Back | Next
Contents

After the Bomb: The Birth of the Bang

Gamow's Nuclear Pressure-Cooker

In 1946, Russian-born George Gamow, who had worked on the theory of nuclear synthesis in the 1930s and been involved in the Manhattan Project, conjectured that if an atomic bomb could, in a fraction of a millionth of a second, create elements detectable at the test site in the desert years later, then perhaps an explosion on a colossal scale could have produced the elements making up the universe as we know it. Given high enough temperatures, the range of atomic nuclei found in nature could be built up through a succession starting with hydrogen, the lightest, which consists of one proton. Analysis of astronomical spectra showed the universe to consist of around 75 percent hydrogen, 24 percent helium, and the rest a mix continuing on through lithium, beryllium, boron and so on of the various heavier elements. Although all of the latter put together formed just a trace in comparison to the amount of hydrogen and helium, earlier attempts at constructing a theoretical model had predicted far less than was observed—the discrepancy being in the order of ten orders of magnitude in the case of intermediate mass elements such as carbon, nitrogen, and oxygen, and getting rapidly worse (in fact, exponentially) beyond those.

Using pointlike initial conditions of the GRT equations, Gamow, working with Ralph Alpher and Robert Herman, modeled the explosion of a titanic superbomb in which, as the fireball expanded, the rapidly falling temperature would pass a point where the heavier nuclei formed from nuclear fusions in the first few minutes would cease being broken down again. The mix of elements that existed at that moment would thus be "locked in," providing the raw material for the subsequently evolving universe. By adjusting the parameters that determined density, Gamow and his colleagues developed a model that within the first thirty minutes of the Bang yielded a composition close to that which was observed.

Unlike Lemaître's earlier proposal, the Gamow theory was well received by the scientific community, particularly the new generation of physicists versed in nuclear technicalities, and became widely popularized. Einstein had envisaged a universe that was finite in space but curved and hence unbounded, as the surface of a sphere is in three dimensions. The prevailing model now became one that was also finite in time. Although cloaked in the language of particle physics and quantum mechanics, the return to what was essentially a medieval worldview was complete, raising again all the metaphysical questions about what had come before the Bang. If space and time themselves had come into existence along with all the matter and energy of the universe as some theorists maintained, where had it all come from? If the explosion had suddenly come about from a state that had endured for some indefinite period previously, what had triggered it? It seemed to be a one-time event. By the early 1950s, estimates of the total amount of mass in the universe appeared to rule out the solutions in which it oscillated between expansion and contraction. There wasn't enough to provide sufficient gravity to halt the expansion, which therefore seemed destined to continue forever. What the source of the energy might have been to drive such an expansion—exceeding all the gravitational energy contained in the universe—was also an unsolved problem.

Hoyle and Supernovas as "Little Bang" Element Factories

Difficulties for the theory mounted when the British astronomer Fred Hoyle showed that the unique conditions of a Big Bang were not necessary to account for the abundance of heavy elements; processes that are observable today could do the job. It was accepted by then that stars burned by converting hydrogen to helium, which can take place at temperatures as low as 10 million degrees—attainable in a star's core. Reactions beyond helium require higher temperatures, which Gamow had believed stars couldn't achieve. However, the immense outward pressure of fusion radiation balanced the star's tendency to fall inward under its own gravity. When the hydrogen fuel was used up, its conversion to helium would cease, upsetting the balance and allowing the star to collapse. The gravitational energy released in the collapse would heat the core further, eventually reaching the billion degrees necessary to initiate the fusion of helium nuclei into carbon, with other elements appearing through neutron capture along the lines Gamow had proposed. A new phase of radiation production would ensue, arresting the collapse and bringing the star into a new equilibrium until the helium was exhausted. At that point another cycle would repeat in which oxygen could be manufactured, and so on through to iron, in the middle of the range of elements, which is as far as the fusion process can go. Elements heavier than iron would come about in the huge supernova explosions that would occur following the further collapse of highly massive stars at the end of their nuclear burning phase—"little bangs" capable of supplying all the material required for the universe without need of any primordial event to stock it up from the beginning.

