Gravity's Engines: How Bubble-Blowing Black Holes Rule Galaxies, Stars, and Life in the Cosmos

by Scharf, Caleb

  It was a long shot that we’d be able to detect anything, but it was a tremendously exciting plan. If we succeeded it would be the most distant detection of this kind of structure yet made, and it really seemed that it was within our grasp. We finally got our chance when we were given the go-ahead after two years of nail-biting anticipation. In late September of 2002, Chandra settled into position to point at our unfathomably distant target. For a total of 150,000 seconds, or about forty hours, it captured and counted X-ray photons streaming in from the universe. Most of these photons were the equivalent of noise on a radio or the fuzzy speckles in an image of a badly tuned TV. X-ray photons, like any other electromagnetic radiation, can traverse the universe. They come from all over; from stars, neutron stars, black holes, great shock waves of supernova explosions, and the hot gas of galaxy clusters. It’s a cosmic forest full of rustling leaves. But in among this barrage of random bits and pieces was a copse of noble trees. A grand total of about 150 of these photons had traversed 12 billion years of cosmic time in a direct path from our mysterious 4C41.17.

  *

  And so here again is the scene where we began, with pixels on a screen. In this case, from a display in the middle of my desk covered in scattered papers and coffee stains in my office in New York, these particular pixels formed an image. They conveyed the message that Chandra, high above the Earth’s surface, had indeed obtained our precious cargo of data. For the last couple of days it had stared at the sky as it silently circumnavigated the Earth. The finest mirrors and instruments that humans could produce had pointed toward a small patch of the cosmos, close to the constellation of Auriga—the Charioteer. In this direction the glorious view across the bows of our Milky Way galaxy took us all the way to 4C41.17, in the deep cosmic past.

  It was mid-morning on the island of Manhattan, and the sound of traffic echoed up through its canyons of rock and steel. I stared at the picture on my screen and squinted at the noisy spread of pixels. This was a preliminary view, before the data were properly processed and massaged to remove spurious features. Fast-moving particles like electrons and protons had ripped through Chandra’s frame high in orbit, spewing energy into its sensitive digital camera. This was nothing unusual, just an occupational hazard of space-based astronomy. But there was a shape in that mess of pixels, and I could see it clearly: a pinpoint of X-ray light, along with something else. I sent the image to a printer down the hall and trotted after to grab the hot paper as it spilled from the machine. Subdued by the heat-fused ink, the noisy features in the image were dissipated. There in stark relief was the extraordinary light of something unknown, a thick streak of brightness poking out from either side of a spot of intensity. It looked like dragonfly wings attached to a compact little body, an entomological picture from a long-forgotten era.

  My delight at finding something crazy and interesting in the data soon turned to puzzlement. At the center of the image was the bright pinpoint of X-ray light, and this was fairly easy to explain. Its spectrum had the fingerprint of intense X-ray emission from around the accretion disk of a supermassive black hole, buried somewhere inside the thick dust that we knew existed in this system. That problem was solved—there was no doubt now about the presence of a monster in the midst. But there was this other mysterious stuff: the thinly spread wings of light. Translating their length on the image into real distances revealed that altogether they spanned over three hundred thousand light-years from end to end. If this was the X-ray light from hot gas, heated as it flopped into a baby galaxy cluster’s gravity well, then it would also have a very particular flavor—that of the bremsstrahlung radiation from hot electrons we encountered before. The spectrum of X-ray photons would obey a particular pattern, and there would be lots of lower-energy photons and few higher-energy ones. Instead, as I wrestled with the data, I found a much more egalitarian spread of energy. That was all wrong—this meant it could not be coming from just hot gas. That wasn’t the only thing that was puzzling. If I computed the total power of this radiation, it was a hundred times greater than the X-ray emission of a normal galaxy cluster. This was also at odds with the radiation originating from hot gas in a baby system, yet here was a vast cloud of something merrily pumping out X-ray photons. I sent a worried message to Smail: things appeared funny, off-kilter, and I didn’t understand.

