Download slac-pub-5910.pdf and more Lecture notes Electromagnetism and Electromagnetic Fields Theory in PDF only on Docsity!
- (^) - .-- (^) SLAC-PUB- September 1992
WEAK-ELECTROMAGNETIC INTERFERENCE(T/E) IN
POLARIZED eD SCATTERING*
Charles Y. Prescott Stanford Linear Accelerator Center, Stanford University, Stanford, CA 94309
; -- ABSTRACT
Observation of parity non-conservation in deep-inelastic scattering of polarized elec- trons from deuterium was reported in an experiment at SLAC in 1978. The events at SLAC and elsewhere leading to the successful search for parity non-conservation in the electromagnetic_. processes are described.
..--. Introduction
In 1978, a team of twenty physicists performed an experiment at SLAC which demon- strated convincingly that the weak and electromagnetic forces were acting together in a fundamental process, the inelastic scattering of polarized electrons. This result showed that the electron was a normal partner in the model of electroweak interactions as spelled out by Wei.nberg in -1967. The work I am about to describe is the summary of work done mostly by other persons as part of the team effort. In this summary I have tried to give credit to the many excellent contributions from this group. I had hoped to point out all of the important individual efforts that were so critical to the overall success. However, looking at what I said and what is written, I feel that this summary falls short of that goal. It is very difficult to-be comprehensive-%c. -; and at the same time concise. Furthermore my perspective on this JCWork Invite-d supported talk presented by Department at the Third of Energy Intern.ationaJ contract Symposium DE-AC03-76SF00515, on the History of Particle Physics, Stanford, California, June 24-27, 1992
- effoi%~is biased toward parts of the work in which I was involved or which I saw going on around me: I surely have missed some of the activities by others on this team, and I apologize for those shortcomings. This talk should be taken as a personal perspective on the work that occurred over a period of eight years at SLAC, Yale University, and other -places. As a part of this talk, the organizers asked that I summarize the work in atomic physics_tp_seekout parity violating effects in atomic levels. I reluctantly agreed to attempt this, even though I had no involvement in those experiments and do not feel qualified to -“discuss historical notes in that field. In the short time allocated to me, summarizing a major piece of experimental work that lies outside my primary topic is not possible. . _. What^ I give^ here^ is a brief^ history^ of the^ search for^ optical^ rotation^ by^ a bismuth vapor, as--reported in the literature. I have not attempted to extend this summary to cover the work. _ eon the other atoms, thallium, lead, and cesium, which came somewhat later. A proper history talk on the subject of parity violation in atomic physics would include those contributions as well. The work with bismuth atoms began in the mid- 19709, so events were occurring during the time work was underway at SLAC. Some of those events had significant impact for our work. With these thoughts, let me begin, Physicists love symmetries. Among the important symmetries, parity (the symmetry of mirror reflection) was assumed to be valid for nature’s forces and fundamental processes until the mid-1950s when the weak force was shown to violate maximally the parity symmetry in P-decay processes. ’ This unexpected result came as a shock and a surprise. The experimental observations were made in charged-current processes (mediated by the-W*, as we now.-- _ know). In those days it was conjectured that weak neutral-current processes%ould-exist, but no experiments had access to such processes, and advances in
was’fhis 4’ experiment that provided the basic information later used for the parity vi- olation work. -Cross sections, counting rates, and backgrounds were measured, and E was the beginning of a learning curve for me; the facility, the equipment, the beams and monitoring, and the people who inhabited the lab (what they do and what they know). In late 1970, Professor Vernon Hughes from Yale visited SLAC and presented a proposal to build a polarized electron source for the linac and to accelerate the electrons to high (^) -..energies.Cd The proposed source was based on a Yale prototype which stripped electrons from a polarized atomic beam of 6Li using an ultraviolet flash lamp.3 The .physics motivation for this proposal was to study the spin of constituents inside polarized protons. That proposal was soon presented to SLAC and was formally accepted and designated E80. The E80 proposal was the beginning of a long and successful program -on spin structure which continues to be active today.
