point, and HD the disruption enhancement factor. An important constraint on the choice of the parameters is the effect of beamstrahlung due to the emission of synchrotron radiation. The average fractional beam energy
damping ring
positron preaccelerator
loss 5E is proportional to Choosing a flat beam size (a* ^> a*) at the interaction point, 5B becomes independent of the vertical beam size and the luminosity can be increased by reducing a* as much as possible. Since cr* oc Jey(5* this can be achieved by a small vertical beta function /3* and a small normalised vertical emittance ey at the interaction point. The average beam power Pbeam = y/snbNefrep = T)PAC is obtained from the mains power PAC with an efficiency r). Equation (1) can then be rewritten as £ oc
VPAC
electron-positron collision high energy physics experiments positron source
aux. positron and 2nd electron source
damping ring
(2)
High luminosity therefore requires high efficiency T] and high beam quality with low emittance ey and low emittance dilution Ae/e oc f%F, which is largely determined by the RF frequency fup of the chosen technology The fundamental difference between the three designs is the choice of technology for
Figure 13. Schematic layout of TESLA
Table 2 compares some key parameters for the different technologies at ^/s = 500 GeV, like repetition rate frep for bunch trains
574
Electron-Positron-Colliders
R.-D. Heuer Superconducting Cavity Performance A 1-GB!I EP
A
Cornell
m 1st test *
$ m
9
TTF
• •
CEBAF test stand
.CEBAF linac 1996
1997
1998
_.
1999
2000
2001
Figure 14. Evolution of superconducting cavity performance. T h e average gradient achieved with TESLA 9-cell cavities produced in industry (first test, no additional processing) is shown as dots.
with Nb bunches, the time AT& between bunches within a train which allows head on crossing of the bunches for TESLA but requires a crossing angle for the other designs. The design luminosity £, beam power Pbeam and the required mains power PAC illustrate that for a given mains power the superconducting technology delivers higher luminosity. On the other hand the lower gradient Gacc requires a longer linac for the same centre-of-mass energy reach. As can be seen from table 2 the X-band machines call for a beam loaded (unloaded) gradient of some 50 (70) MV/m for ^s of 500 GeV. Recently, it has been found that high gradient operation of normalconducting cavities results in surface damage of the structures. Intense R&D is going on in collaboration between SLAC, KEK and CERN in order to understand and resolve the problem. At present it seems that the onset of the damage depends on the structure length and the group velocity within the cavity 16 . The TESLA design requires 23.5 MV/m for A/S = 500 GeV, a gradient which is meanwhile routinely achieved for cavities fabricated in industry as illustrated in figure 14. Table 2 also contains the presently planned length of the facilities 17>16>22>23. An
575
Figure 15. Excitation curves of three electropolished single-cell cavities. Gradients well above 35 M V / m are reached.
upgrade in energy up to around one TeV seems possible for all designs. In the NLC case, more cavities would be installed within the existing tunnel, in the JLC case, the tunnel length would have to be increased to house more cavities. In the TESLA case, a gradient of around 35 MV/m is needed to reach y/s of 800 GeV within the present tunnel length. Higher energies would probably require an extension of the tunnel. Such gradients have repeatedly been reached in tests of single-cell cavities whose surfaces have been electropolished not only chemically treated. The result of this common effort from KEK, CERN, Saclay and DESY is shown in figure 15. In summary, all designs are very well advanced. The TESLA collaboration has presented a fully costed Technical Design Report in March 2001. The other collaborations are expected to provide such reports within the next years. The construction of a linear collider with at least 500 GeV centre-of-mass energy, with upgrade potential to around one TeV, could start soon. 4.2
Multi-TeV Range
To reach centre-of-mass energies beyond the TeV range, up to 3-5 TeV, a two beam accel-
R.-D. Heuer
Electron-Positron-Colliders
Table 2. Comparison of some crucial parameters at 500 GeV for the different technologies under study, see text for details.
Jrep [-H-ZJ Nb
ATb [ns] bunch crossing 7Ve/6wnc/i[1010]
£[10 3 4 cm- 2 s- 1 ] PbeamWW] PAC(linacs)[MW] Gacc [MV/m] Ltot [km]
TESLA 5 2820 337 head on 2 553/5 3.2 3.4 22.6 97 23.5 33
NLC 120 190 1.4 angle .75 245/2.7 4.7 2 13.2 132 48 30
JLC-C 100 142 2.8 angle 1.11 318/4.3 3.9 1.3 12.6 220 36 linac 19
tain a large accelerator facility. The model is based on the experience of large experimental collaborations, particularly in particle physics. Some key elements are listed below:
eration concept (CLIC) with very high accelerating fields is being developed at CERN 24 . The schematic layout of that facility is shown in figure 16. It is optimised for y/s of 3 TeV, using high frequency (30 GHz) normalconducting structures operating at very high accelerating fields (150 MV/m). The present design calls for bunch separations of .67 ns, a vertical spotsize of 1 nm and beamstrahlung 8E of 30%. For this promising concept a new test facility is under construction at CERN which should allow tests with full gradient starting in 2005. 5
JLC-X 150 190 1.4 angle 0.7 239/2.57 5.3 2.64 17.6 141 50.2 16
• it is not an international permanent institution, but an international project of limited duration; • the facility would be the common property of the participating countries; • there are well defined roles and obligations of all partners; • partners contribute through components or subsystems;
Realisation
The new generation of high energy colliders most likely exceeds the resources of a country or even a region. There is general consensus that the realisation has to be done in an international, interregional framework. One such framework, the so called Global Accelerator Network (G AN), has been proposed to ICFA in March 2000. A short discussion of the principle considerations will be presented here, more details can be found in ref. 2 5 . The GAN is a global collaboration of laboratories and institutes in order to design, construct, commission, operate and main-
• design, construction and testing of components is done in participating institutions; • maintenance and running of the accelerator would be done to a large extent from the participating institutions. The GAN would make best use of worldwide competence, ideas and resources, create a visible presence of activities in all participating countries and would, hopefully, make the site selection less controversial.
576
R.-D. Heuer
Electron-Positron-Colliders
Overall Layout at 3 TeV
cue
MW/m RF power at 30 GHz
Figure 16. Overall layout of the CLIC complex for a centre-of-mass energy of 3 TeV.
ICFA has set up two working groups to study general considerations of implementing a GAN and to study the technical considerations and influence on the design and cost of the accelerator. The reports of these working groups can be found on the web 2 6 . Their overall conclusion is that a GAN can be a feasible way to build and operate a new global accelerator, although many details still need to be clarified.
6
Summary
There is global consensus about the next accelerator based project in particle physics. It has to be an electron-positron linear collider with an initial energy reach of some 500 GeV with the potential of an upgrade in centreof-mass energy. The physics case is excellent, only a few highlights could be presented here. There is also global consensus that concurrent operation with LHC is needed and fruitful. Therefore, a timely realisation is mandatory. The technical realisation of a linear collider is now feasible, several technologies are either ripe or will be ripe soon. A fast consensus in the community about the technology is
577
called for having in mind a timely realisation as a global project with the highest possible luminosity and a clear upgrade potential beyond 500 GeV. Acknowledgments The author would like to express his gratitude to all people who have contributed to the studies of future electron-positron linear colliders from the machine design to physics and detector studies. Special thanks go to the organisers and their team for a very well organised, inspiring conference as well as for the competent technical help in preparing this presentation. References 1. J.A. Aguilar-Saavedra et al, TESLA Technical Design Report, Part III, Physics at an e+e™ Linear Collider, DESY 2001-011, ECFA 2001-209, hepph/0106315. 2. T. Abe et al, Linear Collider Physics Resource Book for Snowmass 2001, BNL52627, CLNS 01/1729, FERMILABPub-01/058-E, LBNL-47813, SLAC-R-
R.-D. Heuer
3.
4.
5.
6.
7.
8.
9.
10. 11. 12.
13.
14. 15. 16.
570, UCRL-ID-143810-DR, LC-REV2001-074-US, hep-ex/0106055-58 K. Abe et al, Particle Physics Experiments at JLC, KEK-Report 2001-11, hep-ph/0109166. Proceedings of LCWS, Physics and Experiments with Future Linear Colliders, eds A. Para, H.E. Fisk, (AIP Conf. P r o c , Vol 578, 2001). Worldwide Study of the Physics and Detectors for Future e+e~ Colliders http://lcwws.physics.yale.edu/lc/ ACFA Joint Linear Collider Physics and Detector Working Group http://acfahep.kek.jp/ 2nd Joint ECFA/DESY Study on Physics and Detectors for a Linear Electron-Positron Collider htt p: / /www. desy. de/conferences/ecfadesy-lc98.html A Study of the Physics and Detectors for Future Linear e+e~ Colliders: American Activities http://lcwws.physics.yale.edu/lc/america.html G. Alexander et al, TESLA Technical Design Report, Part IV, A Detector for TESLA, DESY 2001-011, ECFA 2001209. H. N. Brown et al. [Muon g-2 Collaboration], Phys. Rev. Lett. 86 (2001) 2227 J. Drees, these proceedings E.Hinode et al, eds., KEK Internal 954, 1995, eds J.Urakawa and M.Yoshioka, Proceedings of the SLAC/KEK Linear Collider Workshop on Damping Ring, KEK 92-6, 1992 The FFTB Collaboration: BINP (Novosibirsk/Protvino), DESY, FNAL, KEK, LAL(Orsay), MPI Munich, Rochester, and SLAC Proposal for a TESLA Test Facility, DESY TESLA-93-01, 1992 KEK-Report 97-1, 1997. Zeroth Order Design Report for the Next Linear Collider, SLAC Report
Electron-Positron- Colliders
17.
18. 19. 20. 21.
22. 23. 24. 25.
26.
578
474, 1996. 2001 Report on the Next Linear Collider, Fermilab-Conf-01-075E, LBNL-47935, SLAC-R-571, UCRLID-144077 J. Andruszkow et al, TESLA Technical Design Report, Part II, The Accelerator, DESY 2001-011, ECFA 2001-209 O.Napoly, TESLA Linear Collider: Status Report, in ref 4 T.O. Raubenheimer, Progress in the Next Linear Collider Design, in ref 4 Y.H. Chin et al Status of JLC Accelerator Development, in ref 4 A. Chao et al, 2001 Snowmass Accelerator R&D Report, http://www.hep.anl. gov/pvs/dpb/Snowmass.pdf Y.H. Chin, private communication H.Matsumoto, T.Shintake, private communication I.Wilson, A Multi-TeV Compact e+e~ Linear Collider, in ref 4 F. Richard et al, TESLA Technical Design Report, Part I, Executive Summary, DESY 2001-011, ECFA 2001-209, hepph/0106314. http://www.fnal.gov/directorate/icfa/ icfa_tforce_reports. html
N E U T R I N O FACTORY A N D M U O N COLLIDER R & D STEVE GEER Fermi National Accelerator Laboratory, P. O. Box 500, Batavia, IL 60510, USA E-mail: [email protected] European, Japanese, and US Neutrino Factory designs are presented. T h e main R&D issues and associated R&D programs, future prospects, and the additional issues t h a t must be addressed to produce a viable Muon Collider design, are discussed.
1
Introduction
The development of a very intense muon source capable of producing a millimole of muons per year would enable a Neutrino Factory 1 , and perhaps eventually a Muon Collider 2 , to be built. In the last two years Neutrino Factory physics studies 3 have mapped out an exciting Neutrino Factory physics program. In addition, Neutrino Factory feasibility studies 4 ' 5 , 6 have yielded designs that appear to be "realistic" provided the performance parameters for the critical components can be achieved. Some of the key components will need a vigorous R&D program to meet the requirements. Neutrino Factory R&D activities in Europe 7 , Japan 6 , s , and the US 9 are ongoing and have, in fact, resulted in three promising variants of the basic Neutrino Factory design. In the following the various Neutrino Factory schemes are briefly described. The main R&D issues, and the ongoing R&D programs are summarized, and R&D results presented. Finally, the additional issues that must be addressed before a Muon Collider can be proposed are briefly discussed.
2
Neutrino Factory Schemes
In all of the present Neutrino Factory schemes an intense multi-GeV proton source is used to make low energy charged pions. The pions are confined within a large acceptance decay channel. The daughter muons produced from 7r± decays are also confined 579
within the channel, but they occupy a large phase-space volume, and this presents the main challenge in designing Neutrino Factories and Muon Colliders. In the US and European designs the strategy is to first reduce the energy spread of the muons by manipulating the longitudinal phase-space they occupy using a technique called "phase rotation". The transverse phase-space occupied by the muons is then reduced using "ionization cooling 10 ". After phase rotation and ionization cooling the resulting muon phase space fits within the acceptance of a normal type of accelerator. In the Japanese scheme an alternative strategy is pursued, in which the large muon phase-space is accommodated using so called FFAG's, which are very large acceptance accelerators. Finally, in all three schemes the muons are accelerated to the desired final energy (typically in the range from 20 - 50 GeV), and injected into a storage ring with either two or three long straight sections. Muons decaying within the straight sections produce intense neutrino beams. If the straight section points downwards, the resulting beam is sufficiently intense to produce thousands of neutrino interactions per year in a reasonably sized detector on the other side of the Earth 1 !
2.1
US Scheme
In the last 18 months there have been two Neutrino Factory "Feasibility" Studies 4 ' 5 in the US. Within these studies engineering de-
Steve Geer
Neutrino Factory and Muon Collider R&D
,.--•, 166eV ! Proton' Driicer Target'Station , 50m. decay/drift 100m ina .Ulnae.
0 BLA2: 11-60 6eSA 7.5 MeWm average Apeel Fret): 400 Mhz Turns: 5 : 60 m
2)
300m
Arc: 380 m Linac: 2 x 6 0 0 m
60m bunching lAOrri cosimq 1.6OeV,200ftfHz. Linac 0.2 > S.GeV
~
-600m
.w
n U l : 3 1.1 Gs V 1 1
•• j
7.5MeV/m average " " Accel.'Freq:20D'Mhz ^ • • Turns:'.' '4 •; 30 m. Arc. 1 DO m Linac: 2x150.m
' K ^-.'
900m .Storage Ring Circ = 1800 m : Straight = BOOm
1 'i i
. 50G«V muons
180 turns-17e
:
"
-1200m
-1500m
<J v per year
1800m
Figure 1. US Neutrino Factory schematic from design study I [4],
signs have been developed and detailed simulations performed for each piece of the Neutrino Factory complex. Study I was initiated by the Fermilab Director, and conducted between October 1999 and April 2000. The Study I design was for a 50 GeV Neutrino Factory with the neutrino beams pointing down 13° below the horizon, corresponding to a baseline of 2900 km. A schematic is shown in Figure 1. The proton source for Study I n was assumed to be a 16 GeV synchrotron producing 3 ns long bunches, operating at 15 Hz, and providing a 1.2 MW beam (4.5 x 1014 protons per sec) on an 80 cm long carbon target located within a 20 T solenoid. The solenoid radially confines essentially all of the produced ^ . Downstream of the target and target solenoid the pions propagate down a 50 m long decay channel consisting of a 1.25 T super-conducting solenoid which confines the n^ and their daughter muons within a warm bore of 60 cm. At the end of the decay channel 95% of the initial n^ have decayed and, per incident proton, there are ~ 0.2 muons with energies < 500 MeV captured within the beam transport system.
The muon system downstream of the decay channel is designed to produce a bunched cold muon beam with a central momentum of 200 MeV/c (T = 120 MeV). Beyond the decay channel, the first step is to reduce the muon energy spread using a 100 m long induction linac which accelerates the late low energy particles and de-accelerates the early high energy particles (phase rotation). After the induction linac, a 2.45 m long liquid hydrogen absorber is used to lower the central energy of the muons. Muons with energies close to the central value are then captured within bunches using a 201 MHz R F system within a 60 m long channel. Throughout the induction linac, liquid hydrogen absorber, and bunching system the muons are confined radially using a solenoid field of a few Tesla. The buncher produces a string of muon bunches, captured longitudinally, but still occupying a very large transverse phasespace. Downstream of the buncher the muons are cooled transversely within a 120 m long ionization cooling channel. The phase-space occupied by the muon bunches will then fit with the acceptance of an acceleration system consisting of a linac that accelerates the muons to 3 GeV, and two recirculating linear accelerators (RLA's) that raise the muon energies to 50 GeV. The muons are then injected into a storage ring. The ring is tilted downwards at the desired angle (13°), has a circumference of 1800 m, and has two 600 m long straight sections. One third of the injected muons will decay in the downward pointing straight section. A detailed simulation of the Study I design shows that this scheme will produce about 6 x 10 19 muon decays per operational year (defined as 2 x 10 7 seconds) in the downward pointing straightsection, which is a factor of 3 less than the initial design goal for the study. The resulting beam intensity would be sufficient for a socalled "entry-level" Neutrino Factory, but insufficient for a "high-performance" machine. Study II was initiated by the BNL Di-
580
Neutrino Factory and Muon Collider R&D
Steve Geer
4MW Eunch
^^**"8®*^
^ ^ ^ 2^GeV Superc6ftAvctLn4lTLiflae
p*03~lGeV/c
"' 2 € l C 5 S At™a«l«« rat ( 144 6 „ , , „, „,,)
|
r^^***%^ T u ^ P % L - j " 44/88 MHs capture V_ ^^*ifc_Horn ^-^^-P9°l | n B acceleration
p*3£eV(c~10GeV{c
2X10** # j r ^S»^-'- ^ \ 2TOGeV ^*^. -^"S^rcnrculato 10 5u GeV'*' rcc.rculatcc
\
«.
