George Gamow seemed to have gone down a blind alley with his idea that most of the Universe’s heavy atoms were made in a hot Big Bang. But, in the early 1960s, a physicist at Princeton University also concluded that the early Universe had to be hot. He was unaware of Gamow’s work and had come to this conclusion for an entirely different reason. Instead of trying to build up the elements, he was trying to destroy them.
Bob Dicke was a phenomenally prolific scientist. He had trained as an atomic physicist, but had gone on to develop an alternative to Einstein’s theory of gravity and to carry out experiments to prove that Newton’s law of gravity was right to unprecedented accuracy. During the Second World War, Dicke had been one of the key figures in the development of radar at the Massachusetts Institute of Technology’s Radiation Laboratory.
Dicke was also interested in cosmology. But the Big Bang theory unsettled him, particularly its contention that billions of years ago the Universe simply popped out of nothing and started expanding. He wanted to know what happened before the Big Bang. Most scientists simply shrugged their shoulders when asked this question and said science could never answer it, but Dicke thought this a terrible cop-out. He searched for a more satisfying theory – one with fewer loose ends than the conventional Big Bang. And what he came to embrace was the idea of the oscillating, or ‘bouncing’, universe.
To Dicke, the Universe was like a giant beating heart which had been swelling and contracting throughout eternity. The reason that all the galaxies appeared to be rushing away from us, he said, was simply that the human race had appeared on the cosmic stage just when the Universe was undergoing one of its swelling, or expansion, phases.
But even at this moment, the expansion was being braked by the gravitational pull of every galaxy on every other galaxy. In the future, Dicke predicted, the expansion would be slowed to a standstill, then completely reversed. All of creation would embark on a runaway collapse until matter was crushed to the maximum density possible. It was from just such a compressed state – a big crunch – that the Universe around us was ‘rebounding’ today, claimed Dicke.
The great appeal of an oscillating universe was that it dispensed with the creation event and all its unsettling problems. The Big Bang was not unique. It was simply one explosion in a long line of titanic explosions stretching back through the mists of time.
The oscillating universe, like the steady-state universe, neatly sidestepped the sticky problem of what happened before the Big Bang. There had been another Big Bang. And before that, another. The Universe had no beginning. It had been pulsating throughout eternity.
But there was still one loose end that Dicke needed to tie up. Since Gamow’s failed attempt to make the elements in the Big Bang, Fred Hoyle and his co-workers had shown that the Universe’s heavy elements had been built up from hydrogen in the furnaces at the heart of stars. Their theory was so successful in predicting which elements should be common and which should be scarce that few people doubted that it was largely correct. In fact, in the early 1960s, astronomers had found that old stars did indeed contain fewer heavy elements than young stars, which was just what you would expect if as time went on stars made more and more heavy elements deep in their cores.
But if the Universe began as mostly hydrogen, and stars then cooked some of it into heavy elements, what had happened to the heavy elements that had been made during the Universe’s previous cycle of expansion and collapse? There must be a process that destroyed all the Universe’s heavy elements between the Big Crunch at the end of a phase of contraction and the Big Bang at the start of the next expansion.
Dicke realised that extreme heat would do the job nicely. During its compression, the Universe must have been very hot – at least a billion degrees. At such a temperature, the heavy atoms would have been slammed together so violently that they would have disintegrated into hydrogen. Every last trace of the previous era of cosmic history would be erased. The Universe would start the next cycle without any heavy elements.
An unavoidable consequence of such a hot phase in the early Universe was intense radiation. Dicke, like Gamow before him, concluded that the early Universe must have been a brilliantly bright fireball.
Ironically, Dicke had wanted to break down heavy elements, while Gamow had wanted to build them up. It was doubly ironic because both Gamow and Dicke were right about the existence of the fireball radiation – but for the wrong reason.