I attended_- Prof. Hughes’ seminar in 1970. In my mind there still was the interest in searching for parity violation, and perhaps this source could be used to look for s a s terms using the incoming polarized electron beam. I took this idea seriously and began to study the feasibility of a parity violation measurement in the End Station A facility. I remember taking my idea to Richard Taylor looking for support and encouragement and laterto Sid~Drell (which I got from them). Early in 1971, I arranged a visit to Yale to talk to Vernon Hughes and his group. I wished to form an experimental collaboration and felt I needed the ‘involvement of the Yale group. We discussed the physics possibilities and various strategies. We identified three possible approaches: (i) to utilize the planned E experiment and to study parity violation by averaging over the E80 target polarization; (ii) to extend the E80 running to provide dedicated time for a parity violation search; (i-ii)- to propose--.--.L _an independent dedicated experiment for parity violation. We agreed to collaborag and to pursue (i) and (iii).
- The E80 experiment was already planned and item (i) required no action on my part. I focussed my attention on a new proposal, E95, whose objective was solely the search for parity violation. Unlike E80, the E95 target was chosen to be unpolarized hydrogen, eliminating a potentially serious systematic error in E80 from polarized protons. The .E95 target was optimized for the parity violation test. Motivation for E95 was not easy. E95 was (quoting from the proposal) “. ,. not sensitiv&o weak neutral currents... .” We knew that the statistical error on Apv would be too large. Weak-electromagnetic interference required an error AA,, < 10D4Q2/Ml, ..‘and our calculated statistical error was an order-of-magnitude larger. Weak neutral currents were simply not reachable by the techniques we had at hand. (The Yale-SLAC source, called PEGGY, was too low in intensity, 2 x log/pulse at 80% polarization.) Furthermore helicity reversals required reversal of a magnetic field to flip the electron spin. Thea&ion. _ of spin reversal affected the beam parameters and introduced worrisome systematic errors as well. In spite of these limitations and concerns, we proceeded with the E95 proposal. The formalism for inelastic scattering of polarized electrons was not available in the literature..-- We knew- that polarized electron inelastic scattering and v inelastic scattering .. were kinematically very similar. With the aid of a paper by Stephen Adler4 we showed thatrelaxing’parity invariance introduced a third structure function Ws(v, Q2) in addition to the usual IV1 and W2. Furthermore, requiring that current conservation be valid led to TVs(v, Q2) + 0 as Q2 + 0. We argued in the E95 proposal that such parity violating terms may exist and could have escaped detection in former experiments at low energies, for-example,--.--.2 in nuclear_ physics studies which were the most sensitive. We could find no experimexal work which ruled out such terms at the level of sensitivity achievable in E95.
God& in inelastic electron scattering. We proposed to replace the PEGGY source with
a new polarized-electron source that would-be laser-driven. We sought his support. We had discussed more than one type of device, and were considering photoionization of cesium as one possibility. We were also interested in solid materials for a cathode, and .had discussed our needs with Ed Garwin (SLAC). During that year (1974), Garwin went to ETH Ziirich for a visit, and while there proposed with H. C. Siegmann and Dan Pierce the use of negative-electron-affinity gallium arsenide for a suitable cathode material for a high-intensity polarized-electron source. l3 It was their proposal which turned out to be ,a crucial step for success. The combination of a laser (at moderately high powers) and a solid-state cathode material (having high-electron densities) promised to provide the large-electron currents we needed to reach the elusive weak-electromagnetic interference e&t s. Polarization of the photoemitted electrons resulted from circular polarization of the laser exciting valence-band electrons to the conduction-band. Electrons near the surface could^ _~ escape. Polarization values near 50% were expected as a consequence of the angular-momentum selection rules. Polarization reversal was accomplished optically by reversing the circular polarization. Thus in 1974, a new experiment El22 emerged, proposing to test the Weinberg- Salam model via parity violation with a sensitivity AA,, 2 1 x 10v5 near Q2 a 1 (GeV/c)2. The El22 proposal was developed from the experiences E95. Enhancements over the E95 rates would be large:(i) the beam current was up by a factor of 250; (ii) a new spectrometer using magnets from the 8 and 20 GeV/c spectrometers was designed for a large acceptance, an improvement by a factor of 5; and (iii) a 30-cm-long deuterium target was planned, for an additional factor of 3. The overall gain over E95 was the product of-these factors--.--.A _(approximately 3750) w h’ hIC would allow us to reach l-a sensitivity to Weinbergxalamneutral currents in as little as 15 minutes of beam time.
- ‘The proposed El22 experiment was approved in June 1975. During the next two- and-a-half years, work on the PEGGY-II source was underway. In December of 1977, the new source was ready and was tested in a brief run on the SLAC linac. Before describing the El22 experiment, however, I now want to turn to developments in the field of atomic physics which were progressing rapidly at the time.
