WORLD SCIENCE FESTIVAL: ON THE SHOULDERS OF GIANTS

BRIAN GREENE, PHYSICIST: I’d now like to introduce today’s On the Shoulders of Giants, speaker. So it’s my pleasure to introduce Barry Barish. Professor Barry Barish is an experimental physicist at Caltech who was being closely involved in some of the most important physics experiments of modern history. In 1994, Professor Barish became the principal investigator for LIGO, Laser Interferometer Gravitational Wave Observatory that you’ll be hearing more about. Leading the ambitious project that went on to observe gravitational waves this past September, Professor Barish became LIGO’s director in 1997 in the same year, founded the LIGO scientific collaboration and international collective of over a thousand scientists who are dedicated to the study of gravitational waves. And recently, Professor Barish has turned his efforts to another ambitious project, director of the International Linear Collider Global Design Effort from 2006 to 2013. And with that, please join me in welcoming Professor Barry Barish.  

BARRY BARISH, EXPERIMENTAL PHYSICIST: I’m an experimental physicist and what I’m going to try to do today is try to show you by one case example, which is basically gravity leading up to gravitational waves. How science works, physics as an interplay between experiment and theory, how it makes wrong turns, where maybe personalities come in. So I’m going to trace the history as a case example up to and including gravitational waves with a hint at what comes next for us. If I get through all this. So we’re going to start with Newton, but before we start with Newton, and to make sure I get to it, basically this is the reason why several of us are here, which is a picture of a computer simulation of what fits an observation that we made last fall and reported last February of two black holes going around each other and merging into one object. And that’s what we observed.

So what is that? Two black holes, a black hole is when there has been a gravitational collapse of something, of a large star, for example, giving such intense gravity that nothing can get back out. So the simplest explanation of a black hole is that’s what it is. The most important thing to realize is that it’s incredibly small. The size of each of these black holes is about the size of Metropolitan New York, but it weighs 30 times the mass of our sun.  So that’s why the gravity is so incredibly strong. Two of them are going around each other, much like the earth around the sun, only they’re equal in mass pretty much. And slowly, they’re losing energy and merged into each other, as you saw in that, in that picture. While they’re doing that, in order to merge into each other, they’re losing energy and that energy, it comes out in the form of gravitational waves. Which 1.3 billion years later we’ve detected. So this thing happened 1.3 billion years ago. We got our technology good enough to actually see it when it came through the earth last September. So I’m going to come back to that. We’re going to lead up to it. I’m going to start with Newton since this talk was even labeled after a famous saying of Newton’s and where he started with gravity, how it led to Einstein’s gravity and then to us both experimentally and theoretically. So I think we all know Newton’s theory of gravity is probably one of the most successful theories in science. It lasted more than 200 years and explained an awful lot. He recognized, as you kind of see in this picture what happens when you have two massive objects separated from each other. He somehow put together a connection before we know it, that when the apple falls from the tree and the moon goes around the earth, that they’re governed by the same theoretical reason, gravity.

And so that’s the word universal gravity that he developed. And he made a formula which you’ve all seen in school. That is that the force between two massive objects is a product of those two masses. It’s inversely proportional to the square of the distance. You separate them. And it’s governed by some constant G, which I’ll talk about. He did more. He actually, in doing this, he did the calculus and proved that you can take the distance between these two and just take them as if all the mass was at the center of each one. So it’s not some complicated thing that you have to worry about where each mass is, but you can take the center. He proved that if you had an inverse square law like this, that it would lead to elliptical orbits around the sun, which was a thing that anything going to be the planets or the moon you’d have to have.  And that all led to something that he then published eventually in the Principia.

[00:05:19] BARISH: That’s the clean picture. Just like when you read our paper, the picture you get is that a hundred years ago, Einstein predicted gravitational waves. A hundred years later we detected them. The actual story has more complexity than that. And even this one does. It turns out that that’s what we know about Newton. But the problem is actually more complicated. There was another very good scientists at that time. This is in Britain at the Royal Society, named Robert Hooke and he was a physicist. And we know him mostly for kind of his freshmen physics law, which is shown on the right. That is if you have a spring and you put a weight on it and stretch it, it’ll stretch twice as far. If you put twice as much weight on there, twice as much force on it, but he also was a very broad scientist. He got very enamored with the microscope and he studied a lot of geological things, rocks and things. But he got interested also in old fossils and he studied old fossils and actually it was one of the first people that came up with a conclusion that there was biological evolution, so Hooke was actually a pretty sound scientists. However, he ran into Newton. He claimed at one point, this is before the Principia was published, that he invented the inverse square law and that he had written a letter to Newton, which nobody’s ever seen and it became a dispute between them. Newton, well he wrote the Principia. Then he said he wouldn’t publish the Principia because of this dispute. When he wrote it finally and published it, there’s no mention of Hooke anywhere in the Principia, which is kind of a shame, but that’s the fact. So it just starts with confusion, which I’m going to try to emanate. The second thing that happened is that we talked about this formula here, but we didn’t talk about, G, the strength of the interaction. That took a hundred years to figure it out, what G was, and I’ll show you how that’s done. But also, there was no sense of what causes force.

