Science Talk

Large Hadron Collider Backgrounder

Thomas LeCompte of Argonne National Lab was the physics coordinator for the ATLAS experiment at the Large Hadron Collider. He talks about the instrument and its future, as we await the December 13th announcement as to whether the LHC has found the Higgs particle

Podcast Transcription

Steve:          Welcome to the Scientific American podcast Science Talk, posted on December 11th, 2011. I'm Steve Mirsky. Rumors are flying about the search for the long-sought Higgs Boson at Europe's Large Hadron Collider. The Higgs is the elementary particle hypothesized to be the origin of the mass of all matter, so finding it would be a pretty big deal. And our man in Europe, Davide Castelvecchi will be at the press conference on Tuesday, December 13th at which the LHC folks will tell us what, if anything, they've got. For a preview of that event, check out Davide's December 8th article on our Web site at Also to preview the announcement, here's an interview I did months' back with Thomas LeCompte. Only one small section has been previously aired as a 60-Second Science podcast. Tom is with the Argon National Lab and is the physics coordinator for what's called the ATLAS experiment at the LHC. ATLAS is one of the two major detectors at the Large Hadron Collider. So Tom is in the thick of it. We talked about the instrument, the culture at the LHC and how he wound up at the most complex science experiment in history.

Steve:          What about the energies that are currently being produced and where that's going to go? We're still going to get significantly higher energies than are currently being produced there? Is that right?

LeCompte:          Correct. So we're running at seven TEV, trillion electron volts, which is three and a half times more than the Fermilab Tevatron which was previously the instrument that was able to get to the highest energies. We expected double that sometime over the next few years as we inch our way up. But for now, you know, a factor of three and a half is a lot. We're delighted to have this new round to be looking at, and it didn't seem to be worth the risk to really push the device. If you buy a brand new car, you usually don't drive it at 120 miles an hour out of the dealership parking lot.

Steve:          Briefly, the ability to reach these higher levels of energy gives us what?

LeCompte:          It helps us in many different ways. One is it just produces more particles that we'd like to look at, and it also produces heavier particles that we might not have been able to see before. So we win twice. Right, we're producing, we've got more of an energy read, so there are new things to see, but we can also see things much more clearly because we have many more events.

Steve:          So, let's talk about your instrument within the instrument is the ATLAS and the ATLAS—A-T-L-A-S—stands for?

LeCompte:          So, it's one of the worst acronyms ever. It stands for A Large Toroidal Apparatus. The ATLAS experiment is 44 meters long, 22 meters high, and it takes 3,100 people to build and operate. So it's really a scale well beyond anything we've ever done before.

Steve:          And that unit is a detector within the vast ring of the LHC?

LeCompte:          That's right Steve. So, there are actually four large detectors, two very large general purpose experiments. Then there's also two smaller specialized ones: ALICE, which looks at lead collisions, and LHC B, which looks at particles carrying the b-quark, the second heaviest quark. So these four experiments allow us to measure things in a complementary way, so that we don't have one experiment, you're not relying only on one experiment to tell you what's going on.

Steve:          But you're involved in the ATLAS specifically?

LeCompte:          Right, I was formerly the physics coordinator of the experiment, of ATLAS. But there are other things that we've done, you know, for example, in the U.S. one of the things we built is one of the sub-detectors in ATLAS called the calorimeter that measures the energy of these particles.

Steve:          I used calorimeters in chemistry class, bomb calorimeters. Is this more sophisticated?

LeCompte:          We like to think so. I'm not sure if that's the way we would characterize it. The word bomb is never supposed to appear anywhere near the LHC, but the idea is basically the same. You take a particle that contains energy, it gives up this energy in some way, and then you measure that. So in a bomb calorimeter, it's giving up this energy to the liquid that's inside it.

Steve:          Just measure the temperature increase?

LeCompte:          And you measure the temperature increase, and in our calorimeter, it gives up energy and instead of getting the temperature moving up, light is emitted, which we turn into electricity. So the idea is the same. That's why we use the word, even though the technology and the scale is completely different.

Steve:          It's only about, what, 200 years better technology?

LeCompte:          Right, and it's designed to do a slightly different thing, right? It's trying to measure, you know, what we think of is a lot of energy, but of course, you know, it's still microscopic. So to develop a device which can measure a fraction of an erg is really a remarkable thing, particularly when it's spread out over an area the size of a house.

Steve:        So we're using these vast amounts of energy to then accomplish something that gives us a fraction of an erg to look at.