  I knew about hot gas in galaxy clusters, but not enough about strange structures emanating from what had to be a massive black hole on the other side of the visible universe. I started poring over articles in astrophysical journals and cautiously bringing up the data with other colleagues. A few ideas bounced around. Then I noticed two papers that helped shed light on the problem, literally. One of them was recent and written by Dan Schwartz, an astronomer at Harvard. The other one was written in 1966, by Jim Felten and Philip Morrison, two physicists then at Cornell. Felten had played a pivotal role early on in recognizing that galaxy clusters could emit X-ray radiation from their hot gas, and Schwartz was an expert on black hole jets and X-ray astronomy. Despite the time span between these two works, they had something critical in common: both papers talked about the astrophysical manifestations of a phenomenon that I dimly remembered from my undergraduate physics classes, which had the rather nondescript name of “inverse Compton scattering.” I soon realized how important this was in explaining our mysterious object in the distant universe.

  In 1922, the American physicist Arthur Compton had discovered that X-rays could bounce off freely floating electrons and change their energy, or wavelength, in the process. In essence, this phenomenon is simple. Particles like electrons can interact with photons. If an electron is just sitting quietly and a photon comes zooming along and bounces off it, like a pebble off a rock, then some of the photon’s energy will get transferred to the electron, moving it slightly. The photon will carry a bit less energy afterward, shifting to a lower frequency, and the electron will gain a little motion. But if the electron already has a lot of energy, then the outcome is different, and it is the photon that stands to gain.

  Out in the cosmos, there are phenomena that can accelerate particles like electrons to huge velocities that are significant fractions of the speed of light. The jets of matter squirting from black holes are an excellent example. These particles carry an exceptional amount of energy of motion, or kinetic energy. Relativity also tells us that the apparent mass of these particles will increase with speed, and that just further boosts the energy they represent—like a sponge getting more massive as it soaks up water. In this case, when a photon happens to bounce or scatter off one of these speedy electrons, then the inverse effect can occur. The fast and hot electron gives up some of its energy to the photon, which is a process known as “inverse Compton scattering,” and the photon emerges a new, more energetic beast.

  The cosmos is filled with photons. We saw this in our map of forever. Many of these are cosmic microwave photons left over from the early universe, and they make up a large fraction of the background electromagnetic soup of the universe. This means that if fast-moving electrons are generated by a phenomenon like a black hole jet, there’s a pretty good chance that as they zip along they’ll encounter the photons in the cosmic soup and give them a boost of energy. Incredibly, if the electrons move fast enough they can boost a photon from the microwave domain all the way up to the X-ray and even gamma-ray domain. That’s like taking a cup of coffee and boosting its temperature high enough to drive the steam turbines of a power plant.

  Here was a way to generate X-ray photons with just the spectral signature that I was seeing. All I needed were fast-moving electrons and lots of low-energy photons. It already looked like a supermassive black hole lived in the system. This could certainly be spewing out electrons in the form of fast-moving jets. Indeed, 4C41.17 was a potent source of radio emission, just as one would expect from spiraling relativistic electrons as they splashed into the surrounding universe. I knew that Smail had a map of the radio emission. In it there was a central bright point, and there
were two possible dumbbell-shaped zones, almost in line with the wings of X-ray light. It was a good bet that fast-moving electrons were filling these regions.

  But what about the supply of those low-energy photons? In the present-day universe there are about 410 cosmic microwave background photons in every cubic centimeter of space at any given time. We found these earlier on, seeping out of our hypothetical sack full of universe. The problem I was now facing was that this was nowhere near enough to account for the huge output in energy we were seeing. This simply didn’t represent enough photons to be boosted to an appreciable number of X-rays. I twiddled my thumbs and stared out the window, trying to imagine myself in that distant place, and the pieces suddenly started fitting together.