This model also accounted for the observational evidence that stars varied in their makeup of elements, which was difficult to explain if they all came from the same Big Bang plasma. (It also followed that any star or planet containing elements heavier than iron—our Sun, the Earth, indeed the whole Solar System, for example—must have formed from the debris of an exploded star from an earlier generation of stars.) Well, the images of starving postwar Europe, shattered German cities, Stalingrad, and Hiroshima were fading. The fifties were staid and prosperous, and confidence in the future was returning. Maybe it was time to rethink cosmology again.

The Steady-State Theory

Sure enough, Fred Hoyle, having dethroned the Big Bang as the only mechanism capable of producing heavy elements, went on, with Thomas Gold and Herman Bondi, to propose an alternative that would replace it completely. The Hubble redshift was still accepted by most as showing that the universe we see is expanding away in all directions to the limits of observation. But suppose, Hoyle and his colleagues argued, that instead of this being the result of a one-time event, destined to die away into darkness and emptiness as the galaxies recede away from each other, new matter is all the time coming into existence at a sufficient rate to keep the overall density of the universe the same. Thus, as old galaxies disappear beyond the remote visibility "horizon" and are lost, new matter being created diffusely through all of space would be coming together to form new galaxies, resulting in a universe populated by a whole range of ages—analogous to a forest consisting of all forms of trees, from young saplings to aging giants.

The rate of creation of new matter necessary to sustain this situation worked out at one hydrogen atom per year in a cube of volume measuring a hundred meters along a side, which would be utterly undetectable. Hence, the theory was not based on any hard observational data. Its sole justification was philosophical. The long-accepted "cosmological principle" asserted that, taken at a large-enough scale, the universe looked the same anywhere and in any direction. The Hoyle-Bondi-Gold approach introduced a "perfect cosmological principle" extending to time also, making the universe unchanging. It became known, therefore, as the steady-state theory.

The steady-state model had its problems too. One in particular was that surveys of the more distant galaxies, and hence ones seen from an earlier epoch because of the delay in their light reaching Earth, showed progressively more radio sources; hence the universe hadn't looked the same at all times, and so the principle of its maintaining a steady, unvarying state was violated. But it attracted a lot of scientists away from the Big Bang fold. The two major theories continued to rival each other, each with its adherents and opponents. And so things remained through into the sixties.

Then, in 1965, two scientists at Bell Telephone Laboratories, Arno Penzias and Robert Wilson, after several months of measurement and double-checking, confirmed a faint glow of radiation emanating evenly from every direction in the heavens with a frequency spectrum corresponding to a temperature of 2.7ºK. 44 This was widely acclaimed and publicized as settling the issue in favor of the Big Bang theory.

The Cosmic Background Radiation: News but Nothing New

Big Bang had been wrestling with the problem of where the energy came from to drive the expansion of the "open" universe that earlier observations had seemed to indicate—a universe that would continue expanding indefinitely due to there being too little gravitating mass to check it. Well, suppose the estimates were light, and the universe was in fact just "closed"—meaning that the amount of mass was just enough to eventually halt the expansion, at which point everything would all start falling in on itself again, recovering the energy that had been expended in driving the expansion. This would simplify things considerably, making it possible to consider an oscillating model again, in which the current Bang figures as simply the latest of an indeterminate number of cycles. Also, it did away with all the metaphysics of asking who put the match to whatever blew up, and what had been going on before.

A group at Princeton looked into the question of whether such a universe could produce the observed amount of helium, which was still one of Big Bang's strong points. (Steady state had gotten the abundance of heavier elements about right but was still having trouble accounting for all the helium.) They found that it could. With the conditions adjusted to match the observed figure for helium, expansion would have cooled the radiation of the original fireball to a diffuse background pervading all of space that should still be detectable—at a temperature of 30ºK. 45 Gamow's collaborators, Ralph Alpher and Robert Herman, in their original version had calculated 5ºK for the temperature resulting from the expansion alone, which they stated would be increased by the energy production of stars, and a later publication of Gamow's put the figure at 50ºK. 46

The story is generally repeated that the discovery of the 2.7ºK microwave background radiation confirmed precisely a prediction of the Big Bang theory. In fact, the figures predicted were an order of magnitude higher. We're told that those models were based on an idealized density somewhat higher than that actually reported by observation, and (mumble-mumble, shuffle-shuffle) it's not really too far off when you allow for the uncertainties. In any case, the Big Bang proponents maintained, the diffuseness of this radiation across space, emanating from no discernible source, meant that it could only be a relic of the original explosion.