LeCompte:          That's exactly right. And then we do it again and again and again. So there is, you know, we have to measure this energy and then in less than a microsecond, it has to completely clear out, so that we've got the next event and then the one after that and the one after that. So, the total energy is large, but the energy per collision is actually rather small, we just have very many of them.

Steve:          So, the collisions are between protons. We're sending protons in different directions. They're going to smash into each other in the ATLAS detector, and then we're going to look at what happens?

LeCompte:          That's exactly right. Would you like a job as my graduate student?

Steve:          I've tried that already. I'm much better at this. So, what do we hope to see there and what do we—let me channel Donald Rumsfeld—what do we know we don't know and what do you think we might not still even know we don't know.

LeCompte:          The unknown unknowns and the known unknowns.

Steve:          Exactly.

LeCompte:          So the thing we hope for more than anything else is the surprise, the thing nobody expected and could predict. And whenever I say that to reporters their first question is, "Well could you give me an example?" But we also know that the theory that we have now of how particles behave called the standard model is mathematically inconsistent at high energies. It gives you probabilities that are greater than one and less than zero. So, we know that something new has to happen at an energy scale somewhere above the Tevatron and probably within a factor of roughly 10. That's why the LHC has seven times as much energy and 10 times as much delivered beam than past experiments to give us this extra factor of 10 reach to see where things break down. Now people talk a lot about the Higgs mechanism. The Higgs particle is one of the many possible ways that this can happen, but it's certainly not the only one. So we're going to be looking very carefully in this new regime that we can look at to figure out exactly where our present model has to be changed. So that it becomes mathematically consistent. Now the more interesting question is, Will this get us to the true theory? Or will it just put a band aid on it and let us go up another order magnitude before we see where the problem is—you know, the next level. And of course, you know, we could guess that but until we have some real data, you know, speculation is just that—speculation.

Steve:          So, for people who don't know, and I'm one of them, the protons smash into each other and you produce other particles and traceable energy. When you look at all that, what do you then do with it? What does it mean when you look at it?

LeCompte:          So, this is collapsing many years of a PhD into a few moments.

Steve:          That's what we like here.

LeCompte:          Mostly what we do is we look for commonalities. We look for different events that share a common feature. So, one possible common feature would be events that have a large energy imbalance. Everything goes to one side of the detector and not the other. Now getting one or two of these, you expect to get a certain fraction of them from processes we know about, but you know, and also from mismeasurement, you know, we don't have a perfect detector; but if we were to see a large excess of this, this would be a sign of new physics. Another possibility is you see particles that when you combine them, they clump in a particular region; for example, they all have the same mass. That would be indicative that you've discovered a new particle that's decaying into these other guys.

Steve:          You don't mean they physically clump with each other. You mean the qualities of the particle are similar to each other.

LeCompte:          Exactly, that if you look at, you know, a large number of events that you see the kinematics—that is the decay properties, all looking very similar. So if you plotted them in a graph, you might see a bump or if you plot them in a 2-D graph, you'd see a dark spot; or a light spot, right, if you see an area which is unusually depleted of events that would also be something that's interesting. So, we look to find things in events that look like other similar guys, and then we're more convinced that what we're seeing is really something which is produced by nature and not something which is just an artifact of the experimental measurement. That's also why we have multiple experiments, right. We've got two independent teams, so that if one has a really clever analysis idea, you know, you can compare it with what's gone on in the other experiment to make sure that this analysis idea isn't creating new science, it's only discovering what's really there.

Steve:          So, do you like to guess about what you think you might find or are you more sanguine with just doing the experiments and allowing yourself to be surprised?

LeCompte:          Well, I could argue that I have professional theorists whose job it is to guess. But, you know, nobody ever got fired for discovering the wrong thing.

Steve:          As long as you discover something. And, in fact, Steven Lindberg said, if we don't find anything that'll really be interesting.

LeCompte:          Right, you know, and of course, something has to happen because if you ask how often a certain process occurs, and you get a negative probability, you know that can't be it. So we know that even in the world where nothing happens, something somewhere has to change. There has to be a qualitative difference in events, if you like, above the TEV threshold from events below the TEV threshold and by looking for them, we're hoping to be able to see something.

Steve:          If you see popular accounts of what the LHC does, you'll often see described in terms of "trying to approximate the conditions at the big bang". What's wrong with that description?