WOm
HFv./yr 4 . »«" V,/yr
p*U£eWc~20(50)GeV/c
muon storage Figure 3. Japanese Neutrino Factory schematic [8]. ^
V
Muon. dtcav ring Ecvu: on jtbnudplBae
* ^
**'
Figure 2. European Neutrino Factory schematic [12].
rectorate, and built upon the work done during Study I. The Study II was effort focussed on improving the Neutrino Factory design to achieve higher beam intensities. The main improvements came from using (i) a 4 MW proton source, (ii) a liquid Hg target, (iii) an improved induction linac design, and (iv) an improved cooling channel design. Detailed simulations of the Study II design predicted 2 x 10 20 muon decays per operational year (defined as 1 x 107 seconds) in the downward pointing straight-section, thus achieving the initial goal. 2.2
European Scheme
The European Neutrino Factory design 12 has many similarities to the US design. However, the European studies have explored alternative technology choices for several key subsystems. A schematic is shown in Fig. 2. The European studies have focused on a design in which: (i) The proton driver consists of a 2.2 GeV super-conducting linac (rather than a higher energy synchrotron), followed by an accumulator ring designed to produce short proton pulses 13 . The beam power is 4 MW. (ii) In addition to a liquid metal jet, a target consisting of water cooled Ta spheres is also being considered, (iii) The charged pions are focussed using a magnetic horn (rather than a high-field solenoid) with a 4 cm waist ra-
581
dius, and a peak current of 300 kA. (iv) After a 30 m drift, the muon energy spread is reduced using 44 MHz RF cavities (rather than induction an linac). (v) The ionization cooling channel (described later) uses 44 MHz and 88 MHz RF cavities. These are lower frequency cavities than employed in the US scheme, (vi) A bowtie-shaped storage ring (rather than a simple race-track design) has been considered for the final muon ring. Simulations of the European design predict that the resulting neutrino beams will have intensities comparable to the corresponding beams from the US design. A comprehensive design study at the level of those in the US is not yet complete. It seems plausible that, at the end of the day, the optimal European/US-type Neutrino Factory design will inherit some of the technical choices being explored in the European studies, and some of those explored in the US studies.
2.3
Japanese Scheme
The Japanese Neutrino Factory scheme is shown in Fig. 3. The front end of the design consists of a 50 GeV proton synchrotron providing a 4 MW beam (corresponding to an upgraded JHF complex) incident on a target within a 12 T solenoid. This is followed by a large acceptance pion decay channel. However, instead of manipulating and cooling the phase space occupied by the muons exiting the decay channel, very large acceptance FFAG (Fixed Field Alternating
Steve Geer
Neutrino Factory and Muon Collider R&D
Gradient) accelerators are used to raise the beam energy before injecting into a storage ring with long straight sections. The predicted neutrino yield from this scheme seems to be comparable to the corresponding predicted yields from the US/European designs, although more detailed simulations will be needed to confirm this. Although the Japanese Neutrino Factory scheme might benefit from the addition of some muon cooling, in principle the use of FFAGs evades the need to cool the muons before injecting them into an accelerator. This simplification comes at the price of a more challenging accelerator, requiring complicated large aperture magnets and broadband low frequency high gradient RF cavities. Modern design tools have made practical the task of designing the FFAG magnets, and a small proof-of-principle (POP) FFAG accelerator 14 has been built and successfully operated. A second test FFAG, designed to accelerate protons to 150 MeV, is under construction. Furthermore, with some US participation, an R&D program is underway in Japan to develop the required cavities. It is too early to conclude whether the promising Japanese Neutrino Factory scheme will lead to a more cost effective solution than the European/US-type designs. It may even be that the optimum solution consists of a combination of the two concepts, with some phase space manipulation and cooling, but using large acceptance FFAGs for the acceleration.
3
Pion production fc Target R & D
To produce a sufficient number of useful ^ , and hence a sufficient number of muons, all Neutrino Factory schemes begin with a MWscale proton driver, a pion production target, and a collection system optimized to capture as many ^ as possible. To establish confidence in the predicted -K^ fluxes downstream of the target several secondary particle pro-
Figure 4. Proof of Principle (POP) FFAG accelerator [14].
duction experiments are underway 15>16>17. In addition, a target R&D program is being pursued to develop targets that can operate within MW-scale proton beams 18 . 3.1
Pion production
experiments
Figure 5 shows, as a function of the primary proton beam energy, predicted charged pion yields for 7r+ and ir~ captured within a decay channel downstream of a Neutrino Factory target system. Over a broad interval of proton beam energies, at fixed beam power the yields are approximately independent of beam energy. Hence, a wide variety of multiGeV MW-scale proton drivers can be considered when designing a Neutrino Factory. The E910 experiment at BNL 15 has recently measured ir+ and TT~ yields for several targets and different incident proton beam energies. Some of the results are shown in Fig. 6. Note that (i) the measurements are in fair agreement with MARS calculations 15 , and (ii) the pion yields peak in the region 300 - 500 MeV/c. Hence, Neutrino Factory designs tend to have 7r± collection systems optimized to capture particles with momenta in this range. In the next few years we can anticipate further particle production measurements from the HARP experiment 16 at CERN, and the proposed P907 experiment 1 7
582
Steve Geer
Neutrino Factory and Muon Collider R&D 17.5 G e V / c p Cu
•
500
t
7T1"
I
Yield/p 5^=t^=* " IrT
t
\
i -JF-X
t *//
v1
• I
i
\
\
!
i
i
Yield / i
•
;
Cos(fl) 0.6-1.0 ?400 • • E910 - MARS
> TT + / _ + X —x
'••
. .
TT + _-
!
D ^^H^£^L_^ —^. •
i . i . 0
I . . .. 5
! !
i !
i i
10
15
JO
i . . , ,. iI , . . .. i •.
IT" .• .
I . 25
_ J 3S
Proton Energy (GeV) Figure 5. MARS predictions for the yield of pions captured within a Neutrino Factory pion decay channel, shown as a function of the primary proton beam energy [4]. More explicitly, the number of 7r+ + /*+ (filled symbols) and n~ + ji~ (open symbols) within an energy interval 30 MeV < E < 230 MeV is shown 9 m downstream of an 80 cm long 0.75 cm radius carbon target within a 20 T solenoid, and tilted 50 mrad with respect to the solenoid axis. Triangles show the yields divided by the primary proton energy.
at Fermilab. This will enable the relevant -n^ production measurements to be cross-checked and extended to cover the entire region of interest for Neutrino Factory designs. 3.2
Target R&D
Entry-Level Neutrino Factories providing a few xlO 1 9 useful muon decays per year require proton beams with beam powers of ~ 1 MW, and short proton bunches, typically a few ns long. It is believed that carbon targets can operate with these beam parameters. This has been tested at BNL by the E951 Collaboration 19 . Two different types of carbon rod were exposed to the AGS beam and strain gauge data taken. The beam induced longitudinal pressure waves and transverse reflections were both measured. The strains in the two types of rod differed by an order of magnitude, the most promising rod being made from an anisotropic carboncarbon composite.
583
Figure 6. Pion production measurements from the BNL E910 experiment, compared with MARS predictions [15].
High performance Neutrino Factories providing O(10 20 ) useful muon decays per year require proton beams with beam powers of ~ 4 MW. Targets must be developed to operate in these extreme conditions. Solid targets will melt unless very efficient cooling strategies can be developed. Rotating metal bands 20 and water cooled Ta spheres 21 are being considered, but the presently favored solution is to avoid problems with target melting and integrity by using a liquid Hg jet. In the US and Japanese schemes the target is in a high-field solenoid. Hence, the main R&D issues are (i) can a Hg jet be injected within a high-field solenoid without magneto-hydrodynamic effects disrupting the jet, (ii) after the jet has been destroyed by one proton pulse will it re-establish itself before the next pulse, and (iii) can the disrupted jet be safely contained within the target system. To address these questions, R&D has begun at CERN and by the E951 collaboration at BNL. Photographs of a liquid Hg jet injected into a high-field solenoid at Grenoble are shown in Fig. 7. The presence of a 13 T solenoid field seems to damp surface tension waves, improving the characteristics of the jet. This is clearly good news. The E951
Neutrino Factory and Muon Collider R&D
Steve Geer
tfl
I
*itt, '
&»
Figure 9. BNL E951 results [19], T h e sequence of pictures shows the Hg jet at the time of impact (t=0) of the AGS beam, and at t = 0.75 ms, 2 ms, 7 ms, and 18 ms. Time increases from left to right.
liquid Hg jet setup 19 in the BNL AGS beam is shown in Fig. 8. Figure 9 shows a sequence of pictures of the 1 cm diameter Hg jet taken by a high speed camera from the time of impact of 4 x 10 12 protons in a 150 ns long pulse from the 24 GeV BNL AGS. Jet dispersal is delayed for about 40/us, and it takes several ms before the jet, which has an initial velocity of 2.5 m/sec, is completely disrupted. The velocity of the out-flying Hg filaments, which appears to scale with beam intensity, has been measured to be ~ 10 m/sec for a deposited energy of 25 J/g. These velocities are fairly modest, implying that the disrupted jet can be easily contained within the target system. Furthermore, the Hg jet dispersal is mostly in the transverse direction, and after disruption it has been found that the jet quickly re-establishes itself. Figure 7. CERN-Grenoble tests in which a liquid Hg jet is injected into a solenoid providing 0 field (top picture) and 13 T (bottom picture) [13].
Proton Beam
Initial results with Hg jets are promising. However, high performance Neutrino Factories will require Hg jets with much higher velocities (~ 20 m/sec) to be developed and tested. The next steps in the E951 R&D program will require beam tests with a factor of a few higher beam intensities, and finally beam tests in which the Hg jet is injected into a 20 T solenoid. 4
Figure 8. Schematic of the E951 system to test a Hg jet in the BNL AGS beam [19].
Muon Cooling R & D
Before the muons can be accelerated the transverse phase-space they occupy must be reduced so that the muon beam fits within the acceptance of an accelerator. This means 584
Steve Geer
Absorber
Neutrino Factory and Muon Collider R&D
RF
Absorber
RF
Absorber
Figure 10. SFOFO cooling channel design [5]. A 5.5 m long section is shown, consisting of two 200 MHz four-cell cavities interleaved with three liquid hydrogen absorbers.
the muons must be "cooled" by at least a factor of a few in each transverse plane, and this must be done fast, before the muons decay. Stochastic- and electron-cooling are too slow. It is proposed to use a new cooling technique, namely "ionization cooling" 10 . In an ionization cooling channel the muons pass through an absorber in which they lose transverse- and longitudinalmomentum by dE/dx losses. The longitudinal momentum is then replaced using an RF cavity, and the process is repeated many times, reducing the transverse momenta. This cooling process will compete with transverse heating due to Coulomb scattering. To minimize the effects of scattering we chose low-Z absorbers placed in the cooling channel lattice at positions of low-/3j_ so that the typical radial focusing angle is large. If the focusing angle is much larger than the average scattering angle then scattering will not have much impact on the cooling process. 4-1
US Cooling Channel Design
Studies I and II used two simulation tools developed by the Neutrino Factory and Muon 585
Collider Collaboration: (i) A specially developed tracking code ICOOL 22 , and (ii) A GEANT based program with accelerator components (e.g. RF cavities) implemented. Out of these design and simulation studies, two promising cooling channel designs have emerged: (i) The "SFOFO" lattice in which the absorbers are placed at low-/3j_ locations within high-field solenoids. The field rapidly decreases from a maximum to zero at the absorber center, and then increases to a maximum again with the axial field direction reversed. Figure 10 shows the design for a 5.5 m long section of the ~ 100 m long cooling channel. The section shown has 30 cm long absorbers with a radius of 15 cm, within a system of solenoids with a peak axial field of 3.5 T. Towards the end of the cooling channel the maximum field is higher (5 T) and the lattice period shorter (3.3 m). The RF cavities operate at 201 MHz and provide a peak gradient of 17 MV/m. Detailed simulations predict that the SFOFO channel
Steve Geer
Neutrino Factory and Muon Collider R&D
increases the number of muons within the accelerator acceptance by a factor of 3-5 (depending on whether a largeor very-large acceptance accelerator is used). (ii) The "DFLIP" lattice in which the solenoid field remains constant over large sections of the channel, reversing direction only twice. In the early part of the channel the muons lose mechanical angular momentum until they are propagating parallel to the axis. After the first field flip the muons have, once again, mechanical angular momentum, and hence move along helical trajectories with Lamour centers along the solenoid axis. Further cooling removes the mechanical angular momentum, shrinking the beam size in the transverse directions. The field in the early part of the channel is 3 T, increasing to 7 T for the last part. Detailed simulations show the performances of the DFLIP and SFOFO channels are comparable. Earlier less detailed studies 2 have shown that a much larger cooling factor will be required for a muon collider. This will require an extended cooling channel, using higher frequency (e.g. 805 MHz) cavities, higher field solenoids, and possibly liquid lithium lenses 23 . 4.2
MUCOOL R&D
Muon cooling channel design and development is being pursued within the US by the MUCOOL collaboration 24 . The mission of the MUCOOL collaboration includes benchtesting all cooling channel components, and eventually beam-testing a cooling channel section. The main component issues that must be addressed before a cooling channel can be built are (i) can sufficiently high gradient RF cavities be built and operated in the appropriate magnetic field and radiation environment, (ii) can liquid hydrogen
MUCOOL Lab G test area, Fermilab [25].
absorbers with thin enough windows be built so that the dE/dx heating can be safely removed, and (iii) can the lattice solenoids be built to tolerance and be affordable? The MUCOOL collaboration has embarked on a design-, prototyping-, and testing-program that addresses these questions. This R&D is expected to proceed over the next 3 years.
805 MHz RF Tests Early design work for a Muon Collider resulted in a Muon Collider cooling channel design that required 805 MHz cavities operating in a 5T solenoid, with the cavities providing a peak gradient on axis of ~ 30 MV/m. This deep potential well is needed to keep the muons bunched as they propagate down the channel. This requirement led to two cavity concepts: (a) an open cell design, and (b) a design in which the penetrating nature of the muons is exploited by closing the RF aperture with a thin conducting Be window (at fixed peak power this doubles the gradient on axis).
586
Neutrino Factory and Muon Collider R&D
Steve Geer
Measnued $
0
Mo.l •nlculanoh.^
~cLi Power Dissipated Figure 12. Measurements of the deflection of a Be foil in an 805 MHz cavity at LBNL [26]. T h e measured deflections are shown as a function of the heat dissipated in the foil, and are compared with predictions from two finite element analysis calculations.
The MUCOOL collaboration has pursued an aggressive 805 MHz cavity development program, which is now advanced. A 12 MW high power test facility has been built and operated at Fermilab (Lab G). The Lab G facility enables 805 MHz cavities to be tested within a 5T solenoid. The main results to date are: (i) An open cell cavity suitable for a muon cooling channel has been designed, an aluminum model built and measured, and a prototype copper cavity built, tuned, and successfully tested at full power in the Lab G facility (Fig. 11). Dark current produced by the cavity has been identified as an important R&D issue 25 . (ii) A Be foil cavity has been designed at LBNL, a low power test cavity built and measured, and foil deflection studies made 26 to ensure the cavity does not detune when the foil is subject to RF heating. The foil deflection is reasonably well understood for small displacements (Fig. 12). A high power copper cavity with Be-foil windows is under construction at LBNL and the University of Mississippi, and will be tested at Lab G when ready. 201 MHz Cavity Development The cooling channel designs developed for the US Neutrino Factory studies require 201 MHz 587
RF cavities providing a gradient on axis of ~ 17MV/m. Preliminary cavity designs have been made. There are two concepts, both of which close the cavity aperture. The options are to use (a) a thin Be foil, exploiting the work done for the 805 MHz cavity, or (b) use a grid of hollow conducting tubes. Preliminary mechanical tests for both the grid and foil concepts are planned, and should proceed during the next few months. A 201 MHz prototype cavity will then be constructed, and should be ready for high power tests in about 2 years. Absorber Development The cooling channel liquid hydrogen absorbers must have very thin windows to minimize multiple scattering, and must tolerate heating of O(100 W) from the ionization energy deposited by the traversing muons. Absorber parameters for the Neutrino Factory study II cooling channel design are listed in Table 1. To adequately remove the heat from the absorbers requires transverse mixing of the liquid hydrogen. There are two design concepts that are being pursued 27. Forced flow design. The LH2 is injected into the absorber volume through nozzles, and cooled
Steve Geer
Neutrino Factory and Muon Collider R&D Table 1. LH2 absorber parameters in Neutrino Factory design study II [5].
Absorbers Early Late
Length (cm) 35 21
Radius (cm) 18 11
Number Needed 16 36
Heat (kW) Deposited ~0.3 ~0.1
Window Thick -ness (/jm) 360 220
Max. Pressure (atm) 1.2 1.2
Figure 13. KEK prototype absorber. The liquid hydrogen is to be mixed by convection and cooled with a local heat exchanger. A simulation of the convection, performed at IIT, is shown on the right.
using an external loop and heat exchanger. (ii) Convection design. Convection is driven by a heater at the bottom of the absorber volume, and heat removed by a heat exchanger on the outer surface of the absorber. A forced flow absorber prototype is being designed at the Illinois Institute of Technology (IIT) and will be constructed in the coming year. A convection prototype, designed by IIT, KEK, and the University of Osaka, is being constructed in Japan (Fig. 13). Both absorbers will be tested at Fermilab when complete. A first prototype 15 cm radius aluminum absorber window has been made at the University of Mississippi on a CNC milling machine and lathe. The window has a central thickness of 130 /j,m. The window thickness and profile were measured at FNAL and found to be within 5% of the nominal envelope. This verifies the manufacturing procedure. The window has been tested 27 under pressure in a setup at Northern Illinois Uni-
versity in which it was mounted on a backplate and water injected between window and plate. Strain gauge and photogrammetric measurements were made as a function of pressure, and the results compared with FEA predictions. The onset of inelastic deformation was predicted at 29 psig, a pinhole leak appeared at 31 psig, and rupture occurred at 44 psig. The windows required for a cooling channel absorber can be about twice as thick as the first prototype window. The results to date are therefore encouraging. Further window studies and tests are proceeding. 4-3
European Cooling R&D Program
The European cooling channel design 28 is similar in concept to the US design, but is based on 44 MHz and 88 MHz cavities 29 rather than 201 MHz cavities. To minimize the radii of the solenoids used to confine the muons within the channel, the cavities have been designed to wrap around the solenoids.
588
Neutrino Factory and Muon Collider R&D
Steve Geer
Figure 14. Absorber window test, showing an array of dots projected onto the window for photogrammetric measurements of its shape as it deforms under pressure [27].
Figure 15. CERN 88 MHz cavity [29] to be prepared for high power tests.