Like Gamow, Dicke wondered what would have become of the fireball radiation. He realised the expansion of the Universe would have cooled the radiation, continually stretching out the wavelength of its photons and sapping them of energy. Instead of having a temperature of billions of degrees, the relic of the fireball should by now be only a tepid glimmer barely a few degrees above absolute zero. Instead of appearing as gamma rays, it would appear as short-wavelength radio waves.
But Dicke realised something that Gamow, Alpher and Herman had not: that there was a good chance of detecting such radiation in the Universe today.
Working in Dicke’s ‘gravity group’ at Princeton were two young physicists, David Wilkinson and Peter Roll. ‘One day Dicke burst into our lab,’ recalls Wilkinson. ‘He said, “Gee, you know there might be this relic radiation in the Universe.”’
Wilkinson and Roll were intrigued by the possibility of mounting a search for the radiation from the Big Bang. The relic radiation would have two unique and striking characteristics. First, because it permeated every pore of space, it would appear to be coming from absolutely everywhere in the sky. Secondly, it would have the spectrum of a black body.
By now the radiation would be cold. It would appear brightest at short radio wavelengths between about a centimetre and a metre. These are known as microwaves. You did not need a big telescope to see the radiation since it would appear to be coming from everywhere in the sky. All you needed to take the ‘temperature of the Universe’ was a small purpose-built radio telescope.
The first sensitive radio receivers to operate at around about a centimetre in wavelength had been built for radar during the war. Radar equipment needed to be made small to fit in aircraft, so there had been a major effort to make it operate at short, microwave, wavelengths. Dicke was the one who in 1946 invented the instrument that became the standard for measuring microwaves from the sky.1
In the spring of 1964, Wilkinson and Roll started building such an instrument to look for what they had now called the ‘primeval fireball’.
At the same time as Dicke set Wilkinson and Roll looking for the Big Bang radiation, he set a young Canadian theorist thinking about how it might be possible to estimate the present temperature of the fireball.
Jim Peebles had been working as a graduate student in particle physics ever since arriving at Princeton from the University of Manitoba. Undoubtedly, he would have stayed in that field if it had not been for a chance conversation with a fellow Manitoban, Bob Moore, who was in the year ahead of Peebles. ‘Bob told me that the research seminars of a faculty member he was working with – Bob Dicke – were much more interesting than what I was doing,’ says Peebles. ‘I went along to some of them – and Bob was absolutely right.’
Peebles quickly learnt about Dicke’s idea that the early Universe had been a searing hot fireball and that the observable consequence of this might be the detection of the leftover radiation. ‘It was a good idea,’ says Peebles. ‘And like all good ideas it sparked a whole chain of thoughts.’
Peebles immediately went to work on the implications of a hot Big Bang. The first thing he realised was that helium and a few other elements would be produced in abundance in the Big Bang. Soon he had worked out how much helium you would expect to be made and how this amount was related to the present temperature of the Universe.
What Peebles found was that about 25 per cent of the mass of the Universe should be helium. At the time he was unaware that this was precisely the helium abundance astronomers had found in many stars. ‘My knowledge of astronomy was exceedingly limited,’ he says. But earlier Peebles had written a scientific paper on the structure of Jupiter in which he had concluded that about 25 per cent of its mass had to be helium. He looked up the figure for the Sun and found that it, too, was about the same. ‘It was at least reassuring that I could make the numbers for the Big Bang come out consistent with what we know for the Solar System,’ says Peebles.
In fact, he had solved one of the great unsolved problems of astrophysics: why there was so much helium in the Universe. Although Fred Hoyle and his colleagues had proved beyond a doubt that most elements had been forged in the furnaces of stars, helium remained a big puzzle. There was simply no way that stars could have turned 25 per cent of the matter in the Universe into helium since the Big Bang. Even Hoyle was coming around to the idea that the elements must have been made in two places: the heavy elements in stars, and the light elements like helium somewhere else.