It's difficult to follow the insistence on why this had to be so. A basic principle of physics is that a structure that emits wave energy at a given frequency (or wavelength) will also absorb energy at the same frequency—a tuning fork, for example, is set ringing by the same tone that it sounds when struck. An object in thermal equilibrium with—i.e., that has reached the same temperature as—its surroundings will emit the same spectrum of radiation that it absorbs. Every temperature has a characteristic spectrum, and an ideal, perfectly black body absorbing and reradiating totally is said to be a "blackbody" radiator at that temperature. The formula relating the total radiant energy emitted by a blackbody to its temperature was found experimentally by Joseph Stefan in 1879 and derived theoretically by Ludwig Boltzmann in 1889. Thus, given the energy density of a volume, it was possible to calculate its temperature.

Many studies had applied these principles to estimating the temperature of "space." These included Guillaume (1896), who obtained a figure of 5º–6º K, based on the radiative output of stars; Eddington (1926), 3.18º K; Regener (1933), 2.8º K, allowing also for the cosmic ray flux; Nernst (1938), 0.75º K; Herzberg (1941), 2.3º K; Finlay-Freundlich (1953 and 1954), using a "tired light" model for the redshift (light losing energy due to some static process not involving expansion), 1.9º K to 6º K. 47 Max Born, discussing this last result in 1954, and the proposal that the mechanism responsible for "tiring" the light en route might be photon-photon interactions, concluded that the "secondary photons" generated to carry away the small energy loss suffered at each interaction would be in the radar range. The significant thing about all these results is that they were based on a static, nonexpanding universe, yet consistently give figures closer to the one that Arno Penzias and Robert Wilson eventually measured than any of the much-lauded predictions derived from Big Bang models.

Furthermore, the discrepancy was worse than it appeared. The amount of energy in a radiation field is proportional to the fourth power of the temperature, which means that the measured background field was thousands of times less than was required by the theory. Translated into the amount of mass implied, this measurement made the universe even more diffuse than Gamow's original, nonoscillating model, not denser, and so the problem that oscillation had been intended to solve—where the energy driving the expansion had come from—became worse instead of better.

An oscillating model was clearly ruled out. But with some modifications to the gravity equations—justified by no other reason than that they forced an agreement with the measured radiation temperature—the open-universe version could be preserved, and at the same time made to yield abundances for helium, deuterium, and lithium which again were close to those observed. The problem of what energy source propelled this endless expansion was still present—in fact exacerbated—but quietly forgotten. Excited science reporters had a story, and the New York Times carried the front-page headline signals imply a big bang universe.

Resting upon three pillars of evidence—the Hubble redshifts, light-element abundance, and the existence of the cosmic background radiation—Big Bang triumphed and became what is today the accepted standard cosmological model.

Quasar and Smoothness Enigmas:
Enter, the Mathematicians.

At about this time, a new class of astronomical objects was discovered that came to be known as quasars, with redshifts higher than anything previously measured, which by the conventional interpretation of redshift made them the most distant objects known. To be as bright as they appeared at those distances they would also have to be astoundingly energetic, emitting up to a hundred thousand times the energy radiated by an entire galaxy. The only processes that could be envisaged as capable of pouring put such amounts of energy were ones resulting from intense gravity fields produced by the collapse of enormous amounts of mass. This was the stuff of general relativity, and with Big Bang now the reigning cosmology, the field became dominated by mathematical theoreticians. By 1980, around ninety-five percent of papers published on the subject were devoted to mathematical models essentially sharing the same fundamental assumptions. Elegance, internal consistency, and preoccupation with technique replaced grounding in observation as modelers produced equations from which they described in detail and with confidence what had happened in the first few fractions of a millionth of a second of time, fifteen billion years ago. From an initial state of mathematical perfection and symmetry, a new version of Genesis was written, rigorously deducing the events that must have followed. That the faith might be . . . well, wrong, became simply inconceivable.

