For the Love of Physics

by Walter Lewin

  Professor Andy Lawrence at the University of Edinburgh writes an astronomy blog called The e-Astronomer on which he once posted a reminiscence of working on his thesis, staring at hundreds of position plots of X-ray sources. “One night I dreamt I was an error box, and couldn’t find the X-ray source I was supposed to enclose. I woke up sweating.” You can understand why!

  The size of the error box of the X-ray source discovered by Riccardo Giacconi, Herb Gursky, Frank Paolini, and Bruno Rossi was about 10 degrees × 10 degrees, or 100 square degrees. Now keep in mind that the Sun is half a degree across. The uncertainty in figuring out where the source was consisted of a box the area of which was the equivalent of 500 of our Suns! The error box included parts of constellations Scorpio and Norma, and it touched the border of the constellation Ara. So clearly they were unable to determine in which constellation the source was located.

  In April 1963 Herbert Friedman’s group at the Naval Research Laboratory in Washington, D.C. improved substantially on the source’s position. They found that it was located in the constellation Scorpio. That’s why the source is now known as Sco X-1. The X stands for “X-rays,” and the 1 indicates that it was the first X-ray source discovered in the constellation Scorpio. It is of historical interest, though never mentioned, that the position of Sco X-1 is about 25 degrees away from the center of the error box given in the Giacconi et al. paper that marked the birth of X-ray astronomy. When astronomers discovered new sources in the constellation Cygnus (the Swan), they received the names Cygnus X-1 (or Cyg X-1 for short), Cyg X-2, and so on; the first source discovered in the constellation Hercules was Her X-1; in Centaurus Cen X-1. Over the next three years about a dozen new sources were discovered using rockets, but with one important exception, namely Tau X-1, located in the constellation Taurus, no one had any idea what they were, or how they were producing X-rays in such huge quantities that we could detect them thousands of light-years away.

  The exception was one of the more unusual objects in the sky: the Crab Nebula. If you don’t know about the Crab Nebula, it’s worth turning to the photo insert to look at the image of it there now—I suspect you’ll recognize it right away. There are also many photos of it on the web. It’s a truly remarkable object about 6,000 light-years away—the stunning remains of a supernova explosion in the year 1054 recorded by Chinese astronomers (and quite possibly in native American pictographs—take a look here: http://seds.org/messier/more/m001_sn.html#collins1999) as a superbright star in the heavens that suddenly appeared, more or less out of nowhere, in the constellation Taurus. (There is some disagreement about the exact date, though many claim July 4.) That month it was the brightest object in the sky other than the Moon; it was even visible during the day for several weeks, and you could still see it at night for another two years.

  Once it faded, however, scientists apparently forgot about it until the eighteenth century, when two astronomers, John Bevis and Charles Messier, found it independently of each other. By this time, the remains of the supernova (called a supernova remnant) had become a nebular (cloudlike) object. Messier developed an important astronomical catalog of objects like comets, nebulae, and star clusters—the Crab Nebula is the first object in his catalog, M-1. In 1939 Nicholas Mayall from Lick Observatory (in Northern California) figured out that M-1 is the remnant of the supernova of 1054. Today, a thousand years after the explosion, there is still such wonderful stuff going on inside the Crab Nebula that some astronomers devote entire careers to studying it.

  Herb Friedman’s group realized that the Moon was going to pass right in front of the Crab Nebula on July 7, 1964, and block it from view. The term astronomers use for this blocking out is “occultation”—that is, the Moon was going to occult the Crab Nebula. Not only did Friedman want to confirm that the Crab Nebula was indeed an X-ray source, but he also was hoping he could demonstrate something else—something even more important.