Parity Violation in Atomic Physics
Zel’dovich in 19606 was perhaps the first to suggest looking for optical rotation of the plane of linear polarization of light passing through a gas vapor. He concluded that optical rotation by hydrogen would be too small to detect. The subject of optical rotation was revitalized by work of the Bouchiats in Paris in 197414 and by Kriplovich at Novosibirsk.15 They pointed .out that in high-Z atoms, the optical rotation is enhanced by an approximately Z3 factor and that the sought-after parity-violation effects could become m&rable in atomic systems using reasonable laboratory techniques. With this stimulation, several groups at widely separated institutions proposed experiments in 1974. Bismuth (Z=83) was identified as a particularly promising atomic system. Four groups, at Oxford, Seattle (University of Washington), Novosibirsk, and Moscow, proposed generally similar.-- measurements based on optical rotation of the plane of linear polarization in .- bismuth. The specific details of the four proposed experiments differed considerably. At Berkeley, a thallium (Z=81) (^) ex periment was proposed to measure circular dichroism of a light beam. (Circular dichroism, the unequal absorption of opposite-circular polarization, is closely related to optical rotation of linear polarization). The Paris group proposed studying circular dichroism in cesium (Z=55). r” In^ the--.--.---^ bismuth _^ experiments,^ the^ basic^ idea^ starts^ with^ crossed-linear^ polarizers. In a hyp%hetical-ideal experiment with perfect-optical elements, a light beam is not
(iij’molecular species in the bismuth vapor that masked the desired spectral lines; (iii) Faraday rotations induced by stray residual fields; (iv) extra undesired materials in the optical path, such as cell windows; (v) thermal drifts; (vi) scattering and reflections lead- ing to laser-beam interferences; and (vii) undesirable influence on the laser beam due to the scanning or modulation techniques used.
On the theoretical side, considerable work was underway to understand the proper approa&needed to deal with the complicated electronic structure of bismuth. The uncertainties were exacerbated by the lack of a good value of sin20W at the time.
In 1977 Seattle and Oxford completed their first measurements and published ad- jacent articles in Physical Review Letters. l6 Both experiments reported null results on -o$ical rotation._.. -with experimental precision substantially better than needed for the Weinberg&lam- model predictions. These groups announced that the Weinberg-Salam predictions-for the electron neutral current interaction was wrong. In hindsight we know that these experiments were wrong. However the simultaneous publication of two sep- arate groups at that time created considerable turmoil and controversy in the physics community.
During this period when the atomic physics experiments were active and being dis- cussed at conferences and meetings, the work at SLAC had been proceeding steadily. It was at this time, shortly following the publication of null results by the Seattle and Ox- ~ford groups, that the SLAC experiment was ready. A polarized source, suitable for test- ing the Weinberg-Salam effects in deep-inelastic scattering, was completed. The source was tested--.--.- on the_ SLAC linac in December 1977. The El22 experiment was scheduled to begin ZFebruary 1978.
- ‘Before turning to the SLAC experiment, however, I would like to continue the story of atomic bismuth-parity violation. In March 1978, the Novosibirsk group reported seeing evidence for parity violation on the 648 nm line in bismuth. l7 The initial reports were accompanied by somewhat large systematic errors, which were subsequently reduced in 1979 without affecting the reported value. In 1980, the Moscow group reported a null result on the same 648 nm line in;&smuth in their experiment, in agreement with the earlier Oxford and Seattle null results. By 1981 Seattle had improved and repeated their experiment and reported :new results. The Seattle group now reported seeing evidence for parity violation, but somewhat smaller in magnitude than the Novosibirsk result. In 1984 Moscow and Oxford reported results of their improved experiments, which agreed with the 1981 results of Seattle. The Novosibirsk^ group apparently did not report any new measurements in the years after 1979.-~ This history is summarized in Figure 1 where the bismuth results (but not the thallium, lead, or cesium results) are shown. The theory of parity violation in atomic bismuth was sufficiently uncertain in the early years that calculations provided only general guidance. The authors of the papers reporting parity violation all reported agreement with the theory. As the experimental results were improved, the results settled down to approximately l/2 of the value reported by the Novosibirsk group. The theory was refined and remained in agreement with the experiments. Th e values for theoretical expectations as quoted by the authors is also shown in Figure 1.