589
A full engineering design of this concept will be required to understand its feasibility. The initial transverse cooling is performed using 44 MHz cavities with four 1 m long RF cells between each 24 cm long LH2 absorber. The beam is then accelerated from 200 MeV to 300 MeV, and the cooling is continued using 88 MHz cavities with eight 0.5 m long cells between each 40 cm long LH2 absorber. The channel parameters are summarized in Table 2. Simulations of the channel performance with detailed field-maps have not yet been made. However, simulations using simpler field maps yield promising results: the effect of the channel is to increase the number of muons within the acceptance of the subsequent accelerating system by a factor of about 20. Whether or not the predicted increased yield is significantly degraded when full simulations are performed remains to be seen. In the meantime, a prototype 88 MHz cavity is being prepared at CERN (Fig. 15) for high power tests within the coming year.
Neutrino Factory and Muon Collider R&D
Steve Geer
Table 2. CERN ters [28].
cooling channel
Length Diameter Sol. Field RF Freq. RF Gradient Beam Energy
4-4
Channel 1 46 m 60 cm 2.0 T 44 MHz 2 MV/m 200 MeV
design
the first prototype absorbers to be filled. In Phase 2 a linac beam will be brought to the absorber area, and the 5T solenoid will be moved from Lab G so that the absorber can be tested in a magnet whilst exposed to a proton beam. The beam intensity and spot size will be designed to mimic the total ionization energy deposition and profile that corresponds to the passage of 10 12 — 10 13 muons propagating through a cooling channel. In addition, 201 MHz RF power will be piped to the test area from a nearby test-stand, enabling high-power tests to be made of a prototype 201 MHz cooling channel cavity exposed to the proton beam.
parame-
Channel 2 112 m 30 cm 2.6 T 88 MHz 4 MV/m 300 MeV
Cooling Experiments
A sequence of muon cooling-related experiments is being planned. The first, the MUSCAT experiment 30 , is already under way at TRIUMF. The second, the MUCOOL Component Test Experiment, is under construction at the Fermilab Linac. The third, an International Cooling Experiment 31 , is in the planning stage. The fourth, an eventual String Test Experiment, will be planned in the future.
International Cooling Experiment
The goal of the MUSCAT experiment at TRIUMF is the precise measurement of low energy (130, 150, and 180 MeV/c) muon scattering in a variety of materials that might be found within a cooling channel. In a second phase, the experiment will also measure straggling. Scattering measurements for Li, Be, C, Al, CH2, and Fe have already been made. Preliminary results seem to be in good agreement with expectations 30 . Further analysis is in progress. Measurements with LH2 are expected in the future.
A Europe-Japan-US International Cooling Experiment is currently being planned 31 . The goals are to (i) place a cooling channel section in a muon beam, and (ii) demonstrate our ability to precisely simulate the passage of muons confined within a periodic lattice as they pass through LH2 absorbers and highgradient RF cavities. In the envisioned experiment muons are measured one at a time at the input and output of the cooling section, and the precise response of the muons to the cooling section is determined. The main challenge to the design of this type of experiment arises from the prolific X-ray and dark current environment created by the RF cavities. This is currently under study at Lab G and elsewhere. If it is found that single particle detectors can function in this hostile environment, we anticipate a proposal being submitted sometime in 2002.
MUCOOL Component Test Experiment
5
A MUCOOL test area located at the end of the Fermilab 400 MeV Linac was proposed in the Fall of 2000, and is currently under construction. The project is being pursued in two phases. In Phase 1 a LH2 absorber test facility is being built, which will enable
The acceleration system has been identified as one of the cost drivers for a Neutrino Factory. In the US and European schemes the main acceleration systems use SC cavities. The US scheme, for example, uses 201 MHz SCRF delivering gradients of 15 MV/m with
MUSCAT
590
Muon Acceleration and Storage
Neutrino Factory and Muon Collider R&D
Steve Geer Q ~ 5 x 109. Although higher frequency SCRF cavities are no longer novel, 201 MHz is a relatively low frequency and the cavities are therefore large. The associated R&D issues are related to microphonics, fabrication and cleaning techniques and, because of the large stored energy, quench protection. Furthermore, the cavities must tolerate whatever stray magnetic fields they see within the accelerating lattice. To address these issues a 201 MHz SC cavity is being constructed at CERN and sent to Cornell for high-power testing. No major R&D issues have been identified for the final muon storage ring. Building a ring tilted at a large angle raises interesting, but not insurmountable, construction challenges 4 ' 5 . 6
lider will require a cooling channel with stronger radial focusing than is likely to be achieved with affordable high-field solenoids. A new technology (liquid lithium lenses 23 , optical stochastic cooling 32 , ... ) is required. (ii) Cost-effective acceleration: Acceleration is a cost driver for a Neutrino Factory, which requires muons accelerated to 2050 GeV. If a multi-TeV Muon Collider is to be affordable an efficient cost-effective acceleration system must be developed. (iii) Single muon bunches: The Neutrino Factory does not require the muons to be packaged into a small number of bunches. However, to maximize the luminosity, the muons for a Muon Collider should be packaged into one /J,+ and one fi~ bunch per cycle. This will require a more challenging bunching scheme and raises additional issues associated with having more intense muon bunches (e.g. space charge effects).
Muon Collider Issues
A Neutrino Factory, although motivated by its own physics program, would also provide a solid step in developing the technology that would be required to eventually build a Muon Collider. However, additional issues must be addressed before a Muon Collider could be proposed. In particular: (i) Muon Cooling: The 6-D cooling factor required for a Muon Collider is O(10 6 ) compared with the more modest factor of 0(100) for a Neutrino Factory. Muon cooling for a Muon Collider will require additional technology. In particular, present Muon Collider muon cooling schemes require the longitudinal phase-space to be reduced using "emittance exchange" in which some of the reduction in the transverse phase-space is traded for a reduction in the muon energy spread. Although progress towards a viable emittance exchange scheme has been made over the last two years, a convincing design has not yet emerged. In addition, to obtain the final transverse emittances required for a Muon Col-
(iv) Detector backgrounds: Decaying muons within the Muon Collider ring create an intense flux of energetic electrons in the neighborhood of the detector. This has been studied in detailed 33 , and elaborate shielding strategies have been shown to reduce the backgrounds down to levels that appear to be acceptable. Assuming that solutions can be found on paper for these Muon Collider design challenges, I believe it will take many extra years of R&D to develop the additional technology required for a Multi-TeV Muon Collider. However, the best way to eventually build a cost-effective multi-TeV lepton collider is far from clear, and R&D addressing Muon Collider issues is worthy of a significant investment by the community.
591
Steve Geer 7
Summary and P r o s p e c t s
Muon sources capable of delivering a millimole of muons per year seem feasible, and would enable Neutrino Factories, and perhaps eventually Muon Colliders, to be built. Neutrino Factory designs have been developed in Europe, Japan, and the US. Three promising schemes are being studied, and a healthy R&D program is underway to develop the required technologies. The most challenging R&D questions are associated with targets for MW-scale proton beams, and the development of an ionization cooling channel. Additional challenges must be overcome before a Muon Collider can be proposed. With the present level of support, we can expect much progress in Neutrino Factory R&D over the next few years. However, the future level of support is uncertain, and I believe a significant increase (factor of two ?) in the R&D support will be needed, sustained over a handful of years, if we are ever to arrive at a "Technical Design Report". In addition, there must be good international collaboration to enable the most promising design to be eventually chosen, and pursued. There is already a healthy dialogue between the European, Japanese, and US R&D teams, and some cross-participation in the various R&D programs. This international collaboration would greatly benefit from an increase in support to enable, for example, an international ionization cooling experiment. Finally, it should be noted that Neutrino Factory R&D is being pursued by engineers, accelerator physicists, and particle physicists from Laboratories and Universities in Europe, Japan and the US. There are a broad range of interesting sub-projects to be pursued, and with adequate support, the prospects seem bright.
Neutrino Factory and Muon Collider R&D Acknowledgments This talk is based upon the work of many people. I am particularly indebted to material provided to me from my colleagues in Europe and Japan, most notably B. Autin, H. Haseroth, Y. Kuno, and Y. Mori. In addition I am indebted to all the members of the US Neutrino Factory and Muon Collider Collaboration, and those outside of the collaboration who have participated in the two US Neutrino Factory studies. My own humble contribution to the work was supported at the Fermi National Accelerator Laboratory, which is operated by Universities Research Association, under contract No. DE-AC0276CH03000 with the U.S. Department of Energy.
References 1. S. Geer, Phys. Rev. D57, 6989 (1998). 2. G.I. Budker, AIP Conf. Proc. 352 (AIP, New York, 1996) p.4; A.N. Skrinsky, AIP Conf. Proc. 352 (AIP, New York, 1996) p.6; C M . Ankenbrandt et al. (Muon Collider Collabration), V > ~ Collider: A Feasibility Study", Snowmass 96, BNL-52505; ibid. Phys. Rev. ST Accel. Beams 2, 081001 (1999). 3. B. Autin, A. Blondel, J. Ellis (editors), CERN yellow report 99-02., C. Albright et al., (Eds. S. Geer and H. Schellman), "Physics at a Neutrino Factory", Report to the Fermilab Directorate, May 10, 2000, hep-ex/0008064., see also the proceedings of NUFACT99 (Lyon, France), NUFACT00 (Monterey, US), and NUFACT01 (Tsukuba, Japan). There are many other neutrino factory physics papers. Amongst the most cited are: A. De Rujula, M.B. Gavela, P. Hernandez, Nucl.Phys.B547:21-38,1999; V. Barger, S. Geer, R. Raja, K. Whisnant, Phys.Rev.D62:013004,2000., and A. Cervera et al., Nucl.Phys.B579:17-
Neutrino Factory and Muon Collider R&D
Steve Geer 55,2000, Erratum-ibid.B593:731,2001. 4. T. Anderson et al. (Eds: N. Holtkamp and D. Finley), "A Feasibility Study of a Neutrino Source Based on a Muon Storage Ring", FERMILAB-Pub-OO/108-E (2000). 5. M. Goodman et al. (Eds: S. Ozaki, R. Palmer, M. Zisman, J. Gallardo), "Feasibility Study II of a Muon-Based Neutrino Source", BNL-52623, June 2001. 6. NufactJ Working Group (Eds. Y. Kuno, Y. Mori), "A Feasibility Study of a Neutrino Factory in Japan", see h t t p : / / www-prism.kek.jp/ nufactj/ index.html 7. R. Garoby and B. Autin, presentions at the Neutrino Factory Workshop NUFACT01, Tsukuba, Ibaraki, Japan, May 24-30, 2001; B. Autin, "European R&D for Neutrino Factory", CERN Neutrino Factory Note 101, see http://molat.home.cern.ch/molat/ neutrino/nfnotes.html 8. S. Machida, presented at the Neutrino Factory Workshop NUFACT01, Tsukuba, Ibaraki, Japan, May 24-30, 2001. 9. R. Raja et al; "The Program in Muon Physics: Superbeams, Cold Muon Beams, Neutrino Factory and Muon Collider", Contributed to APS / DPF / DPB Summer Study on the Future of Particle Physics (Snowmass 2001), Snowmass, Colorado, 30 Jun - 21 Jul 2001; hep-ex/0108041 10. A.N. Skrinsky and V.V. Parkhomchuk, Sov. J. of Nuclear Physics, 12, 3 (1981); D. Neuffer, Particle Accelerators 14, 75 (1983). 11. R. Alber et al. (Eds. W. Chou, A. Ankenbrandt, and E. Malamud), "The Proton Driver Design Study", FERMILAB-TM-2136, Dec 2000. 12. H. Haseroth (CERN). CERN-PS-2000064-PP. 13. B. Autin (CERN). CERN-PS-2000-067-
593
14.
15.
16.
17.
18.
19.
20.
21.
22.
PP, Workshop on High Intensity Muon Sources, Tsukuba, Japan, 1-4 Dec 1999; B. Autin et a l , CERN-PS-2000-011-AE, Jun 2000. 4pp., 7th European Particle Accelerator Conference, Vienna, Austria, 26-30 Jun 2000. B. Autin et al., CERN-2000-012, Dec 2000. 89pp. Y. Mori, Nuclear Instruments and Methods in Physics Research Section A, Vol. 451 (1) (2000) p. 300. I. Chemakin et al. (E910 Collab.), "Inclusive Soft Pion Production from 12.3 and 17.5 GeV/c Protons on Be, Cu, and Au", nucl-ex/0108007. Proposal to Study Hadron Production for the Neutrino Factory and for the Atmospheric Neutrino Flux", CERN-SPSC/99-35, SPSC/P315, 15 November, 1999 HARP Web Page: http://harp.web.cern.ch/harp/ P907: "Proposal to measure particle production in the Meson area using Main Injector primary and secondary beams", Y.Fisyak et al., R.Raja (spokesperson), http://ppd.fnal.gov/experiments/ e907/e907.htm J. Alessi et al. (E951 Collab.), "Proposal: An R&D Program for Targetry and Capture at a Muon Collider Source", September 28, 1998. A. Hassenein et a l , "An R&D Program for Targetry and Capture at a Neutrino Factory and Muon Collider Source", Proc. Neutrino Factory Workshop NUFACT01, Tsukuba, Ibaraki, Japan, May 24-30, 2001. B. King et al; "A Rotating Inconel Band Target for Pion Production at a Neutrino Factory", MUCOOL Note 199, see http://www-mucool.fnal.gov/htbin/ mcnotel LinePrint P. Sievers, "A Stationary Target for the CERN Neutrino Factory", CERN Neutrino Factory Note 65. See http: / / molat. home, cern.ch/molat/neutrino/nfnotes. h t m l R. Fernow, "Fortran Program to Simu-
Neutrino Factory and Muon Collider R&D
Steve Geer late Ionization Cooling", unpublished. 23. B. Bayanov et al., "Liquid Lithium Lens for Fermilab Antiproton Source", Budker INP 98-23; B. Bayanov et al., "Design and Construction Technology for a 15 cm Long Liquid Lithium Lens", Budker Institute of Nuclear Physics, 1998. 24. The MUCOOL Web Page can be found at http://www.fnal.gov /projects /muon_ collider /cool/ cool.html 25. J. Norem et al., "Preliminary Lab G Dark Current Results", MUCOOL Note 226 at http://www-mucool.fnal.gov /htbin / mcnotelLinePrint 26. N. Hartman, Derun Li, and J. Corlett, "Thin Beryllium Windows - Analysis and Design Status", MUCOOL Note 180, see http://www-mucool.fnal.gov/ fit bin/ mcnotelLinePrint 27. D. Kaplan et al., "Progress in Absorber R&D for Muon Cooling", physics/ 0108027; D. Kaplan et al., "Progress in Absorber R&D 2: Windows", physics/ 0108028. 28. A. Lombardi, "A 40-80 MHz System for Phase Rotation and Cooling", CERN Neutrino Factory Notes 20 and 34, see http://molat.home.cern.ch/molat/ neutrino/nfnotes. html 29. R. Garoby and F. Gerigk, "Cavity Design for the CERN Muon Cooling Channel", CERN Neutrino Factory Note 87, see http://molat.home.cern.ch/molat/ neutrino/nfnotes.html 30. B. Autin, talk given at NUFACT01. 31. K. Hanke, talk given at NUFACT01, and http://hep04.phys.iit.edu/cooldemo/. The cooling experiment was first proposed in the Fermilab P907 proposal: C.Ankenbrandt et al., "Proposal: Ionization Cooling R&D Program for a High Luminosity Muon Collider", April 15, 1998 (see http: / / www.fnal.gov/projects/muon_ collider/cool/ proposal/ proposal.html). 32. A. Zholents, M. Zolotorev, and W. Wan;
Phys. Rev. Special Topics - Accelerators and Beams, 4, 031001 (2001). 33. Muon Collider Collaboration, in Proceedings of the 1996 Snowmass Study, Chapter 9 of the Muon Collider Feasibility Study.
594
S U P E R S T R I N G S , D U A L I T Y , LARGE E X T R A D I M E N S I O N S LISA RANDALL Center for Theoretical Physics and Department of Physics Massachusetts Institute of Technology, 77 Massachusetts Avenue Cambridge, MA 02139
No written contribution received
595
CONCLUDING REMARKS NICOLA CABIBBO Dipartimento di Fisica, Universit di Roma - La Sapienza INFN, Sezione di Roma E-mail: [email protected] Important new results have been presented at this conference. The direct violation of CP in K° —> TT+TT has been firmly established in two independent experiments, NA48 at CERN and KTeV at Fermilab. Both Babar at SLAC and Belle at KeK have determined the CP violation in B^ - B® oscillations through the study of the golden Kg + * decay mode. The observed CP violation agrees with the expectations of the Standard model, based on the quark-mixing phenomenon. The first results of the Sudbury Neutrino Observatory, SNO, suggest that the long-lasting solar neutrino puzzle has been finally solved in terms of neutrino oscillations. Results appeared after the conference which modify the theoretical prediction of the muon anomaly. This new result, if confirmed, would drastically reduce the significance of the discrepancy between the theoretically expected value for the muon anomaly and the recent ressults of the Brookhaven experiment.
1
the cross section for charged current events, e —> v, is many orders of magnitude smaller than that for neutral current events, e —•> e, dominated by e.m. interactions; at high Q2 the two cross sections inch closer and become proportional, as predicted by the standard model. The Hera groups have presented a detailed determination of the scaling violation in deep inelastic scattering, allowing an extensive check on the predictions obtained from perturbative QCD, and an accurate determination of as(Mz). With the advent of more accurate (NNLO) calculations the new experimental results will allow a 1% precision in this important parameter.