This is an empirical formula, but no one conjecture what caused the apple to fall to the earth. There are all kinds of ideas that you read about, for example, planets around the sun which are all wrong, that the sun has some sort of magnetic field that basically did it, which isn’t the reason. So basically this is an empirical formula and the next problem was determining the strength of this and whether it was universal and that was done a hundred years later by Cavendish. And he did a very clever experiment. He built what we call a torsion balance. And that’s a long wire and on the end of the wires is a rod and he put two lead weights on the end of the rod and then he did all the careful work of twisting it, letting it go, how much he had to strain it, to twist it, to calibrate it all. And then he brought up two heavy lead weights next to it and measured how far it moved. And when he did that, he then could determine the strength and the laboratory of this constant G and he got an answer which I have here, 6.75 10 to the minus 11 units that we use. Interestingly, he was almost exactly right. You can see I put just for reference with the number is now. And so this is impressive. And uh, that was kind of the experiment coming into it. The next thing that happened is celestial mechanics. And this is kind of an interesting story about how science proceeds. First of all, this theory was incredibly successful. It aligns with Newton’s. It explained everything from the orbits of the planets, which I talked about already, to the tides, or even the escape velocity from the earth, so everything to do with gravity on a macroscopic scale. And then 200 years or so, there was only one place where it didn’t work and that was the orbit of our smallest planet, the planet closest to the sun, Mercury, which has a orbit around the sun, about every 88 days. And as it gets near the sun, it gets perturbed by the presence of all the other planets that are around.

And if you calculated it using Newton’s formula, you didn’t quite get the right answer. So the answer that was gotten was 500 and can’t read this 532 arcseconds per century from Newton’s theory. And the actual measurements which were good, were 575. So there’s a small discrepancy that existed. And for some people, maybe that is at least part of the motivation or part of the success of Einstein’s theory which came along later. That’s not quite true either. There’s another guy named Verrier, who was a mathematician and did celestial mechanics, and he basically made celestial mechanics a success as a field and respected by doing the following thing. He basically use the formulas of Kepler and Newton. He calculated the orbit of Uranus and found that there were discrepancies in that orbit from those, those theories and those formulas.

[00:10:36] BARISH: And he predicted then that it was because there was a missing planet Neptune, which is the name that he gave it, that it was missing. And he predicted then where it would be. Because otherwise it wouldn’t fix the problem, so he predicted where it would be. The story goes that then he sent a letter to an astronomer in Germany, in Berlin, and it took five days for the letter to get there. He took into account those five days. He knew how long it would take. The astronomer got the letter. This is the story at least. The astronomer got the letter. He then looked up at the sky and the place that was predicted by Verrier. And within one degree, he found Neptune. And at least the part of the story that is true is that he predicted Neptune, I don’t know if all the colors are true, but this makes celestial mechanics a viable subject. He went on. He’s the guy that saw that Mercury had an orbit that wasn’t quite right. So he tried the same idea on Mercury. So whether the Newton theory’s right or not, he said, OK, it’s going to be just like Neptune and Uranus and there’s something missing. He called that Vulcan. And he said it’ll either be a small planet or maybe a series of small objects that were between mercury and the sun and predicted what they were, gave them a name. And they’ve never been found. They’ve been looked for even by NASA in recent years, but they’ve never been seen. Of course they were reinvented for Star Trek. So those of you, so that we know. So that’s basically where the situation was when Einstein came along, there was not a great case, that Newton’s theory didn’t work. In fact, they didn’t find Vulcan, but maybe the experiments were good enough at that point.

So some people think that that was the motivation. Einstein had developed the theory of special relativity in 1905.  And he extended it to include accelerations, which is eventually gravity and 10 years after his three papers that most of us in physics, idolized kind of in 1905, he came out with a theory of- his theory of gravity. And that theory of gravity, forget the formula detail. That theory of gravity basically had the feature that it unified all of space in space time, one universal or one-world space time. Of course, my little laptop doesn’t understand that. Our Mr. Gates because it puts little waves under space time when I typed it in. So you still have to teach them that spacetime is a word. Anyway, Einstein came up with this theory and in addition, it gave the right answer for mercury. But that’s not an enormous triumph obviously, but it was a good start that it gave the right answer for Mercury around the sun. And when we come up with a new theory of nature, one of its ingredients of course, is to explain things that haven’t been explained before, but that can maybe be done non-uniquely. You can make many theories that’ll explain something that you didn’t know before. In this case, the theory was good enough to predict some new things. And the thing that it predicted that’s most famous, was the bending of light. The calculation of exactly how much light would bend if it went near a massive object. Einstein’s theory as the thing that causes gravity is a curvature of spacetime. And so if you get near a massive object, it doesn’t matter whether you have something of mass going by or something that’s not massive like photons going by. They’re going to follow the path of the distortions of space-time and curve as they go by. So he calculated the curvature that it would have. By the way, it also curves in Newton’s theory, but not as much. And he calculated the curvature that it would have. And that was followed up by Arthur Eddington, who in 1919 went to the southern hemisphere on an expedition and looked at the bending of light through an eclipse of the sun where you could see galaxies move behind. And he got the right answer, or basically he got the same answer as Newton-as Einstein had predicted. This was a great triumph. It’s a thing, it is the single thing that made Einstein famous worldwide that hit newspapers all over the world.

[00:15:20] BARISH: And, basically not his wonderful articles in 1905. But the bending of light was what captured the public. In about 1920. I’ve looked at that data from Eddington, which other people probably have now, and let’s say the requirements of what would be convincing to the scientific community we’re a lot different than when we did LIGO. Our big issue in seeing gravitational waves was to produce what seems to be the standard you have to do, which is what’s called 5 sigma. That’s that it’s so insignificantly probable that it happened by some statistical fluctuation that you can ignore that. It has to be that you did something wrong. In the case of Eddington, this was nowhere near five, six by any stretch of the imagination. But it was right and it was repeated a couple years later and of course that is history. So it was right. That same feature of the bending of light can actually be seen if you go to the southern hemisphere and some of you probably have seen it and that’s called the Einstein Cross and it’s a quasar image that appears around a central glow from a nearby galaxy and it’s visible and looking like this cross and that feature so you can see with your naked eyes, but for astronomers, this idea that you basically can have a curvature around objects means you can look at objects that are dark and see the curvature around them and it’s become kind of a cottage industry for, for astronomers, what’s called Gravitational lensing. So that basically came out of this whole thing. I don’t know if that convinces you that Einstein’s theory has anything to do with the real world, but this one will and that is all of you, maybe not as much in New York as we do in Los Angeles, but you drive it around in your car using GPS and GPS relies on general relativity.

