AVS Historical Persons | Raymond G. Herb - 1994

Raymond G. Herb - 1994

Oral History Interview with Raymond G. Herb

Interviewed by Willy Haeberli, December 9, 1994
HAEBERLI: I am Willy Haeberli, Professor of Physics at the University of Wisconsin. As part of the American Vacuum Society Historical Archive Series, I will be talking to Professor Raymond G. Herb1. It is December 9, 1995, and we are in Menominee, near Madison, Wisconsin. 

Let me first briefly introduce Professor Herb, or Ray, as we call him. Ray has had a distinguished career in Nuclear Physics, and is known for his development of the pressurized electrostatic accelerator. He started this work already in 1931, in the very early days of nuclear physics, when he was a graduate student at the University of Wisconsin. The design innovations he introduced form the basis of all such accelerators to this day. At the time, his new machines allowed him pioneering studies in nuclear reactions and experiments on the interaction between protons and protons. He joined the faculty of the Physics Department of the University of Wisconsin in 1939 and was the leader of a very successful nuclear physics group. He later became the Mendenhall Distinguished Professor of Physics. Some years before he retired from the University in 1973, he founded the National Electrostatics Cooperation in Middleton near Madison, Wisconsin. The company builds electrostatic accelerators for a large number of scientific and industrial applications. Professor Herb is active to this day as president of the company. He has received many honors, including membership in the National Academy of Sciences, the Tom W. Bonner prize of the American Physical Society, and several honorary degrees. While his career started in nuclear physics, the members of the American Vacuum Society probably know him primarily for his work on the titanium getter pumps and other contributions to vacuum science. So let us turn now to this topic. 

Well as a first question, Ray, I would like you to recount the history of your interest in vacuum technology and how your work was related to the development of the nuclear physics instrumentation which you have been involved in.

HERB: Thank you, Willy. One little correction, I founded the company 30 years ago, 1965. 

HAEBERLI: So, would you like to elaborate?

HERB: Yes. We are a bit over 30 years old now. Well, it doesn't seem quite that long. As Willy said, I've been associated with high voltage electrostatic accelerators for a long time, longer than I'd like to think. It's quite evident in the early days, vacuum techniques, in fact all vacuum pumps, used oil; the diffusion pumps used oil. Vacuum systems are put together with all sorts of organic materials. I thought these were probably important in limiting the performance, especially in a vacuum, of the accelerating tube. And so sometime, I recall in the '50s, I started a little program. I always had quite a number of undergraduate students working on all sorts of little projects. I thought it was interesting, since all metals tend to corrode, to see if I could incorporate this corrosion into a system for getting a vacuum. So I had a young undergraduate, I'm not sure if he was a physics major or not, David Saxon, go through a series of evaporation. We had a physics department evaporator, and in that we installed a crucible. I got samples of all sorts of metals, and he evaporated them and tried to see if some one of these in the evaporator wouldn't take up oxygen and nitrogen readily and maybe do a little pumping. 

Well, I was told by somebody that a new metal, titanium, was becoming available in pellet form, only in pellet form. I wrote and got a little sample of titanium pellets, and amongst the various things David Saxon tried was using these pellets. He came back to me and said, "Well, something's wrong with the vacuum gauge. They're going to zero." He had evaporated some titanium pellets and they pumped. Most dramatic! I thought this was interesting. We should be able to make a pump out of this, because it was very astonishing how well it did. The pellets were inconvenient for a continuous evaporator. I heard about a new company that was formed, Titanium Metals Corporation: I think it was called Titanium Metals Corporation of America. I wrote to them for twenty mil wire [i.e. 0.020 inch], a pound of it. The company had just been formed. They gave a price of $30 per pound of titanium wire. I ordered it; it came. It was black! It was the first wire, obviously, they produced. Working with titanium is tough business, but it was flexible. We tried it and it pumped. It did quite well, and we carried on with a variety of pumping techniques. Later, I ordered a second pound of twenty mil wire. It came, beautifully clean! The company had advanced at an astonishing rate in the processing of titanium, and before not very long, titanium was a common article of commerce. We combined the evaporation of titanium with an ionizing system to take care of noble gases. We had a general purpose pump! The University patented it, Consolidated Vacuum Corporation took a license, and pumps were used. During the period of this development, I got a letter from Russell Varian, who was interested. And for a reason! You may have heard of Russell Varian, a founder of the Varian Company. We corresponded. He was very interested, and obviously wanted to get in to the business. Then the correspondence stopped. There'd been a transition. A fellow named Lou Hall, who I'd never met, at Varian invented the sputter ion pump. It was simpler; a little egg crate electrode in a magnetic field. No heating, turn on voltage, sputtering did the pumping. That outdid our pump because it was simpler, and so it took over.

HAEBERLI: I do remember, from my early days at the lab in Wisconsin that one time I needed a good fast pump and you gave me a pump which I think you referred to as the "Tirod". Now we called it the "hot rod" because it was pumping incredibly fast, and I wonder whether you have any recollection of this pump and about what the pumping speed might have been of this "hot rod" pump.