Introduction
One of the pleasures of this conference has been the chance to meet Alberto Sirlin after many years. He reminded me of a recipe for the perfect closing lecture I offered him many (30+) years ago: "You have to mention everybody who gave a talk!" I am glad to have the recipe back after such a long time, but I will not be able to follow it. This meeting is rich with important results, among which two new examples of CP violation, the possible solution of the solar neutrino puzzle in terms of neutrino oscillations and a possible discrepancy between the recent measurement of the muon anomaly and theoretical predictions. I will concentrate my attention on these subjects. The discussion of these very hot arguments should not however make us forget many other excellent results presented at the conference; the field is indeed progressing on a very wide front. Among the many experimental results presented at the conference I was particularly impressed by those obtained at Hera by the ZEUS and HI collaborations, which graphically demonstrate the unification of weak and electromagnetic interactions: at low Q2
2 2.1
CP Violation Quark mixing and CP
Violation
The charged-current weak interactions of hadrons are described by the unitary matrix 1 ' 2 V (V"tV = 1). With only two families, e.g. in a world without beauty (or t quarks), V can always be reduced to a real form, so that CP is necessarily conserved. With three families the matrix V can be expressed in terms of four parameters:
596
Nicola Cabibbo
Concluding Remarks
Figure 2. T h e Unitarity triangle in the p - t) plane.
v= \2/2 A
AX3(l-p-ir1)
Vud Vus Vub Vcd Vcs Vcb « Vta Vu Vtb 3
AX (p-ir]) A 1 - A2/2 AX2
-AX2
(1)
where I have used the Wolfenstein 3 parametrization. A non-vanishing value for 77 leads to the violation of CP symmetry. With three quarks CP conservation, 77 = 0, is an exceptional case; CP violation is the norm, obtained for any non-zero value of 77. Two of the parameters, A and A are known with good precisions: A = sin 9, where 9 is my original mixing angle, is determined by Kiz decays, A = sin 9 = 0.2237 ±0.0033
(2)
The rates of the allowed B decays lead a determination of the A parameter. From the analysis by Ciuchini et al. 4 we have: AX2 = Vcb = (41.0 ± 1.6) x IO-
CS)
The problem of determining the two remaining parameters, p and 77, is best seen in the light of the unitarity relation
vudvu*b + vcdv;b + vtdvt*b = 0
(4)
which can be represented as a triangular relation in the complex plane. The unitarity relation is automatically satisfied in the parametrization of Eq. (1), where it reduces, up to terms 0(X2), to the triangle of figure 2. Since the area of the unitarity triangle is 77/2, a non-flat triangle implies CP violation.
597
The form of the unitarity triangle can be determined by measurements of CP conserving quantities. The oscillation of B^ mesons are dominated by graphs with virtual top quarks, so that the mass difference Arrid is proportional to |1 — p — ir]\2, the length squared of one of the upper sides of the triangle. The length of the other side, \p + ir)\, can be extracted from a determination of Vub, e.g from a determination of the rates of the forbidden b —> u leptonic transitions. These determinations point to a non-flat triangle, i.e. to the presence of a certain amount of CP violation. As a first check the values of p, rj so obtained agree well with the observed value of the CP violating e parameter in K° - K° mixing. These different constraints on p and 77 are displayed in figure 1, and lead to the following estimates for p and 77:
p = 0.224 ± 0.038, 77 = 0.317 ± 0.040
(5)
What is perhaps more relevant is the fitted value for sin(2/3), sin(2/3) =0.698 ±0.066,
(6)
since this parameter is directly accessible through a study of CP violation in the "golden decay mode" 5 of B° mesons, (B°d or B°d)^Ks
+V
(7)
In his presentation to this conference, C. Sachrajda has emphasized the central role of Lattice QCD simulations in the determinations of the CKM parameters. Lattice QCD was used for evaluating the B parameter for K mesons, needed in the prediction of e in terms of p and 77, and again for determining the decay and mixing parameters fs and BB for both the Bd and Bs mesons, parameters which are needed for the determination of the two mass differences Arrid and Ams. The present simulations are executed within the quenched approximation, due to the limited computer power available today, while more
Nicola Cabibbo
•FT
Concluding Remarks
1
Am,
0.8
,---'""/
Ams/Amd
^^>i^_^
L~—vy
|Vub| |Vcb| ^ ^
/
>-%
f
0.6 0.4
,3***^^
^-~~
\/jjL*^Sj
0.2 ! , ,
°-l
-0.8 -0.6 -0.4 -0.2
0
Mi
0.2
, ,
,
1
0.4
,
,
,
!
0.6
,
i
,
1
0.8
,
,
,
1
P Figure 1. Constraints on p, r] arising from Vub, e — the CP violating parameter in K° - K° mixing, and t h e B^ — B^ mixing parameter Amj,-
accurate simulations will be possible with the advent of teraflop class computers.
2.2
New results on CP violation in B° decays
Results on CP violation in the "golden mode" of Eq. (7) have been presented at this conference both by the Babar experiment at SLAC and the Belle experiment at KeK. The results are in reasonable agreement among themselves and lead to a value of sin(2/3) which is in good agreement with the prediction of Eq. (6), a remarkable confirmation of the hypothesis that CP violation phenomena arise from complex elements of the V matrix. The "golden" character of B°d ->• Ks + * derives from the fact that the final state is a CP eigenstate, and that this decay mode is dominated by a CP conserving tree diagram. Any CP violation observed in this mode must, to an excellent approximation, be attributed to B°d - B°d mixing. The interpretation of CP violation in Bd — Bd mixing is uniquely simple, since this mixing is dominated by a single diagram whose phase is easily seen to be exp(2i/3). The measurement of CP violation effects in
the decays of Eq. (7) can be directly interpreted as a measurement of the j3 angle in the unitarity triangle of figure 2. The situation in Bd — Bd mixing is very different from that in K° — K° mixing, which is dominated by a CP conserving diagram, CP violation arising from a second smaller diagram. Contrary to Bd case, obtaining information on the mixing matrix from the measurement of CP violation in K° mixing (the e parameter) requires a complex theoretical analysis and one must, as noted above, recur to lattice QCD simulations to obtain an estimate of one of the required parameters. The values of sin(2 /3) presented here by the two experimental groups 6 ' 7 are: sin(2/3) = 0.59 ± 0.14 stat ± 0.05 syst (Babar) sin(2/3) = 0.99 ± 0.14 stat ± 0.06 syst (Belle) (8) These are impressive results: each of them by itself establishes the existence of CP violation in B° decays to many a's. It is remarkable that two rather different experiments at different accelerators and in different Laboratories were able to obtain results of comparable accuracy within a few days of each
598
Nicola Cabibbo
Concluding Remarks
other. Three previous measurements of sin(2/3), obtained by CDF at Fermilab 8 , and by Aleph and Opal at CERN 9 , have larger errors but are generally compatible with the new results, which however supersede earlier preliminary results by the same groups. Combining the five extant results Ahmed Ali a obtains the "world average"
problem is that the direct violation of CP in K° —> 7T7T decays involves "penguin diagrams" which are at present very hard to evaluate in Lattice QCD. Progress is expected in this direction with the advent on the one side of new high-performance parallel computers and on the other of new algorithms for the simulation of low-mass quarks.
3 sin(2/3) =0.79 ± 0 . 1 2 ,
(9)
in excellent agreement with the theoretical prediction in Eq. (6). 2.3
Direct CP Violation in K° —• 7T7T; e'/e
Two experimental groups, NA48 10 at CERN and KTeV n at Fermilab, have presented new determinations of the direct violation of CP in the decays K° -> TT+TT", K° -> 7r°7r°, through a measurement of Re(e'/e): Re(e'/e) = (15.3 ± 2.6)10" 4 NA48 Re(e'/e) = (20.7 ± 2.8)10" 4 KTeV
(10)
The two new results are in good agreement, and each of them is many er's away from Re(e'/e) = 0, so that the presence of direct CP violations in K° decays is firmly established. The new results are in rough agreement with the previous result by NA31 at CERN and with that obtained by he E731 experiment at Fermilab (which was however compatible with e' = 0). The new world average is Re(e'/e) = (17.2 ± 1.8)10~4
(11)
This is in general agreement with the theoretical evaluations, which are however not excessively precise. They are normally quoted as "from a few 10" 4 t o ~ 2 x 10^ 3 ". The a
I am grateful to dr. Ali for providing this result. He commented: "the chi square of the fit is 5.2 to be compared with the expected x = 4. The chi-square is not great but acceptable."
599
Neutrino oscillation: the Solar neutrino puzzle solved?
The solar neutrino puzzle has been with us for over thirty years, since the Davis chlorine experiment 12 detected only about a third of the neutrinos expected on the basis of the current solar model 13 . The deficit of solar neutrinos has over the years been confirmed by the Kamiokande and Super-Kamiokande water detectors, and by the gallium experiments, GALLEX, SAGE and GNO. At the same time the solar model has been refined, and tightened with the help of data on heliosismography, so that we can exclude that the neutrino deficit can find its explanation in some modification of the solar model itself. We can refer the reader to the recent review 1 5 by Bahcall, Pinsonneault and Basu. Already in 1968 Bruno Pontecorvo 14 proposed that a deficit in solar neutrinos could signal the presence of neutrino oscillations. An important theoretical development was the realization that the coherent interaction with the solar matter can modify the neutrino oscillations 16 , and that this can give rise to resonant transformation between neutrino species even for small mixing angles 17 . The coeherent interaction of solar neutrinos with the bulk matter of earth could give rise to day-night effects which could be explored by real-time detectors such as Kamland or Borexino. Many questions remained open: are oscillations real? are the oscillations confined to the known neutrino flavours, or do they in-
Concluding Remarks
Nicola Cabibbo volve also new flavours (called sterile neutrinos) which do not partake of neutral current interactions, and do not therefore contribute to the Z° width? Results obtained at the Sudbury Neutrino Observatory (SNO) 18 , presented at this conference, seem to give a positive answer to this two questions: The solar neutrinos which arrive at the Earth behave as a mixture of ve and other active neutrinos, i.e. v^ and vT. The principle of the SNO experiment is to compare the rate of charged current inverse beta decay events, which can only arise from ive's, and of neutrino-electron scattering events, to which also i/^'s and zVs can contribute. The ratio of the two type of events, { ES ^ ue+ 0.14(i/„ + vT) ' can be used to deduce the total number of neutrinos, which can than be compared with the solar model prediction. The SNO collaboration has obtained an accurate determination of the CC rate, while their determination of the ES rate is not accurate enough to establish the existence, in the solar flux at the Earth, of a fraction of v^ and vT . They can however use the determination of the ES rate obtained 19 at Super-Kamiokande which has the required precision. The two measurements are mainly sensitive to neutrinos in the same energy band, the Bs neutrinos, so that their combination is meaningful. The neutrino fluxes determined through CC and ES events differ by more than three standard deviations, thus giving a strong support for the existence of neutrino oscillations:
g,ES
$^
0
= (0.57 ± 0.17)10bC7rr V (13)
The two measurements, together with Eq. (12), determine the total flux of Bs neutri-
In the next few years we expect important results from both SuperKamiokande and SNO. Two new experiments will give important contributions to the unravelling of the solar neutrino problem: • KamLand will study oscillations in reactor neutrinos with a sensitivity sufficient to confirm or exclude the — now favoured — LM solution for neutrino oscillations. • Borexino will be able to observe in real time the flux of the low-energy Be7 solar neutrinos. This will allow refined studies of day/night and seasonal effects. The new SNO data favour 2 0 ' 2 1 large mixing angle oscillation solutions, which opens the way to the possibility of CP and T violation in neutrino oscillations: CP : (ui —> v2) <-> (Pi —>• v2)
T
: (ux —> v2) <-> (v2 -> vi)
Disappearence experiments (ui —> vi) cannot display CP or T violations Exploring CP or T violations requires superbeams, or better a dedicated neutrino factory, which is also the first step for a muon collider. Both possibilities have been discussed during this conference. I cannot resist quoting from a paper I wrote in 1978 22 : "maximal neutrino mixing requires CP violation" By maximal I mean that all matrix elements of the lepton mixing matrix V L should have equal size. Since V L is unitary, the requirement of maximal mixing has essentially a unique solution which is necessarily complex: 1
„ „ - 22 - 1 $ = (5.44±0.99)10, 66 cm" s
(14)
which is in excellent agreement with the solar model predictions. The solar neutrino gap seems to have closed.
VT =
X X2
X X2
1
1
X
X2
2ni x = exp( — )
We are probably far from this solution, but perhaps not very far. 600
Concluding Remarks
Nicola Cabibbo 4
The muon anomaly: signal for new physics?
James Miller presented here the recent results of the Brookhaven measurement 23 of the muon magnetic anomaly. The new world average, aExP =
( 116 5920.3 ± 1.5) x 10~ 9
(15)
PP; a 2
6924(62) x l O " 1 1
(21)
24
PP;a3
- 1 0 0 ( 6 ) xlO" 1 1
(22)
LL; a3
-85(25) x l O ' 1 1
(23)
disagrees with the theoretical prediction by nearly three standard deviations.
a™ = (1165915.96 ± .67) x 10~ 9 (16) a®xP-alh
= (4.3 ± 1 . 6 ) x 10" 9
(17)
The Brookhaven collaboration expects to be able to decrease the experimental error by nearly a factor three in the near future. Already in its present state the discrepancy seems serious and has stimulated a multitude of theoretical papers which examine different possible implications of this discrepancy, which would clearly be a signal for new physics. The interesting aspect is that the discrepancy is relatively large; by comparison the contribution to the muon anomaly of electroweak effects — diagrams with virtual Z°,W± bosons contribute a correction 5EWail
= (1.51 ± 0.04) x 1CT9
(18)
In order to explain a discrepancy of (4.3 ± 1.6) x 1 0 - 9 one would need new physics at relatively low energies, in other words this discrepancy looks as excellent news for the forthcoming LHC experiments. It is also clear that in view of the importance of a possible discrepancy both the experimental analysis and the theoretical computations must be submitted to to the most careful scrutiny. The theoretical prediction of the muon anomaly is the sum of diagrams with virtual leptons, photons and intermediate vector bosons, QED EW
116584706(3) x l 0 ~ n
(19)
11
(20)
151(4) xlO"
and diagrams which include virtual hadrons, further divided in a2 and a 3 diagrams which include hadron corrections to the photon propagator (PP), and diagrams with hadronic light by light (LL) subdiagrams. These diagrams are the main sources of the theoretical error,
601
The numbers reported in eqs. (19 - 23) are those used by the Brookhaven collaboratin in their analysis. The hadron corrections to the photon propagator can be related to the total cross section for hadron production in electron positron collisions 25 ; their contribution to the muon anomaly can then be expressed as an integral, with a suitable kernel, over the cross section for e + e~ —> hadrons. The most important part of this contribution and of its error arises from the low energy (< lGeV) region. As an alternative to low energy e+ e~ data one can use, via the CVC relation, data on the r decay into hadrons, which are at present more accurate 2 6 . A number of evaluations of the hadronic photon propagator contribution to the muon anomaly have appeared in recent times, with slightly different results 27 and slightly different evaluations of the error. Since the hadron correction to the photon propagator is safely anchored to experimental data on e+e~ collisions and T decays, the error on this contribution to the muon anomaly will improve in the next few years. Particularly promising is the advent of the KLOE experiment at the DA$NE
Concluding Remarks
Nicola Cabibbo this contribution to other measurable phenomena. Waiting for a frontal attack on this contribution using lattice QCD (not easy) we must be satisfied with models whose accuracy is difficult to estimate. The current evaluations of the hadronic light by light contributions to the muon anomaly 28,29 are based on models of the light pseudoscalar mesons and their interactions at low energy - chiral perturbation theory or the extended Nambu - Jona Lasinio model. The dominant hadronic LL contribution turns out to be the one mediated by a single intermediate neutral pion. After the LP01 conference was concluded, a new calculation of the ir° contribution 30 ' 31 to the muon anomaly reached a very suprising conclusion: while previous calculations had found a negative sign, the new result, recently confirmed by an independent computation 32 , found a positive sign. The authors of the two complete evaluations of the LL contributions 28 ' 29 have been able to identify the origin of what they now see as a sign error in the previous computations and have presented new evaluations for the overall LL contributions, which are respectively 33,34 (gg
±
16)
x
10 -11
an(J
(g 3
±
32)
x 1Q
-11
The effect of this sign change is a reduction of the discrepancy to less than 2 standard deviations. Quite apart from the question of sign of the 7T° contribution, which only arose after the conference, we must note a detailed criticism 3 5 by K. Melnikov, who argues that the contribution of quark-loop light by light diagrams has been underestimated in ref. 2 8 , 2 9 In referring the interested reader to Melnikov's paper for the details of his argument, I note that this would further reduce the discrepancy between theoretical evaluations and the experimental result. Although the authors of ref. 29 do not agree with this argument, it is clear that, in order to calculate the muon anomaly with a precision comparable with that expected from the Brookhaven ex-
periment, the hadronic light by light contributions must be carefully re-evaluated. The forthcoming measurements of hadron production in low energy electron-positron collisions should lead to an improved evaluation of the contribution of hadronic corrections to the photon propagator. 5
We are not alone!
While all this has being going on, cosmologists . . . We heard in the talks by Halzen and Turner of the exciting progress our neighbours are making in Cosmology and Astrophysics. With the recent results on the cosmic background anisotropy cosmologists are confirming their own Standard Model. The new results on the cosmic background arise from a serendipitous use of the antartic winds which circulate around the South Pole: a balloon released from an antartic station comes back close to the same spot in obout a month. A balloon circling the South Pole can be very competitive with a satellite: the launch is by far less expensive, and the payload does not need to meet the high standards and associated cost in both money and time that a space launch requires. In the Boomerang flights the cosmic background has been studied with unprecedented resolution. The angular resolution of the recent data corresponds to spherical harmonics of « 1000. In this range three peaks are evident in the power spectrum, which fit very well the expectation for a Big-Bang universe at fi = 1, i.e. a flat universe, whose energy density is much larger than the "observed" baryon density which would correspond to Q w 0.05. This offers futher evidence for the conclusion that most of the matter in the universe is "dark matter", most probably "cold dark matter", i.e. matter constituted of relatively heavy particles, which have decoupled from "normal" matter early in the history of the universe.
