A hot day in holmdel

Today we observe an expanding universe, with distances between galaxies stretching out according to Hubble's law. At the austere summits of remote mountains in Chile and Hawaii and Arizona, giant telescopes slowly gather photons from distant galaxies, building up the evidence for understanding an ancient and remote universe. But another important component of the universe was discovered in Holmdel, New Jersey, near Exit 114 off the Garden State Parkway. In this prosaic setting, Arno Penzias and Bob Wilson found that the universe is full of ancient light: glowing embers from the Big Bang.

More prcciscly, in 1965 they found a hiss of radio emission everywhere they pointed their radio antenna. We now know this emission has the spectrum an opaque object emits at a temperature of 2.725 ± 0.001 kelvins. That's 2.725 centigrade-sized degrees above absolute zero. Today, the universe is transparent, so photons can travel from distant galaxies to us without being absorbed. The light we see from galaxies is a complex mixture of emission from many different stars and gas clouds that carries subtle information about the composition and temperature and motion of the emitters. But the emission Penzias and Wilson discovered was much simpler—it comes almost cxactly equally from all directions and its entire spectrum is described by just one number, the temperature. There are no details.

This gentle bath of low-energy photons is the relic of an earlier time when the universe was hot and opaque, so it behaved like an oven. When you heat an electric oven, the heating element emits infrared light, the cool walls absorb that light, and they warm up until they too begin to glow with infrared light. When an oven has been fully preheated, the thermostat switches off the heating clement. Emission from the walls now fills the oven with an even glow of infrared light. When you put the dough in a baking pan and slide it into the oven, it absorbs energy from the infrared photons bouncing around inside the oven until it, too, approaches the temperature of the walls. Now you're cooking!

That's how you bake bread—dough warms up toward the temperature of the oven walls as it absorbs infrared light. Everything in an oven tends toward the same temperature. Bouncing photons guarantee that this equilibrium is enforced. The spectrum of photons inside the oven is determined only by the temperature, not by the chemical composition of the oven walls or the type of raisins in the dough. Ordinary kitchen ovens don't get hot enough for human eyes to see the walls glow, but a ceramic kiln or a well-kindled charcoal grill does. The red glow of coals in the heart of a fire is radiation of this type, and we all know that the color of the coals tells the temperature of the fire—dull red coals are cooler than bright orange ones, '¡"he cosmic microwave background is the glow from the hot Big Bang—but the temperature of 2.725 degrees above absolute zero means we don't detect this with our eyes: wc need radio receivers like the one that Penzias and Wilson built.

Inside any region of the opaque universe, the same effect is at work—all the objects in an opaque universe come to the same temperature, because photons flying around at the speed of light ensure that any region that is a little cooler gets warmed up, while any region that was a little hotter gets cooled off. Penzias and Wilson detected photons that had their spectrum formed when the universe was opaque. Straightforward calculation shows the universe then had a temperature of at least 4000 kclvins.

So the cosmic microwave background photons observed in New Jersey come from a time when the universe was 1000 times hotter than it is today. These photons have stretched with the cos mic expansion by a factor of about 1000 since they last bounced off matter. Emitted as visible light, they have been degraded by expansion down to the low-cncrgy photons that radio tclescopes detect so well.

These photons fly through a transparent universe, carrying their image of the infant universe in all directions. When they were emitted, the scale of the universe was 1000 times smaller, the density of matter in the universe a billion times higher, and the temperature 1000 times hotter. Those photons show us what the universe was like when it was very young, just at the moment when it changed from being opaque, like the walls of an oven, to transparent, like a window.

This physical change in the universe at large from opaque to transparent results from the microscopic rearrangement of individual electrons and protons. When the universe was hot, the electrons and protons that make up ordinary matter were moving too fast to assemble into hydrogen atoms. Photons bouncing around had plenty of energy to rip apart any atom that did form. But, after about 300,000 years of expansion and cooling the warm post-Big Bang haze of matter and light finally coolcd enough for electrons to give up their freedom. Electrons joined protons to form hydrogen atoms without being harassed by disruptive ultraviolet photons. Free electrons are good at scattering light; hydrogen atoms with electrons in bound orbits are much less effective: the hazy universe turned transparent when hydrogen atoms formed for the very first time.1

The cosmic microwave background CCMB) provides the most dircct cvidcnce that the universe had its origin in a hot Big Bang. This is not just an impression based on the expansion galaxies show, but a real physical change in the universe over time. Hie universe we see today has elaborated over cosmic time from a hot, opaque, evenly distributed soup into a cold, transparent, lumpy universe with galaxies, stars, planets, and people. The early universe was simple and predicable using straightforward physics. But once the universe turned transparent, things began to get interesting, complicated, and unpredictable. That's the messy realm of astronomy.

