After Gamow’s rejection, Weber decided to work on the physics of atoms with Keith Laidler, earning a Ph.D. in 1951 from The Catholic University of America. His thesis led to a paper that he submitted for presentation at an international conference in Canada on “coherent microwave emission.” This paper had many of the key ideas and concepts that would lead to the maser (microwave amplification by stimulated emission of radiation) and ultimately to the laser (where “microwave” is replaced by “light”). Charles Townes was also working on this problem and asked Weber for a copy of his paper, while Nikolay Basov and Aleksandr Prokhorov in the Soviet Union were independently working along the same lines. In 1964, Townes, Basov and Prokhorov were awarded the Nobel Prize in Physics for their construction of the first masers and lasers. Although Weber was also nominated for the Nobel Prize at the same time, he never received it.
According to Kip Thorne, Weber’s interest in general relativity began to grow after the maser discovery, because Weber wished to move into a field of study with less Nobel drama. In 1955 he took a sabbatical at the Institute for Advanced Studies to study gravitational radiation with Wheeler (their presentation at the 1957 Chapel Hill meeting was a result of that work), and then continued his studies at the Lorentz Institute for Theoretical Physics in the Netherlands.
And after these preliminary studies, he set out to do the unthinkable: to detect gravitational waves. Around 1958, he began the project in earnest, first doing the required theoretical calculations to determine just what the physical effects of a passing wave would be on a solid mass, in contrast to a set of disks sliding on a table, and then building an apparatus. By 1965 he had put a simple detector into operation. It consisted of a solid cylinder of aluminum (the reason for aluminum was a mundane one: it was cheap), about a meter in diameter by two meters in length. It weighed about 1.5 tons.
When a gravitational wave passes through the cylinder in a direction perpendicular to its axis, the spacetime distortion in the wave tries to stretch and squeeze the bar end-to-end (see Figure 7.4). There is also stretching and squeezing in the perpendicular directions, but this turns out to be less easy to measure so we will ignore it. As we remarked before, because of the solid material in the bar between the ends, they do not move freely the way two sliding disks would with the same separation, and so the response to the wave can be tiny by comparison.
Figure 7.4 Distortions of a “Weber bar” induced by a gravitational wave traveling perpendicularly into the page that you are reading (vastly exaggerating the scale of the effect). Dashed lines indicate the undistorted cylinder. Top panel: The bar is stretched horizontally, while being squeezed in the other directions. Bottom panel: One half cycle later the bar is squeezed horizontally while being stretched in the other directions.
However, the bar has a property that the sliding disks do not. If you were to hit the bar on the end with a hammer, it would vibrate for a considerable time at a single frequency, called its “resonant frequency.” Every child learns about the phenomenon of resonance at an early age. A playground swing moves back and forth with a characteristic frequency that depends on its length. If you push the swing with exactly the same frequency, achieved most easily by pushing once every cycle right when the swing returns to you (or if you are alone, by kicking your legs once each cycle at an optimal time), you can achieve thrilling amplitudes of swing. What’s more, you continue to swing for a while even after the pushing has stopped, until friction from air or from the ropes brings the swing to a halt.
Weber had very good reasons for choosing a resonant bar as opposed to sliding disks. As a possible source of gravitational waves, he was thinking about supernovae, the only kind of “extreme” event known at the time. He imagined that the gravitational signal from such an event would be a short-lived “burst,” possibly no longer than a small fraction of a second, and that it would be “broad-band,” namely that it would contain waves with a broad range of frequencies as opposed to a single frequency, like a pure sound tone. If some part of the signal was at the resonant frequency of his bar, then the bar would be excited strongly, and in addition it would continue to oscillate at its resonant frequency after the burst had passed, giving more time for his sensors to measure the oscillations. The resonant frequency of his bar happened to be around a thousand cycles per second, or in the kilohertz band, in the same ball park as what one would estimate for the frequency of waves from a supernova. To measure the compression and extension of the bar, Weber bonded devices called “piezoelectric transducers” around the bar at the middle to convert the strains into electrical signals that could be recorded and analyzed. Nevertheless, it was still a daunting prospect: even the crudest estimate of the signal from a supernova in the Milky Way galaxy implied a change in length of his bar of about the diameter of a proton!
Weber devoted a great deal of effort to ensure that his bar was isolated from external disturbances, such as seismic vibrations or the effects of nearby traffic that could set the bar into oscillations and mimic the effect of gravitational waves. This required suspending the bars using the thinnest wires that would support the weight and attaching those wires to supports made of alternating layers of rubber and steel. He also shielded his bars from external electric and magnetic fields.
