MONDAY, 23 MARCH 2020Our night sky is speckled by star light. A rich pattern is formed in our field of view by light that has travelled from distant stars in our galaxy and beyond. Yet, sometimes it is the surrounding darkness of space that draws our curiosity. An exoplanet, a planet orbiting a star outside of the solar system, does not emit its own light. In fact, even starlight reflected from the exoplanet does not provide a clear signal: the exoplanet is completely overwhelmed by the star and appears billions of times fainter, effectively hiding from us in the dark night sky. There is more to this apparent darkness than meets the eye.
We know our place in the universe is not a lonely one: from the gas giant of Jupiter to the small rocky Mercury, we share our sun with a diverse range of planets. Who is to say that such planets cannot exist elsewhere? Until the 1990’s, the detection of exoplanets was hindered by our inability to observe them directly. This changed with the development of indirect detection methods. These asked: if we cannot see exoplanets, can we infer their existence from their effects on neighbouring stars? By answering this question astrophysicists have begun to unveil, one by one, the numerous planetary inhabitants of our rich and busy galaxy.
How can a small exoplanet have a significant enough effect on their neighbouring star that it can be detected here on earth? Although we may think of the force of gravity as pulling the exoplanet into orbit around the star, the star feels an equal pull from the exoplanet. The result of this force is to induce a small variation in the motion of the star, producing a periodic ‘wobble’. Each stellar wobble coincides with an orbit of the exoplanet. If we can measure this motion and how often it occurs, we can learn something about the existence of an exoplanet as well as its mass and orbit.
In 1995, Michel Mayor and Didier Queloz used this to successfully identify the existence of an exoplanet orbiting the star 51 Pegasi at a distance of 50 light years from Earth. Although a few exoplanets had already been detected, this was the first observation of an exoplanet orbiting a sun-like star. This discovery provided tantalising evidence of the possibility that other planetary systems like our own may exist, and earned them half of the 2019 Nobel Prize.
Whether this planetary system was really anything like our solar system is questionable, however. The exoplanet that they found was a hot Jupiter - a large planet of Jupiter-scale mass, orbiting closer to its star than Mercury’s distance from the sun. The formation of such a heavy planet so close to the star was not predicted by existing models of planetary formation, which were based on the only example we had: our solar system. Mayor and Queloz’s discovery led us to the theory that such Jupiter-sized exoplanets can form far from their star before migrating inwards to a short-distance orbit.
The news set fire to the field of exoplanet detection, bringing an influx of brand new observations. Yet, by studying only the stellar wobble, detections were limited to Jupiter-sized exoplanets heavy enough to create a noticeable variation in the star’s motion. It wasn’t until the launch of the Kepler space telescope in 2009 that we began to find exoplanets of our own size. Instead of studying the motion of a star, Kepler watched the brightness of the star’s light. As a planet passed in front of the star, Kepler detected its transit as a slight dimming in this brightness. This method had the potential to detect the presence of planets of much lower masses, and it did: even within the 10-day trial run of Kepler, a candidate Earth-sized exoplanet was found and later confirmed. Since then, Kepler has detected over 2000 exoplanets. Among these is Kepler-22b, a milestone for astronomy: at a distance of over 500 light-years from Earth, this is the first Earth-sized exoplanet found orbiting within the habitable zone of its star.
With over 4000 exoplanets confirmed less than 30 years after the first exoplanet detection, we can confidently say that exoplanets are not rare in our galaxy. The PLANET collaboration in 2012 concluded that in the Milky Way we are more likely to find a star with an orbiting exoplanet than one without. These predictions suggest that our neighbourhood is far busier than we could have imagined.
What if we wanted to look past the Milky Way? What value is there in observing other galaxies when so much of the Milky Way is yet to be studied? This was answered as early as the 1930’s, when astronomers began to study the rotation curves of distant galaxies. These curves tell us how quickly the constituents of galaxies rotate, starting from the matter at the centre of a galaxy and moving out to the galactic halo. From observations it was known that most of the visible matter is clustered at the galaxy’s centre. The gravitational interactions of such a large quantity of mass, however, should lead to high velocities near this centre, while the stars positioned further away were expected to move much more slowly.
