Skip to main content

Full text of "NASA Technical Reports Server (NTRS) 20080040755: The Buffer-Gas Positron Accumulator and Resonances in Positron-Molecule Interactions"

See other formats

Symposium on Atomic & Molecular Physics 


C. M. Surko 

Department of Physics, University of California, San Diego, La Jolla CA 92093 


This is a personal account of the development of our buffer-gas positron trap and 
the new generation of cold beams that these traps enabled. Dick Drachman provided 
much appreciated advice to us from the time we started the project. The physics 
underlying trap operation is related to resonances (or apparent resonances) in positron- 
molecule interactions. Amusingly, experiments enabled by the trap allowed us to 
understand these processes. The positron-resonance “box score” to date is one 
resounding “yes,” namely vibrational Feshbach resonances in positron annihilation on 
hydrocarbons; a “probably” for positron-impact electronic excitation of CO and N 2 ; and a 
“maybe” for vibrational excitation of selected molecules. Two of these processes 
enabled the efficient operation of the trap, and one almost killed it in infancy. We 
conclude with a brief overview of further applications of the trapping technology 
discussed here, such as “massive” positron storage and beams with meV energy 


Since this paper is written in conjunction with the symposium marking the 
retirement of Dick Drachman and Aaron Temkin, I depart from third-person style to 
relate a personal view of the development of the buffer-gas trap and the physics that has 
come from it. I came to know Dick near the beginning of the trap project and because of 
it. I’d like to tell a bit here about the people involved in the development of the trap, 
including Dick, who helped us achieve important perspective and understanding of key 
physics issues that have arisen in the past two decades. 

The story began in a lunchtime conversation with Marv Leventhal at AT&T Bell 
Laboratories in Murray Hill, New Jersey in 1983. Marv asked what one could do with 
positrons in a tokamak plasma. I responded by saying that there were interesting 
problems for sure, such as study of the turbulent transport of electrons out of the hot 
plasma. However, I didn’t know how to get the positrons into the plasma in the first 
place (i.e., across the strong magnetic field). Marv said, “oh, I know how to get them in - 
that’s easy — just convert them to positronium atoms; shoot them in; they’ll ionize and 
you’re in business. ” That started the whole thing off [1, 2], Marv had done positron 
trapping earlier with Ben Brown, and so he was eager to pursue this kind of research. In 
a very nice experiment, they measured the Doppler linewidth of molecular hydrogen [3], 
to compare with Marv’s balloon measurement of this line coming from the galactic center 
[4], Regarding the proposed tokamak application, we did carry it some way forward, but 


Symposium on Atomic & Molecular Physics 

never actually did the experiment. However, this start enabled us to do many other things 
that turned out to be quite interesting and have enabled others to do more. 

The trap was developed to accumulate positrons and store them efficiently. While 
the original goal was to provide an intense, pulsed positron source to study turbulence in 
tokamak fusion plasmas, we quickly realized that it might well have many other 
applications. By now, the trap has contributed to a wide range of scientific problems, 
including many aspects of positron interactions with atoms and molecules [5]; study of 
electron-positron plasmas [6]; commercial-prototype trap-based beams for materials 
characterization [7]; creation in the laboratory of the first low-energy antihydrogen [8]; 
and very recent new work to study the positronium (Ps) molecule, Ps 2 [9]. 


The way to make a nearly ideal “antimatter bottle” was known by the mid ‘80s, 
albeit not fully realized as such. It was called a Penning-Malmberg trap, adding John 
Malmberg’s name to Penning’s [10]. It was Malmberg and colleagues at the University 
of California, San Diego (UCSD) who developed the Penning trap to efficiently confine 
large quantities of single component plasma. Thus it was no accident that I later went to 
UCSD - in large part due to my interests in trapping positrons. 

Malmberg and colleagues worked with electrons, but switching to positrons poses 
no problems from a plasma physics point of view. This nearly ideal “bottle” consists of a 
uniform magnetic field to confine the particles radially, with electrostatic potentials at the 
ends to confine their motion along the field. Confinement of plasmas of a single sign of 
charge in these devices is excellent. A theorem due to Tom O’Neil explains why [1 1]: 
the angular momentum of the plasma is dominated by the electromagnetic term, which is 
proportional to 2 rj 2 , where rj is the radial position of plasma particle j. If you build a 
cylindrically symmetric device (i.e., about the magnetic field axis), there are no torques, 
and hence the plasma can’t expand radially. 

The trick, however, is getting the positrons into the trap efficiently without 
unnecessary losses, since positrons are expensive from the scientific point of view. Marv 
and I thought we might be able to do it using inelastic collisions. Early on, Marv had the 
idea to use vibrational excitation of molecules, since we wanted to avoid positronium 
formation that would be present if we used electronic excitation, for example. Good 
candidates seemed to be H 2 and N 2 , since they had relatively large vibrational energies 
(0.5 and 0.3 eV, respectively). This is where the connection to Dick Drachman began. I 

had no idea who “Drachman” was, but Marv would say, “ Drachman says, ” or “we 

should ask Drachman about that." This happened so many times that, when Marv said 
something to this effect standing in the hall outside his office, I made a note to find out 
who this guy was - he must really be somebody important! 

While I had done a bit of atomic physics some 15 years earlier, the trap project 
provided me with the opportunity to become immersed in the field. I came to know Dick 
well and consulted him frequently. He provided that calm and reasoned voice to lead us 


Symposium on Atomic & Molecular Physics 

through many theoretical minefields (vats of snake oil too). He took the time to patiently 
teach me about theoretical atomic physics and to translate (in language even I could 
understand) various esoteric papers filled with surprising results (alas, not infrequently 
wrong). Dick was an enormous help in keeping us on track. He helped make the venture 
into positron-atomic physics both very productive and most enjoyable. Our first 
conversations related to vibrational excitation. Bailie and Darewych had done a very nice 
calculation of vibrational excitation of H 2 by positron impact [12], and we wanted to 
understand what that would mean for us and how we might extrapolate to other targets. 