602
Concluding Remarks
Nicola Cabibbo While it is clearly the task of astronomers and cosmologist to ascertain the geometry and history of the universe, the task of attempting the detection of these slow particles coasting along in the present universe falls to the high energy community. Many experiments are now underway for the detection of the weak interacting cold matter, and might bear fruit in the coming years. 6
Conclusions and acknowledgements
This conference has been enlivened by many exciting results. Will the next one be even better? It is a tall order, but beautiful things are brewing. After decades in which CP violation was established in a single process, the K° — K° oscillations, we now have two more well established examples, the first in B° — B° oscillations and the second a direct violation in K° —> 7T7T. The first is important for the light it sheds on quark mixing and the Standard Model in general, while the measurement of e'/e has a very special impact because, apart from it being in general agreement with the still imprecise expectations of the Standard Model, it definitely excludes the "Superweak" models of CP violation. We can look forward to new results on CP violation: Babar and Belle should be able to establish new examples of direct violation in B decays and the KLOE experiment at DA$NE should offer a determination of e'/e which is logically independent from those presented by the NA48 and KTeV experiments. The other very exciting development comes fro the SNO results which corroborate the conclusion that the solar neutrino puzzle will find its solution in neutrino oscillations. Here we expect important results in the near future from both SNO and Super-Kamiokande, but also from experiments which are now approaching the datataking phase, in the first instance Kamland
603
and Borexino. The new results by SNO reinforce the proposal that neutrino oscillations are characterized by large mixing angles, and this opens up a very exciting possibility of detecting CP violation effects in neutrino oscillations. The detection of these effects will however require new neutrino beam facilities, which have been discussed during LP01. A first attempt could be carried out with superbeams while more detailed studies will require the availability of full—flegded neutrino factories. The success of this conference is certainly the merit of the many research groups who have contributed important new results, and of the many physicists who have contributed well prepared and well documented presentations of the new data, but LP01 would not have succeeded without the efforts of the organizers and of the many young people who have devoted so much time and efforts to its success. I am particularly grateful to Juliet Lee Franzini and Paolo Franzini, who invited me to give these concluding remarks to a very exciting conference whch has turned out to be a real turning point in the kind of physics I have been working on for many years. To everybody who participated in the conference I would like to present my best wishes that we may all be working very hard, and be ready to surprise each other when we meet in 2003 for the next Lepton Photon Conference.
References 1. N. Cabibbo, Phys. Rev. Lett. 10, 531 (1963) 2. M. Kobayashi and T. Maskawa, Prog. Theor. Phys. 49, 652 (1973). 3. L. Wolfenstein, Phys. Rev. Lett. 51, 1945 (1983). 4. M. Ciuchini, G. D'Agostini, E. Franco, V. Lubicz, G. Martinelli, F. Parodi, P.
Concluding Remarks
Nicola Cabibbo
5. 6.
7. 8. 9.
10. 11. 12.
13. 14. 15.
16. 17. 18.
19. 20.
21.
22.
Roudeau and A. Stocchi JEEP 0107, 013 (2001). I.Y. Bigi, A.I. Sanda Nucl.Phys. B 281, 41 (1987). B. Aubert, et al (The BABAR Collaboration), Phys. Rev. Lett. 86, 2515-2522 (2001). A. Abashian et al (Belle Collaboration) Phys. Rev. Lett. 86, 2509-2514 (2001). T. Affolder et al, Phys. ReV. D 6 1 , 072005 (2000). R. Barate et al (ALEPH Collaboration), Phys. Lett. B 492, 259-274 (2000). K. Ackerstaff et al. (OPAL Collaboration) Eur. Phys. C 5, 379 (1998). L. Iconomidou-Fayard , presented at LP01. R. Kessler, presented at LP01. R. Davis, Jr., D. S. Harmer and K. C. Hoffman Phys. Rev. Lett. 20, 1205 (1968). R. Davis Progr. Part. Nucl. Phys. 32, 13 (1994). J. N. Bahcall, N. A. Bahcall and G. Shaviv Phys. Rev. Lett. 20, 1209 (1968). B. Pontecorvo, Sov. Phys. JETP, 26, 981 (1968) J. N. Bahcall, S. Basu, and M. H. Pinsonneault, Astrophys.J. 555, 990-1012 (2001) L. Wolfenstein, Phys. Rev. D 17, 2369 (1978) S. P. Mikheyev, A. Yu. Smirnov Sov. J. Nucl. Phys. 42, 913 (1985) Q.R. Ahmad et al (The SNO Collaboration), Phys. Rev. Lett. 87, 071301 (2001). S. Fukuda et al, Phys. Rev. Lett. 86, 5651 (2001). G.L. Fogli, E. Lisi, D. Montanino and A. Palazzo Phys. Rev. D 64, 093007 (2001). John N. Bahcall, M. C. Gonzalez-Garcia, Carlos Pena-Garay JEEP 0108, 014 (2001). N. Cabibbo Phys. Lett B 72, 333-335 (1978)
23. H. N. Brown et al (Muon g-2 collaboration), Phys. Rev. Lett. 86, 2227 (2001). 24. A. Czarnecki and W. J. Marciano,.P/iys. Rev. D 64, 013014 (2001). 25. N. Cabibbo and R. Gatto, Phys. Rev. 124, 1577 (1961) 26. M. Davier and A. Hocker, Phys. Lett. B 435, 427 (1998). 27. J. F. de Troconiz and F. J. Yndurain, hep-ph/0106025. F. Jegerlehner, hepph/0104304. S. Narison, Phys. Lett. B 513, 53 (2001). 28. M. Hayakawa and T. Kinoshita, Phys. Rev. D 57, 465 (1998) 29. J. Bijnens, E. Pallante and J. Prades, Phys. Rev. Lett. 75, 1447 (1995); ibid., 75, 3781 (1995), erratum. 30. M. Knecht and A. Nyffeler, Marseille preprint CPT-2001/P.4253, hepph/0111058, submitted to Phys. Rev. D. 31. M. Knecht, A. Nyffeler, M. Perrottet and E. de Rafael, Marseille preprint CPT-2001/P.4260, hep-ph/0111059. 32. I. Blokland, A. Czarnecki and Kirill Melnikov SLAC-PUB-9084, hepph/0112117. 33. M. Hayakawa and T. Kinoshita, hepph/0112102. 34. J. Bijnens, E. Pallante and J. Prades, hep-ph/0112255. 35. K. Melnikov, hep-ph/0105267.
604
Poster Session
This page is intentionally left blank
Posters Session CERN
AD-Experiments ATLAS CMS CP Violation Heavy Ion Physics LEP Tau, b-Physics & Searches The LHC Machine Neutrino Oscillations
CORNELL
CESR to low energies Rare B Decays Charm Physics at CLEO Vub measurement CLEO-c and CESR-c Spectroscopy and QCD Vc5 measurement Superconducting RF
DESY
The HI Experiment The HERA-B Experiment Hermes and the Spin of the Nucleon TESLA The ZEUS Experiment Research at DESY
FERMILAB
Physics at the Energy Frontier Technology at the Energy Frontier Discovery at the Energy Frontier Neutrinos at the Energy Frontier Frontiers in Astrophysics Physics without Borders Future Frontiers 607
IHEP
BES & BEPC Yangbajing Cosmic Ray Observatory BSRF BEPC II
KEK
ATF JLC K2K Collaboration on LHC High Intensity Proton Accelerator
LNF
DA$NE KLOE FINUDA DEAR NAUTILUS
LNGS
The Gran Sasso Laboratory
NOVOSIBIRSK
The CMD2-2M Detector The SND-2000 Detector The VEPP-2000 collider project
SLAC
ARDA ARDB GLAST Next Linear Collider Collaboration PEP-II SSRL The BaBar Detector FTE
608
Contributed Papers
ffii.; i
zm
This page is intentionally left blank
Papers Contributed to LP01 Abada Asmaa
hep-ph/0105221
Adam Wolfgang
Adam Wolfgang Albright Carl H.
hep-ph/0104294
Albright Carl H. Ambrosino Fabio Ambrosino Fabio Ambrosino Fabio
hep-ph/0106157 hep-ex/0107022 hep-ex/0107020 hep-ex/0107024
Anulli Fabio
hep-ex/0106029
Bhattacharyya G. Biebel Otmar
hep-ph/0105057 hep-ex/0012044
Biebel Otmar
hep-ex/0105059
Biebel Otmar
hep-ex/0106066
Cassel David
hep-ex/0105002
Cassel David Cassel David
hep-ex/0102007 hep-ex/0102006
Cassel David Cassel David
hep-ex/0105071 hep-ex/0106060
Cassel David Cassel David
hep-ex/0105086 hep-ex/0101006
Cassel David Cassel David
hep-ex/0104009 hep-ex/0104042
Cassel David Cassel David Chang Ngee-Pong
hep-ex/0107021 hep-ex/0107040 hep-ph/0105153
Cheung Kingman Cheung Kingman
hep-ph/0102238 hep-ph/0103183
Preliminaries on a Lattice Analysis of the Pion Light-cone Wave function: a Partonic Signal? Search for single top production via flavour changing neutral currents: preliminary combined results of the LEP experiments Search for excited leptons: preliminary combined results of the L E P experiments Realization of Large Mixing Angle Solar Neutrino Solution in an SO(10) Supersymmetric GUT Model G U T Implications from Neutrino Mass Detection of —• rfl, 4> -> VI with KLOE Studies of Ks decays with the KLOE detector Detection of 0 ->f0(980)7, cf> -»• a0(980)7 with the KLOE detector Evidence for a narrow dip structure at 1.9 G e V / c 2 in 37r+ 37r~ diffractive photoproduction Can R-parity violation lower sin(2/3)? A Measurement of the QCD Colour Factors using Event Shape Distributions at y/s=14 to 189 GeV Tests of Power Corrections for Event Shapes in e + e ~ Annihilation Measurement of the longitudinal and transverse x-section in e + e " annihilations at v / s=35-44 GeV Rate of DO —> K+ir~ir° and Constraints on DO - DO Mixing First Measurement of P ( D * + ) Mixing and C P Violation in the Decay of Neutral D Mesons at CLEO First Observation of BO —• D*°TT+Tr+-K~n~ Decays Improved Upper Limit on the FCNC Decays B ^ Kl+l~ and B-> K*(892)l+l~ Search for the Decay B+ - • D*+K° Bounds on the C P Asymmetry in Like-Sign Dileptons from BO BO-bar Meson Decays Search for CP Violation in r —> Tr~n°u Decay Experimental Investigation of the Two-Photon Widths of the XcO and the Xc2 Mesons Search for the Decay Upsilon(lS) —> 77/ Measurement of the S c + Lifetime Oscillations of Faster than Light Majorana Neutrinos: A Causal Field Theory Muon anomalous magnetic moment and Leptoquarks Muon anomalous magnetic moment, two Higgs model, and Supersymmetry
611
Cheung Kingman
hep-ph/0104250
Chizhov Mihail
hep-ph/0107025
Chyla Jiri Denig Achim Dobado Antonio Erdmann Martin
hep-ph/0010140 hep-ex/0107023 hep-ph/0107155 hep-ex/0102047
Fajfer Svjetlana Garavaglia Theodore Garavaglia Theodore
hep-ph/0106131 hep-th/0011180 hep-th/0104030
Garavaglia Theodore Gay Ducati Maria B.
hep-ph/0106087 hep-ph/0102069
Goldstein Gary Herrero Maria Jose
hep-ph/0106168 hep-ph/0107147
Herrero Maria Jose
hep-ph/0106267
Hou George W.S.
hep-ph/0005015
Hou George W.S. Hou George W.S.
hep-ph/0008079 hep-ph/0012027
Hou George W.S. Hou George W.S.
hep-ph/0101146 hep-ph/0101162
Hou George W.S.
hep-ph/0103094
Hou George W.S.
hep-ph/0104122
Huang Tao
hep-ph/0102193
Igo-Kemenes Peter Igo-Kemenes Peter
hep-ex/0107029 hep-ex/0107030
Igo-Kemenes Peter
hep-ex/0107031
Igo-Kemenes Peter
hep-ex/0107032
Igo-Kemenes Peter
hep-ex/0107035
Igo-Kemenes Peter
hep-ex/0107034
Sensitivity study of extra dimensions at T E V e+e~ colliders A new mass relation among the hadron vector resonances Heavy quark production in 7 7 collisions Measuring the Hadronic x-section at K L O E Goldstone bosons and solitons on the brane Investigation of Quark-Antiquark Interaction Properties using Leading Particle Measurements in e+e~ Annihilation New Physics in D° —> 7 7 Decay Covariant relativistic quantum theory Covariant Quantum Green's Function for an accelerated particle Jets signal for Higgs particle detection at LHC The Description of F2 at Very High Parton Densities Estimates of the Nucleon Tensor Charge Effective Higgs-quark-quark couplings from a heavy SUSY spectrum Optimal observables to search for indirect SUSY-QCD signals in Higgs bosons decays Phenomenological Consequences of Righthanded Down Squark Mixings Pathways t o Rare Baryonic B Decays Window on Higgs Boson: Fourth Generation V Decays Revisited Glueball States in a Constituent Gluon Model Probing for the Charm Content of B and T Mesons Impact of a Light Strange-Beauty Squark on Bs Mixing and Direct Search Supersymmetric Model Contributions to B°d-B^ Mixing and B —> 7T7I", fry Decays Muon Anomalous Magnetic Moment and Lepton Flavor Violation Search for the SM Higgs Boson at L E P Searches for the Neutral Higgs Bosons of the MSSM: Preliminary L E P Combined Results Search for Charged Higgs bosons: Preliminary LEP Combined Results Searches for Invisible Higgs bosons: Preliminary LEP combined results Searches for Higgs Bosons Decaying into 7's: Preliminary LEP Combined Results Flavour Independent Search for Hadronically Decaying Neutral Higgs Bosons at L E P
612
Kabana Sonja Ko Pyungwon
hep-ph/0104001 hep-ph/0002280
Ko Pyungwon
hep-ph/0103218
Lam C.S. Lam C.S. MacFarlane David
hep-ph/0104116 hep-ph/0104129 hep-ex/0107013
MacFarlane David
hep-ex/0105061
MacFarlane David
hep-ex/0107049
MacFarlane David
hep-ex/0107059
MacFarlane David
hep-ex/0107019
MacFarlane David
hep-ex/0107036
MacFarlane David
hep-ex/0107057
MacFarlane David
hep-ex/0107056
MacFarlane David MacFarlane David
hep-ex/0105001 hep-ex/0107037
MacFarlane David MacFarlane David
hep-ex/0107058 hep-ex/0107025
MacFarlane David
hep-ex/0107060
MacFarlane David
hep-ex/0107026
MacFarlane David
hep-ex/0107044
MacFarlane David MacFarlane David MacFarlane David
hep-ex/0107068 hep-ex/0107075 hep-ex/0107074
Marchetto Flavio Matsuda Koichi
hep-ph/0012357
T h e strange border of the QCD phases Phenomenology of the radion in Randall-Sundrum scenario at colliders Muon anomalous magnetic moment, B —> Xs^ and dark matter detection in the string models with dilaton domination A 2-3 Symmetry in Neutrino Oscillations Neutrino Oscillations via the Bulk Observation of C P violation in the B° meson system Measurement of branching fractions and search for CP-violating charge asymmetries in charmless two-body B decays into pions and kaons Measurement of the B—> J/ipK* (892) decay amplitudes Study of T and CP violation in B° - B ° mixing with inclusive dilepton events Measurement of the B° and B + meson lifetimes with fully reconstructed hadronic final states Measurement of the B° — B° oscillation frequency in hadronic B° decays Measurement of the branching fraction for B° - • D*+D*~ Investigation of B—> D*D*K decays with the BABAR detector Measurement of the decays B—> K and B—> K* Measurement of exclusive branching fractions B° - • r]K*° and B+ - • r]K*+ Measurement of B° decays to 7r+7r~7r° Measurement of branching fractions for exclusive B decays to charmonium final states Measurement of D+ and D* + production in B decays and from continuum e + e ~ annihilations at V« = 10.6 GeV Search for the rare decays B—> Kl+l~ and B ^ K*(892)l+l~ Measurement of J/ip production in continuum e + e ~ annihilations near ^/s = 10.6 GeV Search for B° —» 7 — 7 Search for B° -> a|f (980)7r_ Study of CP-violating asymmetries in r+/" 7r~/ + decays B 33 1 Study of the xo(l -Po) state of Charmonium formed in proton-antiproton annihilations MNS Parameters from Neutrino Oscillations, Single Beta Decay and Double Beta Decay
613
McKay Doug
hep-ph/0011310
McKellar Bruce McKellar Bruce McKellar Bruce Moutoussi Ann Moutoussi Ann
hep-ph/0106121 hep-ph/0106122 hep-ph/0106123
Moutoussi Ann
Moutoussi Ann
Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann
Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann
Moutoussi Ann Moutoussi Ann Moutoussi Ann
Moutoussi Ann
Neutrino Induced Giant Air Showers in Large Extra Dimension Models A See-Saw Mechanism with light sterile neutrinos Neutrino masses or new interactions Neutrino clustering and t h e Z-burst model Search for the SM Higgs boson with ALEPH Search for Supersymmetric Particles in e+e~ Collisions at y ' i up to 202 GeV and Mass Limit for the Lightest Neutralino Search for R-Parity Violating Decays of Supersymmetric Particles in e+e~ collisions at c m . energies from 188.6 GeV t o 201.6 GeV Search for a Scalar Top almost degenerate with the lightest Nuetralino in e + e ~ Collisions at yfs up to 202 GeV Measurement of W-pair production and W branching ratios in e + e ~ collisions up to 208 GeV Measurement of Triple Gauge-Boson Couplings at LEP energies up to 189 GeV Measurement of W-pair production in e + e ~ collisions at 189 GeV Measurement of the W Mass and Width in e + e ~ collisions at 189 GeV Measurement of the Tau Polarisation at L E P I Measurements of BR(6 - • TVX) and BR(& ->
TVD*X)
and Upper Limits on B R ( B —> rv) and BR(b —» svv) A flavour independent search for the Higgsstrahlung process in e + e ~ collisions at c m . energies from 189 t o 209 GeV Search for charged Higgs bosons in e + e ~ collisions at energies up to % /s=209 GeV Searches for neutral Higgs bosons of the MSSM at c m . energies up to 209 GeV with ALEPH Search for an invisibly decaying Higgs boson at L E P at c m . energies up to 209 GeV Search for gamma-gamma decays of a Higgs boson produced in association with a fermion pair in e + e ~ at the highest LEP centre-of-mass energies Measurement of the W Mass and Width in e+e~ collisions at yfs between 192 and 208 GeV Search for scalar leptons in e + e ~ collisions Search for R-Parity Violating Decays of Supersymmetric Particles in e + e ~ collisions at c m . energies between 189-208 GeV Study of the fragmentation of b quarks into B mesons at the Z peak