- In 1987, a group of authors18 published a global analysis of weak-neutral experi- ments.r- In that report regarding the early parity-violating experiments they say: _: .-- _ TV e^ have^ omitted^ the^ early^ null^ experiments,^ the^ Novosibirsk^ bismuth
- ‘&loped and installed to stabilize the beam which had a natural tendency to drift around. Beam-polarization monitors were installed and backup monitors were added to provide a redundancy. A Mott polarimeter at the source and a Meller polarimeter at the experiment were installed. The spectrometer was instrumented with two detectors which operated independently to measure asymmetries. Two independent computer codes were developed-..e4 to check the analysis (ultimately the data were processed in two computers). . (^) This rather elaborate preparation before the experiment reflected our in- ternal concerns that the experiment would be a very difficult one to prove, first to ourselves, but then ultimately to the physics community outside the experi- -- ment, By-February 1978, the El22 experiment was scheduled to run, and checkout of the beam and the spectrometers began using unpolarized electrons from the thermionic gun. The checkout procedures were rather lengthy, involving looking at each component and carefully testing the performance. These tests typically utilized low pulse rates while beams to other experiments were in use. By late -March, the tests were mostly complete. Richard Taylor had arranged, through earlier negotiations within the laboratory, to run El22 without any beams to other experiments. This dedicated mode, that is sole use of the linac for the El -experiment, was exceptional but proved to be very important to the experiment. It contributed to the stability .of the beams. It also contributed to an improved confidence in the crew of experimenters, and to the undivided attention of the _ _.accelerator--..-.L _operators devoted to E122. El22 began dedicated-beam operation in Aprilw.^ 1978 with: polarized beams.
The polarized-electron source delivered longitudinally polarized electrons to the linac at the rate of 120 pulses per second. The source was driven by circularly polarized light from a dye laser at 710 nm wavelength. Circular polarization was achieved by a calcite prism linear polarizer followed by a Pockels cell, as seen in Figure 2. Reversal of the biasing voltage on the cell would reverse the circular polarization. The linac accelerated the beams with little depolarization. The experiment could be operated with eL or eR beams, at the choice of the experimenters.--*e4 Throughout most of the running, eL and eR pulses were mixed .. with^ a randomized^ pattern. The spectrometer looked at forward scattering at 4’ from the 30-cm-long deuterium target. Scattered electrons which entered the spectrometer aperture .~ (^) and fell within the momentum acceptance were detected in the two indepen- dent counters, a gas Cerenkov counter and a lead-glass calorimeter. Up to 1000 electrons per linac pulse were detected in the spectrometer. To analyze the high counting rate, signals were integrated and digitized for the Cerenkov counter and the lead-glass counter. Signals from each of the beam monitors were also digitized for each beam pulse. The data-acquisition computer stored the nor- .-malized^ signals (the^ digitized^ counters^ divided^ by the digitized^ beam charge) for each pulse, sorted by the beam polarization, one for el; and one for eR. After a period of running (typically 1 to 3 hours) the run was ended and summarized. Periodically the incoming linear polarization of the laser light was rotated by mechanically rotating the.axis of the calcite prism by 90’. This rotation had the effect of reversing the circular polarization and hence interchanging _- eL and--.--.:-^ eR. _However,^ the^ data-acquisition^ computer^ was not^ informed^ of these prisr?reversals, but continued summarizing the data referenced to the sign of
- (^) - .-.- by 7~every 3.237 GeV of energy. That is, at 19.42 GeV, the spin precessed 67r before reaching the target, and at 16.18 GeV, only 57r. The on-line asymmetries would be expected to reflect the change in spin orientation, Data were also taken at 17.80 GeV and 22.20 GeV. Figure 5 shows the measured asymmetries for the two counters, the Cerenkov and the lead-glass devices. The asymmetries clearly followed a g-2 curve which was expected if the _ asymmetries-..e4 were not dominated by false effects. The null points, one at 45’ in the prism-rotation curve, and one at 17.80 GeV in the g-2 scan of beam energy, .. were important^ as evidence^ that^ the false effects must^ have been small.^ A^ short