Back at the trap project, we designed what was later to be called a three-stage, 
buffer-gas, positron accumulator, the principle of which is shown schematically in Fig. 1 
[1]. The pressure in stage I is adjusted so that a positron would lose enough energy in 
one transit through the device to be trapped, then subsequent collisions would further 
lower the positron energy, so that it would end up trapped and cool in the lowest pressure 
region in stage III. We realized pretty quickly that, while stage II and III could operate 
on vibrational excitation, stage I required both a large cross section and a larger energy 
loss than vibrations could provide. So we planned that the first stage would operate on 
electronic excitation, even though this would carry with it some loss due to positronium 

Z (cm) 

Fig. 1. Schematic diagram of the three stage buffer-gas accumulator circa 1988. Top: 
electrode structure. Middle and bottom: pressure and electrical potential profiles. There 
is a uniform magnetic field ~ 0.9 kG in the z direction. Stages are labeled I, II and III. 
The trapping mechanism is inelastic collisions with the N 2 buffer gas, labeled A, B and 
B'. In 1989, A was electronic excitation, and B and B' were vibrational excitation of N 2 


By 1985 we had made progress in planning for the trap and the positrons-in- 
tokamak experiment. It so happened that at that time, Allen Mills and Karl Canter were 
arranging a 60 th birthday celebration at Brandeis for Steve Berko, a giant in the field of 


ilar Physics 

Fig. 2. (a) Photo of the attendees at Steve Berko’s 60th birthday symposium at Brandeis in December 1984. 
It’s virtually a who’s who in the U. S. world of positron physics including Dick (upper left). Somehow I 
didn’t hear the announcement about the picture. It is one if the few group photos at meetings that I’m sorry 
to have missed. Reprinted from Ref. [1] with permission from World Scientific Publishing Co. Pte. Ltd, 


Symposium on Atomic & Molecular Physics 

1 . 

J J.Mader 


A, H, Weiss 

21. A. T. Stewart 

31. W.S. Crane 


Takashi Odagaki 


R M. Niemmen 


M, A- Shulman 

22. Martin Deutsch 

32, R. M, Langer 


Y, J. He 


L. O. Roellig 


Mohammad Haghgooie 

23. Stephan Berko 

33. J. H. Terrell 


Eric GuUikson 


P. M. Platzman 


G. Krithivas 

24. A. P. Mills, Jr. 

34. P.H. Carr 


W. S. Farmer 


P. J, Schultz 


R. N. West 

25. P. H. Lippel 

35. K. F. Canter 


Leigh Sneddon 


K. G. Ly nn 


W. H. E. Roeckncr 

26. P. £. Mij ns rends 

36. Steven Chu 


G. M. Beardsley 


Marc Weber 


D. C Schoepf 

27. S. C. Sharma 

37. Christoph Hohenemser 


T. N. Horsky 


D. M- Chen 


B. L. Brown 

28. R. J. Drachman 

38. W F. Huang 


Frank Sinclair 


i T Wejngart 


R. R. Lee 

29. Arun Sareil 

39 R. B. Meyer 


J. F. C Wardle 


Asko Vehanen 


Arthur Rich 

30, H. J. Schiutzer 

40, Stephen Cushncr 


Raoul Tawel 

5 1 . Piero Sferlazzo 

Fig. 2. (b) Roster of names for the Berko symposium picture, December 1984. Reprinted from Ref. [1] 
with permission from World Scientific Publishing Co. Pte. Ltd, Singapore. 


Symposium on Atomic & Molecular Physics 

positron studies of solid-state systems, such as the Fermi surfaces in metals. Allen and 
Karl were planning the proceedings and Allen asked for a paper on our positron-trap 
project [1], This first paper on the trap was also the first of only two papers that I’ve 
coauthored with Allen. Looking back, the paper was anything but modest in its promises: 
we “discussed the possibility” of accumulating positron plasmas containing 10 10 particles 
and holding them for minutes. The latter turned out to be not too hard, given a few years, 
but the former has yet to be achieved some 20 years later (but we do hope to do it soon). 
As shown in Fig. 2, a host of big names attended the Berko symposium, including 
founders of the field, such as the discoverer of positronium, Martin Deutsch, and the 
future Nobel Prize winner, Steve Chu. It was really a great time for the positroners! 
Berko, whom I did not know very well before the meeting, was immensely gracious in 
accommodating the hoopla made over him. Best of all, I had an opportunity to meet 
Drachman, although I didn’t realize at the time his strong connections to Brandeis. 

Shortly after the conference, Fred Wysocki came from the Princeton Plasma Lab to 
Bell Labs as a post doc to spearhead the construction of the trap, and we were off to the 
“positron accumulator races.” Figure 3 shows Marv, Fred, and A1 Passner admiring the 
electrode structure of the first trap (gold plated no less). Al, who was an Associate 
Member of the Technical Staff at Bell Labs, had worked with me for a number of years 
before we began the trap project. He made a great contribution to its early success, 
similar to his contributions to many of our other experiments. 

Fig. 3. Gold-plated copper electrode structure of the buffer-gas positron accumulator as 
assembled in 1986. From left to right, Al Passner, Fred Wysocki, and Marv Leventhal. 
The electrodes sit on laser tables where equipment for Allen Mills’ Ps spectroscopy 
experiment [14] was to be set up. 


Symposium on Atomic & Molecular Physics 


Fred Wysocki designed a very nice device, and so it all looked great. However, 
when we turned the trap on, the results were terrible. We didn’t get anything to speak of 
with vibrational excitation, but we could trap positrons when we increased the positron 
energy in the first stage to where we guessed we were exciting the molecules 
electronically. We tried a large number of molecules, and N 2 seemed to work best for 
reasons then unknown. Actually, we had wanted to avoid electronic excitation, since for 
almost all targets, the Ps formation channel is open there too, and the latter is a potent 
loss process. We had expected 10 6 positrons trapped and minute positron lifetimes, but 
we got about a factor of 100 less, as shown in Fig. 4. So we had not only an excitation 
problem, but also a lifetime problem as well. I was nevertheless pretty excited, somehow 
feeling that, if we weren’t making a mistake, there was likely new physics there 
somewhere. Lots and lots of conversations with Dick followed these early results 
concerning the possible atomic physics processes involved in the trap operation. 

Filling Tine (sec.) 

Positron Counts vs. Storage Time — 8/31/87 

Fig. 4. Data from a trapping experiment in August 1987. Above: number of positrons 
confined as a function of fill time, maximum number ~ 1 x 10 4 . Below: confinement as 
a function of time after the fill was turned off, indicating a 0.7 s confinement time. The 
confinement was limited by annihilation on a very low density of large, molecular 
impurities at pressures < 10' 9 torr. 