I’ll explain that to you. Maybe I use a picture of somebody walking and just because it’s New York. So if you want to walk and guide yourself around, then you have this picture. So what is this? The reason why general relativity matters for GPS and is a good illustration that GPS really, that Einstein’s theory of general relativity is relevant to all of us. First, these  satellites are going fast. They’re going 24,000 miles an hour and we have something we learned as soon as we learned special relativity and that is moving clocks, tick more slowly. So that’s true. We applied special relativity, which is something that’s well established. And if we do that, we find that we have to make a correction. I’ll tell you how big the correction is, but how much it matters in a minute. A correction of minus seven microseconds per day, so just because of special relativity. The fact that they’re going fast, the actual clocks are going to go slower and we have to correct for it somehow. Of course you can put something right in the satellites to do that. But that’s not the whole answer. There is actually a bigger correction that comes from general relativity. And that’s the fact that the satellites are have about a quarter of the gravitational field that we have here on the surface of the earth because they’re way out. And because they have less gravitational field, they have less spacetime warpage around them. And in that case, clocks move faster. And in fact, if you calculate it with general relativity, they gain 45 microseconds a day. So the total is the difference between those two or the GPS correction is 38 microseconds a day. Does that matter? Yes. The accuracy that you need to put us on the road is about 30 nanoseconds. In order to be 10 meter resolution, to be on the road, you have to have something like 30 nanosecond resolution, which is a thousand times better than the 30 microseconds a day. Or if you divide by the number of minutes in a day, every minute or so, you’re going to wander off the road if you don’t make the correction. So we need that correction. And so you should believe that general relativity matters and actually works. This is a regime of general relativity that’s less interesting for us as physicists. It’s what we call the weak field limit. And there it works. It works in other ways, but this is kind of the most dramatic I think for those of us that walk around the streets or drive our cars. We’re interested in what I’ll get to later in LIGO as creating a laboratory where we can study general relativity where the theory is most interesting and most in question. And that is when the fields are incredibly strong, like in these black holes. And that’s why setting black holes become so interesting to us in terms of testing and understanding general relativity.

[00:20:25] BARISH: So Einstein then had put this theory out of general relativity, and a year later, he predicted gravitational waves. He did this, not rigorously, but in analogy, in a sense to electrodynamics where we have electromagnetic waves and the similarity in the equations. And so he predicted gravitational waves in exactly 100 years ago or in the winter of 1916. And he wrote a paper and that paper was badly flawed. But it’s the paper we refer to when we say that a hundred years ago Einstein predicted gravitational waves. That paper has a series of errors in it, but one of them is a factor of two in the magnitude of the gravitational waves. The idea that there were gravitational waves was right. He wrote a follow-up paper two years later in 1918. And in that paper, he fixed the errors, fixed the error, made it a factor of two. It’s still not a rigorous derivation of gravitational waves. But it has an important element. It basically defines the source. So the source of gravitational waves, he shows in that paper is from a quadrupole moment or a quadruple formula. And to remind you, when we have electromagnetic waves, which we compare with, electromagnetic waves come from what we call a dipole moment. If you have a plus charge and a minus charge and you wiggle them to each other. They emit electromagnetic waves depending on how fast we do it, we have microwaves or infrared or ultraviolet or whatever, or radio waves. Same thing here. We have a quadrupole moment instead, that’s four, instead of a dipole moment in order to make gravitational waves. And so we know how to produce them if we wanted to. And again, how fast you do it will depend on what the frequency is. If those ways, just like electromagnetic waves. They’re basically produced, I’ll come to in a second, they’re produced, I want to bring in one other point in first. The gravitational waves, he also shows, travel with the speed of light. So the fundamental difference that matters kind of conceptually between Newton and Einstein, is that Newton’s theory when the apple falls is that you have what we call instantaneous action at a distance. So wherever we are, when it happens, we detected immediately. There’s nothing in between. Well then Einstein’s theory. We produce a messenger if it’s gravitational waves that travels at the same speed as he has in the problem, which we call the speed of light, but it’s basically the speed and the problem. And basically then it takes, in our case, for the example I showed in the beginning, 1.3 billion years for the signal because it was distant to get to us that we’re going to show in a little bit.

And so Einstein has a propagation. That propagation is not, however, like the propagation of light waves, which is particles. It’s actual distortion of space and time. The story doesn’t end there. As I said, the stories become quite contorted. Einstein moved to the US in 1933 from Germany. rote two famous papers, Einstein, Podolsky, and Rosen, which was a famous paper about quantum mechanics. And the second one about with Einstein and Rosen that had to do with what we call wormholes now. He didn’t call it that. And then he wrote a third paper in 1936 with Rosen. That third paper was entitled “Do Gravitational Waves Exist?” You might guess from the title that he had concluded almost 20 years later that they don’t exist after he had proposed them earlier. And he submitted this paper to Physical Review. This is the man that received the paper. John Tate was the editor of Physical review at that time. Physical Review was our main journal certainly at that time. John Tate then took the paper and he send it for peer review. And we know now from the records of Physical Review that he sent it to this guy, Howard Percy Robertson. Howard Percy Robertson was also a general relativist. He was on sabbatical from Princeton. And was at CalTech, my institution. On sabbatical when he got the, the paper to review. And you can see on the second line is Einstein, Rosen and the dates they came and he sent it back. He actually found the mistake in the paper.