HERB: By the way, evaporating titanium you can get almost any pumping speed you want. This was for active gases, and for the pumping of the inert gases, argon and some of the hydrocarbons, the pumping was dependent on what sort of ionization system was developed. It was quite pleasant to see the pumping speed, and actually, Brookhaven ordered quite a large number of these for the first, or maybe the second, of their circular accelerators. The speed was good, but it was complex, and so the market for these things diminished. We at the company still use a few of these, but the sputter ion pump took over.

HAEBERLI: On the other hand, you then later developed the Orbitron. I wonder whether you might recall a bit about the motivation that led you from the getter pump to the Orbitron pump.

HERB: I was looking for something simpler, and a pump for noble gases. The Orbitron was a clever little device, but again, it was never manufactured commercially because it was a bit too complex, even though it was a rather pleasant device. A normal person likes to have as little fussing as possible. So the sputter pump did take over. 

Actually, I would like to spend a bit more time now on the other developments, which were related more to high voltage and the desirability of getting high voltage in a vacuum.

HAEBERLI: By all means, go ahead.

HERB: I wanted to get rid of organic materials, and so I had quite a large number of students working on the bonding of metals to ceramic. I heard of a relatively new alloy called Kovar with a pretty good match thermally to alumina ceramics. After a lot of attempts at hard soldering, we finally put together insulators, hard soldering alumina ceramic to Kovar metal. And so we had the lead-through bushing. The ceramic could be made in a variety of shapes depending on voltages needed. During that period, a former student of mine, Don Benedict, stopped in. He was then with, I think, Midwest Research. He liked the looks of this, took it back to Midwest Research and it spread to other places and the lead-through bushings were now used; and all apparatus requiring high voltage in vacuum used these bushings. We never patented it. 

Then we tried another method which happened to work out. Somehow we got started using titanium more and more for our work in high voltage, and we discovered that we could bond titanium metal to alumina ceramic using, as the glue, aluminum foil, ordinary aluminum foil. Put an aluminum foil washer in between the alumina ceramic, in the form of rings with titanium sheets, and pressed at a certain pressure in a vacuum and the bond was made. A very consistent bond. With these, we had the makings, now, of high voltage apparatus free of organic vapors. And just about then, I and a few of my students, formed a company called National Electrostatics Corporation. We rented a little warehouse space - a shoe string operation. With it, we started bonding and we - well, I probably should digress just a little before carrying on with the bonding, and say that to create high voltage electrostatically, you need a method of carrying charge into the region. I was not satisfied with the belt. And back at the University, with a number of students, including Jim Ferry, we developed this chain [shows a length of metal rings] - pellets on a nylon cord, running around so the chain could carry charge in to a high voltage region. We took this [shows a length of larger metal rings, approximately 2cm diameter, on a cord] and other techniques, using them in our little warehouse space, and we went from this little chain to this: steel cylinders connected by nylon links. This led into a couple of our first machines that we built at the company. This was obviously better. I thought we'd keep going and we did and went, matter of fact, to a larger diameter and got this [shows a length of connected metal cylinders about 4cm diameter]. This was quite a bit better, and now in a great many machines around the world, and has displaced belts in quite a number of the older high voltage engineering machines. 

So we had a charging system, and, at our little warehouse space, we took alumina ceramic rings, titanium sheets, put them together, and had this [shows a cylinder about 15cm diameter and 25cm long]. The first all metal and ceramic -  I believe this was the - well we had one, but different - the first one was developed at our company space, and this would make the whole 1/3 of a million volts; three of these clamped together were meant to hold 1,000,000 volts. This, in almost the identical form we developed way back then, is used in all of our machines, with alumina ceramic rings, titanium metal sheets. In an all metal and ceramic, vacuum-tight container, it'll hold high voltage. For larger machines, support was always a problem. At this little warehouse space, Jim Ferry and I largely had responsibility for the support posts. Well, here again [shows a cylinder about 15cm diameter and 30cm long], alumina ceramic disks, titanium metal sheets, aluminum foil glue, put together, pressed in a vacuum and the bond was made. This was the one million volt support member. These three components largely served as a basis for our business. 

In going back to our origins, in starting the company, and funding... attempting a big machine, I ought to give credit to Oscar Sala at the University of Sao Paulo for a great deal of encouragement and for ordering one of our first machines; he had enough faith in our competence to do this. Also, in carrying on, getting going with our larger machines, certainly Sir Ernest Titterton from the Australian National University in Canberra, visited, liked the looks of our machines, and he was known as saying that he bullied me into building a 14,000,000 volt machine for him - he pushed pretty hard because it was a tremendous undertaking... and we did it.

HAEBERLI: Now this was not yet the biggest machine actually that you built was it?

HERB: At that time it was by far the biggest.