Symposium on Atomic & Molecular Physics 

Now there was another story unfolding here too. It was often said that, in those 
days, Bell Labs wasn’t dollar limited, but was space limited. (Typically the opposite is 
true in universities.) When we started the trap project, we needed someplace to put it. I 
had a lab, but it was full of fluid convection apparatus, so no room there. Marv had only 
a small space, since he did mostly gamma-ray astronomy with groups elsewhere, so we 
had a problem. Fortunately, Allen (as in A. P. Mills, Jr. - the Allen, Allen) just happened 
to have a big, new-to-him lab in the basement that he hadn’t moved into yet. So, we say 
“Allen” (imagine pleasant music in our voices), “errr, could we borrow your lab for just 
18 months; we want to set up a trap and then move it to Princeton to study turbulence in 
tokamak plasmas?” “ Well OK,” (says the cooperative Dr. M., Jr.) “ but remember - just 
18 months /” Well, that was in ’85, and as I recall, by ’87, Allen was unhappy, since (see 
Fig. 5) there were tons of magnet, etc. sprawled all over his space and no signs of our 
departure in sight. So Allen began building his new Ps spectroscopy experiment [14] 
intertwined with ours like spaghetti in a bowl. Then when we first began trapping 
positrons, Allen was really an unhappy camper - well, not really, but he saw the problem 
coming even before we did. Allen was such a good scientist and believed so much in 
everybody’s science that he couldn’t throw us out if we were making progress. If our 
trap didn’t work, we were out on the street (600 Mountain Avenue, Murray Hill NJ, to be 
precise). But it was working, and “so keep your elbows in,” we’re going to be very cozy 
for a while with two “full-bodied” experiments in one underground lab. Great guy, 
Allen; maybe not overjoyed at that moment, but he really did us a huge favor. 

Fig. 5. A1 Passner and me standing in front of the first buffer-gas trap in Allen Mills’ lab 
circa 1987. The source is on the right behind the gas bottle; and Allen’s Ps spectroscopy 
experiment (Helmholtz coils and rails), which was under construction, is “creeping in” at 
the lower left. 


Symposium on Atomic & Molecular Physics 

Back at the trap, we twiddled knobs and got the plasma lifetime to improve a bit at 
a time somehow. The trap had a continuous feed of tungsten-moderated positrons from a 
22 Na source, so longer lifetimes meant more particles trapped. But we really didn’t know 
what was going on. The seminal event occurred one afternoon. I came into the lab, and 
A1 Passner was excited. “Hey Cliff, look at this - 1 think it’s impurities. ” He had a small 
dewar of liquid nitrogen in one hand, and he was looking at the trapped particle signal on 
an oscilloscope as he poured nitrogen on the trap vacuum chamber. Liquid N2 on - 
larger signal; take the LN2 away - the signal decreased. This was a major turn of events. 
Sure enough, in the months that followed, we found that the better we made the vacuum, 
the better the trap worked. 

The problem, it turned out, was large hydrocarbon molecules at the < 10' 9 torr level, 
for reasons no one knew at the time. Basically they had absolutely huge annihilation 
rates. It was good that Wysocki insisted on building such a good vacuum system in the 
first place, or the experiment would have never worked. So on the hardware side, we 
installed an in situ liquid nitrogen dewar in the vacuum system next to the final trapping 
stage to pump impurities, and we began switching out the turbopumps on the system (i.e., 
which required conventional oil mechanical pumps behind them) in favor of cryopumps 
that didn’t require any backing. 

On the physics side, the first major paper was a 1988 Phys. Rev. Letter (PRL) on 
(no surprise) annihilation on large molecules [15]. As shown in Fig. 6, we found that 
annihilation rates increased exponentially with molecular size. Plotted here is the 
conventional normalized annihilation rate, Z e ff, which is the measured annihilation rate, 
r, relative to that expected for positrons in a free electron gas of the same density, 
namely [16], 

Z cff = T /nr 0 2 cn m , (1) 

In Eq. 1, r G is the classical electron radius, c is the speed of light, and n m is the molecular 
number density. We had the intuition to point out in the PRL that the large Z e ff values 
were likely due to vibrational resonances and invoked the “RRKM” formalism from the 
chemistry literature to explain it. We were familiar with Heyland’s work on small 
hydrocarbons (see Fig. 6) [17], but we were unaware of the seminal work of Paul and St. 
Pierre who studied annihilation in dense gases for hydrocarbons as large as butane 
(C4H10) [18], We agreed quantitatively with the previous measurements and extended 
them to molecules as large as C16H34 finding annihilation rates orders of magnitude 
larger. More significant really was the ability to study positron interactions with 
molecules in a vacuum environment. There was now no question that the large rates 
were due to a two body effect, and we could also make independent measurements of the 
positron temperature [confirmed to be the electrode temperature of 300 K (i.e., 25 meV)]. 

We were on the road now. The next year we published another PRL announcing 
the first positron plasma in the laboratory [13]. It contained a modest 3 x 10 5 e + , with a 
Debye screening length, Xd, which was a similarly modest 1/4 the plasma radius (i.e., a 
good measure of the plasma regime is the degree to which Xd « r p ). Later, John 


Symposium on Atomic & Molecular Physics 

Malmberg and Hans Dehmelt told me that they were referees for the paper, exceedingly 
pleasing, because both were giants in the trap field. We realized that the annihilation on 
molecules might result in a spectrum of ions. At Allen Mills’ suggestion, A1 Passner 
looked and sure enough, there was a story to tell there too. We found that annihilation 
with a 300 K thermal distribution of positrons left a broad spectrum of ions [19]. I’m 
sorry to say that this is the second of only two papers I coauthored with Allen, during all 
our time at Bell Labs. 


12 3 4 5 7 9 12 16 


Fig. 6. Normalized annihilation rate Z e ff/Z for alkanes (C n H 2 n+ 2 ) as a function of the 
number of molecular electrons, Z: Open circles are from an experiment using 
atmospheric pressure test gas [17], and (•) are data taken with a low-pressure of the test 
gas in the positron trap. From Ref. [15], 

Les Hulett and collaborators at Oak Ridge followed up on this positron-induced 
ionization effect and produced a series of interesting papers [20, 21], Initially, there was 
a disagreement about the significance of our experiment, which resulted (amicably) in a 
joint publication [22]. Oakley Crawford from the Oak Ridge group wrote a nice 
theoretical paper that explained the basic phenomenon [23], saying that the incoming 
positrons annihilate with equal probability on any valence electron, not just the highest- 
lying molecular orbitals. Later we confirmed this prediction with Doppler broadening 
measurements [24, 25]; it was only then that I realized the full implications of Crawford’s 

At that point (Fall, ’88), I moved from Allen’s lab to UCSD in La Jolla, much to 
his relief. Sadly the collaboration with Marv and interactions with Allen tailed off as the 
miles separated us and other interests and obligations got in the way. In La Jolla, we 
continued the studies of annihilation in molecules and continue them still, many 
generations of experiments later. 