[00:25:13] BARISH: It had to do with the coordinate system. The fact that Einstein used a single coordinate system to cover all of spacetime, which is very difficult and found singularities that he thought basically were false and where the reason that they were assuming they were gravitational waves, but there weren’t. And Robertson reformulated it in a different coordinate system, cylindrical coordinate system, and showed that the singularity went away. Did his review, sent it to Mr. Tate, who was the editor of PhysRev and Mr. Tate then send a letter which I’ve seen, but I’m not duplicating here, back to Einstein that was very mild. He basically said the part I took in quotes that he would be glad to have Einstein’s reaction to the referees comments and criticism. Nothing about whether you publish it or not. Einstein’s response. And he actually wrote a written response and that written response, which you wrote in German, but it’s translated was, “Dear sir, we Mr. Rosen,” who had gone back to the Soviet Union by this time, “and I had sent you our manuscript for publication, but had not authorized you to show it to specialists before it’s printed. I see no reason to address the in any case, erroneous comments of you’re anonymous expert. On the basis of this incident, I prefer to publish the paper elsewhere. Respectfully, Einstein.” There’s a P.S saying that Rosen agrees with it even though he’s gone back to the Soviet Union. So he went on just as he said here, and he sent it to a more obscure journal. It was actually the journal for the Franklin Institute in in Pennsylvania. In those days, we didn’t have kind of a high tech way that we publish now. The publications had to be set up and presented and read and it took months.

And so this was percolating through the system at the Franklin Institute. They accepted the article as was written, which was exactly like the one sent to Physical Review. In the meantime, Rosen had gone back to the Soviet Union and Einstein had a new assistant who was an early important cosmologist, Leopold Infeld. And Infeld then was important in the final resolution. So the sabbatical of Robertson, although these are anonymous reviewers, so nobody knows what happens to the reviews that are done, returned to Princeton. And when he returned to Princeton, he had no idea what had happened in this. But he happened to befriend Mr. Infeld. So after he met Infeld, they in talking physics, he mentioned to Infeld that he didn’t believe this conclusion. And Infeld showed them the proof that he had of Einstein’s and Rosen’s idea. And  Robertson continued to then basically show him where the error was. So Infeld went back and discussed this with Einstein. And Einstein responded saying he had already found the error. In the meantime, the article came back from the Franklin Institute, got rewritten. He changed the title, change the substance. And basically if you read the first sentence, it says basically that he had rigorous solutions of the cylindrical gravitational wave. And he presents some in his paper. So it became a paper that carried the proof of gravitational waves further. So he wasn’t embarrassed by this. He’s never officially come out against gravitational waves. However, Einstein never published anything again on gravitational waves and he never published anything again in physical relativity. To be fair, peer review is a system that was invented not long before that. It’s probably true. He had moved to the US in 1933 that he had never experienced peer review. So in his articles published in Germany, he probably never had peer review. His two famous papers that I mentioned as far as we can tell, both were accepted by the editor at that point who had the- he could send these out to peer review on discretion at that point. And so the two articles, the one with Podolsky and Rosen and the one on wormholes that he published weren’t reviewed. So this is probably the first review that he ever experienced. So there’s his side of the story as well. But anyway, all’s well that ends well. The next thing is as an experimentalist, we’d like to control all the variables. So if we do a measurement that’s hard or easy or whatever we do, you really want to have everything in your control so the shoe can understand all the aspects of the source, the ends, and in our case, the analogy here is to go back to Hertz in the 1880’s, who demonstrated for the first time electromagnetic waves and he did this in a laboratory by starting with a source of making electromagnetic waves, going far enough away so that you can see the wave nature, and prove that there were electromagnetic waves.

[00:30:37] BARISH: And this is a famous experiment. If you wanted to do this in the laboratory for gravitational waves, what would it take to do that? And why don’t we do that? So let me just give you an imaginary experiment. So just like these two objects that go around each other that we’re going to talk about later, black holes, we could sit in our laboratory and make something similar by making a big barbell and putting a lot of weight on each end and rotating it just like that. And it’s going to emit gravitational radiation. So I’ve put in numbers here just to give you a feeling of what that will take. You can ignore the little formula, just tell us what the strength of the effect is. And so I’ve put kind of unreasonable but trying to make the numbers larger or unreasonable numbers in. We took masses of a thousand kilograms for the two masses, separated them by a meter, spun it a thousand times a second, get out of the lab because it’s pretty dangerous, and go away about 300 meters. You actually want to go much further than that to see the wave nature. But I’ve tried to make the effect as big as we can. Then, so we spin it, then the effect is in, just remember the number 10 to the minus 35. So this is the fractional change that’s created by a gravitational wave. The numbers that I’m going to show you, from the object that we, that we’ve detected are more like 10 to the minus 21. 14 orders of magnitude. That’s zero, zero, 14 times larger than this. And we can barely detect it. As you can see, I’ve spent years developing the technology to do this. So we’re nowhere near the ability to do what an experimentalist wants, which is to carry the whole problem from start to finish inside our laboratories. So instead we’re forced to look out in the universe, of course, that in this case, turns out to be very fortunate because what we see out in the universe turns out to be so interesting in its own right, so don’t just measure gravitational waves. So here’s the same numbers just for completeness, for the object that we saw that I’m going to talk about that because something like 10 to the minus 21, again, their weight, their mass is roughly 30 times the mass of the sun. Their radius is a  hundred kilometers. And the frequency where we detected is roughly a hundred Hertz. And if we do that, then, and the distance away, which we’ve measured. If we do that, we get a strength of something like 10 to the minus 21. About 14 orders of magnitude stronger than this barbell, which is good. So that’s why we go out into space.