Detecting the cosmic microwave background was a major event for cosmology. A hot Big Bang had been contemplated by George Gamow and his students I lermann and Alpher decades earlier as a possible site for the synthesis of elements, but this never led to a search, and the site of manufacture for heavy elements was later identified in stars and supcrnovac. Even though Pcnzias and Wilson were not intending to find out anything about the universe, their measurement was so important that they received the Nobel Prize in Physics for 1978.2

But there is something curious about the uniformity of the CMB. Hie fuzzy horizon of the CMB is off in the distance in all directions 14 billion light-years away. And the temperature we see in any direction is 2.725 kelvins. But, spin on your heel, and you can also see 14 billion light-years in the opposite direction, where the temperature is also 2.725 kelvins.1 Now, in an oven, things come to the same temperature because the photons from a warm region sap energy from the hot places and heat up the cool regions. But photons can only travel at the speed of light, and when they are bouncing around in a fog, they propagate even more slowly. The regions wc sec on opposite sides of the sky have never been able to exchange photons to even out differences. Why do they have the same temperature?

There's something odd about this. It's as if you traveled a billion light-years at 99.999% of the speed of light, landed on a planet, and found the inhabitants playing baseball. By exactly the rules of major-league baseball: no aluminum bats. It would make you wonder, if you were really our first emissary to this distant place, how they knew to play by the same rules as the Red Sox. So the question is, "How did the universe get so uniform?

One idea that sounds wild and fanciful, but that is taken seriously by thoughtful people, is that the entire patch of the universe that we see in all directions was once small enough for photons to establish a single temperature. Then, due to an energy associated with empty space, the universe underwent a tremendous exponential expansion in which the scale of the universe increased by a factor of something like 10 w during the time around 10 seconds after the Big Bang. In this picture, during the "inflation era," the observable universe grew from a region so small that photons could cross it in the time available into something the size of a grapefruit. The prccisc numbers depend on the details of how particles and fields behave at energies that have never been observed by any particlc accelerator on Farth, but the basic idea does not depend on these details. In this picturc, today's expanding cosmic-horizon is once again encountering regions that were once before in contact.

In other words, before inflation, the material in the observable universe was oncc in good thermal contact, like the interior of an oven. Then, during the inflationary era, the universe expanded exponentially, placing regions that were once in touch out of contact. Inflation ended somewhere around 10"* seconds, then a lor of time (1017 seconds—a Hubble time!) passed. For cach placc, the observable patch of the universe grows—now we can see other parts of the universe billions of light-years away. When regions say hello again, 14 billion years later, they have the same temperature because they were in touch long, long ago, in the fraction of an instant before inflation got rolling.

This "inflation" idea sounds crazy. The fact that it is taken seriously by people who sit firmly in endowed chairs doesn't automatically make it right. Bur it has strong roots in the quantum world of particle physics and it does more than just resolve the "horizon problem" of a uniform temperature in parts of the universe that are just now getting in touch. Inflation makes this a neonatal ward reunion instead of a first-time meeting. Inflation also makes some firm predictions about departures from absolute smoothness and about the geometry of the universe. These predictions can be subjected to observational tests. If the predictions are not borne out, then the simplest version of the inflation idea can't Ix? right.

If the predictions arc confirmed, that doesn't necessarily mean inflation is the right picturc. After all, there could be some other iclea we haven't thought of yet that would also make these predictions. But if inflation keeps passing observational tests, it's not just sloppy logic to think we might be on the right track. It could have been shown wrong!