This is not what astronomers observed. The galaxy rotation curves flattened unexpectedly at distances far from the centre, suggesting that something was causing these galaxies’ outer parts to move quicker than anticipated. By the 1970’s enough data had been gathered to postulate that this phenomenon may be caused by the presence of additional undetected matter distributed throughout the galaxy. Among the many contributors to this conclusion was James Peebles, recipient of the second half of the 2019 Nobel Prize. The missing piece of the puzzle was dark matter. Unlike exoplanets, which appear dark next to their brightly shining stars, dark matter appears completely invisible; a ray of light passes straight through as if it weren’t there. This is not as foreign a concept as one might think: our own Milky Way is home to vast quantities of dark matter.
Earth is not alone in the universe - we are not even in a quiet neighbourhood. The confirmation of this simple fact has relied on decades of development of a diverse range of experimental techniques, uncovering clusters of dark matter in galaxies and planetary systems orbiting the stars in our sky. Not only are we not alone, but we are not even particularly unique. And although our place in the universe may not be as special as we would like, it is a place that provides a spectacular view of the structures that surround us. It is by looking away from our place in the universe that we have begun to truly understand it, placing it in context with the rich and exciting variety of the universe nearby.
What more can we say about this wider context? How much is there to discover, even beyond the plethora of exoplanets?
From planets, zoom out, and we see stars arranged into galaxies. Taking an even broader view, we see galaxies arranged in clusters, which are themselves lined up in superclusters, collections of tendrils. But there is an end to this hierarchy. Zoom out even further, to the point where galaxies are only dots, and we see a universe that is roughly the same from region to region, with no structure, center or edge.
Have our familiar starry skies, and these structures they’re embedded in, always existed? The most successful model of the universe that we have, the ”Big Bang” model, answers with an emphatic no. This model has been subject to rigorous and precise tests, and the successes of these tests are what gives us confidence in its grand story, a story that James Peebles played a central part in developing. This is not just the story of our place in space, but also of our place in time.
Stories have characters, and our first character is space itself. Here it has a central role, it expands. As the space between galaxies expands, the average distance between galaxies increases and the wavelength of light travelling between them stretches. In 1929, by observing the reddened light of distant galaxies, Edwin Hubble showed that this was true of our sky, and any astronomer on any exoplanet in the universe would find the same result.
After this discovery, some physicists attempted to reconcile this expansion with an unchanging universe; a steady state, whose skies were always filled with stars. This is a simple idea, but it makes concrete, testable predictions. The light we collect in our telescopes has to travel to reach us, so by looking further away, we see the universe as it was in the past. Steady state models predict that this doesn’t matter. If the universe was always the same, even looking back in time, we should see a universe that matches what we see nearby.
That does not happen. An important clue leading to this conclusion was announced by Penzias and Wilson in 1965. They had discovered a mysterious radiation, seemingly perfectly uniform in temperature across the sky. The uniformity meant that it had nothing to do with our sun or even our galaxy; it had to be cosmic in origin.
Enter James Peebles and his collaborators. Publishing simultaneously with the Penzias–Wilson result, they showed that a changing universe could naturally explain this mysterious radiation, unlike the steady state universe. They explored the idea that this cosmic radiation was left over from an earlier epoch of the universe, one that was much hotter and densely packed. As the universe expanded, the radiation would have cooled until it reached its current temperature of a mere three degrees above absolute zero.
The Penzias–Wilson discovery was cosmic background radiation at a particular microwave frequency. Making the simple assumption that the early epoch was in equilibrium, physicists could then make a precise prediction: they could calculate the intensity at other frequencies. This would be confirmed to spectacular precision as more measurements of this “cosmic microwave background” (CMB) radiation were made.