- run on hydrogen was also taken and showed evidence with parity violation in agreement with the deuterium results. .~ (^) Evidence (^) .for parity-violation in deep-inelastic scattering of polarized elec- trons (^) - was-~ announced at a colloquium at SLAC in June 1978 and a week later in Europe at Trieste. ” The Weinberg-Salam model agreed for a value of sin2 8w = 0.20 f 0.03. In the summer and fall of 1978, plans for further measurements were made. During this period, many talks were given at many places. I would like to tell one story which occurred at CalTech. Richard Feynman was in the audience and listened to the talk I gave on the careful tests and checks that were done. At the end he asked a typically astute question, “How do you know that the -detectors respond equally to eL and eR beams?” He sought an experimental test we had done to exclude that possibility. I explained the usual arguments, that soft-electromagnetic processes were responsible for light produced in the _ -- Cerenkov-and--.--.L lead-glass counters,^ and these processes were known^ to be helicity- indegndent.. We had not performed experimental checks because we did not
- ‘i; (^) ave t-he facility to do so. He was not satisfied. He preferred to see checks with experimental tests. Upon returning to SLAC I looked into the question of the spin in the detectors. The spectrometers deflected the scattered electrons an additonal 14’ bending at a lower energy E’. The spin at the detectors precessed even faster than the g-2 curve (see Figure 5, the dotted curve). I argued in a letter to Feynman that the dotted curve showed there was no evidence for his; -- conjectured systematic effect. In a subsequent conversation he told me he believed our results, even without that argument, but felt we should have made . tests to^ rule^ out^ experimentally^ that^ possible^ systematic^ error. .- (^) By the fall of 1978 we resumed our running of El22 to extend our data sample. The goals of the spring 1978 running were mostly met. Existence of -- 1 parity violation. in deepiinelastic scattering had been demonstrated. However, questions- -~ regarding the Weinberg-Salam model remained open, and we used the extend-edzfall running to pursue the answers. Let me explain. In a parity-violating process such as deep-inelastic scattering where the interaction is mediated by a vector boson (the Z’), there are two electron cou- plings, one for eL and one for eR. These couplings, gL and gR, are necessar- ily different for parity non-conservation to exist. The vector and axial-vector couplings gv and gA are defined to be the sum and difference of gL and gR; SV = (SR +gL)/% and gA = (SR -iU)/2. It t urns^ out^ that^ while^ deep-inelastic -scattering is sensitive to both coupling terms, the atomic-parity violation in bismuth is sensitive only to g;4. Could it be that both the SLAC and the Ox- ford/Seattle results were valid (at that time the dual results of Oxford/Seattle _- (^) had not been proved to be wrong)? Perhaps gA = 0 and gv # 0, thus agreeing &f. -; with the experiments (but not the Weinberg-Salam model). The purpose of the
- ‘essentially all the goals for E122; (ii) the experiment had occupied nearly six months of SLAC’s beam time and was-a heavy hit on other experiments trying to run; (iii) significant improvement over El22 would be hard, requiring new developments in the source and experimental apparatus; and (iv) SLAC was beginning to embark on the SLC program and it seemed best to participate in that project to understand better the physics of the Z”.
-..e4Thus^ ended^ the^ eight-year^ search at^ SLAC^ for^ parity^ violation^ in^ the^ elec- tromagnetic processes. .
_ -. (^) _: .-- *. -;
REFERENCES
1 T. D. Lee and Cn. N. Yang, Phys. Rev. 104, 254(1956), and C. S. Wu et al., Phys. Rev. 105, 1413(1957). 2 S. Rock et al., Phys. Rev. Lett. 24, 748(1970). 3 M. J. Alguard et al., Nucl. Inst. and Meth. 163, 29(1979). 4 ;S-: L. Adler, Phys. Rev. 143, 1144(1966). 5 Ya; B. Zel’dovich, JETP 33, 1531 (1957). 6 Ya. B. Zel’dovich, JETP 36, 964 (1959). 7 S. Weinberg, Phys. Rev. Letters 19, 1264 (1967). -~ I 8 G.t’Hooft, (^). Nucl. Phys. B35, 167 (1971). 9 M-J:~Alguard et al,, Phys. Rev. Lett. 37, 1261 (1976).
10 W. B. Atwood et al., Phys. Rev. D18, 2223 (1978).
11 F. J. Hasert et al., Phys. Lett. 46B, 138 (1978).
12 J. Blietschau et al., Nucl. Phys. B114, 189 (1976).
1.3 E. L.. Garwin et al., Helv. Phys. Acta 47, 393 (1974). 14 M. A, Bouchiat and C. C. Bouchiat, Phys. Lett. 48B, 111 (1974). 15 I. B. Kriplovich, JETP Lett. 20, 315 (1974). 16 L. L. Lewis et al., Phys. Rev. Lett. 39, 795 (1977), and P. E. G: Baird et al., Phys. Rev. Lett. 39, 798 (1977). _ -17 L..-M. Barkov and M. S. Zolotorev, Pis’ma Zh. Eksp. Teor. Fiz. 27, 379 ($78). --