Symposium on Atomic & Molecular Physics 

Dick’s 1989 symposium at Goddard was, from my point of view, something of a 
coming-of-age party for the trap. We presented a paper on our annihilation results that 
was well received [26]. I was quite gratified that the trap-based results were embraced by 
the positron community as “mainstream,” something that is frequently not easy coming 
into a new field with a different technique. The meeting was a great opportunity to meet 
people who would later play key roles in the science that the trap was to enable. 


More superb Princeton plasma talent came with Tom Murphy, who joined us at Bell 
Labs in time to disassemble the trap at Bell and move it to La Jolla. Even while a grad 
student at Princeton, Tom wrote a great paper on using a Ps beam to study transport in 
tokamaks [27]. Immediately upon our arrival in La Jolla, Gene Jerzewski joined the 
effort. His technical expertise and ability to teach students and post docs about hardware 
were invaluable to our efforts in the years to follow - wonderful contributions on a level 
with A1 Passner’s contributions at Bell. 

In La Jolla, Tom Murphy focused on understanding the problem, namely 
annihilation in large molecules [28], He also made great strides in understanding how the 
three-stage buffer-gas trap actually worked [29]. A key discovery was that the optimum 
trapping potential difference between stages was ~ 9 - 10 eV per stage for each of the 
three stages. As described below, this is the energy window in N 2 where the electronic 
excitation cross section is larger than that for Ps formation due to a resonance in the 
excitation channel not yet understood. The result is efficient trapping when all stages of 
the trap are tuned to operate in this regime. 

On the annihilation front, Tom cleared up a long-standing ambiguity in the data for 
annihilation in xenon for a thermal distribution of positrons at 300 K, confirming Z e fr« 
400 [30], This value would be doubted by theorists for another decade, but is now the 
accepted number [5]. Tom also made new systematic studies of Z e fr for a range of 
compounds [28], Along the way, there were innumerable conversations with Dick 
Drachman about low-energy positron interactions with atomic and molecular targets in 
our attempt to get some degree of theoretical understanding to match the results provided 
by our new experimental capabilities. He was kind enough to look over any paper that I 
sent him, including providing very useful comments and suggestions on our papers when 
we sent them to him in the draft stage. 

In the early 90’s, Shengzhang Tang joined us from Ken Roellig’s shop, and Rod 
Greaves came from the plasma community. Koji Iwata did his thesis on annihilation ( “he 
did all this work for a thesis ?” one member of his doctoral committee asked!), and Mark 
Tinkle did his thesis on mode diagnostics of positron plasmas. Shenzhang took the last 
steps that we would take toward a Ps beam fusion diagnostic showing that one could 
make a pretty good, variable-energy Ps beam by charge exchange on H 2 [31]. We got a 
very important new tool from Shengzhang, namely our first Doppler broadening 
measurements [32], an example of which is shown in Fig. 7 [33]. Later this technique 
allowed us to determine the site of positron annihilation in large molecules [24, 25]; all 

Symposium on Atomic & Molecular Physics 

valence electrons seem to do the trick, as per Oakley Crawford’s prediction a few years 
earlier. During that period, we also made quantitative studies of inner shell annihilation 
[34] and studied annihilation in polycyclic aromatic (PAH) molecules, which are 
important constituents of the interstellar medium [35, 36]. 


Fig. 7. (open circles) Doppler-broadened gamma-ray spectrum from positron annihilation 
on helium atoms and comparison with ( — ) theoretical predictions using a variational 
wavefunction, and ( — ) a Gaussian fit. From Ref. [33]. 

Fig. 8. Progress in trapping positrons with buffer-gas accumulators using 22 Na sources 
with strengths (~ 100 mCi), beginning with the first trap at Bell Labs. All except the last 
point (ATHENA collaboration [39]) are from the traps at Bell and UCSD. 


Symposium on Atomic & Molecular Physics 

Around ’95, Rod Greaves pushed us to develop further the neon moderator for use 
with the trap [37]. Invented much earlier by Allen Mills [38], neon was not used as a 
moderator much, because the energy spread was considerably larger than that of tungsten 
(i.e., ~1.5 eV, FWHM, as compared to - 0.5 eV for tungsten), even though neon is more 
efficient by an order of magnitude. Rod realized that this was not a disadvantage for the 
trap. The buffer-gas trap has an energy acceptance window — 2 eV; and once trapped, the 
positrons cool to 25 meV, which is far superior to the energy spread of the conventional 
tungsten-moderated beam. Rod designed a compact system, which is now standard fare 
for traps and other applications. 

In ’97, we designed and built a second-generation three-stage positron trap that was 
somewhat more compact. By then we had improved the overall efficiency of the system 
by more than a factor of 10 4 as compared with the first results shown in Fig. 4. Figure 8 
shows this progress and includes later results from the ATHENA antihydrogen 
collaboration at CERN. As discussed below, they used a buffer gas accumulator to 
accumulate positrons, then stacked them in a high-magnetic-field Penning-Malmberg 


In ’95, Chris Kurz came from MIT to join our group as a post doc, and Steven 
Gilbert joined to do a thesis and then a short post doc. With Chris, we made the first 
annihilation measurements as a function of positron temperature, heating them in situ in 
the trap with rf radiation [40]. This was a precursor to Rod Greaves, Chris and Steven 
realizing that we could make a cold beam by simply dumping the trap slowly [41]. The 
results, shown in Fig. 9, were spectacular and turned out to be a big advance for us. The 

Fig. 9. Retarding potential curve of the cold trap-based positron beam indicating a 
parallel energy spread of 18 meV, FWHM at 1.7 eV [41]. The inset illustrates the beam- 
formation technique, whereby the potential of stage III of the positron trap is raised 
slowly to force the positrons out of the well. 