Of course, Gamow had already located that place – the fireball at the start of the Universe – but because his theory was unable to produce the rest of the elements it had been discredited. It turned out nature was not simple. The elements were not built either in stars or in the Big Bang: they were made in both places. When Gamow’s work had been tossed aside, the baby had been thrown out with the bath water.
At his first ever colloquium on the subject of the hot Big Bang, Peebles told his audience that if the whole thing hung together the temperature should be about ten degrees above absolute zero. ‘I didn’t realise that Alpher and Herman had got a similar answer from a similar line of reasoning 16 years earlier,’ says Peebles.
But despite the enthusiasm with which Peebles had explored Dicke’s idea, he did not have high hopes that Wilkinson and Roll would actually find the Big Bang radiation. ‘I’m never optimistic,’ he says. ‘The hot Big Bang was simply an interesting thing to play with. I suppose I was counting on them not finding anything and considering the implications of a negative result.’
While Peebles theorised, Wilkinson and Roll got on with the job of building a telescope to look for the cooled remnant of the Big Bang fireball. They had decided to look for the radiation at a wavelength of three centimetres. Equipment was readily available because this was a common radar wavelength, known as X-band. The wavelength had the added advantage that it was one at which water vapour in the atmosphere would not be glowing too brightly. Also, the tenuous halo of gas which was known to surround the Galaxy and fill most of our sky with a background glow would not be too much of a problem either.
Wilkinson and Roll built their apparatus on the cheap, buying most of the parts they needed from army surplus stores in Philadelphia, a short 45-minute drive from Princeton. They even made use of vacuum tubes, which glowed when electricity throbbed through them. ‘It was just at the end of an era,’ says Wilkinson. ‘Transistors hadn’t quite come in. Neither Peter nor I knew anything about microwaves. But Dicke knew a lot, of course. We would chat with him, go off and build something in the lab, then show him what we’d done.’
‘Essentially, they were building the same kind of instrument I had built at the Massachusetts Institute of Technology [MIT] during the war,’ says Dicke. ‘I gave them advice, and they went and did all the work of soldering.’
For the site of their experiment, the two astronomers selected the roof of Guot Hall, Princeton’s geology building. ‘It was fine for our purpose because, apart from a few towers, its roof was flat,’ says Wilkinson. They began assembling the antenna on a piece of plywood in a disused pigeon coop.
The heart of the apparatus was the ‘antenna’. An antenna is simply the name given to any device that collects radio waves from the sky. For instance, a television aerial is an antenna: it collects radio waves from a TV transmitter. Other examples of antennas are the giant bowl-shaped dishes used by astronomers to pick up faint radio signals from distant galaxies.
When radio waves impinge on an antenna, they drive tiny electrical currents in its metal structure. It is by recording these currents that a radio telescope measures the strength of the radio waves.
The best type of antenna for collecting microwaves is simply a metal funnel, commonly known as a ‘horn’. Wilkinson built his from four sheets of copper, which he soldered together. It looked rather like a square trumpet, six feet long. The microwaves from the sky were collected by the flared opening, which was about a foot square. The horn then funnelled them down to a ‘receiver’, the complicated electronic bit which actually detects radio waves. All TVs have receivers built into them. Wilkinson and Roll’s receiver was where all their glowing vacuum tubes went.
The design of the antenna was crucial to Wilkinson and Roll’s experiment. All antennas are designed so that they pick up radio waves coming from only a small area of the sky while ignoring everything else. For instance, a TV antenna must pick up radio waves from the TV transmitter it is pointed at and not radio waves from other places – for instance, other TV transmitters.
But though most of the radio waves an antenna picks up come from where it is pointing, some radio waves from other directions always manage to leak in. These get into an antenna because they are able to bend round corners, just like sound waves.2 The corner in this case is the sharp metal edge at the horn’s flared opening. Unwanted radio waves come from sources such as the ground, the Earth’s atmosphere and the components of the radio telescope itself. Any material that is above absolute zero naturally produces radio waves. The common denominator is electrons. All materials – even blocks of ice – contain electrons jiggling about inside, and jiggling electrons give out radio waves. In fact, the hotter a material is, the faster its electrons are jiggling and the stronger the radio waves it broadcasts are.
