The Tevatron collider, the giant marvel accelerator built at the Fermi National Accelerator Laboratory in the eighties and operated there for over thirty years, until its demise in 2012, lost one of its fathers the other day, as Alvin Tollestrup passed away.
Tollestrup was maybe the key contributor to the project of putting together a 1-TeV particle accelerator, solving many issues for the construction of its superconducting magnets. But he also was for a long time the spokesperson of the CDF experiment, and the driving force behind the collective effort that led CDF to see a first evidence of the top quark in 1994, and then claim discovery for the long sought sixth quark in 1995.
In my book "Anomaly! Collider Physics and the Quest for New Phenomena at Fermilab" I devote a chapter to tell the story behind one of the first important measurements that the CDF collaboration pulled off, at a time when competition with the European laboratories at CERN and with the Stanford Linear Accelerator Center was fierce. I think a good way to remember Alvin is to paste below a few excerpts of that chapter. But before I do, I need to give some context.
CDF in 1987 was just starting to fiddle with its small first bounty of proton-antiproton collisions, seeing for the first time processes of energy higher than ever before. Among the signals that were popping into view were largish sets of W and Z boson decays, which had only seen in handfuls at the CERN SppS collider until then. But SLAC was also just turning on, and the electron-positron collider led by Burton Richter there promised to sweep the floor with very precise measurements of the Z boson -before CERN, in turn, would claim the business to itself in a few more months.
Alvin Tollestrup, then the spokesperson of CDF, pushed hard the collaboration to produce a measurement of the Z boson mass before it became irrelevant. By doing so, CDF proved to be a competitive experiment, capable of doing both searches at very high energy (the bread and butter physics of proton colliders) and electroweak parameter measurements of high precision.
So, thank you Alvin - I was little more than a student when we met in CDF, but it's been great to work with you.
Photo: Cindy Arnold, FNAL
Below are a few clips from my book, taken from chapter 3, titled "Revenge of the Slimeballs".
Why Don’t You Do It?
A few days later, the issue of Stanford's bad-mouthing of Fermilab became the topic of a lunch time discussion in the cafeteria of the Hirise. Tollestrup was sitting in the company of a dozen colleagues around the two big round tables they had set side by side. Steve Errede, an assistant professor from the University of Illinois who convened the Electroweak working group, was sitting across from Tollestrup. (In CDF, the progress of physics analyses was discussed within several independent working groups, led by pairs of conveners that remained in charge for two years; the Electroweak group managed analyses that targeted measurements of standard model properties in processes with electroweak bosons.) Everybody agreed that the SLAC scientists had gone too far with their denigrating statements. One of them was reported publically referring to as "slimeballs" the CDF experimenters, making fun of their efforts to extract meaningful physics results from a collider unworthy of its fame.
The CDF scientists knew that from an objective standpoint the alleged imprecision of hadron collider experiments was baloney. For instance, they had already measured quite accurately the mass of the J/psi meson using its decays into muon pairs. This showed that the tracking was capable of determining particle momenta excellently despite the "messy" hadronic environment and the large number of particles that proton-antiproton collisions generated in each interaction. And if SLAC scientists considered "precision physics" solely the measurement of the mass and properties of electroweak bosons, then again, that kind of precision physics was certainly also in CDF's cards. Using the couple of hundred Z decay candidates collected in the 1988-1989 run, a value much more precise than the one until then known appeared achievable, especially once Wicklund had showed how the electron energy could be calibrated to sub-percent accuracy.One of the graduate students at the table said "We should just prove that the SLAC guys are wrong!", but that looked like wishful thinking. The members of CDF appeared resigned to the fact that in a very short time Mark II, the detector of SLC, would sweep the table with a Z mass measurement that from then on only the LEP collider at CERN would be capable of challenging. The first Z boson had been detected by Mark II on April 11, so a first measurement of the Z mass was considered imminent. At some point Tollestrup, who was sitting on the other side of the big table, looked in Errede’s direction with his gorgeous, penetrating blue eyes, and said:
"So why don’t you do it?"
Errede did not answer. He was taken by surprise; "Is he talking to me?" he thought. He had a great respect for Tollestrup as a scientist. In particular, he was deeply impressed by the transformation that Tollestrup had wrought on the construction of the superconducting magnets for the Tevatron after assuming the leadership of the project. Errede was also a bit intimidated by the fixed stare of those deep blue eyes and that snow-white thatch: "He was one of those guys, you wouldn't like to be in their gun sight," he recalls. And the Illinois professor could not be sure that Tollestrup was really singling him out from among all the colleagues sitting around the table; others could have been the target of the enquiry. A couple of seconds went by; then Tollestrup repeated his question insistently:
"Why don't you do it? Why don't you do it?"