Symposium on Atomic & Molecular Physics 

beam was tunable from ~ 100 meV upwards and had a parallel energy spread of 18 meV. 
It was superior to conventional beams used for atomic physics studies by more than a 
factor of 10 in energy resolution. 

Steven and Rod then decided to try a scattering experiment - vibrational excitation 
of CF4. It was supposed to be a preliminary experiment, but turned out so well we sent 
the results to PRL [42]. What we had done in just two years was to advance the state of 
the art of positron beams for positron-atomic physics by more than an order of magnitude 
in energy resolution. We had also developed a new scattering technique that advanced 
the state of the art, even in electron scattering, particularly for measuring integral 
inelastic scattering cross sections. 

In ’98, Rod Greaves moved on to First Point Scientific, Inc., in Agoura Hills CA. 
There he has developed commercial positron traps and compact neon moderator systems 
that have aided greatly in allowing people around the world to exploit positron trapping 
technology [7, 43]. On the science side, he continues to collaborate with our group, and 
as discussed below, more recently with Allen Mills, who went from Bell Labs to the 
University of California at Riverside in the late ‘90s. 

Positron cooling is a crucial issue for both plasma and scattering experiments. Rod 
Greaves had measured gas-cooling times for positrons and found that CF 4 and SF 6 were 
the best for rapid cooling [44], This was confirmed by Gilbert’s direct measurement of 
the vibrational cross section for CF 4 . Giving the nod to its small annihilation cross 
section, we subsequently used small amounts of CF 4 in stage III of the trap for rapid 
cooling. This allowed us to cycle our pulsed positron beam rapidly for scattering and 
annihilation experiments. 

Around that time, the results flowed in. Steven Gilbert made the first energy- 
resolved measurements of positron annihilation on molecules, discovering huge Feshbach 
resonances and measuring the first positron-molecule binding energies [45], Joan Marler 
did her thesis work under the tutelage of post doc James Sullivan who joined us from the 
Australian National University in Canberra, thoughtfully bringing his advisor Steve 
Buckman along for the second year. Steve Buckman and James gave our program an 
enormous boost, pushing us to measure every conceivable cross section and hunt for 
every resonance imaginable (at least that’s what it seemed like to me) [46-50]. One of 
their “gifts” was the technique of unfolding electronic excitation cross sections in 
molecules using the known Franck Condon factors [47] - good physics and great fun. 

Turning back to key features of the operation of the positron trap, there are three 

• Why is N 2 the best trapping gas? 

• Why did CF 4 work so well for cooling positrons? 

• What’s the story with the large annihilation rates in molecules? 


Symposium on Atomic & Molecular Physics 

Energy (eV) 

Fig. 10. Positron-impact cross section for the excitation of the a'll state of N 2 [47, 51]. 
Shown for comparison (open circles) are electron data for the same cross section [52, 53], 
Also shown by the lines are calculations by Marco Lima et a!., for the positron- impact 
process, using three different basis sets for the target. See Ref. [51] for details. The 
sharp rise at onset and relatively small near-threshold Ps formation cross section make N 2 
the molecule of choice for buffer-gas trapping. 

The cold beam told it all, more or less - at least all experimentally that is - the 
theory is about one and a half for three at this point. Shown in Fig. 10 is the positron- 
impact electronic excitation cross section for N 2 [47], The unexpected, sharp rise at 
threshold (some kind of resonance?) opens up faster than the positronium formation cross 
section, which is a loss process in the positron trap. This resonance is a very efficient 
way for positrons to lose energy and drop into successive stages in the trap. In CO, there 
is a similar sharp resonance, but the Ps formation cross section is even larger, so CO is 
not as good as N 2 for trapping [51]. The case is closed for the experimentalists - the 
theorists still find this N 2 resonance difficult to explain , 1 as shown by the theory curves in 
Fig. 10. 

The CF 4 vibrational excitation story has a similar ring to it. Gilbert’s early 
measurement of the CF 4 cross section turned out to be inaccurate due to all the 
machinations we had to go through to measure it, so Joan Marler repeated the 
measurement, as shown in Fig. 1 1 [54]. We found a sharp rise at onset there too, and the 
largest vibrational cross section measured to date. Joan went on to make the same 
measurement for electron impact - the first in situ comparison of state-resolved electron 
and positron inelastic cross sections [54]. The fact that the cross sections were the same, 
both in magnitude and shape, provided the needed clue (kindly delivered to us by Gleb 
Gribakin) to compare the measurements with the predictions of the Bom dipole model. 

Indeed, the equality of the electron and positron scattering cross sections could be 
explained quantitatively by long range, electrostatic dipole coupling. Infrared absorption 

M. A. P. Lima, private communication, 2005. 


Symposium on Atomic & Molecular Physics 

measurements provide the strength of the dipole matrix element, and the theory fits quite 
well with no adjustable parameters. So in this case, the sharp rise in the cross section is 
not a resonance but arises naturally from the long-range electrostatic coupling. The 
remaining mystery is that all of the molecules and modes studied to date, except the one 
homopolar molecule, H 2 , have positron-impact vibrational cross sections with a very 
similar energy dependence, even though the magnitude of the Bom dipole coupling is too 
small by as much as a factor of five for everything except CF 4 [55]. So there’s more to 
be learned here. 

Fig. 11. Positron- and electron-impact cross sections for excitation of the v 3 asymmetric 
stretch mode of CF 4 . This is the largest positron-impact vibrational cross section 
measured to date. Also shown for comparison (-) are the predictions of the Bom dipole 
model for this cross section, with no fitted parameters. From Ref. [54], 

The final question led to the most spectacular result. As discussed above we, and 
others before us, had suspected that the large annihilation rates in molecules are due to 
vibrational resonances [15, 56], but it was hard to nail down experimentally with thermal 
distributions of positrons. Once we had the cold beam, Steven Gilbert and Levi Bames 
(who had just joined the group as a Ph.D. student) took on the very ambitious project of 
studying annihilation rates in molecules as a function of positron energy. They built a 
wonderful apparatus such that they could cycle 10 9 positrons with only one background 
count. This is needed because, while the resonances are quite large, the basic cross 
section [i.e., the Dirac cross section,, Eq. (1)] is miniscule, and so without a very 
careful experiment, it is still difficult to distinguish the annihilation from extraneous 