So we’re looking then driven this way for astrophysical sources that can give us signals in the detectors that were, I’ll talk about now a little bit, that we build. So what is the effect? The effect is that if a gravitational wave comes, it distorts spacetime in a way shown in this picture at the frequency that is coming through you or me or our detectors. It’ll make in one direction a stretch and at the same time a squash in the other direction and then go back and forth at the frequency of the gravitational wave. So in this case, it’s once a second or so, but and we try to measure that. This is just on this side, just the picture of that statically. But it basically goes back and forth, and our problem then is to make an instrument, if we want to detect these, that can measure this difference in the two directions. Fortunately, it’s one of the best kinds of instruments we have, what’s called an interferometer. So we want to measure this with interferometer metrics techniques. But that’s not where the problem started. So the problem of doing it experimentally began with a man named Joe Weber. And began in the 1960’s. And he had the idea that you could take, a very sensitive way to measure it, would be to take a great big bar of aluminum in this case, instrumented so that any changes in its shape could be detected easily. And it could be detected even if they’re very, very small. And if a gravitational wave came through, it would distort the-distorted. And we know that the amount of distortion or what happens in a big object like this is it has a resonant frequency. If you hit it, it rings at some frequency. So the advantage and good thing is that it has a lot of sensitivity near its resonance frequency.

[00:35:15] BARISH: The bad thing is it doesn’t have very much width. If you get off the resonant frequency, you can’t see anything very much. It won’t have any sensitivity and the instrument that we’re going to talk about in a while, the interferometer, we have a broadband ability to see, so that’s an improvement. But the concept was his and he pursued this. He was at the University of Maryland. He had been at Princeton before and had come up with the idea. He never detected gravitational waves. I’ll show you that, but we have a long legacy of what he’s done other than in addition to basically beginning the field experimentally. So Joe Weber really was the pioneer that began the experimental efforts to see gravitational waves. But if we look at LIGO, which I’m going to come to in a while, we’ve actually benefited from some of the techniques that he invented. I have three listed: sensitivity calculation and noise analysis that he did is similar to what we do. That is understand what it is that gives you signals besides the thing you’re looking for. That he did a coincidence for background rejection, just what we do, we have two detectors, one in Louisiana, one in Washington, and we asked that they be coincident within the time resolution of the gravitational waves. And lastly, he even explored how to measure what the backgrounds were by looking at things that had the same timing and things that don’t have the same timing. All three of those are basically legacies that we’ve inherited and use as part of the basic scheme for detecting gravitational waves even though the technique is quite different. The sensitivity of the bars, as I said, it’s limited by their narrow width and that practical size that you can make it.

I showed you. It’s a fractional changing in stretching or squashing and so the bigger you make it the better. And you can only make a bar so big. This effort though to do bars went for a long ways because clever people decided you can make it better by cooling it. And until we got LIGO working in the early part of this century, these bars were running as one of the main search elements to look for gravitational waves. Now they’re in the storehouse. There’s another thing that came from him and that is student Robert Forward, was one of the first people, so in the US to actually look at the idea of using interferometers. He worked with Weber and then he went to Hughes Research Labs in California and actually built a little interferometer. So he had that history. This idea was then picked up by my colleague, Ray Weiss, who developed the real ability to what it would take to make a real interferometer that would be sensitive enough a few years later. So Weber, though, unfortunately as great as he was in bringing all this, was a much better technologist and visionary than he was as the scientist. In 1969, he published, even though Einstein couldn’t, he published in Physical Review the discovery of gravitational waves, and this is the figure from that paper. So the first, this is his first discovery. And that’s what he said is he saw a blip, one at Argonne lab near Chicago and one at University of Maryland. This proved to be wrong. And unfortunately this happened more than once in his career, so he somewhat spoiled his reputation by having wrong science results, which is devastating as a scientist, despite the fact that he made such an incredible impact on the field. We have another way that gravitational waves have been observed and that is indirectly. So if you look at the picture here, we have two objects called neutron stars, one going around the other, and again in this elliptical orbit, radiate some gravitational radiation.

[00:40:17] BARISH: This was Taylor and Hulse and they were studying this fast rotating one on the left here, which was rotating at 17 times a second. Notice that it had a modulation of about once every eight hours. And that was due to the fact that they were a pair. Once they discovered there was a pair, then in subsequent years, it was the period of the long period, the eight hour period or almost eight hour period was measured over a period of time. And as it radiates gravitational radiation, that period gets a little faster and a little faster and in a million years or so we were detect it in LIGO.  But at this point, the data that they had is published, is shown on the picture on the right, and in that picture you see a bunch of dots and the vertical scale. It’s years. I’m sorry, in the vertical scale, it’s seconds and that’s how many seconds faster or shorter the period got. And on the bottom scale is years. So over about 20 years, the actual period of almost eight hour period shorter by about 20 seconds. And they measured that very accurately as you can see by all those little dots. The line that’s shown on top of the dots is the line that’s calculated from general relativity. So knowing all the features of that system which they had measured independently, you can calculate the line. It’s not a fit-the impressive thing about this to me is it’s not a fit to the data, it’s a line drawn from the parameters that they have, which is unbelievably impressive, that you can actually calculate independently and it falls right on the line. So this is a very strong indication that gravitational waves exist and caused this effect. If we wanted to measure this system this way, we’d wait about a million years and it will be in our frequency band. OK, so now the direct detection. And I’m going to do this in a kind of qualitative way. We have the same problem. What we want to do is see the squashing and stretching of free masses as they move through time. So I’ve drawn here the technique that we have superimposed on a circle of free masses. Imagine masses that are free to move as they’re there. So an interferometer measures basically the time it takes for light to go down the arms parallel- horizontal, the light going vertical. We basically send the light, both split the light from a laser, send it in both directions, bring it back. And if it comes back at the same time, we can orient things so they cancel each other. But if one arm gets a little longer than the other because a gravitational wave came through, then they won’t cancel at that time and we measure that in time evolution. The details of how that’s really done, which is very important and very clever and very difficult were explained by Ray Weiss, my colleague yesterday in a lecture and you just have to go onto the website here whenever they put it up and see how the interferometry really works to do this. I don’t have time to do that today, but it stretches and squashes like this. So that’s basically what we’re doing is changing the length of the arms and then measuring the signal that we get. We have to have free masses for two reasons. One is we want free masses like the picture conceptually, but we also want to isolate the masses from the earth itself. And we do that by hanging the masses from wires. It’s a pendulum basically, and it wants to move and shake itself, move itself. And it has a natural frequency, but that frequency is lower than the frequency band we work in. So we’re able to have them hanging from a wire, use all kinds of isolation and make them as isolated from the ground as possible.