The physical mechanism for inflation has its roots in the weird world of quantum physics. One idea that has proved very fruitful in the quantum realm has been to think about the properties of empty space: the vacuum. In the subatomic realm, ordinary common sense ideas turn out to be worse than inadequate—they are just plain wrong. In the big world of tilings we can sec with our own eyes, objects like a thrown baseball have a definite location at every instant, and motion that we can measure with a radar gun. But on the small scale of electrons and protons and below, these commonsense ideas of position and motion are replaced by a kind of intrinsic vagueness: the Heisenberg uncertainty principle says that you cannot know both the exact position and motion of something at the same time.

For big objects, this is not a practical issue, but on the subatomic scale, it is of the essence. The human scale is as big compared to the atomic scale as a star is compared to a human. You can't really expect to have a good feel for what things are like for an electron. We can't say an electron orbiting the proton in a hydrogen atom is exactly "there," with precisely such and such a motion, but arc driven to more subtle formulations describing the probability of finding an electron in a given state.

For inflation, the weird idea is that the vacuum of empty space may have an energy associated with it. You may think empty space must have zero energy, but physics does not tell us that empty space must have zero energy. It's a little like kx>king at a topographic map of the Earth—the heights are given as the distance above sea level, but that leaves out the radius of the Earth. In the same way, physical events tell us about energy differences, but they don't tell us if there's an underlying floor of vacuum energy. There could be, either for a brief moment, or for a longer time, an energy of the vacuum that is not quite zero that lurks below all the measurements of energy differences that we make.

The effect of a vacuum energy in general relativity would be a "negative pressure" that makes the expansion of the universe accelerate. If the energy in the vacuum stays constant or just declines slowly enough, the rate of expansion is proportional to the size: it is literally an exponential growth, just like compound interest, and just like currency inflation. In December 1979, Alan Guth, a not-so-young postdoc in a temporary job (now Wcisskopf Professor at MIT), was not thinking about carecr advancement as he rode his bicycle to work at the Stanford Linear Accelerator. lie was thinking about what might happen if the universe got into a state where the vacuum energy wasn't zero. I Ic was so eager to get to work that morning, to check out the consequences of his wild idea, that he set his personal best cycling time of 9 minutes, 32 seconds. After a few years of bruising price rises in the late 1970s, inflation was in the back of everybody's mind, even a mind as busy with other ideas as that of the other-worldly Guth. That's why this runaway expansion of the universe in the first 10"^ seconds is called inflationary cosmology.4

Physicists like this idea for the origin of the Big Bang. First, it comes from their turf: the world of theoretical particle physics, not the messy world of astronomical observation. "Scalar fields," like the field that produces inflation, are their bread and butter. Scalar fields give masses to the quarks that make up neutrons and protons. Particle physicists do not regard inventing such entities out of whole cloth as a strange way to think. They do this before breakfast. Second, it is mathematically elegant, and if truth is beauty, then beauty is truth and inflation must be the right model. Or, to put it more seriously, this is a powerful and attractive theoretical idea. Third, it accounts for known facts like the expanding universe and the uniform microwave background. But most important, it makes some predictions, at least in its most straightforward forms, that observers can test. Inflation spans the microscopic and the cosmic— it is audacious, esthetically appealing, and, test of all, we can find out if it is wrong.

One prediction of the simplest version of inflation is that the universe will have the geometry of flat spacc: that £2 = 1. Even if the universe started out with some curvature, the tremendous expansion of the inflationary era would increase the radius of that curvature and force the geometry to become the geometry of flat space. If you take a region the size of a grapefruit and expand it to the size of the universe, the rind will be very, very flat. Or, as Guth says, "The value of omega will tie driven to one with exquisite pre cision." So, if we can measure the effects of f2, we can test whether this is true and find out if this version of inflation is wrong/

A more subtle feature of inflation is that you can compute the character of variations in density f rom place to place in the universe. If quantum mechanics rules the first instants of the universe, then quantum uncertainty predicts there must be a range of values for the density of matter and energy that you measure when you sample different chunks of the universe. What this means is that the universe should contain a variety of density variations that resemble waves ranging from tiny little ripples to the longest waves that could fit into the cosmic horizon at every instant of the inflation era. These variations in energy density will leave an imprint on the cosmic microwave background that we can detect as subtle temperature differences from place to place in that smooth background, like a watermark on otherwise smooth bond paper.