The results, as advertised above and illustrated in Fig. 12, were spectacular indeed 
[45, 57]. Alkane molecules were a favorite target of ours for studying annihilation, 
because they are conveniently available in a variety of sizes and exhibit huge 
enhancements in annihilation rates. We found very large enhancements in alkanes with 
energy spectra that closely mimic the spectra of the molecular vibrational modes, with 
particularly large resonance associated with the C-Fl asymmetric stretch mode. The 


Symposium on Atomic & Molecular Physics 

added bonus was that the annihilation spectra are downshifted from the vibrational 
spectra. We interpret these data in the context of Gleb Gribakin’s vibrational Feshbach 
resonance model [60] as evidence that positrons bind to alkanes. The downshift is a 
measure of the positron binding energy, which ranges from ~ 40 meV in butane (C 4 H 10 ) 
to > 200 meV in Ci 2 H 26 - Levi Barnes and Jason Young have now carried these 
experiments further, most recently finding evidence for a second, positronically excited 
bound state in the very large alkanes C 12 H 26 and C 14 H 30 [57, 58]. 

energy (eV) 

Fig. 12. Z e ff for butane (•) as a function of positron energy: (a) 0 to 5 eV, and (b) 0 to 0.5 
eV. From Refs. [57] [58], The arrow on the abscissa in (a) is the threshold for 
positronium formation. Shown in (b) is the vibrational-mode spectrum of butane (— , 
arbitrary vertical scale), with each mode broadened by 25 meV [59], The downshift, ft,, 
represents the positron-molecule binding energy which is ~ 40 meV for butane. Arrows 
on the ordinate indicate values of Z e{{ for a 300 K Maxwellian distribution of positrons. 

In (a), Z e ff at energies > 0.5 eV is ~ 100, comparable to the value of Z = 34 for this 
molecule, which is expected in the absence of the vibrational resonances. 


I’ve carried the positron trapping saga and positron-atomic physics story along from 
the mid 80’s to the present. There is a parallel story for positron plasmas that is too 
lengthy to tell here in any detail [6]. Mark Tinkle did a thesis on the development of 
mode diagnostics for positron plasmas [61, 62], Rod Greaves and I did the first electron- 
positron plasma experiment, studying the instability generated when an electron beam is 
passed through a positron plasma [63, 64], Greaves used a rotating electric field to 
compress positron plasmas radially (the so-called “rotating wall” effect) [44, 65], a tool 


Symposium on Atomic & Molecular Physics 

recently developed further by James Danielson [66, 67], that has enormous potential for 
tailoring positron plasmas and beams. James is now continuing this research and 
developing new methods to trap more positrons (i.e., a “multicell trap” [68]), and to 
produce much colder (the goal 1 meV, FWHM) positron beams [6]. 

At First Point Scientific Inc., Rod Greaves developed a commercial prototype 
buffer gas trap for materials characterization and positron research [43]. The basic 
buffer-gas trap design was used by the ATHENA collaboration to make the first low- 
energy antihydrogen [8]. The trap gave them a significant advantage vis a vis their 
competitors in the ATRAP collaboration, who also created low-energy antihydrogen very 
close to the same time in a set of complementary experiments [8, 69]. James Sullivan 
and Steve Buckman have now built a trap in Canberra for positron atomic physics 
research. Mike Charlton has one at Swansea [70], as does Igor Meshkov in Dubna [71], 
who got the design from Mike and is making a new Ps beam facility for fundamental 
physics studies. 

Very recently, Allen Mills, David Cassidy, Rod Greaves, and collaborators used 
Rod’s version of the buffer gas trap to observe effects they attribute to creation of the first 
Ps 2 molecules in the laboratory [9]. This line of experiments has enormous promise, and 
is likely to lead to the creation and study of BEC Ps, which is a long term goal of Allen 
and Phil Platzman [72, 73]. 

So it’s been a very productive and rewarding time. The Drachman-Temkin 
symposium is a great place to tell the story. Dick’s guidance and counsel was much 
appreciated by all of us involved in the positron trapping effort. As I finished writing 
this, one thought occurred to me: What if Marv and I had sat at different lunch tables that 
day at Bell? I may well have not had the opportunity to work with positrons or meet 
Dick - it was truly good luck the way it turned out! 

I thank Rod Greaves, Marv Leventhal and Allen Mills, Jr., for careful reading of the 


1 . Surko, C. M., Leventhal, M., Crane, W. S., et al., The Positron Trap - A New 
Tool for Plasma Physics, in Positron Studies of Solids, Surfaces, and Atoms; a 
Symposium to Celebrate Stephan Berko's 60th Birthday, edited by J. A. P. Mills, 
W. S. Crane and K. F. Canter (World Scientific, 1984), p. 222. 

2. Surko, C. M., Leventhal, M., Crane, W. S., et al.. Use of Positrons to Study 
Transport in Tokamak Plasmas, Rev. ofSci. Instrum. 57, 1862, 1986. 

3. Brown, B. L. and Leventhal, M., Laboratory Simulation of Direct Positron 
Annihilation in a Neutral-Hydrogen Galactic Environment, Phys. Rev. Lett. 57, 
1651, 1986. 


Symposium on Atomic & Molecular Physics 

4. Leventhal, M., Callum, C. J. M., and Stang, P. D., Detection of 51 1 Kev Positron- 
Annihilation Radiation from the Galactic-Center Direction, Astrophys. J., LI 1, 

5. Surko, C. M., Gribakin, G. F., and Buckman, S. J., Low-Energy Positron 
Interactions with Atoms and Molecules, J. Phys. B: At. Mol. Opt. Phys. 38 , R57 

6. Surko, C. M. and Greaves, R. G., Emerging Science and Technology of 
Antimatter Plasmas and Trap-Based Beams, Phys. Plasmas 11 , 2333, 2004. 

7. Greaves, R. G. and Moxom, J. M., Recent Results on Trap-Based Positron Beams, 
Mat. Sci. Forum 445 - 446 , 419, 2004. 

8. Amoretti, M., Amsler, C., Bonomi, G., et al., Production and Detection of Cold 
Antihydrogen Atoms, Nature 419 , 456, 2002. 

9. Cassidy, D. B., Deng, S. H. M., Greaves, R. G., et al.. Experiments with a High- 
Density Positronium Gas, Phys. Rev. Lett. 95 , 195006, 2005. 