Steve finally emerged from his momentary hypnotic state. Alvin was really asking him to direct the Electroweak group’s forces toward a measurement of the Z mass. It had to be competitive with the one which Mark II was going to soon produce, and crucially it had to be published first. It was a fantastically tough challenge! But Steve knew he could count on quite a few strongly motivated colleagues to take it on. He quickly decided that there was only one possible answer, and he gave it.
"Yes, I think we can do it. Yeah, of course! Let's do it!"
Of course, being a fundamental parameter of the standard model, the Z boson mass did not need spite as a motivation to be measured. It was clearly in the interest of CDF to invest significant efforts to obtain a precise result. To Alvin and many others, the fact that Mark II was close to publishing their first Z mass measurement, whose foreseen precision CDF could not hope to attain, made the matter an urgent one. CDF had to produce a precise measurement before Mark II could publish its first result. Only thus would the work receive precious citations in all future physics papers discussing the Z boson. Otherwise the CDF members would be beaten by a nose, turning the small but precious bounty of Z bosons captured in Run 0 into a largely irrelevant dataset. That same afternoon Steve assembled a group of willing and able collaborators, and identified and distributed the different required tasks to the crew. Then special weekly meetings were scheduled to closely monitor the progress of the activities. The ongoing activities were better not discussed at their otherwise natural place, the Electroweak group meeting: it was considered unsound to publicize too much the Z mass measurement effort within the collaboration. In particular, a few of the collaborators from the Berkeley group also participated in the Mark II experiment. They needed to be kept in the dark about the ongoing effort, as the possibility of leaks to the competitors had to be minimized.
The weekly cadence was soon sped up to two meetings per week, and by the beginning of July the participants ended up meeting multiple times every day, as it was realized that the analysis effort would otherwise not converge to a final result in time. The meetings were not a waste of time: there were many different analysis activities progressing steadily in parallel, and it was crucial that they could provide feedback to one another. First of all, there were two separate measurements: one based on the sample of 132 Z->mu mu decay candidates isolated in data collected by a trigger selecting events with high-momentum muons; and a second one using the 73 Z->ee candidates coming from data triggered by electrons of high energy. Those two measurements used the same calibration procedures for particle momenta and modeling of the physics of Z production and decay, and specific inputs related to the peculiarities of the measurement of electrons and muons. A multitude of subtle physics effects playing a role in the measurement required a detailed study and the full dedication of one or two scientists each. During those weeks, Steve felt like he was providing more psychological support than physics advice. He kept running up and down the trailer corridors from one office to the other to pass on information and provide specific input to his crew members. And he often found himself encouraging them and alleviating their discomfort for the previous commitments that they had been forced to set aside.
One PRL a Day
Toward mid-July, the CDF members finally realized they had in their hands a very competitive result. The Z mass was measured to be 90.9 GeV. The statistical uncertainty of the combined fit to di-electron and di-muon Z decay candidates produced by Keutelian was coming out in the ballpark of 300 MeV, and the dominant source of systematic uncertainty was the energy scale, amounting to 200 MeV. Overall, the CDF data allowed a fourfold reduction in the Z mass uncertainty over the current world average! It was time to wrap it up.
The article was written on the night of July 18, just hours before a scheduled collaboration meeting, a two-day event organized every three or four months where CDF members could listen to a detailed overview of the status of the experiment and the analyses that were being carried out. The publication was going to be a "Letter" -a short article to be sent to the prestigious Physical Review Letters (or "PRL" as everybody called it). As short as it needed to be in order to fit within the strict standards of PRL, writing a paper overnight was a challenging task for its editors. It was 6 AM on July 19 when the article was finally finished. Exhausted but happy, Steve went for breakfast, then drove to the Hirise. There the measurement could finally be presented to the collaboration. Sleep-deprived and physically drained, Steve had the task to convince his colleagues, who for the large majority had not had a chance to follow the progress of the measurement, that the result was solid enough to warrant an approval to be published without further scrutiny. The request of an approval without review was an unusual procedure. Normally, an internal review process was required before a publication could be sent out to a scientific journal. Fortunately, the response of the collaborators was very positive: this was indeed a careful, precise measurement deserving to be promptly sent to PRL.