These random variations would be the ultimate origin of the large density differences we see today in the distribution of galaxies. 'Ihe action of gravity in the past 14 billion years amplifies those initial seeds into the jungle of cosmic ecology we observe today. We stan from random fluctuations, gravity organizes matter to form galaxies and stars, nuclear physics elaborates the elements inside stars, and then the universe begins to get interesting, eventually making planets and people. So another test of inflation is to see whether people on a planet (Earth!) can see the predicted fluctuations in the microwave background.

Early measurements of the cosmic emission showed that the microwave background is smooth. Unlike the high-contrast galaxy distribution we see today, with dense clusters and yawning empty voids, at the time when the universe cooled and turned transparent, matter in the universe was almost exactly evenly distributed everywhere. Almost exactly, but not quite. Very careful measurement of the CMB from satellites, balloons, and ground-based instruments at very dry sites like the Atacama desert in Chile and the South Pole shows definite signatures of subtle variations in the brightness of the background.

The lumpiness in this cosmic soup is about one part in 100,000—that's like having a scoop that digs out $1000 in pennies

From a tremendous penny jar, and getting the same answer every time, to the penny. That's really smooth. A baby's tottom is the colloquial standard of smcxithness. Hands-on observations of my own children showed that a bottom has bumps of 0.1 millimeters on a span of 10 centimeters, so it's only smooth to one pan in 1000—a human infant's skin is a hundred times rougher than the infant universe. And that's without diaper rash.

A map of these tiny variations reveals some important clues to the physical state of the universe when it was young. It shows the dense regions, destined to grow denser as gravity magnifies inequality and the low-density regions that are fated to lose out as time goes by. 'Ihcse tiny variations are the seeds that flower into the high-contrast bouquet of clusters and voids that we see kxlay in galaxy surveys. Just as the rich get richer, the dense get denser through ruthless cosmic unfairness as gravity makes contrast grow.

The first map of these fluctuations in the early universe was made in 1992 by the Cosmic Background Explorer (COBE) satellite. Those early observations smeared together the measurements to an angular scalc of about 6 degrees, about the angle your fist covers on the sky when you hold it at arm's length. Even in this blurry image of the sky, COBE definitely detected fluctuations of the general sort predicted by inflation. While this did not prove that the inflation model was right, it was a test that the model could have failed.6

We see an expanding universe, with the distance between galaxies stretching out over time. We see the relic glow of a time when the universe was yoting and smooth and hot. There is another picce of evidence that the universe wc see today is the result of a hot Big Bang 14 billion years ago. That is the ubiquitous presence of helium, the second-simplest element, in stars of all ages. Helium is produced alter inflation ends (if inflation really happens) in the hot, expanding universe.

There arc degrees of audacity. Inflation is an extrapolation far beyond anything we're ever measured in a terrestrial laboratory. While it is an intriguing idea, it is a speculation. The inflation era corresponds to energies 1013 times larger than have been produced

figure 7. f. Th» growth of structure. Once baryons recombined. they could move under the farce of gravity. Matter that could form galaxies, stars, planets, and people drained into the valleys that dark matter formed, as shown in these computer simulations The distribution of luminous matter tracts the presence of dark matter. Courtesy of The VIRGO Consortium figure 7. f. Th» growth of structure. Once baryons recombined. they could move under the farce of gravity. Matter that could form galaxies, stars, planets, and people drained into the valleys that dark matter formed, as shown in these computer simulations The distribution of luminous matter tracts the presence of dark matter. Courtesy of The VIRGO Consortium in the most powerful particle accelerator on Earth, Ah we learn more about the subatomic world, as we continue to journey inward Toward measuring the properties of the very small, inflation may or may not seem like such a great idea twenty years from now. But the world at 4000 kelvins or 40,000 kelvins or 40 million or even 40 billion kelvins is well within the scope of today's experimental physics.7 We're not guessing about how electrons and protons and neutrons and neutrinos interact at these temperatures. This is the low-energy realm of nuclear reactions and, for good and for ill, we know how those reactions work in stars and in bombs. Thinking about a 100 billion degree opaque universe in the seconds after the Big Bang is not nearly such a big extrapolation as speculating what happened in Lhe first 10~35 second! Our knowledge of the time when the universe was as hot as the inside of an exploding star is really quite secure.