The early epoch they pictured, which would become known as the “Big Bang fireball”, was a smooth soup. There were no stars, and even if there had been, there was no empty space for the starlight to sail through. The CMB radiation is made up of particles of light, or photons, which in this epoch had extremely high energies, causing them to strip electrons from atoms, turning the matter (that would eventually make us) into a plasma that pushed and pulled against the radiation.
The distribution of the photons and plasma was extremely uniform, but not perfectly so. Some regions started off packed slightly denser and thus had a stronger gravity, pulling the matter and radiation of nearby regions inwards. However, this gravitational collapse was halted by the compressed radiation’s pressure, and the contest between the two caused the radiation to ripple with pressure waves. Another component of this primordial soup was dark matter. It had no pressure to stall the gravitational collapse of its dense regions, so instead of rippling, it began to form clumps.
Eventually, the radiation’s temperature fell so low that electrons and protons in the plasma could safely combine to form neutral atoms, no longer pushing and pulling against radiation. The radiation began to travel freely through the newly transparent universe, leaving visible matter behind and releasing the CMB that we see today. Peebles, in a 1970 paper, explored the transition between these two very different eras: the first few hundred thousand years when photons could travel only very short distances (before interacting with electrons and protons), to the last few billion years, where photons can travel extremely long distances (stopping only when they collide with an astronomer, for example).
The exciting realisation was that the billions of years that have passed since that epochal transition might not have erased this evidence. The visible matter dropped by pressure waves in the earlier era seeded the formation of some of the galaxies we can see today, and could have left an imprint on their distribution. Indeed, in 2005 the SDSS collaboration presented a detection of such an imprint. Along with the measurements of the CMB, this incredible detection formed another pillar of evidence for the hot Big Bang model.
While the foundations of this story are set, there is still much more to understand. After the fireball, the formation of structure began in earnest with the first stars and galaxies. While signals of early galaxy evolution have been detected since the 1960s in the form of quasars, there are so far no uncontroversial detections of the much fainter signals of the first stars. As they first lit up the dark universe, their radiation output is thought to have left an imprint on the CMB, but this tiny signal is very difficult to extract.
Even if it is not yet on solid ground, the work of those teams aiming to detect the universe’s first stars is a pioneering addition to a proud lineage of cosmological research. It is one example of modern theoretical and experimental work that continues to add to our understanding of a distant history filled with plasma, whose non-uniformities eventually seeded the galaxies, stars and planets in our familiar transparent universe. The diverse array of confirmations of this cosmic story, thanks to Peebles and many others, teach us that the humility of the steady state theory was not justified. Emerging from the exotic, mysterious, very early universe comes our place, with stars and sunlight.
We have not yet reached the start of the story...
So far we’ve talked about how we look at the Milky Way to find planets different from our own, and how we look away from the Milky Way to probe even further back in time, up until the oldest light, the CMB. If this is the oldest light, how can we go further? It turns out imagination is quite powerful, and although we can’t physically see it, we can use our minds to find traces of what came before the CMB.
The main idea behind the “Big Bang Model” does not surround some mystifying point in time where everything came from (you may have heard the word “singularity” thrown about) but something we have already explored: in the past the universe was denser and hotter. Therefore, the radiation floating about was so energetic that, right before the CMB was formed, it could rip the electrons out of an atom. What if we push this idea further back? Surely there must have been a time when the universe was SO hot that the photons could even rip a nucleus apart. We are talking of temperatures around 10 billion Kelvin (energies of about 1 MeV), nearly 300 thousand years before the CMB was emitted.
If we start here and run time forward we should see the first nuclei, i.e. the first elements being formed. This is what Peebles called the primeval fireball. And yes, this is the same Peebles as before, an instrumental figure in our understanding of the universe once again. Let’s see how the story goes.
Initially there were protons and neutrons floating about in space interacting with each other and the radiation. At these temperatures/energies, protons and neutrons are spontaneously converted into each other; everything is in equilibrium.