  We already knew that the region we were looking at had a great output of lower-energy infrared photons from hot dust. That could certainly contribute to the reservoir being boosted to higher energies. But there was something else that I’d overlooked, and it was a huge contributing factor—indeed, it made all the difference. It hinged on the fact that the universe itself was very different 12 billion years ago. At that epoch it wasn’t yet 2 billion years after the Big Bang, and back then the cosmos was a much smaller place. Spacetime was quite literally more compact; there was much less space between everything. The distance between the baby forms of galaxies was certainly growing, but it was less than a quarter of what it is today. This also meant that the cosmic background photons had not yet been stretched as much as they would be over the next 10 billion years. At that stage in the history of the universe they were almost five times more energetic than they are today, their little wavelengths that much smaller. If I combined these factors I found that these cosmic photons had provided an electromagnetic ocean around 4C41.17 that was more than five hundred times richer than it is now. It was quickly clear that this could be the answer! The thick sea of photons would bounce and scatter off the electrons pouring out from around a black hole, get boosted in energy, and would light up the region with X-rays. It was like shining a searchlight into a fog—the volume of the beam itself would glow with scattered light. The only difference was that this was a beacon we could see across the known universe, and it would only get more efficient the further back in cosmic time we went.

  This meant something else, too, something wonderful. If this was indeed the birthplace of a galaxy cluster, then we were witnessing a central black hole blowing bubbles just as its descendants do in the present-day universe. Except the bubbles were not dark voids, they were lit up as they boosted photons into the X-ray band. Push the cosmic clock back far enough, I realized, and you could invert the color scheme of bubbles in a structure, and a negative could become a positive. It was a beautiful and elegant manifestation of basic, fundamental physics. I called Smail in the U.K., bursting with excitement. We might not have seen the hot gas of a baby galaxy cluster directly, but we’d found the stuff inside that gas. We’d found the glowing bubbles driven by the massive heart of a black hole.

  Now there were some big questions that we had to tackle. We wanted to know exactly what was happening in this extraordinary environment. Our measurements told us there was lots of warm dust. Altogether this dust represented a hundred million times the mass of the Sun and was made up of tiny, microscopic grains. It was spread out across a hundred thousand light-years. Something was heating it up, perhaps scores of young bright stars, perhaps the supermassive black hole as it wolfed down more matter. The hole was squirting out relativistic particles and inflating bubbles within an unseen medium of gas, and this gas was being gathered up in the growing gravity well of the system. But it also had to be getting dense and cool enough to produce all the big and short-lived stars that were in turn making all the dust, and it had to be feeding the black hole. We were missing a crucial piece, something that would tie it all together, something that would show us exactly where the rest of the matter was in this system.

  A few weeks went by and serendipity raised its head again. This time it was in the form of a chance meeting with the Dutch-born astronomer Wil van Breugel, from the Lawrence Livermore National Laboratory and the University of California. Van Bruegel’s specialty and passion was tracking down the most distant and massive galaxies. He also happened to have access to the two great Keck telescopes that are perched on Mauna Kea in Hawaii. At an altitude of thirteen thousand feet, these enormous hunks of steel and glass could gulp down photons from across the universe. They were the perfect tools for capturing visible light from ancient structures. Mentioning our investigation provoked a strong response in van Breugel, who told us he had something we’d want to see.

  Wil van Breugel and his colleagues had used custom-built light filters to sniff out the photons that came from very specific changes in the energy hierarchy of electrons within hydrogen atoms. Like any element, hydrogen has a number of electromagnetic scents, and this was one of the key ones. If you could hold these atoms in your hand, they would glow with a distinct ultraviolet light. Place them across the universe, though, and the expansion of spacetime would stretch that light out to visible wavelengths. To exploit this, Wil van Breugel’s special filters were tuned to catch precisely these photons from various cosmic objects whose distance was already known. What he showed us took our breath away.