  By 1964 a renewed interest had emerged among astronomers in a type of stellar object whose existence was first postulated during the 1930s but that had never been detected: neutron stars. These strange objects, which I discuss more fully in chapter 12, had been conjectured to be one of the final stages in a star’s life, possibly born during a supernova explosion and composed mostly of neutrons. If they existed, they would be of such great density that a neutron star with the mass of our Sun would only be about 10 kilometers in diameter—about 12 miles all the way across, if you can imagine such a thing. In 1934 (two years after the discovery of neutrons), Walter Baade and Fritz Zwicky had coined the term “supernova” and proposed that neutron stars might be formed in supernova explosions. Friedman thought that the X-ray source in the Crab Nebula might be just such a neutron star. If he was right, the X-ray emission he was seeing would disappear abruptly when the Moon passed in front of it.

  He decided to fly a series of rockets, one after the other, right as the Moon was going in front of the Crab Nebula. Since they knew the Moon’s exact position in the sky as it moved, and could point the counters in that direction, they could “watch” for a decline in X-rays as the Crab Nebula disappeared. As it happened, their detectors did indeed pick up a decline, and this observation was the first conclusive optical identification of an X-ray source. This was a major result, since once we had made an optical identification, we were optimistic that we would soon discover the mechanism behind these enigmatic and powerful X-ray sources.

  Friedman, however, was disappointed. Instead of “winking out” as the Moon passed over the Crab Nebula, the X-rays disappeared gradually, indicating that they came from the nebula as a whole and not from a single small object. So he hadn’t found a neutron star. However, there is a very special neutron star in the Crab Nebula, and it does emit X-rays; the neutron star rotates about its axis about thirty times per second! If you want a real treat, go to the Chandra X-Ray Observatory website (http://chandra.harvard.edu/) and call up images of the Crab Nebula. I promise you, they are stunning. But forty-five years ago we had no orbiting imaging X-ray telescopes in space, so we had to be much more inventive. (After the 1967 discovery of radio pulsars by Jocelyn Bell, in 1968 Friedman’s group finally detected X-ray pulsations—about thirty per second—from the neutron star in the Crab Nebula.)

  Just as Friedman was observing the occultation of the Crab, my friend (to be) George Clark at MIT was in Texas preparing for a high-altitude balloon night flight to search for high-energy X-rays from Sco X-1. But when he heard about Friedman’s results—even without the Internet, news traveled fast—he completely changed his plans and switched to a day flight in search of X-rays in excess of about 15 keV from the Crab Nebula. And he found them too!

  It’s hard to put into words just how exciting all this was. We were at the dawn of a new era in scientific exploration. We felt we were lifting a curtain that had been hiding these amazing realms of the universe. In reality, by getting our detectors up so high, by getting into space, by getting to the top of the atmosphere where X-rays could penetrate without being absorbed by air, we were removing blinding filters that had been on our eyes for all of human history. We were operating in a whole new spectral domain.

  That has happened often in the history of astronomy. Every time we learned that objects in the heavens emitted new or different kinds of radiation, we had to change what we thought we knew about stars, about their life cycles (how they are born, how they live, and how and why they die), about the formation and evolution of clusters of stars, about galaxies, and even about clusters of galaxies. Radio astronomy, for instance, showed us that the centers of galaxies can emit jets hundreds of thousands of light-years long; it has also discovered pulsars, quasars, and radio galaxies and is responsible for the discovery of cosmic microwave background radiation, which radically changed our views of the early universe. Gamma-ray astronomy has discovered some of the most powerful and (fortunately) distant explosions in the universe, known as gamma-ray bursts, which emit afterglows in X-rays and visible light all the way down to r
adio waves.

  We knew that the discovery of X-rays in space was going to change our understanding of the universe. We just didn’t know how. Everywhere we looked with our new equipment, we saw new things. That’s not surprising, perhaps. When optical astronomers started getting images from the Hubble Space Telescope, they were thrilled, awestruck, and—maybe this isn’t so obvious—hungry for more. But they were basically extending the reach of a centuries-old instrument, in a field dating back millennia. As X-ray astronomers, we were experiencing the dawn of a whole new scientific field. Who knew where it would lead, or what we would discover? We surely didn’t!

  How fortunate for me that Bruno Rossi invited me to MIT in January 1966, just as this field was taking off, and that I immediately joined George Clark’s group. George was a very, very smart physicist, a really impressive guy with whom I became friends for the rest of my life. Even now, I can hardly believe my good luck—a great friend and a new career, both in the same month.