In June 1969, Weber made the stunning announcement that he was detecting signals simultaneously in two detectors spaced 1,000 kilometers apart, one in Maryland, and the other at Argonne National Accelerator Laboratory near Chicago. The reason for using two detectors is simply that any one detector is often in oscillation because of disturbances from the environment that leak in despite his best attempts to isolate the bar from such noise, and because of the inevitable random internal motions by the atoms inside the bar produced by heat energy. Therefore, in a single bar it is difficult if not impossible to distinguish a disturbance from a gravitational wave from a disturbance of an environmental or thermal origin. As early as 1967, Weber had reported disturbances in a single bar, but he could not reliably claim that they were from gravitational waves. However, with two detectors separated by such a large distance, a disturbance that appears simultaneously in both detectors would be unlikely to be environmental or thermal because the probability of such a random coincidence is very small. Coincident events would therefore be good candidates for gravitational waves. Even more remarkable than the 1969 report of coincident events was his announcement in 1970 that the rate of such events was highest when the detectors were oriented perpendicular to the direction of the center of the galaxy, implying that the sources were indeed extraterrestrial, perhaps concentrated near the Galactic Center. These reports caused a sensation both in scientific circles and in the popular press.
There were two problems, however. The observed events occurred with a disturbance size and at a rate (around three times per day) that shocked theorists, for it implied a rate of gravitational wave bursts at least a thousand times what they predicted. This in itself was not necessarily bad, for often the mark of an important experimental discovery in physics is the degree to which it upsets theoretical sacred cows.
The second problem was more devastating, however. The main body of Weber’s coincidence results were reported between 1969 and 1975. But by 1970, independent groups worldwide had built their own detectors with claimed sensitivities equal to or better than Weber’s, yet between 1970 and 1975, none of these groups saw any unusual disturbances over and above the inevitable noise. By 1980 there was a general consensus that gravitational waves had not been detected by Weber. Weber never accepted any of this and he continued to work on the detection of gravitational waves with resonant bars until his death in 2000.
So, is this a story of a tragic failure or of a great success? It is certainly a good example of the self-correcting nature of science. The acceptance of new results always requires their external confirmation, typically done by carrying out the experiment again in a different setting and perhaps with more sophisticated instruments. In Weber’s case, his results
could not be replicated, so his claim was not accepted.
But as John Wheeler put it in 1998: “No one else had the courage to look for gravitational waves until Weber showed that it was within the realm of the possible.”
Over time, a more nuanced view of Weber’s legacy has emerged. Prior to Weber, the field of general relativity was almost completely dominated by theorists. The field was often called a “theorist’s paradise and an experimentalist’s purgatory.” During the Chapel Hill conference, Feynman complained that the problem with general relativity is the lack of experiments. Weber’s announcement induced experimentalists from other branches of physics to get involved in general relativity, such as William Fairbank from low-temperature physics, Ronald Drever from magnetic resonance work, Vladimir Braginsky from precision measurements, Heinz Billing from computer science, Edoardo Amaldi from elementary particle physics, J. Anthony Tyson from astronomy, and others. It also piqued the interest of a young MIT professor named Rainer Weiss, who would soon lay the foundations of the LIGO instrument. These experimentalists helped to transform the landscape of the field into one with a healthy synergy between theorists and experimentalists.
Weber’s work also helped transform theory. As we have seen, when Weber started building his detectors, the main concern of theorists with regard to gravitational waves was their physical reality. After his announcement, the direction shifted dramatically as theory groups around the world starting thinking about plausible (and also implausible) astrophysical sources for the enormous signals he was claiming. Although no scenario was ever found that could explain Weber’s signals, the insights gained and the techniques developed during this period helped to advance the growing interactions between general relativists and astrophysicists that had begun with the discoveries of quasars, pulsars and the cosmic background radiation during the 1960s.
Weber’s work also had a personal impact on one of us. At the time of Weber’s first announcement, Cliff had completed his first year as a Caltech graduate student and was thinking about doing a summer project in Kip Thorne’s research group. Thorne told him:
I am worried that if Weber is correct, then general relativity itself might be wrong. I want you to spend the summer finding out all there is to know about the current experimental support for the theory and to think about what might be done in the future to prove the theory right or wrong.
That launched Cliff’s fifty-year-long career in general relativity!
The consensus that Weber had not found waves did not end the effort to detect them, of course. Many groups continued to develop advanced bar detectors. One strategy involved cooling the entire bar and associated sensing devices to one or two degrees above absolute zero, in order to reduce the size of the disturbances due to the thermal motions of the atoms inside the bar. Some groups replaced the piezoelectric crystals used by Weber with sophisticated sensors attached to the ends of the bars, leading to greatly improved performance. Some fabricated their bars out of different materials such as sapphire that might have an improved response to the gravitational wave excitations.
None of these groups reported credible detections, but instead established increasingly stringent upper limits on the strength of gravitational waves bathing the Earth. Although by 1979, as we saw in Chapter 5, the measurements of the binary pulsar had verified the existence of gravitational waves, the actual waves emitted by that system were far too weak and of too low a frequency to be detectable by resonant bars. Work continued on bars for another 25 years, but gradually declined for lack of funding until the last “Weber bars” ceased operations for gravitational wave detection around 2008. But important advances were made in the course of this research, in new techniques for isolating the bars from things like seismic noise, in the control of thermal noise and in data analysis techniques. Many of these lessons would be used in helping to develop an alternative detector concept that began to emerge during the 1970s, the laser interferometer.
