Dudley Herschbach Interview: Conversations with History; Institute of International Studies, UC Berkeley
Page 3 of 5
In your own work, as you emphasized yesterday in your lecture, you were, in a way, "standing on the shoulders" of scientists who came before you. Let's talk now a little about your work. Help us understand what you did in layman's terms to merit a Nobel Prize and how it reflected a lot of what you've just described, namely working in a community but at the same time being, as you say, somewhere on the lunatic fringe of chemistry.
Yes, that's what it was called at the time.
Talk a little about that.
It's characteristic, in fact, of many, many episodes or ventures in science that I've witnessed in my time, and read about in history because I love history, and it's certainly generally true that, as I mentioned before, science is intrinsically cooperative. Every scientist knows how much they owe to their predecessors and even to their contemporaries, even if nominally they might look to be competitors, since you have to announce to the world whatever you find in science. Your nominal competitor may actually help you more than they help themselves because what they find and report may, from your perspective, advance your development of the problem more than their own.
At any rate, what happened in my case was, as I mentioned earlier, Harold Johnston, my mentor, was doing a classic kind of research on how fast can reactions go, but [it was] all in the bulk, that's all people could do then. People thought it was not feasible at all to carry out experiments where you could study things at a level of individual collisions, get that detail one would love to have on what actually happens in these collisions, how much energy goes into separation of products or vibration of the new bonds or tumbling rotational motion. And then, fundamentally understanding the forces that govern such things, how the electronic structure, reactants and products, determine what's going to happen. It's the most fundamental thing in chemistry.
It was difficult to get at that in a conventional way, because as I always like to say, suppose you could only study human psychology by flying a blimp over a stadium, listening to the murmuring and shouts of the crowd. The chemist is in more or less that condition by having to work with zillions of molecules in the test tubes. Most of what you'd like to know is averaged down. The method we developed was pretty obvious and naïve: you [create] a high vacuum, so you pump out molecules that otherwise would be there; you can form two streams, very dilute streams, of molecules which are spaced far enough apart that they don't interact with each other in either stream and they'll cross, a few of them will collide. If you have enough sensitivity, you can detect the newly formed products that fly through the vacuum to your detector. So, single-collision chemistry.
It turned out that for initially a simple class, a special class of reactions that people thought was non-representative, involving so-called alkali reactions like sodium with reagents that included halogens (sodium is willing to give up an electron to a halogen-containing reagent) it was very easy, so it's a rather facile, more rapid reaction than most. So, you could hope to get a bigger yield, and therefore even under these severe conditions, dilute beams interacting, maybe you could detect something.
As I mentioned yesterday, in our first experiments our yield was equivalent to a monolayer of molecules a month. It's pretty puny by chemical terms, but because of these properties, we could detect these special molecules at that level with not terribly elaborate apparatus. We knew we could do that by virtue of previous work of other people that didn't quite get to the point of doing this. This is what happens in every new generation. I tell students sometimes, you may think from the textbooks and from all you read that "Gee, previous scientists had it easy, they just get all these lovely results. What's left for me to do?" And what you may not recognize is that you have a legacy of new tools, concepts as well as instruments, that open up new eyes literally, that you can use to see things your predecessors could not, and to think of doing probes that nobody thinks of until the technical feasibility of actually doing it is somewhere on the horizon, at least. If it's too far over the horizon, people don't imagine it.
That's why in the early days people said that's lunatic fringe to think you could study chemical reactions that way. Basically what we did was just show that that's not the case. When it came to going to a broader range of reactions, that was technically a lot harder, and people thought that was out of the question, but by then we'd developed a lot of experience with the easier ones. Especially with the genius of Yuan Lee as a superb experimentalist, we were able to go on and do it, and many other people soon found they could do it, sort of like the four-minute mile -- for so long people thought you couldn't do it and once it was done, very soon other people were doing it too.
And as you recounted yesterday, there's experimentation, there's tinkering, there's sort of building apparatus that will make this possible, but you're also looking back in time because this came from a line of experiments done by a physicist named Stern.
He was a chemist by trade.
Oh, chemist. Okay.
Yes. Remember, he got his Ph.D. in chemistry for dissolving CO2 in various solvents -- a beginning [that] you would hardly imagine led ultimately to the work he did.
Let me ask this question, and remember, I'm not a scientist. There are analogies with what you did to what Townes had done in physics to discover the maser in terms of the way the apparatus was set up to make discoveries.
They had a kinship. If you think of it, it's an evolutionary thing. What Otto Stern did was ancestor both to Towne's discovery of the maser and then the laser, and to the work we did, and to much other work. So, all these kind of evolved. What happens is things get combined and out of the combination new things grow. It's a kind of biology we're talking about now -- this is the biology, if you like, of the intellectual development of physical science. The ideas are terribly important, and they propagate and combine and spawn progeny. That's really what happens.
What are the implications for a general audience of the work that you did? I gather it can help us in understanding global warming. Is that fair?
Yes, there are lots of practical applications -- so many, in fact, that it's a little embarrassing if you start listing them because it sounds like you're taking credit for everything under the sun. The reason is simply that it's so basic.
This is the way I try to explain it. Nature speaks to us in many tongues and they're all alien. Alien, not foreign, alien. What the research scientist, in a basic sense, is trying to do is decipher one of Nature's dialects in some area -- the vocabulary, some of the grammar. To the extent you succeed, you discover she's left messages all over the place we couldn't read before. So in that sense, basic research is just trying to understand. It's the most practical possible [approach] because it enables you to read the messages.
There are plenty of historical sagas we can trace where a huge effort in human and financial resources were devoted with no result, and later you see the reason was you couldn't read the messages that were there all the time because you didn't have the basic science that made it possible to go ahead with the practical. Often with the basic sciences there, zingo! Take the development of the atom bomb. That was a very short time after the discovery of fission, which was motivated entirely by basic science. Nobody was looking for that. Another example that's often cited is the discovery of x-rays. If you'd tried to ask what you could do for surgery, no one would've gone in a direction that would've led to discovering x-rays. On and on. The whole business of nano chemistry -- you hear about so much work and fascinating new materials -- all that came from playing around with molecular beams and discovering that gee, you've got a big peak at carbon-60 when you expanded carbon vapor through one of those supersonic nozzles I mentioned yesterday. What could that be? Well, they played around and said, "Gee, it must have a structure something like a soccer ball." That touched it all off.
So, you don't know what's going to grow from it, but you do know that with basic research, since you're learning something in Nature's language, there's a whole literature there that becomes accessible to you. That's the nature of the work we did. It turns out even though we studied a certain range of reactions, because we could do it in this fundamental level of single collisions and therefore figure out what it meant in terms of how the electrons govern what happens, which we'll be talking about today, it's transferable to a huge range of reactions. I don't know, DNA, anywhere, if you have a certain bonding situation, you can say, oh yeah, that's what's going to happen when they react. See what I mean? So, it turns out relevant to practically, well, really all of chemistry. That's the answer.
Next page: Science and Society
© Copyright 2006, Regents of the University of California