  CHAPTER 11

  X-ray Ballooning, the Early Days

  When I arrived at MIT, there were five active balloon groups in the world: George Clark at MIT, Ken McCracken at the University of Adelaide in Australia, Jim Overbeck at MIT, Larry Peterson at UC San Diego, and Bob Haymes at Rice University. This chapter is largely about my own experiences with X-ray ballooning, which was at the center of my research in the decade between 1966 and 1976. During these years I made observations from Palestine, Texas; Page, Arizona; Calgary, Canada; and Australia.

  Our balloons carried our X-ray detectors to an altitude of about 145,000 feet (about 30 miles), where the atmospheric pressure is only 0.3 percent of that at sea level. When the atmosphere is this thin, a good fraction of X-rays with energies above 15 keV get through.

  Our balloon observations complemented the rocket observations. Rocket-borne detectors typically observed X-rays in the range from 1 to 10 keV and only for about five minutes during an entire flight. Balloon observations could last for hours (my longest flight was twenty-six hours) and my detectors observed X-rays in the range above 15 keV.

  Not all sources that were detected during rocket observations were detectable during balloon observations, since the sources often emitted most of their energy at low-energy X-rays. On the other hand, we were able to detect sources emitting largely high-energy X-rays invisible during rocket observations. Thus, not only did we discover new sources and extend the spectra of known sources to high energies, but we also were capable of detecting variability in the X-ray luminosity of sources on time scales of minutes to hours, which was not possible with rockets. This was one of the early successes of my research in astrophysics.

  In 1967 we discovered an X-ray flare from Sco X-1—that was a real shocker—I’ll tell you all about this later in this chapter. My group also discovered three X-ray sources, GX 301-2, GX 304-1, and GX 1+4, never seen before during rocket observations, and all of them showed changes in their X-ray intensity on time scales of minutes. GX 1+4 even showed periodic variability with a period of about 2.3 minutes. At the time we had no idea what could be the cause of such rapid changes in the X-ray intensity, let alone the 2.3-minute periodicity, but we knew we were breaking new ground—uncovering new territory.

  For some astronomers, though, even in the late 1960s, the significance of X-ray astronomy hadn’t yet sunk in. In 1968, I met the Dutch astronomer Jan Oort at Bruno Rossi’s home. Oort was one of the most famous astronomers. He had been an incredible visionary; right after World War II, he started a whole radio astronomy program in the Netherlands. When he came to MIT that year, I showed him the balloon data from our flights in 1966 and 1967. But he said to me—and I’ll always remember this—“X-ray astronomy is just not very important.” Can you believe it? “Just not very important.” He couldn’t have been more wrong. This was one of the greatest astronomers of all time, and he was completely blind to its significance. Maybe because I was younger, and hungrier—to be fair, he was sixty-eight by then—it was obvious to me that we were harvesting pure gold, and we were only just scratching the surface.

  I remember in the 1960s and 1970s I would read every single paper that came out on X-ray astronomy. In 1974 I gave five lectures in Leiden (Oort was in my audience), and I was able to cover all of X-ray astronomy. Nowadays thousands of papers on X-ray astronomy are published every year, in a multitude of subfields, and no one can grasp the entire field. Many researchers spend their entire careers on one of dozens of specific topics such as single stars, accretion disks, X-ray binaries, globular clusters, white dwarfs, neutron stars, black holes, supernovae remnants, X-ray bursts, X-ray jets, galactic nuclei, and clusters of galaxies. The early years were the most fantastic years for me. They were also demanding, in just about every way: intellectually, physically, even logistically. Launching balloons was so complicated and expensive, time-consuming, and tension producing, I can hardly describe it. I’ll try, though.

  Getting Aloft: Balloons, X-ray Detectors, and the Launch

  Before a physicist can do anything (unless, that is, you’re a theorist, who may need only a piece of paper or a computer screen), you have to get the money to build equipment and pay students and sometimes to travel very far. Lots of what scientists really do is write grant proposals, in highly competitive programs, to get supported to do research. I know it’s not sexy or romantic, but believe me, nothing happens without it. Nothing.