Complex elements such as oxygen or iron are prcxluced when stars generate energy or erupt in supernova explosions. We know from spectra that the oldest stars in our Milky Way galaxy have only 1/1000 as much of these elements as the sun does. This means that the abundance of these elements has been building up over time, like old shoes in the back of the closet. The exception to this is the element helium—the second element in the periodic table. Although helium is produced in stars as they fuse hydrogen, even the oldest stars have about as much helium as the sun. When we look at gas clouds in other galaxies, as Wal Sargent and Leonard Searlc were doing in the early 1970s, following up lists of strange objccts that Fritz Zwicky was compiling in his basement workrexim at Caltech, they found that there are some galaxies with very little oxygen. Presumably these are the places where stars have done the least to enrich the mix, and these galaxies are closest to the composition that came out of the Big Bang itself. But even the most pristine gas cloud seems to have a dollop of about 25% of its mass in helium. This is a powerful clue that helium has not been building up over time in the same way as other elements. How did helium get a head start?

The answer to this riddle lies farther back in time than the era of stars, in the hot, dense Big Bang. The microwave background shows us an image of the universe when it was 1000 rimes cozier than it is today. If we dare to push back another factor of 1000 in cosmic scale, beyond the time we observe directly, the universe would have been a million times hotter than it is today. We can't see into that era, because the universe was opaque, but we do understand how things work at these temperatures and densities. We can't sec into the center of the sun, but we know what's going on in there, and this is similar, if more remote. Going back another factor of a million beyond that stage is still within the realm of well-tested terrestrial physics. The universe would have been a nuclear furnace, fusing the lightest particles into helium. Or, more precisely, since the universe was on a one-way trip from hot to cold, a nuclear freezer, in which nuclei froze out oncc the temperature was low enough,8

At the end of the first few minutes after the Big Bang, as the temperature sank low enough for the simplest nuclei to stick together without being broken up by high-energy photons, there must have been a universal game of musical chairs. Every proton would have grabbed a neutron to form deuterium, and then in a few steps, the deuterium nuclei would form helium. A helium nucleus has two protons and two neutrons, so by computing the number of neutrons present when the universe was cool enough for deuterium to stick together, we can figure out how much helium would form in the expanding Big Bang. This works out to be about 25 pcrcent of all the mass of ordinary matter. And that's just about what wc see. When the numbers come out this close, the ideas have the ring of truth.

Even the first generation of stars would start out with a dowry of helium from the Big Bang. George Gamow started out with the aim of cooking the elements in the fireball of the Big Bang, but this source stumbles at the gap from helium to lithium that stars bridge by whacking three helium nuclei together to make carbon. Wc inherit carbon and oxygen and iron and gold from previous generations of stars, but helium is a legacy directly from the Big Bang itself.9

So we have gcxxl observational reasons to think the universe began about 1A billion years ago as a hot, dense Big Bang. After a brief early epoch of exponential expansion, the universe was a simple, hot, nearly uniform place. The element helium formed as that oven cooled. Before (recombination, the growth of contrast, at least for ordinary matter like protons and electrons, was checked by the interactions of matter with light, which would act to smooth out any lumps. After recombination, hydrogen atoms made up most of the ordinary matter, and oncc the universe was transparent, gravity could begin to make ordinary matter grow lumpy. There must have been a first generation of stars in which nuclear reactions generated energy and made a start on the elements of the periodic-table. Galaxies began to form out of the uneven distribution of matter, and big galaxies formed by gobbling up their little neighbors.

Our Milky Way would be the product of a long tree of mergers going back 13 billion years into the past. The sun and the Earth formed from the gas rich in iron and silicon and calcium and oxygen and carbon accumulated in our galaxy after 8 billion years of stellar burning. And here we are, living things made of carbon and calcium and iron, breathing oxygen, and looking back up the river of time toward our origins. This is a beautif ul and simple picture of where we came from.

Of course, beautiful and simple are not always the same as "correct." If you look closely enough at the luminous fresco by Michelangelo that decorates the ceiling of the Sistine Chapel, you can begin to see the cracks, smudges, and gaps in the picture. In the same way, if you take a close look at this picture for the expanding universe, you can see places that need more work. This doesn't necessarily mean the framework is wrong, but it does mean we need to understand better what is in the universe and how the laws of physics play out to make the world around us.