10. Malmberg, J. H. and deGrassie, J. S., Properties of Nonneutral Plasma, Phys. Rev. 
Lett. 35, 577, 1975. 

11. O'Neil, T. M., A Confinement Theorem for Nonneutral Plasmas, Phys. of Fluids 

12. Bailie, P. and Darewych, J. W., Collisions Entre Un Positron Et Une Molecule 
D'hydrogene Avec Excitation De Vibrations, J. de Phys. Lett. 35 , 243, 1974. 

13. Surko, C. M., Leventhal, M., and Passner, A., Positron Plasma in the Laboratory, 
Phys. Rev. Lett. 62, 901, 1989. 

14. Fee, M. S., Chu, S., A. P. Mills, Jr., et al.. Measurement of the Positronium 13S1 - 
23 SI Interval by Continuous-Wave Two-Photon Excitation, Phys. Rev Lett. 70 , 
1397, 1993. 

15. Surko, C. M., Passner, A., Leventhal, M., et al., Bound States of Positrons and 
Large Molecules, Phys. Rev. Lett. 61, 1831, 1988. 

16. Fraser, P. A., Positrons and Positronium in Gases, Adv. in At. and Mol. Phys. 4 , 
63, 1968. 

17. Heyland, G. R., Charlton, M., Griffith, T. C., et al.. Positron Lifetime Spectra for 
Gases, Can. J. Phys. 60 , 503, 1982. 

18. Paul, D. A. L. and Saint-Pierre, L., Rapid Annihilation of Positrons in Polyatomic 
Gases, Phys. Rev. Lett. 11, 493, 1963. 

19. Passner, A., Surko, C. M., Leventhal, M., et al.. Ion Production by Positron- 
Molecule Resonances, Phys. Rev. A 39 , 3706, 1989. 

20. Hulett, L. D., Donohue, D. L., Xu, J., et al.. Mass Spectrometry Studies of the 
Ionization of Organic Molecules by Low-Energy Positrons, Chem. Phys. Lett. 

216 , 236, 1993. 

21. McLuckey, S. A., Hulett, L. D., Xu, J., et al., Gas-Phase Ionization of Polyatomic 
Molecules Via Interactions with Positrons, Rap. Comm. Mass Sp. 10 , 269, 1996. 

22. McLuckey, S. A., Glish, G. L., Donohue, D. L., et al., Positron Ionization Mass 
Spectrometry: Ionization by Fast Positrons, Int'l. J. of Mass Spectrometry and Ion 
Processes 97 , 237, 1990. 

23. Crawford, O. H., Mechanism for Fragmentation of Molecules by Positron 
Annihilation, Phys. Rev. A 49 , R3147, 1994. 


Symposium on Atomic & Molecular Physics 

24. Iwata, K., Greaves, R. G., and Surko, C. M., Gamma-Ray Spectra from Positron 
Annihilation on Atoms and Molecules, Phys. Rev. A 55 , 3586, 1997. 

25. Iwata, K., Gribakin, G. F., Greaves, R. G., et al., Positron Annihilation on Large 
Molecules, Phys. Rev. A 61, 022719, 2000. 

26. Leventhal, M., Passner, A., and Surko, C. M., Positron-molecule Bound States 
and Positive Ion Production, in Annihilation in Gases and Galaxies, edited by R. 

J. Drachman (National Aeronautics and Space Administration, Washington, DC, 
1990), p.272. 

27. Murphy, T. J., Positron Deposition in Plasmas by Positronium Beam Ionization 
and Transport of Positrons in Tokamak Plasmas, Plasma Phys. and Controlled 
Fusion 29 , 549, 1987. 

28. Murphy, T. J. and Surko, C. M., Annihilation of Positrons on Organic Molecules, 
Phys. Rev. Lett. 67, 2954, 1991. 

29. Murphy, T. J. and Surko, C. M., Positron Trapping in an Electrostatic Well by 
Inelastic Collisions with Nitrogen Molecules, Phys. Rev. A 46 , 5696, 1992. 

30. Murphy, T. J. and Surko, C. M., Annihilation of Positrons in Xenon Gas, J. Phys. 
B: At. Mol. Opt. Phys. 23 , L727, 1990. 

31. Tang, S. and Surko, C. M., Angular Dependence of Positronium Formation in 
Molecular Hydrogen, Phys. Rev. A 47 , 743, 1993. 

32. Tang, S., Tinkle, M. D., Greaves, R. G., et al., Annihilation Gamma-Ray Spectra 
from Positron-Molecule Interactions, Phys. Rev. Lett. 68, 3793, 1992. 

33. VanReeth, P., Humberston, J. W., Iwata, K., et al., Annihilation in Low-Energy 
Positron-Helium Scattering, J. Phys. B: At. Mol. Opt. Phys. 29 , L465, 1996. 

34. Iwata, K., Gribakin, G., Greaves, R. G., et al., Positron Annihilation with Inner- 
Shell Electrons in Noble Gas Atoms, Phys. Rev. Lett. 19, 39, 1997. 

35. Surko, C. M., Greaves, R. G., and Leventhal, M., Use of Traps to Study Positron 
Annihilation in Astrophysically Relevant Media, Hyperfme Interactions 81 , 239, 

36. Iwata, K., Greaves, R. G., and Surko, C. M., Positron Annihilation in a Simulated 
Interstellar Medium, Can. J. Phys. 51 , 407, 1996. 

37. Greaves, R. G. and Surko, C. M., Solid Neon Moderator for Positron Trapping 
Experiments, Can. J. Phys. 51 , 445, 1996. 

38. Mills, A. P., Jr. and Gullikson, E. M., Solid Neon Moderator for Producing Slow 
Positrons, Phys. Lett. 49 , 1121, 1986. 

39. Jorgensen, L. V., Amoretti, M., Bonomi, G., et al.. New Source of Dense, 
Cryogenic Positron Plasma, Phys. Rev. Lett. 95, 025002, 2005. 

40. Kurz, C., Greaves, R. G., and Surko, C. M., Temperature Dependence of Positron 
Annihilation Rates in Noble Gases, Phys. Rev. Lett. 77 , 2929, 1996. 

41. Gilbert, S. J., Kurz, C., Greaves, R. G., et al.. Creation of a Monoenergetic Pulsed 
Positron Beam, Appl. Phys. Lett. 70 , 1944, 1997. 

42. Gilbert, S. J., Greaves, R. G., and Surko, C. M., Positron Scattering from Atoms 
and Molecules at Low Energies, Phys. Rev. Lett. 82, 5032, 1999. 