  He and his colleagues had captured our object, 4C41.17. It had taken one of the Keck telescopes, with its more than thirty-foot-diameter mirror, over seven hours of exposure time to produce an image. That alone was quite an achievement, but it was the view that stunned us. There was the hydrogen gas, recovering from some as yet unknown buffeting, cooling off by emitting photons of ultraviolet light. It was a colossal structure, and in its center were bright clumps and specks, each thousands of light-years across. But they were inside an even larger canopy, stretching out across the same space that our X-ray light came from. And that canopy had sweeping shapes and forms, cusps and spurs of light. What looked like a huge band of dust seemed to obscure part of the hydrogen gas, as if a belt were tightly adjusted around an overflowing midriff. There was an hourglass shape to the whole thing, hundreds of thousands of light-years across. It gave the impression of matter both coming and going, flowing inward, but also being propelled outward. It was a picture of a tempest from 12 billion years ago. Smail and I knew that we now had the other piece of the puzzle. We just had to put it all together.

  Figure 15. The incredible structure of warm hydrogen gas seen around the distant object 4C41.17. End to end, it spans over three hundred thousand light-years. Near the center great bands of darker, dust-rich material seem to wrap around like a belt. The gas coincides with the spread of X-ray light seen by the Chandra Observatory. Also a source of radio emissions and intense infrared light (seen as submillimeter radiation), this remote place is a gigantic, busy construction site for a galaxy and its billions of stars.

  We needed to form an accurate mental picture of this distant, primordial environment. A few phone calls later, and over in the city of Leiden in the Netherlands, one of van Breugel’s collaborators, Michiel Reuland, found himself tasked with making a visual representation of all our data combined. It was like constructing a painting in translucent layers, each representing a different part of the electromagnetic spectrum, and each overlapping with the others. Experimenting again and again with different color schemes, and by morphing the maps into smooth and simplified forms, Reuland finally came up with a portrait.

  It wasn’t pretty. In fact, with its artificial neon colors and overlapping shapes, it looked like a colossal mess. I nicknamed it “cosmic roadkill.” But despite the aesthetic issues, it was wonderfully informative. Here was just about everything we knew about this remote object in one single view. In the core was a thick shroud of obscuring dust, tens of millions of times the mass of our Sun. It betrayed the intense formation of new stars and new elements hidden within. The vast spread of ultraviolet light from hydrogen gas looked like an unfortunate spill of curdling milk, splattered across three hundred thousand
light-years. On a similar scale was the fearsome X-ray glow from cooling but still relativistic electrons as they powered up cosmic background photons and the infrared photons from the warm dust. This X-ray light seemed intertwined with the ultraviolet glow of gas. In a line following the main axis, the main thrust of all this radiation and matter were the dumbbell clouds of radio wave emission. Here the very highest-energy electrons were flowing and corkscrewing outward, glowing with the synchrotron radiation that we see in today’s systems. They were still quite fresh, just a few million years old. Their origin could only be from a young supermassive black hole at the very core of this chaotic space. And unseen, but hinted at by the very edges of these structures, there had to be an outer cocoon of gas cooling and flowing inward from the cosmic web.

  Particularly intriguing to me was the transfer of energy from the jet-driven particles of the black hole to the cosmic photons in order to produce the glowing X-ray forms. Some of this inverse Compton energy had to be absorbed by the very same hydrogen gas that we were seeing all across the structure in van Breugel’s images. That energy could help heat and strip electrons away from this material, ionizing the hydrogen gas and slowing down its cooling. This would in turn put the brakes on the condensation of raw material into stars. As the atoms sought to regain those particles, they would glow with ultraviolet radiation when the captured electrons settled back into the atomic energy levels. This was a new mechanism for a black hole to reach out and tweak the cosmic structure. It would only happen in the youthful universe, where spacetime was still compact enough for the soup of photons to efficiently seed this transfer. If this was correct, then we had found a new way for black holes to sculpt and mold the world around them, to disrupt the formation of new stars and structures.