LeCompte:          So I don't think it's wrong. You've noticed I haven't used that analogy. One of the things that's different is this is a much simpler situation, right. We have a lot of energy in the small space, like you did at the big bang, but we don't have a whole universe in a small space. And indeed when we run protons, we only start with two particles, rather than the, I don't know, 1050, or however many particles there are in the visible universe. So it's a much simpler system, and it's a few-body system. If you really want to understand what's going on with the early universe, you have to understand how all of these interactions are going on. That said, it's not a completely bad analogy either. We know that the universe is expanding and cooling, so earlier on it was smaller and hotter, and we're studying the properties of small, hot things. But I don't like saying that because it gives people the idea that the only thing we're doing is really trying to, you know, turn back the clock. That's one direction we're looking at. We're also trying to just in general study the behavior of matter, energy, space and time on small scales.

Steve:          What is it about this field that attracted you?

LeCompte:          How much tape do you have here? (laughs)

Steve:          It's digital—there's no tape at all.

LeCompte:          So, my background is actually little bizarre. As an undergraduate, I went to M.I.T. and I was going to be some kind of engineer, and I really didn't know what kind of engineer. And the institute is very good about letting you explore. And then after a couple of years, I realized I had to graduate, and I looked at the classes that I took, and they were all physics classes. So it wasn't that I, you know, at one point decided, "Today I am a physicist." It was something that I was just, sort of, you know, slowly drawn to. "Ooh! This is interesting!" It's like getting a page-turner, you know? You really need to see: What's the next thing? What's the next thing? What's the next thing? I think this particular kind of physics attracts me because it's a very basic, it's a very fundamental sort of physics, but it also requires you to know an awful lot about applied physics. I am an experimenter; I need to make the detector work. So, I had to know a lot about the material properties of what I am looking at. The electronics, I need to know exactly what it's going to be doing to the signal. You were told that ATLAS is sitting, you know, many tens of meters, almost a hundred meters, underground, right? That signal has to get propagated up, right. Well I need to know what it looked like when it was produced, not a hundred meters later. So it's a real mix of the most fundamental plus the most prosaic. One of the things that we had to worry about is what kind of paint to use.

Steve:          That's really interesting. Why was that an issue?

LeCompte:          We have a detector made out of iron, and of course, you don't want it to rust, and you know, that would be bad, and it has scintillating tiles slid into it, which do the actual read out.

Steve:          They'll let you know when some particle hits them.

LeCompte:          Right they produce light. When the particle comes out, we can then collect the light. Well, it turns out, and of course, you want to have minimum amount of gap between the tile and the iron because you could have put more tile or more iron in, right. The air doesn't do anything for you. Well as it turns out, the paint we were using was a little bit too thick and the tiles wouldn't go in right. So we had to have a crash course in paint dynamics to understand the proper kind of paint to use.

Steve:          So, some chemists actually got some work.

LeCompte:          I don't think it was so much chemists, but a lot of technicians painted a lot of things, and we went through to measure the amount of paint that we had added. So these are the sorts of really prosaic kind of physics without which the experiment just never runs.

Steve:          That's a fascinating story. Has that been covered? I don't remember seeing that anywhere.

LeCompte:          I think this may be an exclusive, you know. Usually, we don't talk about paint when we're talking about these experiments.

Steve:          You needed paint that would cover the surface, but in a thin enough layer with one coat.

LeCompte:          Exactly. So that we could slip in the scintillating tile without damaging it.

Steve:          Do you remember what you actually wound up with?

LeCompte:          No, I don't remember what it was.

Steve:          That's fine.

LeCompte:          There is a sociological aspect, which I think is often missed, and that is if you have an experiment with a hundred people, your top 10 people are in the top 10 percent. If you have 3000 people, the top 10 people are in less than the top 1 percent. So I have had the opportunity to work for some extraordinarily talented physicists, and I think this is the sort of thing which, you know, you see 3000 people, and you think you will never make a difference on this sort of thing. But from someone whose seen every physics analysis go through, it's really remarkable how, you know, a few grad students who've got really good ideas can get them through a collaboration of this size. And, you know, not only does the cream float do the top, but we all learn from each other. The way you advance in this field is you collaborate with people who are smarter than you and you learn from them. So this is, you know, the sort of thing that the people who study the sociology of science, I think, really should be taking a close look at; because, you know, people, it's easy to see the organizational difficulties of getting 3,000 people to push in the right, or at least the same direction, the right direction might be even harder. But what's often missed is the chance to work with such really, really sharp people, and I think the experiment as a whole has gained more from that than we ever would have guessed at the beginning.

Steve:          So you can talk to the theorists, but you wouldn't actually want to be one.