  You could have a wonderful idea for an experiment or an observation, and if you don’t know how to transform it into a winning proposal, it goes nowhere. We were always competing against the best in the world, so it was a cutthroat business. It still is, for just about any scientist in any field. Whenever you look at a successful experimental scientist—in biology, chemistry, physics, computer science, economics, or astronomy, it doesn’t matter—you are also looking at someone who’s figured out how to beat the competition over and over again. That does not make for warm and fuzzy personalities, for the most part. It’s why my wife, Susan, who’s worked at MIT for ten years, is fond of saying, “There are no small egos at MIT.”

  Suppose we got the funding, which we usually did (I was generously supported by the National Science Foundation and NASA). To send a balloon up nearly 30 miles, carrying a 2,000-pound X-ray telescope (connected to a parachute), which you had to recover intact, was a very complex process. You had to have reliably calm weather at launch, because the balloons were so delicate that a gust of wind could sink the whole mission. You needed to have some infrastructure—launch sites, launch vehicles, and the like—to help get the balloons way up into the atmosphere and to track them. Since I wanted to observe in the general direction of the center of the Milky Way, which we call the galactic center, where many X-ray sources were located, I needed to observe from the Southern Hemisphere. I chose to launch from Mildura and Alice Springs, Australia. I was very far away from my home and family—I had four children by then—usually for a couple of months at a time.

  Everything about the launches was expensive. The balloons themselves were enormous. The largest one I flew (which at the time was the largest balloon ever flown, and it may well still be the largest ever) had a volume of 52 million cubic feet; when fully inflated and flying at 145,000 feet, its diameter was about 235 feet. The balloons were made of very lightweight polyethylene—one-half of one-thousandth of an inch thick, thinner than Saran Wrap or cigarette paper. If they ever touched the ground during launch, they would tear. These gigantic, beautiful balloons weighed about 700 pounds. We usually traveled with a backup, and each one cost $100,000—forty years ago, when that was real money.

  They had to be made in immense plants. The gores, the sections of the balloon that look like tangerine skin segments, were made separately and then put together by heat sealing. The manufacturer only trusted women to do the sealing; they said it was well known that men were too impatient and made too many mistakes. Then we had to ship the helium to inflate the balloons all the way to Australia. The helium alone cost about $80,
000 per balloon. In current dollars that was more than $700,000 for just one balloon and its helium, without even considering the backup balloon, our transportation, lodging, or food. That’s right—here we were trying to ferret out the secrets of deep space, living in the middle of the Australian desert, utterly dependent on the weather. And I haven’t even told you about Jack. I’ll get to Jack in a bit.

  But the balloons were cheap compared to the telescopes. Each telescope, an extremely complicated machine weighing about a ton, took roughly two years to build and cost $1 million—$4 million in today’s dollars. We never had enough money for two telescopes at a time. So if we lost our telescope—which happened twice—we were out of luck for at least two years. We couldn’t even start building a new one until we’d gotten the funding. So it was a catastrophe if we lost one.

  And not just for me, not at all. This would cause a major delay for my graduate students, who were all deeply involved in building the telescopes, and whose PhD theses were about the instruments and the results of the observations. Their degrees went up in the air with the balloons.

  We needed the cooperation of the weather, too. There are intense winds in the stratosphere, flowing east to west at about 100 miles per hour for about six months of the year, and west to east the other half of the year. Twice a year these winds reverse direction—we call it the turnaround—and as they do, the wind speeds at 145,000 feet become very low, which would allow us to make observations for many hours. So we needed to be in a place where we could measure these winds and could launch during the turnaround. We probed every other day with weather balloons that we tracked by radar. Most of the time they would make it to about 125,000 feet, about 24 miles up, before they popped. But predicting the atmosphere isn’t like pushing ball bearings down a track in a lab demonstration. The atmosphere is so much more complex, so much less predictable, and yet everything we did depended on making good forecasts.