Being able to distinguish between the signal from space and other unwanted signals is the major problem which radio astronomers face.
It is not a serious problem with a TV aerial, because the unwanted radio waves are so much weaker than those from the transmitter. But Wilkinson and Roll were wanting to measure the coldest thing in the Universe, so picking up unwanted radio waves from hotter bodies near by was an enormous source of worry.
The Big Bang radiation would be only a few degrees above absolute zero, whereas everything else in the vicinity of the experiment would be very much hotter – at least several hundred degrees above absolute zero.3 If a substantial amount of radio waves from any of these objects got into the antenna, they would utterly swamp the tiny signal from the background.
The Big Bang radiation might make up 99.9 per cent of the radiation flowing through the Universe, but at microwave wavelengths it was 100 million times fainter than the heat emitted by the Earth. If you had microwave glasses, you would be able to see it – well, as long as they were sensitive and could exclude the light from the ground. But it would be like trying to make out the faint uniform glow of the sky while the ground beneath you and every object around you was shining with white heat.
So Wilkinson and Roll had a formidable task ahead of them. They had to design their antenna so that when it was pointing at the sky, as little radiation as possible found its way in from the ground and other hot objects near by. The trumpet-shaped microwave horn was good but not good enough. Wilkinson and Roll supplemented it with a sort of upside-down metal skirt that surrounded the antenna. This ‘ground shield’ made it very difficult for radiation from hot objects near by, particularly the ground, to get into the antenna.
However, in addition to having a well-designed antenna, there was something else that was absolutely crucial to the experiment: a special device known as a ‘cold load’. This was needed because the antenna was trying to see the coldest thing in the Universe, and no antenna could do that if it operated like a conventional radio telescope.
So how does a conventional radio telescope operate? Essentially, the radio waves picked up from a star or a galaxy generate ‘static’ in its receiver, rather like the background hiss from a radio tuned between stations. Unfortunately, lots of other things produce a similar static in the receiver. For instance, static is produced by radio waves coming from the Earth’s atmosphere, and even by electrons jostling about inside both the metal of the antenna and inside the electronics of the receiver.
So how do astronomers tell the astronomical static apart from the spurious static? They use a simple trick. First they point their antenna at the star or distant galaxy they are interested in and note down the strength of the radio waves. Then they point the antenna at a piece of background sky near by and take another reading. In both cases, the unwanted static created by the antenna, the receiver and the atmosphere will be the same. So, if they subtract one reading from the other, they will be left with the strength of the radio waves coming from the star or galaxy. The unwanted static will have cancelled out neatly.
Of course, all the radio astronomers will have measured is how much brighter their star or galaxy is than the background sky. But in practice the background sky will be giving out almost no radio waves, so it won’t matter very much.
This ‘on source/off source’ trick works perfectly when astronomers want to look at a source of radio waves which covers only a small area of the sky – a star or distant galaxy, for instance. Then it is easy to point an antenna at a piece of background sky away from the source. But Wilkinson and Roll were planning to observe a source of radio waves which covered the entire sky. The Big Bang radiation was the background sky, so it would be impossible to look away from it.
But if it was impossible to compare the Big Bang radiation with the background sky, then it would have to be compared with something else. Wilkinson and Roll realised they would have to make an artificial source of radio waves called a ‘cold load’. They could then point their antenna at the sky, note the strength of the radio waves, and then point it at their artificial source and take another reading. By subtracting one reading from the other they would discover how much hotter the sky was than their artificial source.
If they knew the temperature of their artificial source well, then they would know the precise temperature of the Big Bang radiation. In the jargon, their artificial source of radio waves would enable them to make an ‘absolute’ measurement: rather than simply comparing a radio source with the sky, as most radio astronomers did, they would be able to measure the true temperature of what they were looking at.