I mean, let me just show you in a little video that I didn’t make this idea just to emphasize that light comes from a source, which is the laser, goes through a mirror, gets split, goes out the two arms, comes back, almost cancels, but if any light goes to the receivers, then we see that light. This is just showing the same idea with waves themselves because the light comes in different waves and how they come back and they cancel or nearly cancel. And then what we see in the detector. So that’s the basic idea. And I said it as a more quantitative explanation, just go to Ray’s talk when it gets posted. He does it in nice detail. OK. So how small is this change that we want to measure? That’s the technique. And you can learn more about the technique from Ray, but how small is it? We ultimately want to go to something like 10 to the minus 19 meters. We talk about 10 to the minus 18, but it’s 10 to the minus 19 is kind of our goal. And how small is that? Let me remind you how small it is. A meter, we all know how big a meter is, basically that long. The human hair, some of us have some left, is about a hundred microns in width. So that’s 10,000 times smaller than a meter. OK. We can all handle that. The wavelength of light, like a light in the laser beam, is 100 times smaller than that. That’s basically one micron. The atomic diameter now getting to things that are a little harder for us to visualize, but the atomic diameter is 10,000 times smaller than that. And that’s 10 to the minus 10 meters. A proton getting into something that’s basic and used in particle accelerators for example. It has the size of 10 to the minus 15 meters. That’s 100-a thousand times smaller than the atom. And we want to go a factor of 10,000 smaller than that. So that’s the magic and maybe the reason that people thought this would never work, but it does. So that’s the goal and you have to do a lot better than a simple picture I drew of the interferometer to make it that good.  

[00:45:24] BARISH: So we built these instruments that are large for the reasons that I said, the amount of effect depends on how big you make it. So the largest practical size at the time we built this was to make the arms, the two arms on the interferometer, about four kilometers long. So they’re four kilometers long. This picture is LIGO in Livingston, Louisiana. And you’ll notice the terrain. It’s all, it’s a commercial forest and the way we build it is to build it up on a berm to get it high enough above the surface so it doesn’t flood. And so we build it to the 500 year flood plain. That’s about 15 feet up in Louisiana. Added another meter for safety so we don’t get flooded in the lifetime of LIGO. And in doing that, it makes-we borrow the dirt from a channel which immediately fills with water and that’s what you see on the right side. And then alligators, fish and everything else you can imagine is. So that’s the terrain we live with in Louisiana. There’s no bedrock. So trying to make a stable, a physical structure is very difficult, is basically floating on water. The second LIGO Observatory, which is identical for our purpose, is totally different from a standpoint of this picture and reality. It’s on high desert in Hanford, Washington. So it’s basically on sand. You have to go very, very deep to find any water. And yet the two instruments we make and their sensitivity are basically identical. Well we try to keep them identical. So that’s the two instruments. We have two in order for confidence, like that was done wherever it was trying to do, but also to tell where this gravitational wave when we detect it, comes from. So once in Hanford, Washington, once in Livingston, Louisiana, they’re 3000 kilometers apart. At the speed of light, if we have a signal or gravitational wave goes through Livingston and then goes to Hanford and goes at the speed of light, which is speed of gravitational waves, it’ll take 10 milliseconds to go from one to the other. If we go the other direction, that’ll take 10 milliseconds to go from Hanford to Livingston. And if we happen to have a gravitational source that was directly overhead between the two, they would come at exactly the same time. So that’s the scheme. And we ask that they both happen and we tell something about the direction by the difference in the two.

The event that I’ll talk about or show you, it was 6.9 milliseconds different coming first in Livingston and then to Hanford Washington. And I’ll talk about what we use that for a minute. This is just some more of the internal workings of LIGO. It has a great big vacuum pipe. It, I think is the largest high vacuum system in the world. It’s 1.2 meter diameter pipes, a total of 16 kilometers of vacuum in the two sites. And these are the working parts of LIGO, big chambers where we put the optics and mirrors to guide the laser beam. And these have a lot of ports and things because we have a lot of test equipment, side beams and all kinds of things that we do, which I’m not going to talk about today. So what limits us? Well we can make it big enough, we can calculate that we’re going to try to do as well as I said. But in reality, we get limited by things that get in our way. And these are the most common ones. First, if the vacuum isn’t good enough and that’s why we make very high vacuum, we get residual scattering off the vacuum itself. So the light goes down, scatters off the vacuum. You can imagine a lot of ways you can get in trouble with that. You can have particles that scatter then off the wall, photons and back, have a different path length and are out of time and so forth. So we need to keep the scattering down as much as we can. Second, as much as we think of a laser beam has very, very stable, it’s not stable enough for us. So we do a tremendous amount of work to stabilize the laser in its wavelength and in its amplitude test. The second thing we have to do to keep from being able to do this. So third is that we have to isolate ourselves well enough from the ground. This is a huge problem. We basically are living on the earth where the ground shakes a lot and we do that in two ways. As I’ll show you. The next one is that we work at room temperature and at room temperature, protons and molecules move around in any substance. So the mirrors that we have are made out of molecules and they move around, something we call Brownian motion. And lastly, there’s more subtle things that we like to increase the light level as high as we can to get the best measurements. But if we do that, we have other problems that come in. In this case, what I’ve shown here is something we call quantum noise, and that is we increase the light level by making more and more powerful lasers, we produce a pressure on those test masses themselves that we have to worry about.