614
Moutoussi Ann Moutoussi Ann
Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann
Moutoussi Ann
Moutoussi Ann
Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann
Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann
Measurement of the e + e ~ —> ZZ production cross section Single and multi-photon production and a search for slepton pair production in GMSB topologies in e + e ~ collisions at yfs up to 208 GeV Leptonic decays of the Ds meson Study of Bs oscillations Measurement of the Michel parameters and the T neutrino helicity in the r lepton decays Search for rjb in Two-Photons Events Inclusive Production of LO and r\ Mesons in Hadronic Z Decays Measurement of the forward-backward asymmetries in Z—>bb and Z—+cc decays with leptons Simultaneous Measurement of the Strong Coupling Constant and the QCD Colour Factors from 4-jet hadronic Z decays Search for charginos nearly mass-degenerate with the lightest neutralino e + e ~ collisions up to v ^ = 209GeV Searches for single top production at yfs between 202 and 208 GeV Measurement of A ^ B (b) using inclusive b decays Fermion pair production in e + e ~ collisions at high energies and limits on Physics beyond the SM Limits on anomalous neutral gauge couplings using data from ZZ and Z7 production between 183-208 GeV Measurement of Triple Gauge-Boson Couplings in e + e ~ collisions up to 208 GeV Measurement of the Single W Production Cross Section at energies up tyb fs = 209GeV Inclusive semiteptonic branching ratio of b hadrons produced in Z decays Further studies on Bose-Einstein correlations in W-pair decays Searches for single sneutrino production via R-parity violation at fs=189-209 GeV Constraints on Anomalous Quartic Gauge Boson Couplings Search for Scalar Quarks in e + e ~ collisions aty/s up to 208 GeV Lower Mass Limit for the MSSM selectron and sneutrino obtained with the ALEPH detector Search for weak dipole moments of the T A study of the impact of stau mixing on the chargino and neutralino limits Color reconnection studies in W-pair events
615
Moutoussi Ann Moutoussi Ann Moutoussi Ann Moutoussi Ann
Nasriddinov Komjlon Nelson Charles
hep-ph/0001276 hep-ph/0106138
Obrant Gennady Ots Ilmar
hep-th/0106137
Probst Franz Ronga Francesco
hep-ex/0106049
Roudeau Patrick Rubbia Andre
hep-ph/0110397 hep-ph/0010308
Sanchez-Colon G. Sato Joe Sassot Rodolfo Schleper Peter Schleper Peter
Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter
hep-ph/0012333
Search for Gauge Mediated SUSY Breaking topologies with ALEPH up to Vs=208 GeV QCD measurements in e + e ~ annihilations at c m . energies between 91 and 206 GeV Limit on the LSP mass in R P V SUSY scenario Search for Charginos and Neutralinos in e + e ~ collisions at *Js up to 208GeV and Mass Limit for the Lightest Neutralino On the r —> a\vT decay Three Numerical Puzzles and the Top Quark's Chiral Weak-Moment A study of multiple jet production at transverse energy near 20 GeV "Dynamical" non-minimal higher-spin interaction and gyromagnetic ratio g = 2 Results and Plans of CRESST dark matter search Matter Effects in Upward-Going Muons and Sterile Neutrino Oscillations DELPHI Tau and Charm physics highlights The flavor of neutrinos in muon decays at a neutrino factory and the LSND puzzle An estimate of the lower bound on the masses of mirror baryons Neutrino Masses and Lepton Flavour Violation in Supersymmetric Models On the d/u Asymmetry and Parton Distributions Dijet Production in Charged and Neutral Current e+p Interactions at High Q 2 Measurement and QCD Analysis of Jet Cross Sections in Deep-Inelastic e + p Collisions at T/S of 300 GeV Diffractive Jet Production in Deep-Inelastic ep Collisions at HERA Deep-Inelastic Inclusive ep Scattering at Low x and a Determination of as Measurement of Charged and Neutral Current x-Sections in e-p Collisions at high Q 2 at HERA Measurement of Deeply Virtual Compton Scattering at HERA Measurement of the Photoproduction x-Section with a Leading Proton at HERA Inclusive Measurement of DIS at high Q 2 in ep Collisions at HERA Measurement of the Beauty Photoproduction cross section Beauty Production in DIS
616
Schleper Peter
Schleper Peter Schleper Peter
Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter Schleper Peter
Schleper Peter
Schleper Peter Schleper Peter Schleper Peter
Measurement of D* Meson Production and the Charm Contribution to the Proton Structure F2c in Deep Inelastic Scattering at HERA Diffractive *(2s) Photoproduction at HERA Investigation of Pomeron and Odderon Induced Photoproduction of Mesons Decaying to Pure Multiphoton Final States at HERA Measurement of density matrix elements for p meson production at large t Three-jet production in DIS at HERA Measurements of 2-jet cross sections in photoproduction at HERA A New Measurement of the DIS Cross Section and of FL at Low Q 2 and Bjorken-x at HERA Measurement of the Proton Structure Function Using Radiative Events at HERA A Measurement of the Rise of F2 towards Low x Radiative Charged Current Interactions at HERA Diffractive J / * vector Meson Production at high t at HERA Measurement of the diffractive structure function F?(3) Measurement of semi-inclusive diffractive DIS with a leading proton at HERA Photoproduction of p Mesons with a Leading Proton Dijet cross-section in photoproduction and DIS with a leading neutron at HERA Measurement of single inclusive high E-r jet cross-sections in photoproduction at HERA The Photoproduction of Protons at HERA A Search for Excited Fermions at HERA Searches at HERA for Squarks in R-Parity Violating Supersymmetry Search for Compositeness, Leptoquarks and Large Extra Large Dimensions in eq Contact Interactions at HERA Observation of events with isolated leptons and missing P y and comparison to W production at HERA A Search for Leptoquark Bosons in e~p collisions at HERA A Search for Excited Neutrinos in e~p collisions at HERA Search for Single Top Production in e + ~ p collisions at HERA
617
Schrempp Fridger Sirlin Alberto Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S.
Soldner-Rembold S. Soldner-Rembold S.
Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S.
Soldner-Rembold S. Soldner-Rembold S.
hep-ph/O012241 hep-ph/0106094
Zooming-in on Instantons at HERA Novel Approach to Renormalize the Electroweak Sector of the Standard Model A Measurement of the Branching Ratio for D s —» rvT Decays A Search for a Radial Excitation of the D*+~ Meson A Simultaneous Measurement of the QCD Colour Factors and the Strong Coupling Searches for Higgs Bosons in extensions to the SM in e + e ~ collisions at the highest L E P energies New Particle Searches in e + e ~ collisions at v/s = 200-209 GeV Measurements of Standard Model Processes in e+e~ collisions at ^ = 2 0 3 - 2 0 9 GeV Precise Determination of the Z Resonance Parameters at LEP: Zedometry Precision Neutral Current Asymmetry Parameter Measurements from the Tau Polarization at L E P Limits on Low Scale Quantum Gravity in Extra Spatial Dimensions from Measurements of e+e~ -> e+e~ at LEP2 Measurement of triple gauge boson couplings from W W production at L E P energies up to 189 GeV Measurement of W Boson Polarisations and CP-violating Triple Gauge Couplings from W + W ~ Production at L E P Search for the Standard Model Higgs Boson in e + e " Collisions at Vs=192-209 GeV Searches for Charginos with Small Delta M in e + e ~ Collisions at ^fs = 189 - 202 GeV Search for Single Leptoquark and Squark Production in e-7 scattering at LEP2 Precision Neutral Current Asymmetry Parameter Measurements from the Tau Polarization at L E P Measurement of the Vector and Axial-Vector Spectral Functions in Hadronic Tau Decays A Study of Bs meson oscillation using D s -lepton correlations Investigation of the Decay of Orbitally-Excited B mesons and first measurement of the Branching Ratio BR(B} -> B*ir(X)) Charged Multiplicities in Z Decays into u, d, and s Quarks Determination of the b Quark Mass at the Z Mass Scale
618
Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S.
Soldner-Rembold S.
Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S.
Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S.
Soldner-Rembold S.
Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S. Soldner-Rembold S.
QCD Studies with e + e ~ Annihilation Data at the highest LEP-2 Energies A high-Q 2 measurement of the photon structure function F ] at LEP2 Di-jet Cross-sections in Photon-Photon Collisions at LEP Search for anomalous Z-Z-7 and Z-7-7 couplings Measurement of Open Beauty Production in Photon-Photon Collisions at ^/s e e =189-202 GeV Inclusive Production of charged D* Mesons in Photon-Photon Collisions at e + e ~ c m . energies from 183 to 209 GeV First Measurement of the Charm Structure Function of the Photon in Deep Inelastic e-7 scattering at LEP Contribution to the Study of Four-Jet Events from Hadronic Decays of the Z Boson Angular Analysis of the Muon Pair Asymmetry at LEP 1 Search for Doubly Charged Higgs Bosons decaying into Two Tau Leptons at LEP Measurement of V u b using b hadron semileptonic decay Determination of S-matrix parameters from fermion-pair production at OPAL Two Higgs Doublet Model interpretation of Neutral Higgs Boson Searches up to the highest LEP Energies Determination of the L E P Beam Energy Using Radiative Fermion-pair Events Search for Single Top Production using the OPAL detector at LEP Searches for Intermediate Lifetime Signatures in GMSB Models with a Slepton NLSP in e + e " Collisions at ^ = 1 8 9 - 2 0 9 GeV Study of Z Pair Production and Anomalous Couplings in e + e ~ Collisions at y/s between 190 and 209 GeV Search for Yukawa Production of a Light Neutral Higgs Boson at L E P Genuine Correlations of Like-Sign Particles in Hadronic Z0 Decays Determination of the W mass in the leptonic channel using an unbinned maximum likelihood fit Measurement of the Branching Ratio for the process b—>
Soldner-Rembold S.
TVTX
Fermi-Dirac Correlations Between Protons in Hadronic Z Decays
619
Soldner-Rembold S.
Soldner-Rembold S. Soldner-Rembold S. Sopczak Andre
Sopczak Andre Sopczak Andre Steinhauser Matthias
hep-ph/0106135
Steinhauser Matthias
hep-ph/0104044
Steinhauser Matthias
hep-ph/0012002
Timmermans Jan
Timmermans Jan Timmermans Jan
Timmermans Jan
Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan
Measurement of Triple Gauge Couplings using Photonic Events with Missing Energy at v^=189-202 GeV Measurement of Z/7* production in Compton scattering of quasi-real photons Search for Technicolor Particles with the OPAL detector at LEP Study of scalar top quarks in the neutralino and chargino decay channel at a future e + e ~ linear collider A direct determination of tan(/3) at a future e + e ~ linear collider Comparison of Higgs boson mass and width determination of the LHC and a linear collider Nonabelian ^ / ( m j L ) heavy quark-antiquark potential J/tp plus dijet associated production in two-photon collisions Three-loop non-diagonal current correlators in QCD and NLO corrections to single-top-quark production Determination of the e+e~ —> 77(7) cross section using data collected with the DELPHI detector Search for an LSP Gluino at LEP Measurement of the W-pair Production cross section and W Branching Ratios at ^ = 2 0 5 - 2 0 7 GeV Updated results on the Search for R-parity Spontaneous Violation up to a c m . energy of 202 GeV Update at 202-209 GeV of the analysis of photon events with missing energy Search for Charged Higgs Bosons in e + e ~ Collisions at LEP Search for supersymmetry with R-parity violation at y i = 1 9 2 to 208 GeV Results on Fermion-pair production at L E P running in 2000 Search for supersymmetric particles in light gravitino scenarios Update on Single-W production at yfs= 205 and 207 GeV Measurement of helicity components of the fragmentation function A Measurement of the r Topological Branching Ratios Update of the ZZ cross-section measurement in e + e ~ interactions using data collected in 2000
620
Timmermans Jan
Timmermans Jan
Timmermans Jan
Timmermans Jan
Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan
Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan
Timmernjans Jan Timmermans Jan Timmermans Jan
A Study of the energy evolution of event shape distributions and their means with the DELPHI Detector at L E P QCD Results from the DELPHI Measurements of Event Shape and Inclusive Particle Distributions at the highest LEP energies Measurement of the strong coupling as and its energy dependence from the four jet rate of hadronic events with the DELPHI Detector at L E P A Measurement of the Cross Section Ratio Rf, and the Forward-Backward Asymmetry A^B for BB events with the DELPHI Detector at L E P 2 Update of Single Resonant Neutral Vector Boson Production at i / i = 2 0 5 and 207 GeV A precise measurement of the T lifetime Symmetric double radiative return withZ —> qq final state Searches for neutral Higgs bosons in e + e ~ collisions from 191.6 to 201.7 GeV Measurement of inclusive fl(1285) and fl(1420) production in Z decays with the DELPHI detector Measurement of the production of the four-fermion final states mediated by neutral current processes Search for Single Top Production at L E P via Four Fermion Contact Interactions at ^fs = 189-208 GeV Search for a Fermiophobic Higgs at LEP 2 Searches for invisibly decaying Higgs bosons Searches for supersymmetric particles in e + e ~ collisions up to 208 GeV, and interpretation of the results within the MSSM Limits on Higgs Boson Masses from a MSSM Parameter Scan Multiplicity in Hadronic Three Jet Events Search for excited leptons with the DELPHI detector at LEP Searches for neutral supersymmetric Higgs bosons in e + e ~ collisions up to 209 GeV Investigation of Colour Reconnection in W W pairs using particle flow Search for Technicolor with DELPHI Search for AMSB with the DELPHI data Measurement of the forward-backward asymmetry of bottom quarks at the Z0 peak using charged kaons for quark-charge tag Spin Density Matrix analysis of the reaction e + e " - • W+W~ at 189 GeV Particle correlations in e+e~ —> W+W~~ events Preliminary determination of E;, eorn at LEP2 using radiative 2-fermion events in DELPHI
621
Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Timmermans Jan Vovenko Anatoli
hep-ex/0104013
Ward Bennie
hep-ph/0006357
Ward Bennie
hep-ph/9806310
Ward Bennie
hep-ph/0104049
Ward Bennie
hep-ph/0103163
Ward Bennie
hep-ph/0007012
Ward Bennie
hep-ph/9907436
Ward Bennie
hep-ph/0006357
Ward Bennie
hep-ph/9912214
Yamauchi Masa
Yamauchi Masa Yamauchi Masa
Inclusive b Decays to Wrong Sign Charm Mesons Searches for Higgs Bosons in a general Two Higgs Doublet Model Search for Single Top Production at LEP via FCNC at v7^ = 189-207 GeV Measurement of the proper lifetimes of B+ and B° mesons Search for Ba — Bs oscillations in inclusive samples Search for single Leptoquark production in e + e ~ collisions up to ^ = 208 GeV with DELPHI Measurement of B*-B mass difference isospin splitting Generalised search for hadronic decays of Higgs bosons with the DELPHI detector at LEP-2 Measurement of the W mass and width at cms energies of 192-209 GeV Study of Trilinear Neutral Gauge Boson Couplings Study of Bs — Bs oscillations using inclusive leptons with large PT Determination of AbbB using inclusive charge reconstruction and lifetime tagging at LEP1 Determination of the high-twist contribution to the structure function xF£N C P Violation in Exclusive B Decays: Recoil Phase Effect Size of Penguin Pollution of the CKM C P Violating Phase in Bs —> pKs The Monte Carlo Program KoralW version 1.51 and The Concurrent Monte Carlo KoralW Y F S W W 3 The Monte Carlo Event Generator YFSWW3 version 1.16 for W-Pair Production and Decay at L E P 2 / L C Energies Precision Predictions for (Un) Stable W + W ~ Pair Production At and Beyond LEP2 Energies Final State Radiative Effects for the Exact O ( a ) YFS Exponentiated (Un) Stable W + W ~ Production At and Beyond LEP 2 Energies Coherent Exclusive Exponentiation For Precision Monte Carlo Calculations The Precision Monte Carlo Event Generator KK for Two-Fermion Final States in e + e ~ Collisions Measurements of the branching fractions of B meson decays to final states including charmonia at Belle Measurements of the B meson decays B —> D*D* and B -> D*D*K at Belle Search for color-suppressed B decays at Belle
622
Yamauchi Masa Yamauchi Masa Yamauchi Masa Yamauchi Yamauchi Yamauchi Yamauchi
Masa Masa Masa Masa
Yamauchi Masa Yamauchi Masa Yamauchi Masa Yamauchi Masa Yamauchi Masa Yamauchi Masa Yamauchi Masa Yamauchi Masa Yamauchi Masa Zukanovich Funchal R.