But in fact, serious disagreements were developing between these idealized realms of thought and what astronomers surveying reality were actually finding. For one thing, despite all the publicity it had been accorded as providing the "clincher," there was still a problem with the background radiation. Although the equations could be made to agree with the observed temperature, the observed value itself was just too uniform—everywhere. An exploding ball of symmetrically distributed energy and particles doesn't form itself into the grossly uneven distribution of clustered matter and empty voids that we see. It simply expands as a "gas" of separating particles becoming progressively more rarified and less likely to interact with each other to form into anything. To produce the galaxies and clusters of galaxies that are observed, some initial unevenness would have to be present in the initial fireball to provide the focal points where condensing matter clouds would gravitate together and grow. Such irregularities should have left their imprint as hot spots on the background radiation field, but it wasn't there. Observation showed the field to be smooth in every direction to less than a part in ten thousand, and every version of the theory required several times that amount. (And even then, how do galaxies manage to collide in a universe where they're supposed to be rushing apart?)

Another way of stating this was that the universe didn't contain enough matter to have provided the gravitation for galaxies to form in the time available. There needed to be a hundred times more of it than observation could account for. But it couldn't simply be ordinary matter lurking among or between the galaxies in some invisible form, because the abundance of elements also depended critically on density, and increasing density a hundredfold would upset one of the other predictions that the Big Bang rested on, producing far too much helium and not enough deuterium and lithium. So another form of matter—"dark matter"—was assumed to be there with the required peculiar properties, and the cosmologists turned to the particle physicists, who had been rearing their own zoo of exotic mathematical creations, for entities that might fill the role. Candidates included heavy neutrinos, axions, a catch-all termed "weakly interacting massive particles," or "WIMPS," photinos, strings, superstrings, quark nuggets, none of which had been observed, but had emerged from attempts at formulating unified field theories. The one possibility that was seemingly impermissible to consider was that the reason why the "missing mass" was missing might be that it wasn't there.

Finally, to deal with the smoothness problem and the related "flatness" problem, the notion of "inflation" was introduced, whereby the universe began in a superfast expansion phase of doubling in size every 10-35 seconds until 10-33 seconds after the beginning, at which point it consisted of regions flung far apart but identical in properties as a result of having been all born together, whereupon the inflation suddenly ceased and the relatively sluggish Big Bang rate of expansion took over and has been proceeding ever since.

Let's pause for a moment to reflect on what we're talking about here. We noted in the section on evolution that a picosecond, 10-12 seconds, is about the time light would take to cross the width of a human hair. If we represent a picosecond by the distance to the nearest star, Alpha Centauri (4.3 light-years), then, on the same scale, 10-35 seconds would measure around half a micron, or a quarter the width of a typical bacterium—far below the resolving power of the human eye. Fine-tuning of these mathematical models reached such extremes that the value of a crucial number expressed as a part in fifty-eight decimal places at an instant some 10-43 seconds into the age of the universe made the difference between its collapsing or dispersing in less than a second.

But theory had already dispersed out of sight from reality anyway. By the second half of the 1980s, cosmic structures were being discovered and mapped that could never have come into being since the time of the Big Bang, whatever the inhomogeneities at the beginning or fast footwork in the first few moments to smooth out the background picture. The roughly spherical, ten-million-or-so-light-year-diameter clusters of galaxies themselves turned out to be concentrated in ribbonlike agglomerations termed superclusters, snaking through space for perhaps several hundred million light-years, separated by comparatively empty voids. And then the superclusters were found to be aligned to form planes, stacked in turn as if forming parts of still larger structures—vast sheets and "walls" extending for billions of light-years, in places across a quarter of the observable universe. The problem for Big Bang is that relative to the sizes of these immense structures, the component units that form them are moving too slowly for these regularities to have formed in the time available. In the case of the largest void and shell pattern identified, 150 billion light-years would have been needed at least—eight times the longest that Big Bang allows. New ad-hoc patches made their appearance: light had slowed down, so things had progressed further than we were aware; another form of inflation had accelerated the formation of the larger, early structures, which had then been slowed down by hypothetical forces invented for the purpose. But tenacious resistance persisted to any suggestion that the theory could be in trouble.

Yet the groundwork for an alternative picture that perhaps explains all the anomalies in terms of familiar, observable processes had been laid in the 1930s.

Back | Next
Contents
Framed