43. Greaves, R. G. and Moxom, J., Design and Performance of a Trap-based Positron 
Beam Source, in Non-Neutral Plasma Physics V, edited by M. Schauer, T. 
Mitchell and R. Nebel (American Institute of Physics, 2003), p. 140. 


Symposium on Atomic & Molecular Physics 

44. Greaves, R. G. and Surko, C. M., Inward Transport and Compression of a 
Positron Plasma by a Rotating Electric Field, Phys. Rev. Lett. 85, 1883, 2000. 

45. Gilbert, S. J., Barnes, L. D., Sullivan, J. P., et al., Vibrational Resonance 
Enhancement of Positron Annihilation in Molecules, Phys. Rev. Lett. 88 , 

46. Sullivan, J., Gilbert, S. J., and Surko, C. M., Excitation of Molecular Vibrations 
by Positron Impact, Phys. Rev. Lett. 86 , 1494, 2001. 

47. Sullivan, J. P., Marler, J. P., Gilbert, S. J., et al., Excitation of Electronic States of 
Ar, H 2 , and N 2 by Positron Impact, Phys. Rev. Lett. 87, 073201, 2001. 

48. Sullivan, J. P., Gilbert, S. J., Buckman, S. J., et al.. Search for Resonances in the 
Scattering of Low-Energy Positrons from Atoms and Molecules, J. Phys. B: At. 
Mol. Opt. Phys. 34, L467, 200 1 . 

49. Sullivan, J. P., Gilbert, S. J., Marler, J. P., et al., Positron Scattering from Atoms 
and Molecules Using a Magnetized Beam, Phys. Rev. A 66 , 042708, 2002. 

50. Marler, J. P., Sullivan, J. P., and Surko, C. M., Ionization and Positronium 
Formation in Noble Gases, Phys. Rev. A 71, 022701, 2005. 

51. Marler, J. P. and Surko, C. M., Positron-Impact Ionization, Positronium 
Formation and Electronic Excitation Cross Sections for Diatomic Molecules, 

Phys. Rev. All, 062713, 2005. 

52. Campbell, L., Brunger, M. J., Nolan, A. M., et al.. Integral Cross Sections for 
Electron Impact Excitation of Electronic State of N 2 , J. Phys. B: At. Mol. Opt. 
Phys. 34, 1185,2001. 

53. Mason, N. J. and Newell, W. R., Electron Impact Total Excitation Cross Section 
of the A 1 n g State of N 2 , J. Phys. B 20, 3913, 1987. 

54. Marler, J. P. and Surko, C. M., Systematic Comparison of Positron and Electron 
Impact Excitation of the N 3 Vibrational Mode of Cf 4 , Phys. Rev. A 72, 062702, 

55. Marler, J. P., Gribakin, G., and Surko, C. M., Comparison of Positron-Impact 
Vibrational Excitation Cross Sections with the Bom-Dipole Model, Nucl. Instrum 
Methods A, in press, 2006. 

56. Smith, P. M. and Paul, D. A. L., Positron Annihilation in Methane Gas, Can. J. 
Phys. 48, 2984, 1970. 

57. Bames, L. D., Gilbert, S. J., and Surko, C. M., Energy-Resolved Positron 
Annihilation for Molecules, Phys. Rev. A 67, 032706, 2003. 

58. Bames, L. D., Young, J. A., and Surko, C. M., Studies of Energy-Resolved 
Positron Annihilation Rates for Molecules, Phys. Rev. A, submitted, 2006. 

59. Gribakin, G. F. and Gill, P. M. W., The Role of Vibrational Doorway States in 
Positron Annihilation with Large Molecules, Nucl. Instrum. Methods B 221, 30, 

60. Gribakin, G. F., Mechanisms of Positron Annihilation on Molecules, Phys. Rev. A 
61, 22720, 2000. 

61. Tinkle, M. D., Greaves, R. G., Surko, C. M., et al., Low-Order Modes as 
Diagnostics of Spheroidal Non-Neutral Plasmas, Phys. Rev. Lett. 72, 352, 1994. 

62. Tinkle, M. D., Greaves, R. G., and Surko, C. M., Modes of Spheroidal Ion 
Plasmas at the Brillouin Limit, AIP Conference Proceedings 331, 229, 1995. 


Symposium on Atomic & Molecular Physics 

63. Greaves, R. G. and Surko, C. M., An Electron-Positron Beam-Plasma 
Experiment, Phys. Rev. Lett. 75, 3846, 1995. 

64. Gilbert, S. J., Dubin, D. H. E., Greaves, R. G., et al., An Electron-Positron Beam- 
Plasma Instability, Phys. of Plasmas 8, 4982, 2001. 

65. Greaves, R. G. and Surko, C. M., Radial Compression and Inward Transport of 
Positron Plasmas Using a Rotating Electric Field, Phys. Plasmas 8, 1879, 2001. 

66. Danielson, J. R. and Surko, C. M., Torque-Balanced High-Density Steady States 
of Single Component Plasmas, Phys. Rev. Lett. 95, 035001, 2005. 

67. Danielson, J. R. and Surko, C. M., Radial Compression and Torque Balanced 
Steady States of Single-Component Plasmas in Penning-Malmberg Traps, Phys. 
Rev. A, in press, 2006. 

68. Surko, C. M. and Greaves, R. G., A Multi-Cell Trap to Confine Large Numbers of 
Positrons, Rad. Chem. and Phys. 68, 419, 2003. 

69. Gabrielse, G., Bowden, N., Oxley, P., et al., Background-Free Observation of 
Cold Antihydrogen with Field-Ionization Analysis of Its States, Phys. Rev. Lett. 

70. Clarke, J., vanderWerf, D. P., Charlton, M., et al.. Developments in the Trapping 
and Accumulation of Slow Positrons Using the Buffer Gas Technique, 

Nonneutral Plasma Physics V, edited by M. Schauer, T. Mitchell and R. Nebel 
(American Institute of Physics, 2003), p. 178. 

71. Meshkov, I. N., Lepta Project: Generation and Study of Positronium in Directed 
Fluxes, Nucl. Instrum Methods B 221, 168 2004. 

72. Platzman, P. M. and Mills, A. P., Jr., Possibilities for Bose Condensation of 
Positronium, Phys. Rev. B 49, 454, 1994. 

73. Mills, A. P., Jr., Positronium Molecule Formation, Bose-Einstein Condensation 
and Stimulated Annihilation, Nucl. Instrum. Methods B 192, 107, 2002.