hep-ph/0105196
Study of B- -> D°K~ decays at Belle Measurement of radiative B decays at Belle Search for direct C P violation in B —> Kn decays Study of 3-body charmless B decays at Belle Search for B —> pir, pK and K*TT at Belle Search for neutrinoless T decays into e//x K° Determination of Vc(, from the semileptonic decay W ^ D*+l~v Measurement of B(B° —> D+l~v) and determination of Vcf, Measurement of inclusive semileptonic B meson decay at Belle Study of kaon-lepton correlations in B decay at Belle Search for purely leptonic decays of B meson Measurements of charmed meson lifetimes and the D° — D° mixing parameter yep at Belle Observation of B -> J/ipK\(U7Q) Measurement of branching fractions for B —> 7T7T, KIT and KK decays A measurement of the branching fraction for the inclusive B —> Xsy decays with Belle Observation of B+ —> \coK+ decay Probing FC neutrino interactions using neutrino beams from a muon storage ring
623
This page is intentionally left blank
List of Participants
>.:.^ •
35F-
•j
!!-•-
a
4
-•il- :
ii*.-"?-
# < & : . •
This page is intentionally left blank
LP01 Participants Abada Zeghal Asmaa Abbrescia Marcello Abe Kazuo Abe Koya Adam Wolfgang Adzic Petar Aglietti Ugo Aihara Hiroaki Ajduk Zygmunt Al-Aithan Thamer Alam Mohammad S. Albright Carl H. Albrow Michael Alfonsi Leonardo Ali Ahmed Aliev T. Altarelli Guido Ambrosino Fabio Amelung Christopher Ammar Ray Anderson Gregory Antonelli Antonella Antonelli Mario Anulli Fabio Aoki Shigeki Aoyama Hideaki Apollinari Giorgio Appelquis Thomas Arfaei Hessamaddin Arpaia Veronica Artamonov Andrei Aubert Jean - Jacques Aydin Zekeriya Bacci Cesare Balagura V. V. Baldini Roberto Baranov S. P. Barbieri Riccardo Barker Anthony Baroncelli Antonio Barsuk Sergey Bartl Alfred Basu Rahul
L P T H E Orsay INFN, Bari KEK Research Center for Neutrino Science Austrian Academy of Science VINCA Institute of Nuclear Sciences CERN University of Tokyo University of Warsaw King Fahd University State University at New York State Fermilab Fermilab INFN DESY METU CERN INFN, Napoli CERN University of Kansas Northwestern University
INFN, LNF INFN, LNF INFN, LNF University of Kobe Kyoto University Fermilab Yale University Sharif University of Technology INFN, LNF CERN IN2P3 University of Ankara INFN, Roma III ITEP, Moscow INFN, LNF FIAN NPAD Scuola Normale Superiore, Pisa University of Colorado, Boulder INFN, Roma III ITEP, Moscow University of Vienna CIT, Chennai
627
Behnke Ties Beier Eugene W. Bella Gideon Bellettini Giorgio Bellini Giampaolo Benitez Manuel Aguilar Berger Edmond L. Berkelman Karl Bernido Christopher Casenas Bertolini Stefano Bertolucci Sergio Bertrand Daniel Besancon Marc Besinger Jim R. Bethke Siegfried Bhattacharyya Gautam Bianco Stefano Bijnens Johannes Bini Cesare Blair Ratcliff N. Blaising Jean - Jacques Blazey Gerald Blinov V. Bloch - Devaux Brigitte Blocker Craig A. Blucher Edward C. Blumenfeld Barry J. Bodek Arie Bohm Manfred Bohrer Armin Bondarenko Nikolai Bonissent Alain Bonneaud Gerard Boos E.E. Bortoletto Daniela Boscolo Manuela Bossi Fabio Botella Francisco J. Bozovic-Jelisavcic Ivanka Bozzi Concezio Braccini Saverio Braibant Sylvie Branco Gustavo Brandt Siegmund Brau James E. Briere Roy Brock Raymond Buchholz David Buesser Karsten
DESY-FLC University of Pennsylvania Department of High Energy Physics INFN, Pisa INFN, Milano CIEMAT Argonne National Laboratory Cornell University Central Visayan Institute INFN,Trieste INFN, LNF Universite' Libre de Bruxelles DAPNIA, Saclay Brandeis University Max Planck Institut, Mnchen Saha Institute of Nuclear Physics INFN, LNF University of Lund Universita' di Roma La Sapienza SLAC LAPP, Annecy Northern Illinois University BINP SD RAS DAPNIA, Saclay Brandeis University University of Chicago Univ, Johns Hopkins University of Rochester Wurzburg University of Siegen Kharkov Institute C P P M Marseille LPNHE, Ecole Polytechnique NPI MSU Purdue INFN, LNF INFN, LNF University of Valencia DESY, Zeuthen INFN, Ferrara INFN, LNF CERN Institute Superior Tecnico Univesity of Siegen University of Oregon Carnegie Mellon Michigan State University Northwestern University DESY 628
Bukhari Masroor Buran Torleiv Buras Andrzej J. Burchat Patricia R. Buschhorn Gerd Cabibbo Nicola Cacciari Matteo Caceres Elena Caffo Michele Calaprice Franck P. Caldwell Allen Calvetti Mario Campagnari Claudio Capon Giorgio Carboni Massimo Carli Tancredi Carloni Calame Carlo Michel Carlsmith Duncan Carnegie Robert Carrera Bias Cartlidge Edwin Casalbuoni Roberto Cashmore Roger Cassel David G. Castellani Marzia Cavalli Matteo Cavoto Gianluca Centioni Rossana Chakanov Sergei Chang Darwin Chang Lay N. Chang N g e e - Pong Chaussard Lionel Chauveau Jacques Chen Hesheng Cheng Ta-Pei Cheung Kingman Chigashike Yuichi Chizhov Mihail Chkareuli Jon Choi Kiwoon Choi Seong You Chomeiko L. Nikolai Chowdhury Abul Mansur Chwastowski Janusz Chyla Jiri Ciafaloni Marcello Cianchi Alessandro Clare Robert B.
University of Manchester Oslo University University Munchen Technische Stanford University Max Planck Institut, Munchen Roma La Sapienza INFN, Milano I C T P , Trieste INFN, Bologna Princeton University Columbia University INFN, Firenze UCAL, Santa Barbara INFN, LNF GarrBI DESY INFN, Pavia University of Wisconsin Carleton Stanford University Physics World INFN, Firenze CERN Cornell University Universita' Roma La Sapienza Insto. Fisica de Altas Energias Universita' Roma La Sapienza INFN, LNF DESY Tsing-Hua University Virginia Tech City College of CUNY Claude Bernard University LPNHE-Paris IHEP, Beijing University of Missouri - St. Louis Tsing-Hua University Seikei University CERN Georgian Academy of Sciences KAIST Chonbuk National Minsk University of Chittagong Institute of Nuclear Physics Academy of Sciences of the Czech Republic INFN, Firenze INFN, Roma II UCAL, Riverside
629
Clavelli Louis Close Frank Coan Thomas E. Cochran James H. Colangelo P. Conetti Sergio Conrad Janet Contreras Nuno J. G. Correa dos Reis Alberto Cotti Gollini Umberto Cousins Robert D. Cowen Douglas Cox Bradley B. Crabb Donald Crittenden James A. Cudell Jean - Rene Cundy Donald Curatolo Maria Cvach Jaroslav D'Amato M. Cristina Dagan S. Dai Yuanben Dallapiccola Carlo Danilov Mikhail Dasu Sridhara Davier Michel Davies Gavin J. De Lucia Erika De Rafael Eduardo De Sanctis Enzo De Zorzi Guido Debu Pascal Del Aguila Francisco Del Duca Vittorio Del Re Daniele Dell'Agnello Simone Demina Regina N. Demortier Luc M. Denig Achim Denner Ansgar Desch Klaus Deschamps Olivier Devoto Alberto Di Ciaccio Anna Di Donato Camilla Di Giacomo Adriano Di Nezza Pasquale Diemoz Marcella
University of Alabama Oxford University Southern Methodist University Iowa State University INFN, Bari University of Virginia Columbia University CINVESTAV Centro Brasileiro de Pesquisas Fisicas Universidad Michoacana de San Nicolas de Hidalgo UCLA University of Pennsylvania University of Virginia University of Virginia DESY Universite' de Liege CERN INFN, LNF Academy of Sciences of the Czech Republic INFN, LNF Tel Aviv University I T E P , Beijing University of Massachusetts I T E P , Moscow University of Wisconsin LAL , Orsay Imperial College Universita' di Roma La Sapienza Centre de Physique Theorique, CNRS INFN, LNF Universita' di Roma La Sapienza DAPNIA, Saclay Universidad de G r a n a d a INFN, Torino Universita' Roma La Sapienza INFN, LNF Kansas State University Rockefeller University INFN, LNF PSI Institut fuer Experimentalphysik Univ. Blaise Pascal INFN, Cagliari Univ. Roma Tor Vergata INFN, Napoli INFN, Pisa INFN, LNF INFN, Roma I
630
Dikansky Nikolai Dissertori Gunther Dobado Antonio Dodd Jeremy Dominguez Cesareo Augusto Dong Liaoyuan Dore Ubaldo Dorfan Johnatan Dracos Marcos Drees Jurgen Drell Persis S. Du Dong- Sheng Dubinin M. Ducati Maria B. Gay Eigen Gerald Eisenberg Yehuda Ellis John Ellis Keith Elsen Eckhard Elsing Markus Erdem R. Erdmann Martin Ereditato Antonio Ernst Jesse A. Errede Steven M. Eschrich Ivo E. Escribano Rafel Esposito Bellisario Etzion Erez Extermann Pierre Faber Manfried Faccini Riccardo Fajfer Syjetlana Fanchiotti Huner Fayard L. Iconomidou Federici Claudio Federici Dolores Feldman Gary J. Felsmann Dubois G. Ferrari Anna Ferrari Pamela Ferreira Erasmo Ferroni Fernando Filip Peter Finn John M. Firestone - Alexander Flaminio Vincenzo Fleischer Manfred Floreanini Roberto
Budker Institute of Nuclear Physics CERN Universidad Complutense de Madrid Columbia University University of Cape Town IHEP, Beijing Roma La Sapienza SLAC IRES Strasbourg Univ. Bergische Wuppertal Cornell University IHEP, Beijing NPI MSU Universidad Federal do Rio Grande do Sul University of Bergen Weizmann Institute of Science CERN Fermilab DESY CERN Izmir Yuksek Teknoloji Enstitusu Karlsruhe University INFN, Napoli University of Illinois at Urbana-Champaign University of Illinois at Urbana Imperial College INFN, LNF INFN, LNF Tel-Aviv DPHNC - Univ. Geneve Univ. Technical Vienna INFN, Roma I Univ. Ljubljana Universidad Nacional de La P l a t a LAL , Orsay INFN, LNF INFN, LNF Harvard Caltech INFN, Roma III Indiana University UFRJ Universita' di Roma La Sapienza Max Planck Institut, Munchen College of William and Mary National Science Foundation INFN, Pisa DESY INFN,Trieste 631
Foa Lorenzo Forcella Piero Ford William T. Formanek Jiri Fortugno Fabio Foudas Costas Fraas Hans Frank Marianne Franzini Paolo Frekers Dieter Fried Herb Frisch Henry J. Fritzsch Harald Furtjes Andreas Fuster Verdu Juan A. Gambini Rodolfo Ganis Gerardo Gao Yongsheng Garavaglia Theodore Garvey John Gatti Claudio Geer Stephen Geiser Achim Genchev Vladimir I. Gennari Stefano Georgi Howard Gerard Jean-Marc Giacomelli Paolo Gidal George Gilchriese Murdock G Gilman Frederick Giorgi Marcello Giovannella Simona Giromini Silvia Vannucci Girone Maria Godbole Rohini Goerlach Ulrich Goerlich Lidia Gold Michael S. Goldberg Marvin Goldstein Gary Golob Bostjan Golubev V. Golutvin A.I. Goodman Jordan A. Goshaw Alfred T. Gratta Giorgio Graziani Enrico Greenlee Herbert
INFN, Pisa INFN University of Colorado, Boulder Charles University INFN, LNF Imperial College Institut fuer Theoretische Physik und Astrophysik Concordia University Universita' di Roma La Sapienza Institut fuer Kernphysik, Mnster Brown University University of Chicago Munchen CERN Universidad Valencia Universidad de la Republica Universita' di Roma Tor Vergata Southern Methodist University DIAS-STP Birmingham INFN, Pisa Fermilab DESY-ZEUS Bulgarian Institute for Nuclear Research Nuclear Energy Compaq Harvard University Universite' Catholique de Louvain INFN, Bologna Lawrence Berkeley National Laboratory Lawrence Berkeley National Laboratory Carnegie Mellon University INFN, Pisa INFN, LNF INFN, LNF CERN Indian Institute of Science IRES Strasbourg Institute of Nuclear Physics University of New Mexico National Science Foundation Tufts University University of Ljubljana BINP I T E P - Inst, of Theoretical and Experimental Physics University of Maryland Duke University Stanford University INFN, Roma III Fermilab
632
Grenier Philippe Greub Christoph Grifols Gras Josep Antoni Grigoriev Dimitry Grindhammer Guenter Groom Donal E. Grossmann Nancy Grunhaus J. Guler Murat Gupta V.K Gutay Lazlo J. Guyot Claude Hagopian Vasken Halzen Francis Hamacher Klaus Hambye Thomas Hanson Gail Happacker Fabio Hara Yasuo Harari Haim Harris Fred A. Harvey B. Newman Hauer Thomas Hauschild Michael Hawkings Richard Hayashi Takemi Hazumi Masashi Hearty Christopher Heintz Ulrich Hemingway Richard Hernandez Javier M. Hernndez Rey Juan Jose' Herquet Philippe Herrero Maria Jose' Herten Gregor Heuer Rulf Dieter Heusch Clemens A. Hiller Gudrun Hinchliffe Ian Hirai Shiro Hirata Kohji Hitlin David G. Hofmann Ralf Hollik Wolfgang Hong Deog Ki Honscheid Klaus Hoogland Walter Hou George W. S. Hoyer Paul
CERN Bern Universidad Autonoma de Barcelona BINP Max Planck Institut, Munchen Lawrence Berkeley National Laboratory FERMILAB Tel Aviv University Middle East Technical University University of Jammu Purdue DAPNIA, Saclay Florida State University University of Wisconsin Univ. Bergische Wuppertal Centre de Physique Theorique, CNRS Indiana University INFN, LNF Teikyo-Heisei University Weizamnn Institute of Science University of Hawaii Caltech CERN CERN CERN Kogakkan University Osaka University of British Columbia Boston University Carleton University Univ. Aut. Puebla Valencia Univ. Mons-Hainaut Univ. Autonoma de Madrid Freiburg Hamburg University UCAL, Santa Cruz SLAC Lawrence Berkeley National Laboratory Osaka Electro-Communication Junior College Graduate University for Advanced Studies, Miura Caltech Max Planck Institut, Munchen Karlsruhe Pusan National University Ohio State University Amsterdam University of Taiwan The Niels Bohr Institute 633
Huang Tao Hughes Richard Hulth P - 0 Huo Wujun Httman Kay Hylen James Iacobucci Giuseppe Iarocci Enzo Iddir Farida Ghislaine Imlay Richard L. Incandela Joseph R. Ishii Takanobu Ishikawa Kenzo Isidori Gino Jacquet Marie Jaffe Robert L. Jain Vivek Janot Patrick Jarlskog Cecilia Jaros John A. Jawahery Abolhassan Jegerlehner Fred Jikia Georgi Jin Changhao Jowett John Jung Chang Kee Kabana Sonja Kagan Harris Kalmus P. I. P. Kang Joo Sang Karlen Dean Karshon Uri Kaseman Matthias Kawamoto Tatsuo Kawarabayashi Ken Kayser Boris Kephart Robert Kernel Gabrijel Kessler Richard Khan Akram Khodin Alexandre A Khoze Valeri A. Kim Young Kee Klein Joshua R. Kleingrothaus H. V. Klapdor Kluit P. Kluth Stefan Kneringer Emmerich
IHEP, Beijing Ohio State University University of Stockholm IHEP, Beijing Max Planck Institut, Mnchen Fermilab INFN, Bologna INFN Universite' d'Oran Louisiana State University UCAL, Santa Barbara KEK, IPNS Graduate School of Science, Sapporo INFN, LNF LAL , Orsay MIT Brookhaven National Laboratory CERN CERN SLAC University of Maryland DESY, Zeuthen Universitat Freiburg University of Melbourne CERN SUNY at Stony Brook University of Bern Ohio State University Queen Mary and Westfield College Korea University Carleton University Weizmann Institute of Science Fermilab University of Tokyo University International Tokyo National Science Foundation Fermilab Institute Jozef Stefan Chicago University University of Edinburgh Belarus National Academy of Sciences University of Durham Lawrence Berkeley National Laboratory University of Pennsylvania MPI Heidelberg NIKHEF Max Planck Institut, Munchen University of Innsbruck
634
Kniehl Bernd A. Ko Pyungwon Kobayashi B Shigeharu Kobayashi Tetsuro Koenigsmann K a y Kofler Richard R. Kolanoski Hermann Kolb Edward Kolodziej Karol Komamiya Sachio Kopp Sasha E. Krner Jrgen Kowalewski Robert Kozanecki Witold Kramer Robert Kreuzer Peter Krishnaswamy M.R. Kroeger Robert Kroll Peter Kruchinin S. P. Kubota Takahiro Kulikov Slava Kundu R. Kuno Yoshitaka Kusenko Alexander Kuze Masahiro Kuzmin A. Kvatadze Ramaz Lacava Francessco Lafferty George Lam C. S. Lami Stefano Lanfranchi Gaia Lanou Robert Lashin Elsayed I Lau Kwong A. Laurelli Paolo Laveder Marco Layter John G. Lebrun Patrice Lee - Franzini Juliet Lee Roman Leemans Wim Leibbrandt George Leith David W. Lemaire Marie-Claude Lemonne Jacques Leroy Olivier
Institut fuer Theoretische Physik, Hamburg KAIST Saga University University Fukui Freiburg Stanford University Humboldt-Universitaet Berlin Fermilab Silesia University Tokyo University University of Texas Mainz University of Victoria DAPNIA, Saclay Carnegie Mellon University DESY Tata Institute University of Mississippi Univ. Bergische Wuppertal I T E P , Moscow Osaka University I T E P , Moscow DrVrije Universiteit Osaka UCLA KEK BINP Tbilisi State University INFN,Roma I University of Manchester McGill University Rockefeller University INFN, LNF Brown University Ain Shams University University of Houston INFN, LNF INFN, Padova UC Riverside IPNL, CNRS-IN2P • INFN, LNF Accelerator and fusion research University of Guelph SLAC DAPNIA/SPP University Vrije Brussel C P P M Marseille