HAEBERLI: But since then you - 

HERB: And I said, just go a little farther, saying that putting these big machines together, we had our support posts, you see this [shows a photograph] - four support posts in between a pair of aluminum castings, performing 1,000,000 volts support. These were stacked up and these, fourteen of these, were stacked up around the terminal, a 14,000,000 volt machine, and keeping on the terminal upper end because you'd have tandems. And through them, the accelerating tubes, of course, were arranged between a pair of aluminum castings to form a 1,000,000 volt support module. These were stacked up, the 14,000,000 volt tandem having 14 of these around the terminal, another 14 from the terminal to upper end. And you have a question in regard to...?

HAEBERLI: Yes, I had a question about what the biggest machine was that you ever built using these modules.

HERB: Again, we tackled building a bigger machine. And the biggest is shown here. This is the 25,000,000 volt tandem machine at the Oak Ridge National Laboratory. There you see the building housing it [shows photograph of building], and if you go to Oak Ridge, it's a landmark. Beautiful building. Of course, all these machines are in pressure tanks. I should have said the pressure tank was filled with high pressure sulfur hexafluoride. And here you see the 25,000,000 volt machine. You see a dome here [shows photo looking down on the dome], 14 feet in diameter, resting on a column. And, if you look carefully, you'll see three of the little dots at the bottom; these are three men on a service platform. For this structure, 16 of these posts are arranged in a circle between a pair of aluminum castings, like this, only bigger in diameter, and those modules are stacked up to form the support structure for this machine. The machine operates with a tube, 25,000,000 volts, which is by far the world's highest. Without the accelerating tube inside, it could be taken up to almost 32,000,000 volts, which happens to be the world's highest.

HAEBERLI: Now we mostly talked at the moment about these very large machines which are used for nuclear physics research, in particular for heavy ions and so on, but you might give us some insight about other devices your company has produced for other purposes, often smaller machines for interesting, special applications.

HERB: Yes, the plan initially was mainly to form larger machines and the larger, higher in voltage, the better. We did quite well at that. Go back just a moment to say, this large machine was in the Guinness Book of World Records. And attempts to go higher we have not actually tried; others have tried and it's quite difficult. It seems to me, at the moment, pretty firmly anchored at the world's highest. The end of the Cold War brought a fairly rapid change to funds available and interest in nuclear physics, and interest in other things developed, and before too long we had a pretty good line of smaller machines, based on these components, and our smaller machines went from 1,000,000 volts up to, well fairly compact design going all the way up to 5,000,000 volts. They did a variety of things. 

Well, the 1,700,000 volt machine at the University of Oxford, England was used for the investigation of Alzheimer's disease. There'd been the belief that aluminum was involved in the development of this plaque that appears in the brains of Alzheimer's patients. Two people at Oxford used this machine using a technique called PIXE [Proton Induced X-ray Emission] where protons bombard a target, X-rays are emitted depending on the composition; X-rays of course depend on the elements. They focus the beam down to about a micron, and with it, could examine, in this case, plaque from an Alzheimer's corpse, looking for aluminum. It's extremely sensitive. They looked, found none - no question of it. It so happens that a group at the University of Tokyo used the 5,000,000 volt machine and, using a different method, looked at this plaque, and found some. Then they looked farther and found that the aluminum they saw was contained in the ultra pure reagent that was used for washing the tissue! This finding was verified. 

HAEBERLI: Well these are certainly very interesting applications of your new techniques. But I wonder whether you might give us some thoughts on the future of vacuum technology and of applications of such high voltage electrostatic accelerators.

HERB: Yes, I would like to say a little bit about this. All of my life I have struggled with attempts to make something of electrostatics. It has been quite frustrating because of the limitations, especially high voltage breakdown in vacuum, which is limited but could be done. Our results in regard to eliminating organic materials, so that, yes, it was a fine thing to do but it did not solve all of problems of high voltage breakdown in a vacuum, and so there is much more to be done. I could say also this that this machine, the 25,000,000 volt machine, functions really very dependably, and I am forced to say, though, that it rests on a foundation of ignorance. We build only by learning what doesn't work, picking and choosing - We really don't understand the fundamentals of limitations, and I believe that something can be made of electrostatics much beyond this if we learn the fundamentals. And I'm still quite stubborn about carrying on and trying to do this, and I would like to make this machine obsolete.

HAEBERLI: Yes, it is interesting to hear you say that after your long career of studying vacuum breakdown, you feel that the fundamentals are still eluding us to a large extent.

HERB: Oh! Completely! We're ignorant.

HAEBERLI: It certainly is a challenge to another generation of investigations to learn more about that. Ray, do you have anything you'd like to add at this time?

HERB: No, this is fine.

HAEBERLI: Well thank you very much. If that's the case, I would like to thank you for being here today and for contributing to the American Vacuum Society Historical Archives.

1. For a biography, see http://www.nap.edu/readingroom/books/biomems/rherb.html

return to top