[00:50:34] BARISH: If we put that all together, we get a curve that looks like this. Now what to look at in this is the shaded area. This is an old picture, so it’s not the advanced LIGO, but it shows the concept. If we look in the shaded region, that’s the highest level that we’re limited by something. All the lines below are the things that we try to control and make sure they’re below and we’re eventually limited by three things. At the lowest center, at the lowest level on the left, very steep is seismic noise. That’s the shaking of the earth. At the highest level, we’re limited by what’s called shot noise, which is really fundamentally how many photons you have. So it’s photon statistics. And in the middle we’re limited by this thermal noise, the fact that we’re working at room temperature. So nasty problem for experimentalists because we don’t have one problem that we have to deal with, we have three different ones. The total a sensitivity region might look familiar to you. It’s essentially the same as the earth. So the audio-so we talk about being in the audio band. We basically have ears that respond to tens of Hertz and go up to thousands of Hertz just like this does. And it’s really for the same physical reasons we live on the earth. That’s where the earth is quietest. And so our laboratory for LIGO is on the earth and although we’re not dealing with audio things, we have the same basic things that limit us. So we work in what’s called the audio band and it comes from the fact that that’s the laboratory that we have, which is here on earth. So that’s our sensitivity. And it looks like this curve here. This is a real curve now. The lines that go up, you can ignore. Those are little residences that we have in our system and just ignore them for now. You want to learn about them in more detail talk. The colored lines are the data from our earlier measurements that we  didn’t detect gravitational waves. And then we went through a big rebuilding program to improve in three directions. To improve the power of the laser, to be better at high frequencies, it’s the number of photons or how fast you can sample at the low frequencies to do better seismic isolation. And the middle levels to make bigger and better test masses and suspension systems to try to control that region. We did those improvements and completed them about a year and a half or so ago. And technically, these are just some pictures. I’ll just show you quickly. The laser that we have, only message to take away here from here is just a lot more complicated than the little one I would use as a pointer if I didn’t have glass up in. And it has all these stabilizing elements in it. The mirrors themselves are fantastically wonderful objects. These are made out of silica. They’re 20 centimeters thick, 35 centimeters or 34 centimeters across. They weigh 40 kilograms. So they’re big things. The optical surfaces that the best, the best we can make. And they’re coated so that they basically to the eye, they look perfectly clear, but they’re coated so that they reflect at the wavelength of our laser, which is 10, 24 nanometers.

This is the suspension system as we hang from wires. We do it in a series of four steps, and this is done so that the bottom layer level is as quiet as we can make it. We do all the controls at the higher levels. And lastly, and you can’t see this very well in the picture, we have a very complicated system of trying to isolate ourselves from the ground, both passively by using basically very advanced shock absorbers like you have in your car, but more recently, by doing it with a feedback system that does it actively. That is by measuring the motion, just like sound. Things that you were using in your ears on an airplane drowned out the sound in the background. We correct for the motion. A combination of those has given us the improvements that we needed to make this measurement. So the curves that I showed before where the top curve here and we started improving with the physical improvements we made on our way to get to this very bottom line and we stopped in between because of the fact that as we increase the sensitivity, say a factor of two, that means that we look out into the universe a factor of two further. And the number of galaxies, stars and so forth in the universe as we go out a factor of two further are a factor of eight bigger because it goes as the volume. So having increased our sensitivity a factor of three at the high frequencies and where we put in active isolation at the low frequencies, it’s even more.

[00:55:35] BARISH: We were able to make this measurement. So we turn the apparatus on last September. This is a familiar curve that people have seen. One is from LIGO Livingston on the left and one on the right is from Hanford Observatory and what they show is they’re seven milliseconds apart. What they show is something that’s getting bigger and higher frequency. And that’s the motion as it’s going in, it’s going faster around each other, the two objects, and then they merge. And they merge at this merging and finally bring down a little bit. And that’s the picture which I’ll show you what we calculate. So drawn through there then, and the last is a line, which comes from our best fit using general relativity. If we now look at that, this is the picture that we have going backwards now from what looks like the data. The top is the physical picture of the different parts of the graph. We see that picture here. That is basically the picture that is extracted from the raw data where the frequency gets higher and higher. Uh, the signal gets bigger and bigger. These are the two objects getting closer and closer together. These very compact objects and then merging. And then there’s a little bit of ring down, which we’re not able to detect very well at this point. We fit all that with general relativity and we’re able to tell the masses of the two objects. We’re able to tell the amount of energy that was released into gravitational radiation about three times the mass of our sun, equivalent of the mass of our sun radiated away, and other things about it. How far away it is and so forth. If you look at the bottom picture here, to me, it’s very revealing of what’s happening. First, look at the black curve on the top and you see that the separate, it’s the scale on the right, which is in units of what we call Schwarzschild radii.