One crack in the fresco is the measured amount of matter, and our curious inability to say precisely what the matter of the universe is made of. While inflation suggests that Q = 1, direct attempts to measure the matter of the universe indicate something different. Fritz Zwicky, irascible but prescient, showed in 1933 how to measure the mass associated with galaxies by measuring the speed of galaxies as they swarm in galaxy clusters. The more mass in a cluster, the faster the galaxies will move. Measure the motions of galaxies relative to the cluster redshift and infer the mass. This technique, and other effects that depend on mass that have been developed in recent decades, like the emission of X-rays from gas in clusters, or gravitational leasing in clusters, all point to the same result—the total mass that is clumped with galaxies is much larger than the mass of the stars emitting visible light, but much too small to give a gravitating mass density, £2„;, equal to one. The best estimates give values of iim closer to 0.3 ±0.1.

A common approach to this problem of the contents of the universe, employed regularly over the last decade, but familiar since Biblical times, has been a heady mixture of skepticism mixed with flattery and a dash of pride. More than one theorist has said to me, with a thin-lipped smile, "Well, Bob, measuring the matter density and the expansion rate of the universe are very difficult things done by talented, but, let's be frank, fallible observational astronomers. Astronomers have been wrong before and may well be wrong now. Not all observations are correct. Since we know, from our highly developed esthetic sense, that £2m equal to one is the right answer, you observers should just go back and do the measurements again until you get it right."

We bring data down from the mountains on magnetic tape, not stone tablets, and there have been many false steps in building the observational picture of the universe. What has changed, but only in the last five years, is that the observations have become more certain, more telling, and the conclusions cannot be ignored even when the implications are quite uncomfortable. This has led to a surprising new synthesis of theory and observation, but only by inviting one of the old skeletons out of the closct: A the cosmologi-cal constant.

What makes the measurement of the matter content of the universe especially interesting is that even £2m of 0.3 demands that most of the matter in the universe is invisible and unfamiliar stuff. Put another way, ftm = 0-3 ± 0.1 is 7a low compared to QIT. = 1, but big compared to the density you'd get by adding up the masses of all the visible stars that make galaxies shine. If you do that, you get only CI = 0.005. More generously, when you add in the mass of hot gas we see emitting X-rays and all the other matter we can detect directly, the sum is still only about one-tenth of the total mass we know is present in galaxy clusters. We know the mass is present because we see its gravitational effects, but we don't see light of any form being emitted or absorbed by this material. So we conclude that most of the matter in clusters, and presumably in the universe at large, is dark. Zwicky named this "dunkle Materia" dark matter. "Matter" because we know it is there. "Dark" because we can't see it. But having a name for something doesn't necessarily mean you know what it is. Or as Zwicky said in 1957, 'it is not certain how these startling results must ultimately be interpreted."10

There is an even more curious problem with the nature of the dark matter, based on a combination of observation, reasonable physical theory, and current understanding of helium cooking in the Big Bang. That confluence of evidence suggests that mast of the dark matter is not made of the neutrons and protons and electrons that make up our bodies, the Earth, and all the stars wc see, but is mostly "matter" that is very different from the material world we know.

The argument is a bit subtle, but it leads to a very interesting conclusion. During the nuclear cooking that synthesizes helium in the first minutes of time, deuterium, the dclicate isotope of hydrogen that has one neutron and one proton, plays a special role. Deuterium sets the moment when helium synthesis can begin. Helium gets assembled only after the universe cools enough that deuterium can survive the bath of gamma rays that is the cosmic background radiation in the early universe. Most of the deuterium nuclei then get locked up into helium nuclei, but a little is left over. The moment of helium synthesis passes as the universe expands and cools. Some stragglers of deuterium survive to bccome pan of the gas in the universe we see today.

The amount of deuterium that survives the mad dash to assemble helium is small, but detectible. The leftover amount is very sensitive to the density of neutrons and protons in the universe at the time of helium assembly. So the amount of deuterium delivered by Big Bang cooking depends on ft—the density. More precisely, it depends on ftt„ the fraction of the universe that is made of baryons. "Baryon" comes from the Greek word for heavy—and this is apt since neutrons and protons are heavy compared to the leptons (from the Greek word for "light"), like the electron and the neutrino. Here's the curious fact: measurement of the amount of deuterium, seen in absorption lines formed in intergalactic gas clouds, shows that the amount of deuterium (several parts in 105) left over from the era of helium cooking is more than 10 times larger than you'd compute for £2b = 1. The best estimate for fth based on the residual deuterium is about 0.04 ± 0.01.

