
DURING the first split second of existence, an extraordinary force stretched the universe from a cramped sub-microscopic speck into the forerunner of the spacious cosmos we now inhabit. Or at least that’s what cosmologists would have us believe. This theory, known as inflation, can explain a number of puzzling features about the cosmos, such as the fact that one side of the universe looks much the same as the other. Yet despite the enthusiasm, experimental backing for the idea has been thin on the ground.
Last week, however, cosmologists unveiled powerful new support for the theory from observations made by NASA’s Wilkinson Microwave Anisotropy Probe. “We have evidence that the universe suddenly grew from sub-microscopic to astronomical size in the blink of an eye,” says WMAP team leader Charles Bennett of Johns Hopkins University in Baltimore, Maryland. If Bennett and his team are right, they have glimpsed what the universe was like less than a billionth of a billionth of a second after the beginning of time.
“We have evidence that the universe grew from sub-microscopic to astronomical size in the blink of an eye”
Advertisement
Even so, the findings by no means prove that inflation took place – one rival theory could also fit the data. Other critics point to anomalies in the results that inflation theory struggles to explain. The triumphant tone of the WMAP team masks a debate that is by no means over.
Inflation was proposed 25 years ago to explain a number of cosmological puzzles. For example, on the largest scales the universe is remarkably uniform, and space seems to be almost perfectly flat. Inflation theory proposes that the universe we now observe grew from a patch of the initial fireball that was so small it was effectively uniform. The expansion also stretched space so that it is now flat, and distributed the seeds of future galaxies.
Those seeds were planted by quantum fluctuations in the energy field that powered inflation. These variations were blown up to structures of astronomical size and then frozen in place when inflation ended. Sure enough, these structures appear as subtle variations in the microwave background radiation – a faint microwave glow across the sky, which dates from 380,000 years after the big bang. Although at that time the universe was an almost uniform soup of hot atoms, the slight variations in density left over from the inflationary period show up in the microwave map as brighter and dimmer spots. They were first measured in 1990 by the NASA probe COBE and WMAP was launched in 2001 to vastly improve on COBE’s data.
The pattern of spots allows cosmologists to deduce many properties of the universe. To start with, they used the WMAP data to pin down the age of the universe and the proportions of its main constituents: ordinary matter, dark matter and dark energy (èƵ, 15 February 2003, p 12). Now, with three extra years of measurements to play with, the team believe they can see more deeply into the past.
The key is that WMAP has revealed a slight bias towards large-scale fluctuations. In other words, bigger spots are brighter than smaller ones, which is just what the theory of inflation predicts.
Discovering that bias has been tremendously difficult. The microwave radiation has been scattered by ionised gas in the later universe, roughening the texture of the microwave map and masking the imprint of inflation. The WMAP team has managed to remove this distortion by looking at another property of the microwaves, their polarisation. The microwaves become polarised when they scatter off ionised gas, in the same way sunlight becomes polarised when it is reflected off the sea. But the effect on the microwave background is small, which is why the WMAP team had to accumulate three years of data to see the effect clearly.
The resulting polarisation chart gives the team a separate measure of the ionised gas distortion. Subtracting its effect on the main microwave map produces a sharper image of the early universe, which reveals the tilt towards larger clumps. “It is a real landmark,” says cosmologist Max Tegmark of the Massachusetts Institute of Technology. “Is it ironclad proof of inflation? Absolutely not. But inflation has just passed another test.”
One difficulty with understanding inflation is that it occurred so early in the universe – some time between 10-35 and 10-20 seconds after the big bang – that we don’t yet know what physical processes were at work. This gives theorists license to hypothesise, which they have clearly relished. The result is hundreds of different versions of inflation theory, each positing an inflationary force that behaves in a slightly different way.
The latest map means cosmologists can prune this thicket of models for the first time. For example, some complicated versions of inflation called hybrid models predict an especially coarse texture for the microwave sky, which WMAP now rules out.
Even so, there is still much pruning to do. What’s more, WMAP doesn’t yet tell us much about the fundamental nature of inflation because the class of models WMAP favours can be produced by several different kinds of underlying physics. “It’s a little bit disappointing,” says cosmologist Peter Coles of the University of Nottingham in the UK.
There are also some niggling anomalies, first detected in WMAP’s earlier data set. Two large-scale components of the pattern called the quadrupole and octopole ought to be randomly oriented, but instead are suspiciously aligned with each other and perpendicular to the plane of the solar system. “It could just be a fluke,” says Tegmark, “but it is important to be modest as cosmologists, and keep in mind that we might be wrong.” Glenn Starkman of Case Western Reserve University in Cleveland, Ohio, takes a stronger line. “The alignments remain bad news for generic models of inflation,” he says.
Meanwhile WMAP’s results are also consistent with an entirely different cosmological model called cyclic universe theory, developed by Paul Steinhardt of Princeton University and Neil Turok at the University of Cambridge. The idea is that our universe repeatedly collapses in a big crunch and bounces back again in a new big bang.
Steinhardt and Turok have worked out that the process can generate fluctuations in the microwave background that look just like those in the new map, with large spots being more prominent. But David Spergel, an astrophysicist at Princeton University and one of the WMAP team, thinks the theory is not well-enough developed. “We don’t really understand what happens when the universe goes through the transition from big crunch to big bang,” he says.
The argument might soon be settled though, as astronomers have discovered a way to distinguish between the two. The violence of inflation should generate gravitational waves – travelling distortions in space-time – which would leave their own faint mark on the microwave map in the form of swirling patterns of polarisation. The cyclic universe model predicts no such waves.
“The simple inflation favoured by the WMAP data should produce a gravitational-wave signal strong enough to be detected soon”
Encouragingly, the simple inflation favoured by the latest WMAP data should produce a gravitational-wave signal strong enough to be detected in the near future, possibly by WMAP but much more likely by balloon-borne telescopes or the European Space Agency’s upcoming Planck satellite due for launch in 2007. If a gravitational-wave signal emerges, it would all but prove inflation and really begin to tell us what the universe was like when it was only 10-32 seconds old.
The mystery of the megasuns
When did the first stars light up the universe? The latest results from the Wilkinson Microwave Anisotropy Probe put the date at 400 million years after the big bang. That’s a relief for many astronomers who had been puzzled by earlier WMAP data indicating that the first suns formed after only 200 million years.
During the long dark age before the first suns, the gas that filled space was cool and electrically neutral. Eventually the first generation of stars began to form – monsters more than a hundred times the mass of the sun and a million times as bright. When enough of them switched on, their light tore electrons from the atoms of gas left behind.
The primordial light that we now see as the cosmic microwave background scattered off those loose electrons, becoming slightly polarised in the process. It is by measuring that polarisation that the WMAP team has managed to work out when the first generation of stars began ionising the universe.
When WMAP reported its first results in 2003, the team could only use a cruder measure of stellar genesis, which suggested a date just 200 million years after the big bang. Yet if early star formation had been so rapid and easy, there should have been a huge population of these ancient megasuns, which would have given the sky a diffuse infrared glow that is not observed.
“The WMAP year-one results caused quite a bit of consternation,” says Brian O’Shea of the Los Alamos National Laboratory in California, who models these early stars. The new date for the end of the dark age, 400 million years, solves the problem. “It really is a relief,” O’Shea says.