But that’s about a hundred kilometers each. They start in our band about 400 kilometers apart and when they’re finally merging there only about a hundred kilometers, which is their size. And that’s what you saw. And this all happens in about a quarter of a second or less. The other curve going up is one that’s astounding. These things, although it didn’t look like that in the first movie that I showed you, are highly relativistic, they’re going at the beginning, at 30 percent the speed of light and by the end when they’re merging, they’re up to about 60 percent of the speed of light. So it’s a tremendously energetic system and fast system. We compared the times when it arrived, I kept saying they’re seven milliseconds apart. If you remember how you tell where something is from using triangulation on a boat or something, this should tell us a ring in the sky basically. And if we had a third detector, which we will have soon in Italy, then you can do better. You can tell two points, two positions in the sky. What I show here is how will we actually are able to tell the detection and. We can tell it comes from the southern hemisphere and it’s just a banana shaped thing fits for us, not very good. It’s about 500 square degrees. But it’s not the whole ring. And the reason it isn’t the whole ring is that the two detectors don’t have spherical acceptance. They have an acceptance, that depends whether you’re overhead or inside and so forth, and they’re separated from each other about 16 degrees on the earth’s surface. And so looking at how big the signal is in Livingston compared to Hanford we’re actually able to rule out that pretty much ruled out the northern hemisphere. And so this is what we up with. It’s very important to us to get this localization much better because we’re now able to for the first time, look at the sky with gravitational waves and you’d like to see that whatever you see is it also seen in some sort of electromagnetic radiation? Whether it’s visible or infrared or some other object. And of course this huge swath is not good enough. So the way we’re going to do that is to add other detectors. There’s one that’s been built and will soon be hopefully in Italy, French Italian collaboration near Pisa. And it should be operational hopefully by the end of this year. And that will give us much better resolution, which I’m not going to show you, but much better. And finally, we’ve been approved by the Indian government to put a third interferometer. We are contributing the interferometer and them the infrastructure in India, and that should be finished in about five years. So that will in the end give us much better position. We’ve just opened, and that’s why this is so important and exciting, a totally different way of looking at the sky. We know what we don’t know is probably the most important, that there are things that give gravitational signals that we don’t see electromagnetically. In fact, this very first signal that we receive, these 30 solar mass black holes weren’t known to be there by any electromagnetic experiments that had been done. So that’s already.

[01:00:54] BARISH: The known sources that we expect to see. I’ve just listed here as the possible targets that we’re going to be looking for in the next few years. On the upper right, I show burst. That’s like the collapse of a star, Supernova or a gamma ray burst. All these things that are seen optically, but the information that we can get if we add information, gravitationally can give us great insight, for example, into the collapse of a star and the collapse of a star is gravitational, the physics is gravitational. The signal that seen optically is kind of a shockwave that comes out to. So to really understand it, the gravitational signals should be added to the other. So we hope to see the collapse of the star, gamma ray sources eventually as our sensitivity improves. What I call continuous sources in the lower right, that’s spinning, spinning objects that aren’t totally spherically symmetric. Examples are what we call pulsars or neutron stars, and if they’re young enough, they’re probably not so spherically symmetric, and we look for those as a continuous wave at some frequency that they’re rotating. At the lower left is a dream. We’re not able to do that very well in the LIGO bandwidth, but that’s to see signals from the early universe. Lastly, we’re going to improve the technologies in LIGO. I showed you we’re only partway to where we’ve designed to go. We’ll finish that the next few years. We know how to go beyond that and we’ll keep improving. We gain as a cube of every improvement we make. But there’s other improvements that can go beyond that. There’s being made in Japan now that has two improvements that will be important in second or third generation detection in the sense that LIGO advanced LIGO, second generation, I would call this two and a half generations. That is, it includes an advantage of being deep underground by in deep underground, there’s much less seismic noise to worry about. Things don’t shake and they’re making an attempt that we looked at but gave up in our time scale before and that is cooling the system, cooling the mirrors so that we don’t, they don’t have the thermal noise that we have.

There’s also been look both in the US but primarily in Europe so far at how to make a detector on the earth’s surface that can go well beyond LIGO. And in Europe through a study, they’ve done one that would be deep underground, 10 kilometers instead of four kilometers. A triangle, which helps you do the polarization which exists that I didn’t talk about that is cryogenic and that it has to optical configurations, one that will work at very low frequencies and one that would look at work at much higher frequencies. We’re looking at some alternatives to this, but the point is that this. We’re not at the end of the road at all. We can develop this technology over the coming years on the earth’s surface to do much better than we’re doing now. That’s not the end of the story. We know how to do that, but just like in astronomy, where the big one of the big gains in the last decades has been the ability to look at different wavelengths. Looking at optical, looking at astronomical things that happened, but looking at them in the infrared and the ultraviolet and the visible in different wavelengths has been able to pull together the dynamics. Similar to that, gravitational waves are going to be at different frequencies. We’re working at the very highest frequencies, which means we’re working where things take milliseconds to happen because they’re very violent things that only take milliseconds and we’ve seen one that’s this object coming together. If we want to look at and there’s many possibilities, at things that take longer time scales and if it’s minutes or hours we go in space. You can’t do that for the reasons that I said on the earth. We go in space. There’s a program to go into space. It’s called LISA. The Europeans are supporting it. It was supported by NASA, but because of budget problems, it was pulled back. We’re hoping that it’ll be a put back into the NASA program, possibly stimulated some by our success. But that’s a very important way to get to longer timescales. Yet longer timescales are done by a pulsar timing, an array of timing very accurately where the timing gets affected by the passage of gravitational waves. And that can get two years and decades. And finally, we can look at the experiments that look at the early universe signals of billions of years away, and look at the effect of what they measure by the presence of gravitational wave patterns that they look at are the polarizations that they see. So the future in our mind is really exciting and it will take a long time, but opens up a totally new area of physics and astronomy. And with that I’ll close. Thanks.