Quantities matter. If the amount of matter, ftn., is about 0.3, and the baryon density Qh is 0.04, 7 times smaller, then most of the matter in the universe cannot be baryons. Even if measuring errors and systematic errors have thrown both of these numbers off by a factor of two, we would still conclude that most of the dark matter in the universe cannot be anything made from neutrons and protons—the stuff of all the chemical elements, and of our own bodies, if we take this conclusion seriously, then we are not made of the kind of stuf f that makes up most of the universe.

What's more, when we use our baryonic brains to try to think what most of the matter in the universe could be, there is one conspicuous candidate. We know of elusive panicles that don't emit or absorb light and are not baryons: neutrinos. Neutrinos seem like a very gcxxl candidate for the dark matter, except for one thing. The problem with neutrinos as the gravitating matter that makes up most of the mass in the universe is that they have too little of precisely the one thing dark matter must have: mass. Lack of mass is a real drawback for something that is supposed to outweigh all the stars in the universe! There is evidence now from underground neutrino detectors that the mass of a neutrino is not quite zero, so neutrinos do make a small contribution to the total of about 0.003-A neater universe crafted by Occam's razor might have just one form of dark matter, but our extravagant universe apparently must have at least three: some dark baryons, a pinch of neutrino mass, but mostly something else, instead of a minimalist universe, we seem to live in a rococo one: wc have everything you can think of, and more than you can think of. Perhaps wc should not be so quick to use Occam's razor to reject wild ideas: we need even wilder ones to interpret these startling results.

If we follow this chain of argument, most of the universe is in a form of dark matter that isn't baryons and isn't neutrinos. We know what it isn't but we don't know what it is. Theoretical particle physics has produced some possible candidates with whimsical names like the axion and the ncutralino. These particles may have the right properties to be the dark matter, but at present they have the distinct disadvantage that they have not yet been discovered! Particle physicists are rightfully proud of the role that powerful theoretical ideas have played in predicting the existence of particles that have later been found (like Dirac's prophccy of the positron— the antimatter clone of the electron). But it doesn't seem unreasonable to wait f or terrestrial experiments to show that these particles actually exist and have the right mass before asserting Ux> confidently that they make up most of the universe.

If the dark matter is something like a neutrino, only with a lot more mass, those particles would be everywhere. Since they don't interact by the strong force that glues nuclei and they don't interact by the clectrical force that makes it hard for people to walk through walls, these "weakly interacting massive particles" (WIMPs to the wags who dub these things) would be present in the room where you read this book. As the Earth orbits the sun, the sun orbits the center of the Milky Way, and M31 tugs the Milky Way in its direction we would be drifting through a mist of WIMPs just as we are drifting through the photons of the cosmic microwave background. You can detect the microwave background from anywhere, and you could find the dark matter just by catching one of these particles as it drifts through your laboratory.

Now, just as the academic prestige of the inflation theorists doesn't prove they are right, the fact that people have built experiments to detect WIMPs doesn't prove that most of the mass in the universe is in this weird form. But it does show that competent people take these arguments seriously enough to test the ideas by observation. As a scientist, you really have control of only one resource: your own time. When professors and postdocs and graduate students spend years to build a delicatc WIMP-catching apparatus, and set it up, not at a beautiful mountaintop in Chile, or even off the Garden State Parkway in New Jersey, but deep in an oppressive abandoned iron mine in the middle of nowhere Minnesota, you know they arc serious about trying to find our what the world is made of.

Cosmic timescales pose the most difficult problem for a universe with equal to one. Gravitation slows cosmic expansion, but the amount depends on In the low-£2m case, you can correctly compute the cosmic age from the present rate of expansion, i„ = 1 ///„. Recall our mythical marathoner Eddie, who computed the time elapsed in the Boston Marathon without a watch. He measured distance and velocity for various runners assuming all of them ran at a steady speed from the start in Hopkinton to the finish line on

Boylston Street. This is just like a low-density universe, where gravitation doesn't slow the expansion.