Ideally, the artificial source of radio waves should be close to the expected temperature of the Big Bang radiation – between three and ten degrees above absolute zero. Wilkinson and Roll therefore decided to cool their artificial source with liquid helium, which boils at about 4.2 degrees above absolute zero (–269°C). This is why the artificial source of radio waves was called a cold load.
Nowadays liquid helium is readily available and there is a lot of experience in handling it, but back in 1964 it was a pretty novel substance to be playing around with.4 It was Peter Roll who took on the task of designing and building the cold load. The important thing was to make sure that it absorbed all the radio waves that fell on it and did not reflect any back. This was because, when the antenna was pointed at it, the cold load had to appear to be precisely 4.2 degrees above absolute zero. But if it reflected any radio waves at all, radio waves emitted by the metal of the antenna would bounce off the cold load straight back into the antenna. It would see the cold load plus its own reflection, causing Roll and Wilkinson to overestimate its temperature. They would assume that the cold load was at 4.2 degrees – it was, after all, their temperature reference – but its temperature might in fact be higher, say six degrees. Since they would be comparing this with the Big Bang radiation, they would underestimate its temperature, and the whole experiment could be screwed up.
It might seem a silly thing to worry about, but every possible source of unwanted radio waves has to be thought about when you are attempting to measure the coldest thing in the Universe, and by definition everything in existence is hotter. ‘You really have to understand every detail of your instrument,’ says Dicke.
Roll made the cold load non-reflecting by using a length of silver-plated X-band ‘wave guide’ – basically, just a hollow metal tube with a rectangular cross-section. This dipped down into a vacuum flask of liquid helium. So when the antenna looked at the cold load, it saw a source of radio waves at precisely 4.2 degrees.
Wilkinson and Roll arranged their instrument so that it switched from looking at the sky, then at the cold load and then back at the sky again, and this was repeated very rapidly. The electrical device that made this possible was called a ‘Dicke switch’. Dicke, it seemed, had invented virtually everything in the field of microwave astronomy.
In fact, it was Dicke who, in 1946, introduced the standard convention of measuring the brightness of a radio source in terms of an equivalent temperature. So when radio astronomers turn their telescopes on an object in the sky and say they measure a temperature of, for instance, 100 degrees above absolute zero, what they mean is that their instrument registers the same signal as it would if a body at a temperature of 100 degrees was stuck right in front of the antenna. It is just a convenience. Wilkinson and Roll expected the cosmic background radiation to be between five and ten degrees above absolute zero.
As Wilkinson and Roll worked on the roof of Princeton’s geology building, few people walking around the campus realised that the six-foot trumpet sticking out of a pigeon coop above their heads was designed to see into the fireball at the beginning of the Universe. ‘Our experiment didn’t attract a lot of attention on campus,’ admits Wilkinson. ‘But then we didn’t go out of our way to let people know what we were doing.’
From time to time, even Wilkinson and Roll thought that maybe they were just a little mad. ‘It wasn’t obvious from the beginning that this was a good way to spend a few years,’ says Wilkinson. ‘Most people at the time believed in the steady-state theory, not the Big Bang.’ But at other times Wilkinson was quite optimistic about their search. ‘I thought we had a fifty–fifty chance of finding it,’ he says.
The telescope Wilkinson and Roll were assembling had two unique features: a cold load and an antenna carefully designed to reject radio waves from the ground. No other instrument in the world was capable of detecting the microwave background radiation from the Big Bang. Or so the two astronomers thought.
1. To this day, astronomers refer to it as a ‘Dicke radiometer’.
2. It is only because sound waves do bend around corners – for instance, buildings – that we can hear people shouting even when they are out of sight.
3. Room temperature is about 300 degrees above absolute zero.
4. Liquid helium is probably the most bizarre liquid in nature. It can behave as a so-called superfluid, defying gravity by running uphill and squeezing through tiny holes that no other liquid can squeeze through.