LeCompte:          You know, some of my best friends are theorists. I don't, yeah, they work very hard, but I think that there's a difference between being attracted by coming up with possible ways that the world might be arranged, and being the guy who tells you, "This is the way the world actually is arranged" and that's what really drove me towards experimental physics. There's this idea that, you know, the good physicists are theorists and the bad ones are experimenters. I think part of this is because the books are written by theorists. You know, they're written in one of two ways: Either the theorists comes out with a brilliant idea and then the experimenter goes out and measures it, or the experimenter makes a bunch of measurements and the theorist explains it. But I think that it doesn't really do justice to the experimental side of things. This really does tell you how nature behaves, and there have been many times on ATLAS where I was the first guy to make a plot, and I realized I was the only person on the planet who knew exactly what nature was doing in one particular region.

Steve:          Well that's pretty cool. I mean, probably the best example I can think of anyway, for what you are talking about, is the Michelson–Morley experiment.

LeCompte:          Right, so what they were trying to do was to measure something called the ether wind. The idea was that light was carried by this material called ether, and they recognized that the Earth would be going through this in different directions as it went around the sun. If they go towards the ether into the wind, things would slow don, if they went away from the wind, things would speed up, and they were shocked to find out that, in fact, there was no ether wind and the speed of light moved in the exact same velocity, no matter what direction that you looked. And for that brief moment, everybody else in the world thought they knew how the world behaved, but in fact, Michelson and Morley realized that it was completely different.

Steve:          And they wound up informing Einstein, but Michelson and Morley were not in the dark about what they had found. They also appreciated that it was overturning one of the big assumptions in physics.

LeCompte:          Exactly. You know, they expected that they were going to get a number, and the only question is what that number was going to be. And the fact that number was identically zero and, you know, as good experimenters did, the first thing they did is they went out and measured it again, right? You know, ripped apart the apparatus, started from scratch to make sure that they didn't build into the zero, and they got exactly the same number out of it. There's a certain sense of responsibility, for lack of a better word, when you get something that's that surprising to make sure that you've double and triple checked it before you sent the word out. This is why the LHC, you know, some people are saying, "Well you're little bit slow with the results", but we don't want to produce a wrong result because that doesn't do anybody any good. As soon as we're sure that it's right, it goes out the door.

Steve:          The difference between what you're doing in the Michelson–Morley work is they were surprised by what they found, but you hope to be surprised by what you find.

LeCompte:          Right. I hope to be surprised, you know, I'm almost guaranteed some surprise, right. But what I'm really hoping for is the unpredictable surprise, you know, when you look at it and you say, "Wow, I would never have guessed that."

Steve:          And how long do you think we're going to wait before we maybe see one of those unpredictable surprises?

LeCompte:          That's a question a lot like "What kind of unpredictable surprises do you expect to see?"

Steve:          True, but we're not going to just keep running it for 50 years without seeing anything.

LeCompte:          Right. So, these experiments tend to have a lifetime of about 20 years. And what ends up happening in the early phase of the experiment, you know, the doubling time for the data is short. It's two weeks now, that's going to get longer and longer as the machine works better and better, and there's less room to grow. And at a certain point, you're going to be in a point, where it's linear, right. So if you run 10 percent longer, you get 10 percent more data. And it usually doesn't make a whole lot of sense to run much beyond that, right. At that point, you've seen what you're going to see, and you have to start thinking about is there some other way, some other instrument; which might be a bigger accelerator, it might be a more sensitive detector, it might be looking in, you know, some completely different region of—you know, for example, particle astrophysics didn't exist, you know, 10 or 15 years ago, but now it's one of the most active fields. So you have to decide after about 20 years or so, you have to do something that's different.

Steve:          So that's pretty exciting, because you're a young enough guy where you'll be around to see whatever we're going to see.

LeCompte:          That's right. I'm glad somebody is calling me young, but I'll be around for the full term of the LHC and be involved at least in the design of the next big thing after that, even though it will be up to my students to be the people who actually execute that. If you look around at the LHC, you'll see some old people, but you'll see a remarkable number of young people. So, it's not something which is being run by a bunch of 70-year-old bald men; you know, they're telling the younger folks what to do. The people whose careers are the LHC are the ones who are really pushing it and driving it.

Steve:          Remember to check back into on Tuesday, December 13th for Davide Castelvecchi's coverage of the announcement from the Large Hadron Collider. For Science Talk, I'm Steve Mirsky. Thanks for clicking on us.

Share this Article:


You must sign in or register as a member to submit a comment.

Starting Thanksgiving

Enter code: HOLIDAY 2015
at checkout

Get 20% off now! >


Email this Article