As with any scientific publication, the title and abstract of the article were as important as the article itself and caused a lengthy discussion. On the title Errede managed to have it his way: "Measurement of the mass and width of the Z boson at the Fermilab TEVATRON." The capitalized name of the hadron collider was one of the points on which Errede stood his ground. The other was the explicit mention of the fact that CDF had measured the mass as well as the natural width of the resonance. The latter was something which electron-positron guys expected to be exclusively their own business, thanks to the already mentioned energy scan of the production rate.
The natural width of a particle can be determined by the distribution of observed mass values around the most frequent value. The width is inversely proportional to the particle's lifetime and is directly connected to the intensity of the interactions responsible for the particle decay. The stronger the interactions, the larger is the range of mass values that the particle may take, as the decay proceeds more quickly, giving no time to the particle to "settle" to its nominal mass. Arguably, the natural width of the Z boson is no less important than is the Z mass itself. Compared with model predictions, the measured width allows one to determine whether there may be decay modes of the particle that are not seen experimentally. A larger-than-predicted width implies the existence of unknown particles into which the Z can disintegrate, as mentioned in Chapter 1.
After the meeting Errede, still in a state of sleep deprivation from the past weeks and the frantic final night of work, met Alvin Tollestrup and Barry Wicklund. Together, they took the stairs to the second floor of the Hirise and went straight to John Peoples' office: they needed something from the Fermilab director. As instructed by Alvin, that morning Errede had sent to Peoples a copy of the article, along with a special request. The laboratory director was asked to undersign the paper before it got sent to PRL. The journal would ordinarily take several weeks to review an article. Suitable reviewers had to be found and be given sufficient time to comment on the results, require clarifications, propose modifications, or reject the article. The exchange between authors and PRL reviewers could easily push the publication date to September or October. Furthermore, the journal editors might choose a member of Mark II as reviewer, bringing in a conflict of interest. A SLAC reviewer might fall in the temptation of purposely slowing down the publication of the CDF paper. However, an article signed by the director of a major American physics laboratory allowed the peer review process to be waived. The article would go to press in the very next issue of the magazine.
The special no-review procedure had been instituted in 1976 by Physical Review Letters in what was called "An Editorial Experiment." The April 26 issue of the magazine featured a letter bearing that title, where the editors explained that the time to publication of even carefully crafted articles could be long, so they had cooked up a special procedure to bypass the regular review process:
"To receive this special treatment, the Letter must be forwarded for publication by a division leader, department chairman, or director at the author's institution [...] This person, who may not be a co-author, must explain why special handling is appropriate. [...]. The editors reserve the right to evaluate the "extraordinary circumstances" and to refuse the request. [...] The published Letter will bear, immediately above the abstract, a note reading, "Published without review at the request of (person forwarding the Letter) under policy announced 26 April 1976.""
Peoples agreed to sign the paper and write a cover letter to PRL: it was certainly in the interest of the laboratory that CDF proved its worth in the competition with the Stanford experiment. Those were years when SLAC and Fermilab were competing on the amount of funding that they received from the Department of Energy. Richter had the reputation of being the one who won those Washington battles. The perception was that he was more persuasive and more charismatic than Peoples; the Nobel Prize he had won in 1976 also played a role. A joke in fact circulated among American experimental physicists on the matter: Richter and Peoples would both go to Washington to get money for their labs, and Peoples would be the one getting in the revolving doors at the entrance first, but the one who would get out of the doors first would always be Richter! However, the truth was different: Fermilab had actually a larger budget than SLAC at the time, but was starting to suffer from the ramping up of the funding to the Superconducting Supercollider (SSC). The Texas machine was a hadron collider, hence it fished in the same funding pond of the Tevatron. Peoples did more than undersign the paper and the cover letter. That same day, he called the editor-in-chief of PRL David Lazarus, forewarning the arrival of the Z mass paper and asking whether the editor agreed to publish it in the next issue of the magazine. The editor denied the request, on the grounds that it would be difficult to organize the special handling of the article in such a short timescale. As he hung up, Peoples was left with an unpleasant suspicion. After discussing the matter further with his predecessor Lederman and with Tollestrup, Peoples called the editor again. Finally, the situation was clarified: Lazarus explained that SLAC had already preannounced the submission of a Mark II paper on the Z mass, so he had set up an internal review panel to handle it. It was only through the stubborn insistence of Peoples that a similar panel was formed for the CDF paper.