After a while, a small instability builds up. The thing is, neutrons are ever so slightly heavier than protons, and, because E=mc^2, this means it costs slightly more energy to make a neutron than a proton. If the universe is hot enough this doesn’t make a difference because the photons lying around can easily provide that excess energy. But, as the universe expands and cools down, this starts to matter and the number of neutrons start decreasing. The cosmic clock is ticking fast, and in 15 minutes there won’t be any neutrons left, which is a huge problem because we need them if we want to build up nuclei. Once inside a nucleus they can sit happily, safe from decay.
So what do we need to form a nucleus? Well the simplest thing we can do is form Deuterium, which is just a neutron and a proton stuck together, a heavier version of Hydrogen. By itself, Deuterium is not very useful, but it can be used to form Helium-4 (which has 2 protons and 2 neutrons), and from there we can go on to form heavier elements in a chain reaction (to form elements heavy enough to make up planets we actually require stars but that’s a story for another time). Going from Deuterium to Helium is practically instantaneous.The real hardship is forming enough Deuterium in the first place.
You may think that all we have to do is wait until the photon’s average energy is low enough that it won’t break Deuterium apart. This would occur at around 5 minutes in. However, there is an important subtlety: there are A LOT of photons. Roughly, we have about 10 trillion photons per matter particle so even if the average is quite low, there are still quite a few sneaky outliers that could ruin everything, meaning we need to wait 10 minutes so not even a sneaky outlier can destroy our Deuterium nucleus. After that, there were still enough neutrons to make 1 Helium nucleus for every 3 Hydrogen nuclei in the universe. Actually measuring these abundances is extremely hard. The first few measurements were made in the 60s. They have since been continually improved upon. And lo and behold, within experimental error, they agree with the theoretical prediction.
Notice how delicate a balance this is. We have the weak interaction controlling the decay of neutrons, the strong interaction controlling the energy difference between protons and neutrons and the energy required to produce Deuterium, all alongside gravity to control the expansion of the universe. Had any of these forces been different this balance would change quite dramatically. It is an astonishing consistency test of all these disparate areas of physics, a test we have passed with flying colours.
Alright, we’ve gone past when the first nuclei formed, surely this is it, right? We can’t possibly go further back than this. In fairness, there are a few things we can infer from carefully looking at the CMB and the galaxy distribution about what happened at even earlier times, but no direct measurement. We can only explore this level of energy by smashing things together at incredibly high speeds, as we do at the Large Hadron Collider in Geneva, Switzerland. So what do we expect?
At some point, the universe gets so hot that even protons and neutrons are ripped apart into their constituent quarks. Not even the strong force is strong enough to beat these photons at approximately 1 trillion Kelvin (~100 MeV). The hot soup that’s formed goes by the name of quark-gluon plasma and it has been successfully recreated at the LHC.
Going even further back in time, the temperature is so high that the Higgs boson “melts” and can no longer do its job of giving mass to the fundamental particles. At this temperature of a quadrillion Kelvin (~100 GeV) every particle we know and love is massless, so they behave more like radiation than actual matter. The universe sure does look different from what we’re used to.
Now this is definitely it. Beyond these energies we haven’t tested our theories so we can’t know if they stand or not. But, come on, aren’t you curious to see what they say? It turns out they seem to say that at some point in the past the density and temperature of the universe becomes infinite! People called this the “Big Bang” and interpreted it as the beginning of time. You may be uneasy about extending theories beyond their validity (and you should be), but the singularity seemed to always ‘appear’ when making a priori reasonable assumptions such as the existence of regular matter. The real problem was that this “Big Bang” does not fit the data. Now what most physicists believe is that there was some phase of extremely accelerated expansion called inflation, whose precise details are still a matter of heated debates.
One beautiful aspect of the idea of inflation is that it provides an origin for the tiny non-uniformities we mentioned earlier. In this story, these non-uniformities are actually quantum fluctuations that get stretched by extreme expansion, in turn seeding the gravitational collapse that begins our journey from a featureless universe to stars, galaxies, planets and exoplanets, and eventually humans to wonder at our place in it all.
Maeve Madigan, Philip Clarke and João Melo are PhD students studying theoretical physics in the Faculty of Maths. Artwork by Marzia Munafò.