If the universe does have an appreciable mass density, the relation between the present rate of expansion and the actual elapsed rime since the Big Bang is a little less simple. Gravitation slows expansion, making the Hubble time an overestimate for the age of the universe. Estimating the age of the universe from the local expansion rate, the Hubble constant we measure in the local patch out to 1 or 2 billion light-years, is equivalent to looking only at the last miles of the Boston Marathon. You don't know what the runners were doing earlier, so you just assume that the present is like the past, and make your best estimate. But it ain't necessarily so. If the runners arc actually slowing down, but you watch them only over the last mile, you will overestimate how long they've been running the course. If some poor footsore devil limps the last mile in 10 minutes, you might think they've been out on the course for 26 miles x 10 minutes/mile = 260 minutes = 4 hours and 20 minutes. But maybe they were churning along fine at 7 minutes per mile until they hit the wall on Heartbreak Hill and they've been slowing down ever since. Observing those aching survivors only at the end of their travail will lead you to overestimate the actual time they've I)cen suffering out on the course.

Similarly, if mass has been decelerating the cosmos, then the universe, like somebody who turned gray at 35, is younger than you think from a first glance (always check the eyebrows!). If you start out with closer to one, the slowing-down effect gets larger. The boundary of this ever-slowing expansion is = 1.000000. . . . In that case, when you computc the effect of expansion and deceleration, the age of the universe turns out to be exactly two-thirds of the age you would infer from the present rate of expansion. The real elapsed time since the Big Bang is just two-thirds of the Hubble time. In symbols, we could write, tlt = 2/3 (1 /HJ.

If gravitation has been slowing cosmic expansion, the real age of the universe would be younger than 14 billion years. Deceleration would reduce this to 9 billion years—significantly shorter than the 12 billion years estimated for ages of the oldest globular clusters or white dwarfs. This would be embarrassing, liven taking into account the uncertainty in the ages of the oldest stars of 1 billion years, this would be a 3o discrepancy. Gauss says that only happens by chance one time in 370, so, if the numbers arc right, there's a 99-7 percent chance that there's a real problem with the cosmic ages. Globular clusters should not be older than the universe in which they reside! Common sense suggests that this much deceleration can't be present, even though ilm - 1 apparently demands it. This is definitely a crack in the fresco! Or, to put it in a more positive light, what we know about the ages of stars helps separate the one real universe we actually live in from the many that are mathematically possible.

Appeals to common sense are not good enough. We should look for effects to measure from direct observation, not esthetics, or even logic, whether the universe has or has not been decelerating. The best way to do this is to use powerful tclcscopes to look deep into the past to sec how cosmic expansion has changed over lime. In recent years, we have used supcrnovae, detected halfway back to the Big Bang, to trace the history of cosmic expansion and measure its change.

A value of = 1 is the razor's edge. If is even slightly more than (me, say 1.001, then the expansion will eventually stop, reverse, and become a contraction. If the universe started out in a Big Bang, then a universe with £2m greater than one will eventually end up in a gnaB giB, back in that unimaginably hot and dense state. All the elaboration of the universe would be reversed—stars would evaporate back into gas, nuclei eventually melt back into the simple particles out of which they are made, and the wonderful complexity of the world would be erased. It's not a pretty thought, but we shouldn't expect the universe to care what we think.

Although the cosmoiogical constant was exiled to a theoretical leper colony after the 1930s, it is worth exploring how A affects cosmic ages. Einstein invented A to balance out gravitation to produce a static, eternal universe. Eternal is an age. Infinitely old. De Sitter noticed that A would make a massless universe accelerate, and Eddington suspected that Slipher's observations of the reces-

Figure L4

The spiral galaxy pair NGC 220T and fC 2143. Distances between galaxies are not always large compared to the sizes of galaxies.These two are coflfding. Note the absorption of light from one galaxy by dust lanes in the other. Courtesy of NASA and the Hubble Heritage Team (STSd/AURA)

Figure 1.6

The Hubble Deep Field.

Composed from 342 images taken over 10 days at the end of 1995, the Hubble Deep fiedd represents the Bmit of present methods for observing faint, distant, and young objects- Almost every dot and smudge m this picture is a galaxy, with light from the most distant ones traveling 12 billion light years to reach us Courtesy of R WUrtamsfNASA/STScl/ AURA

0 0

Post a comment