The laser interferometer is based on an apparatus devised by US scientist Albert A. Michelson originally to measure the speed of light very accurately, but then famously used by him and Edward W. Morley in 1887 to try to detect the motion of the Earth through the “aether,” the hypothetical medium through which light supposedly propagated. Schematically, Michelson’s interferometer consists of two straight arms set at right angles to each other (Figure 7.5). Each arm has a mirror at one end. At the intersection where the arms are joined, a half-silvered mirror splits a light beam into two, each traveling down one arm, each reflected back by the mirror at the end of each arm. When the two beams recombine, they interfere to produce a characteristic pattern of fringes that depends on the difference in time required for the two beams to make the round trip. Michelson and Morley failed to detect any effect of motion through the aether, and the conundrum inspired by that failure ultimately led to Einstein’s special theory of relativity.
Figure 7.5 Schematic version of a laser interferometer. Light from a laser is split by a half-reflecting mirror and travels along two perpendicular arms. Mirrors at each end reflect each beam back. The beams are brought together at a sensor. If the waves interfere constructively, a bright spot is seen; if they interfere destructively, a dark spot is seen. A “plus”-polarized gravitational wave that travels perpendicularly into the page you are reading will stretch one arm while compressing the other, thus altering the interference between the two light beams.
Michelson’s interferometer is an excellent tool for measuring distance precisely, because a change in the length of an arm of a quarter of the wavelength of light will turn a bright spot at the output into a dark spot. Since the wavelength of light is measured in millionths of a meter, it is easy to see the potential, at least in principle, of measuring the tiny changes in separation between objects induced by a gravitational wave. The reader might again ask: the gravitational wave changes the distance between the mirrors and the beam splitter, but doesn’t it also affect the propagation of the light ray? The answer is yes, but just as in the example of the disks sliding on the table, the two effects do not cancel each other out, and there is a true, measurable change in the brightness of the output beam.
The first to think about such a scheme were the Russian physicists M. Gerstenshtein and V. Pustovoit in 1963, but their work was not recognized until many years later. Weber and his student Robert Forward independently considered this possibility, and Forward actually built the first prototype for an interferometric detector in 1972. But the person generally credited with showing how to turn this idealized concept into a large-scale device that might actually detect gravitational waves is Rainer Weiss, or Rai, as he is known to all his colleagues.
Rai’s family escaped Nazi Germany and relocated to New York City in 1939. A brilliant tinkerer in anything electronic, he started his studies in electrical engineering at MIT, but left Boston in his junior year to pursue a romantic relationship in Chicago that would eventually fizzle out. Upon his return to MIT, he discovered he had been expelled because he had gone AWOL. Undaunted, he managed to convince MIT physicist Jerrold Zacharias to give him a job as a technician in his lab. Around that same time, Zacharias was working on the first practical version of an atomic clock, based on cesium atoms. With the help of Zacharias, Rai was readmitted to MIT, finished his undergraduate degree in 1955 and finally obtained a Ph.D. under Zacharias in 1962.
After a two-year stint as a postdoc with physicist Robert Dicke at Princeton University, where he began to develop experiments to test general relativity, he returned to MIT in 1964 as an assistant professor. At the urging of radio astronomer Bernard Burke, he took an interest in the cosmic microwave background radiation, recently detected by Penzias and Wilson. This radiation is the remains of the hot electromagnetic radiation that would have dominated the universe in its earlier phase, now cooled to a few degrees above absolute zero by the subsequent expansion of the universe. Cosmological theory suggested that the strength of this radiation should have a very specific de
pendence on the wavelength of the radiation, known as a “black-body spectrum,” but measurements made using sensors on rockets threatened to discredit this prediction. Rai and his graduate student flew a device on a high-altitude weather balloon that showed convincingly in 1973 that the spectrum was that of a black body, and also measured its temperature to be 3 degrees above absolute zero (the modern measured value is 2.725°). Rai would later be a leader on NASA’s Cosmic Background Explorer satellite, which would make even more precise measurements of the properties of this radiation between 1989 and 1993.
But he had never truly forgotten about his first love, gravity experiments. Around 1968, MIT asked him to teach a class on general relativity, but not being an expert on the theoretical side of the subject, he chose to focus on the experimental side, using a small 1961 primer on general relativity and gravitational waves that had been written by Weber. But when he studied Weber’s discussion of using resonant bars, he could not understand how Weber could achieve the sensitivity needed to detect the waves. So he assigned a homework problem in his course: find a way to measure gravitational waves by sending light beams between things. The students were stumped, so not much came out of that homework set. Soon Weber was claiming detections, to Rai’s skepticism, and so he decided to think more seriously about the interferometer idea. He analyzed in detail every source of noise or disturbance he could think of and concluded that, with a large enough interferometer, it might be possible to beat Weber’s sensitivity by a factor of a thousand. He also recognized a key difference between Weber’s resonant bars and an interferometer. The bars respond strongly only to the part of the gravitational wave whose frequency is close to the resonant frequency of the bar, whereas if one suspended the mirrors in the interferometer on long pendula, they would respond fully to the gravitational wave, no matter what its frequency was, just like our frictionless hockey pucks.
Is Einstein Still Right? Page 23