635
Letts James Lhallabi Touria Li Ling-Fong Li Weiguo Liang Zuo - Tang Limon Peter Linnemann James Lipkin Zvi Lisi E. Litchfield Peter J. Littenberg Laurence Liu Chun Liu Qiu - Yu Loewe M. Lohmann Wolfgang Lopez Castro Gabriel Losty Michael Love Sherwin T. Lu Gongru Lu Jizong Lucha Wolfgang Luci Claudio Lucio Martinez Jose Luis Ludwig Jens Lusignoli Maurizio Luth Vera MacDonald Art Macfarlane David B. Machefert Frederic Macri' Mario Maharana J. Maharana L Mahlke - Kruger Hanna Maiani Luciano Majerotto Walter Maki Akihiro Maksimovic Petar Mallik Usha Maltoni Michele Mandelkern Mark Mandula Jeffrey E. Mankel Rainer Mannelli Italo Manz Andreas Maor Uri Marage Pierre Marciano William Margoni Martino Marini Giuseppe
Indiana University Universite' Mohammed V Carnegie Mellon IHEP, Beijing Shandong University Fermilab Michigan State University Weizmann Institute of Science INFN, Bari Rutherford Appleton Laboratory Brookhaven National Laboratory I T E P , Beijing USTC, Hefei Universidad Catolica de Chile DESY, Zeuthen CINVESTAV TRIUMF PurdXe University Xinxiang Normal University University Shanghai Teachers Austrian Academy of Science Universita' di Roma La Sapienza Universidad Guanajuato University of Freiburg Universita' di Roma La Sapienza SLAC Queen's University UCAL, San Diego LAL , Orsay INFN, Genova Institute of Physics Utkal University Cornell University CERN Austrian Academy of Science KEK Johns Hopkins University University of Iowa Universidad de Valencia UCAL, Irvine U.S. Department of Energy DESY INFN, Pisa Max Planck Institut, Munchen Department of High Energy Physics Universite' Libre de Bruxelles Brookhaven National Laboratory INFN, Padova Universita' di Roma La Sapienza
636
Marks Joerg Marlow Daniel L. Martinelli Guido Mateev Matei Matsuda Koichi Matsuda Masahisa Matsuda Satoshi Matteuzzi Clara Mazzitelli Giovanni Mc Bride Patricia Mc Carthy Bob Mc Kay Doug Mc Kellar B. H. J. Medin Gordana Melanson Harry Meloni Davide Menary Scott Menichetti Ezio Merenkov Nikolav Merk M. Merola Leonardo Mes Hans Mescia Federico Metzger W. J. Meyers Peter Mezzorani Giuseppe Migliozzi Pasquale Mikenberg Giora Mikuz Marko Miller David H. Miller James Minard Marie - Noelle Minkowski Peter Miscetti Stefano Mishra Shekar Mohapatra Rabindra Monteil Stephane Moreau Francois Morii Masahiro Mork Kjell Moroni Luigi Morse William Moshe Moshe Moulson Matthew Moutoussi Ann Murayama Hitoshi Muresan Raluca Murtas Fabrizio
Heidelberg Princeton University Universita' di Roma La Sapienza Univ. Kliment Ohridski Kyoto University Aichi University Kyoto University INFN, Milano INFN, L N F Fermilab SUNY at Stony Brook Kansas University University of Melbourne Univ. Podgorica Fermilab Universita' di Roma La Sapienza York University INFN Torino NSC K I P T NIKHEF INFN, Napoli Carleton University Universita' di Roma La Sapienza Univ. Katholieke Nijmegen Princeton University INFN, Cagliari INFN, Napoli Weizmann Institute J. Stefan Institute Purdue University Boston University LAPP, Annecy University of Bern INFN, L N F Fermilab University of Maryland Universite' Blaise Pascal LPNHE, Ecole Polytechnique Harvard The Norwegian University INFN, Milano Brookhaven National Laboratory Technion INFN, LNF CERN UCAL, Berkeley The Niels Bohr Institute INFN, LNF
637
Mutchler Gordon Nagashima Yorikiyo Nagy Miroslav Nakamura Kenzo Namkung Won Nandi Satynarayan Napoly Olivier Narain Meenakshi Nash Jordan Nason Paolo Nasriddinov Komiljon R. Nassalski Jan Nauenberg Uriel Nelson Charles A. Neubert Matthias Nicholson Howard Niczyporuk Bogdan Nikolaidou Rodanth Nishiura Hiroyuki Nisius Richard Nowak Grazyna Nozaki Tadao O'Dell Vivian O' Donnell Patrick J. O' Fallon John R. Oakes Robert Oakham Gerald Odyniec Grazyna Oh Benedict Y. OhC H Oh Sun Kun Ohshima Takayoshi Olesen Poul Olsen Stephen Olsson Jan Onofre Antonio Oreglia Mark J. Oren Yona Orr Lynne H. Osland Per Ots Ilmar Paar Hans P. Padley Paul Paganoni Marco Pakvasa Sandip Palutan Matteo Pan Yibin Pancheri Giulia
Rice University Osaka University Slovak Academy of Sciences KEK, IPNS University Pohnag Oklahoma State University DAPNIA-SEA Boston University Imperial College INFN, Milano Uzbekistan Academy of Sciences CERN University of Colorado, Boulder SUNY at Binghamton Cornell Mount Holyoke College Thomas Jefferson National Accelerator Facility DAPNIA, Saclay Osaka Institute of Technology,Junior College CERN Polish Institute of Nuclear Physics KEK Fermilab Toronto U.S. Department of Energy Northwestern Uiversity Carleton University Lawrence Berkeley National Laboratory Penn State University National University of Singapore Konkuk University Nagoya University The Niels Bohr Institute University of Hawaii DESY CERN Chicago Tel Aviv University University of Rochester University of Bergen Tartu University UCAL, San Diego Rice University INFN, Milano University of Hawaii INFN, Roma III University of Wisconsin INFN, LNF
638
Panella Orlando Paolini Gabriella Paolone Vittorio Paramatti Riccardo Parsons John Partridge Richard Parzefall Walter Schmidt Paschos Emmanuel A. Pasqualucci Enrico Passeri Antonio Paterson Ewan Paul Ewald Paul Stefan Paus Christoph Pavlenko Oleg Pedrini Daniele Pellegrini Claudio Penarrocha Jose' Penev Vladimir N. Perez Angon Miguel A. Perries Stephane Perrino Roberto Perroud Jean - Pierre Pesen E. Peskin Michael Pestieau Jean Peterson Daniel Peterson Earl Petkov Serguey Todorov Petrolo Emilio Petronzio Roberto Pham Tri - Nang Phillips Hywel Phua K. K. Picciotto Charles Pichl Hannes Pilaftsis Apostolos Pinto Eboli Oscar Jose' Pohl Martin Poling Ron Ponce Gutierrez William Pontecorvo Ludovico Poppitz Erich R. Pospelov Gennady Possanza Pina Prepost Richard Price Lawrence E. Pullia Antonino Puis Philipp
INFN, Perugia GarrBI University of Pittsburgh Universita' di Roma La Sapienza Columbia University Brown University Institut fuer Experimentalphysik, Hamburg Institut fur Theoretische Physik III INFN, Roma I INFN, Roma III SLAC Bonn University Munchen Technische Massachusetts Institute of Technology BITP INFN, Milano University of California, Los Angeles Universidad de Valencia Joint Institute for Nuclear Research- JINR CINVESTAV Universite' Claude Bernard INFN, Lecce IPHE CERN Stanford Linear Accelerator Center Universite' Catholique de Louvain Cornell University University of Minnesota INFN,Trieste INFN, Roma I Universita' di Roma Tor Vergata Ecole Polytechnique Royal Holloway University National Singapore University of Victoria INFN, LNF Universitaet Wwrzburg Universidade de Sao Paulo DPHNC Universite' de Geneve University of Minnesota Universidad de Antioquia INFN, Roma I Yale University BINP INFN, LNF University of Wisconsin Argonne National Laboratory INFN, Milano University of Vienna 639
Raczka Piotr Randall Lisa Reay Neville W. Rebelo M. N. Reidy James Reina Laura Rekalo Mikhail Repko Wayne Reucroft Stephen Riazzuddin S. Richard Francois Richter Burton Rijssenbeek Michael Rith Klaus Rizzo Marco Roberts B. Lee Rodrigo Teresa Roe Byron Roe Natalia A. Romero Alessandra Roney Michael Roodman Aaron Rosier - Lees Sylvie Rosner Jonathan L. Rossi Antonio Maria Rotelli Pietro Roth Markus Roudeau Patrick Rubinstein Roy Rubio Juan Antonio Rckl Reinhold Ruhlmann Vanina Rusack Roger Sabatini Lia Sabinov Vladimir Sachrajda Chris T. Sadoff Ahren Sadrozinski Hartmut F. Saito Naohito Sakai Norisuke Sakuda Makoto Salvati Andrea Sanchez - Colon Gabriel Sanda Antony Sander Heinz Georg Sankar S. Uma Santangelo Paolo Santoro de sa A. F. Saphiro Marjorie D.
Warsaw University MIT Kansas State University Instituto Superior Tecnico University of Mississippi Florida State University NSC K I P T Michigan State University Northeastern University Pakistan National Centre for Physics LAL , Orsay Stanford Linear Accelerator Centre SUNY at Stony Brook Physikalisches Institut INFN, LNF Boston University Universidad de Cantabria University of Michigan Lawrence Berkeley National Laboratory INFN, Torino Victoria University SLAC LAPP, Annecy University of Chicago INFN, Bologna INFN, Lecce Leipzig LAL , Orsay Fermilab CERN Wurzburg DAPNIA/SPP University of Minnesota INFN, LNF University of Pittsburgh University of Southampton Cornell UCAL, Santa Cruz RIKEN BNL Research Center Tokyou Institute of Technology KEK, IPNS GarrBI CINVESTAV Nagoya University Mainz Indian Institute of Technology INFN, LNF LAFEX.CBPF Lawrence Berkeley National Laboratory
640
Sassot Rodolfo Sato Joe Satta Alessia Saxon David Schael Stefan Schmidt Carl R. Schmidt Michael P. Schopper Andreas Schrempp Fridger Schroeder Henning Schubert Klaus R. Schulte Reiner Schultz Jonas Schwartz Morris Schwarz Andreas Sciascia Barbara Sciolla Gabriella Semertzidis Yannis K. Serednyakov S. Serin M. Sfiligoi Igor Shaevitz Michael Shanidze Revaz Sharma Vivek A. Sheldon Paul Shen Benjamin C. Shepard Paul F. Shipsey Ian P. Shrock Robert Silverman Dennis J. Singer Paul Sirlin Alberto Sirunyan Albert M Sissakian A. S. Sitenko Yuri Skrinsky Alexander Skuja Andris Skwarnicki Tomasz Sliwa Krzysztof J. Smith A. J. Stewart Smith Jack Smith James G. Sobie Randall Son Dongchul Song Hee Sung Soper Davison Sotnikova N. A. Spada Francesca
Buenos Aires University Kyushu Universita' di Roma La Sapienza Glasgow University RWTH Aachen Michigan State University Yale University CERN DESY Universitaet Rostock Institut fuer Kern und Teilchenphysik, Dresden RWTH Aachen UCAL, Irvine Johns Hopkins University DESY Universita' di Roma La Sapienza MIT Brookhaven National Laboratory BINP Middle East Technical University INFN, LNF Fermilab Universitaet Erlangen UCAL, San Diego Vanderbilt University UCAL, Riverside Pittsburgh University Purdue University SUNY at Stony Brook University of California, Irvine Technion University of New York Yerevan Physics Institute JRNR BITP Budker Institute of Nuclear Physics University of Maryland Syracuse University Tufts University Princeton SUNY at Stony Brook Syracuse University University of Victoria Center for High Energy Physics University Seoul National University of Oregon NPI MSU Universita' di Roma La Sapienza
641
Spadaro Tommaso Spengler Joachim Spooner Neil Sridhar K. Stahl Achim Stavropoulos Georgios Steinhauser Matthias Stodolsky Leo Stoesslein Uta Stone Sheldon Straumann Ulrich Strom David M. Strowink Mark W. Stroynowski Ryszard A. Struczinski Wolfgang Stugu Bjarne Su Dong Su Jun-Chen Sugawara Hirotaka Sun Luorui Surrow Bernd Sykora Tomas Szklarz Gerszon Tagasuchi Eiichi Tajima Hiroyasu Takasaki Pumihiko Takeuchi Tatsu Tantalo Nazario Tarem Shlomit Tata Xerxes Tauscher Ludwig Taylor G. N. Teh Rosy Chooi Gim Terranova Francesco Teubert Frederic Timmermans C. W. Timoshin Sergei Toback David Tokushuku Katsuo Tolun Perihan Tomalin Ian R. Toohig Timothy E. Tornkvist Ola Tracas Nicholas D. Treuman T.L. Tully Chris Tung Wu-Ki Turala Michal
Universita' di Roma La Sapienza MPI Kernphysik University of Sheffield T a t a Institute DESY, Zeuthen CERN Institut fuer Theoretische Physik, Hamburg Max Planck Institut, Munchen University of Colorado Boulder Syracuse University Zurich University of Oregon Lawrence Berkeley National Laboratory Southern Methodist University RWTH Aachen University of Bergen SLAC Jilin University KEK Zhengzhou University DESY Charles University LAL , Orsay University of Osaka University of Tokyo KEK Virginia Tech INFN, Roma I Technion University of Hawaii University of Basel University of Melbourne Mara Institute of Technology INFN, LNF CERN Univ. Katholieke Nijmegen Gomel State Technical Texas A&M University KEK Middle East Technical University Rutherford Appleton Laboratory U.S. Department of Energy World Scientific NTU Athens Brookhaven National Laboratory Princeton Michigan State University Polish Institute of Nuclear Physics
642
Turnau Jacek Turner Michael Tyurin N. E. Urner David Valencia German Valente Enzo Valente Paolo Van Bibber Karl A. Van Middelkoop G. Van Soa Dang Verderi Marc Vincent Pascal Voena Cecilia Volkas R. R. Volobouev Igor Voss Rudiger Votano Lucia Vovenko A. S. Wadhwa Maneesh Wagner Albrecht Wagner Gregor Wagner Stephen R. Wahl Heinrich Walenta Heinrich Wali Kamesh Wappler Frank S. Ward Bennie Ward C. Pat Ward David R. Webb Robert C. Weber Alfons Weiler Thomas J. Weilhammer Peter Weinstein Alan J. Wengler Thorsten Wermes Norbert Westerhoff Stefan Whitmore Jim J. Wickens Fred J. Wickens John Williams Philip K. Willocq Stephane Wilson Peter Wimpenny Stephen J. Wise Mark Wisniewski William Wiss James Witherell Michael
Polish Institute of Nuclear Physics University Chicago IHEP Cornell University Iowa State University INFN, Roma I INFN, LNF LLNL NIKHEF Vietnam Institute of Physics LPNHE, Ecole Polytechnique LPNHE-Paris Universita' di Roma La Sapienza University of Melbourne Lawrence Berkeley National Laboratory CERN INFN,LNF IHEP Basel DESY University of Rostock SLAC CERN Gesamthochschule, Siegen Syracuse University SUNY at Albany University of Tennessee Cambridge Cambridge University of Texas A&M NAPL Vanderbilt University CERN Caltech CERN Bonn Columbia University Penn State University Rutherford Appleton Laboratory Universite' Libre de Bruxelles U.S. Department of Energy University of Massachusetts Fermilab UCAL, Riverside Caltech SLAC University of Illinois at Urbana Champaign Fermilab
643
Wojcicki Stanley G. Wotschack Jorg Wright Douglas M. Wu Sau Lan Wuerthwein Franck Yabuki Tetsuo Yaffe Laurence G Yager Philip M. Yamada Sakue Yamamoto Hitoschi Yamamoto Richard K. Yamanaka Taku Yamauchi Masanori Yan Tung Mow Yilmazer A. U. Yoshimura Motohiko Yu Hoi - Lai Yuta Haruo Zacek Josef Zanello Lucia Zeller Michael Zepeda Arnulfo Zeppenfeld Dieter Zhang Zhao-xi Zhang Changchun Zheng Zhi - Peng Zhu Shouhua Zhu Yongsheng Zielinski Marek Zito Marco Zuber Kai Zukanovich Funchal Renata Zumerle Gianni Zwirner Fabio
Stanford University CERN LLNL University of Wisconsin MIT University Rakuno University of Washington UCAL, Davis KEK University of Hawaii MIT University of Osaka KEK Cornell University University of Ankara University Tohoku Academia Sinica, Taiwan Aomori University Academy of Sciences of the Czech Republic Universita' di Roma La Sapienza Yale CINVESTAV University of Wisconsin IHEP, Beijing IHEP, Beijing IHEP, Beijing Karlsruhe IHEP, Beijing Rochester University DAPNIA, Saclay Lehrstuhl fuer Experimentelle Physik IV, Universidade de Sao Paulo INFN, Padova Universita' di Roma La Sapienza
644
Author Index Kim, Young-Kee, 183 Klapdor-Kleingrothaus, H. V., 515 Klein, Joshua R., 470
Aoki, Shigeki, 485 Barbieri, Riccardo, 141 Bijnens, Johan, 240 Bossi, Fabio, 106
Maiani, Luciano, 555 Miller, J. P., 408 Murayama, Hitoshi, 495
Cabibbo, Nicola, 596 Carli, Tancredi, 273 Cassel, David G., 29 Close, Frank E., 327
Nash, Jordan, 60 Nason, Paolo, 209 Neubert, Matthias, 14
Dorfan, Jonathan, 3 Drees, J., 349
Odyniec, Grazyna, 191 Olsen, Stephen L., 4
Ellis, John, 374 Erdmann, Martin, 259
Randall, Lisa, 595 Roudeau, P., 119
Geer, Steve, 579 Goodman, J. A., 445
Sachrajda, C. T., 226 Stosslein, Uta, 308
Halzen, Francis, 526 Hanson, Gail G., 426 Heuer, R.-D., 565
Tajima, H., 45 Turner, Michael S., 540 Wise, Mark B., 150
lacobucci, Giuseppe, 292 Iconomidou-Fayard, Lydia, 92 Isidori, Gino, 160
Zwirner, Fabio, 390
Jung, C. K., 456 Kessler, R., 79
645
XX INTERNATJONAI Syiviposiuivi ON
LEpTON ANCJ PHOTON INTERACTIONS AT Hiqln ENERGIES
LEPTONPHOTON
0/
This important book covers topics
that are of major interest to the high energy physics community, including the most recent results from flavour factories, dark matter and neutrino physics. In addition, it considers future high energy machines.
www. worldscientific. com 4903 he
