CERN Press Office
PR.02.2000 - 10.02.2000
Welcome to the weird and wonderful world of quarks! As far as we know, they are the tiniest components from which much of matter is made. All of our theories about matter rest on the assumption that many particles, for example the protons and neutrons that we find inside atoms, are actually made up of quarks. But nobody has succeeded in isolating a quark. We think that normally they are so tightly bound up inside other that it is impossible for them to break free and exist independently.
But our theories also say that our universe was created 12-15 billion years ago in a massive explosion that we call the Big Bang. A tiny fraction of a second after this Big Bang, when the temperature and concentration of energy were immensely high, we believe that all of the quarks must have existed freely in a kind of soup, before clumping together and condensing into the kinds of particles we know today as the universe expanded and cooled (rather like steam condensing into water droplets).
While nobody had ever observed free quarks, let alone their transformation into ordinary matter, this fascinating story remained just a theory, albeit a very strong one. So here at CERN, a programme was launched to try to recreate the conditions that existed immediately after the Big Bang (a project affectionately known as the "Little Bang") to see whether we could make free quarks, and watch their transition into complex particles as the system cooled. To follow this story we need to know a little more about these enigmatic characters called quarks.
Before 1955, nobody had ever heard of quarks. We thought that atoms were simply made up of protons, neutrons and electrons, and that these were the simplest components of matter you could get. Then Gell-Mann and Zweig proposed their theory of quarks. He got the name from James Joyce's book "Finnegans Wake", and pronounced it to rhyme with "pork", rather than "park". Particles made up of quarks are called hadrons (there are also other particles called leptons, for example electrons, which have no internal structure). There are two different types of hadrons - baryons, such as protons or neutrons, are composed of three quarks, while other particles called mesons are made of a quark and an anti-quark. Up until 1974, we could explain all the hadrons we knew about by various combinations of three different types of quarks, called up, down and strange. All hadrons had a logical explanation and all of the expected combinations were known, apart from one - three strange quarks. Gell-Mann predicted that this combination must exist, although it had never been seen, and he called it the Omega baryon. When the Omega was subsequently discovered in 1964, this pretty much convinced everyone that quarks really did exist.
In 1974, a new particle called the J-psi was discovered. This gave theoreticians some work to do, but they quickly explained that this was a meson made from a new kind of quark, which they called charm. A few years later, a fifth quark was discovered, called bottom and in 1995, a sixth, called top. And that's it, as far as we know!
Recreating the Big Bang
But how to generate temperatures and energies high enough to release these quarks from their hadron cages? Physicists at CERN do it by smashing heavy ions together as hard and fast as they can. The heavier the ions are, the greater the energy that is released in the crash. The ions of choice are lead, their atomic weight is a hefty 208 and they collide at energies around 3.5 TeV. Large amounts of energy are squeezed into a very small space (a high "energy density. This frees the quarks (and the particles which carry the strong force, called gluons) into a "quark-gluon plasma". The resulting fireball is like a hot gas at high pressure. It rapidly explodes, simultaneously expanding and cooling. As the tiny inferno cools, the quarks condense into hadrons (a process known as "hadronisation"). These hadrons continue to interact with each other for as long as the particle density is high enough. But these interactions, mediated by the strong force, occur only at very short range. As further expansion occurs, these interactions cease, and the particles that are left survive to travel onwards and outwards towards the detectors, which are placed all around.
The CERN lead beam programme was opened in 1994. It consists of several specially built systems, for example a new linear accelerator, as well as pre-existing, interconnected accelerators (the Booster synchrotron, the Proton Synchrotron and the Super Proton Synchrotron), which were upgraded for the project. Seven large experiments were involved, measuring different aspects of lead-lead and also lead-gold collisions. They were codenamed NA44, NA45, NA49, NA50, NA52, NA57 and WA98. The project is a feat of collaboration as well as engineering - scientists from institutes in almost twenty countries are participating in the experiments, including Italy, Germany, France, Russia, Finland, Poland, Greece and the US. Due to specialised nature of the programme, building the machinery for producing the lead ion beam involved an international collaboration.
Looking for quarks
How does one recognise quark-gluon plasma? And what do the physicists want to know about it? One thing they want to measure is the critical temperature (called To) and energy density (Eo) which are needed for quark-gluon plasma to form. They have calculated these from theory (To is predicted to be 150-200 MeV, or 1000 times the temperature at the centre of the sun, and Eo is predicted to be 1 GeV/fm3, or seven times the energy density of normal nuclear matter) but they want to check if particles really do melt under these conditions. The problem is that the physicists can only see the particles that escape from the fireball and reach their detectors. From these signals they have to reconstruct what happened before, to work out whether the quarks and gluons were produced in a dense enough state to form a free plasma. The experiments were all optimised for measuring different signals which might indicate if and how a quark-gluon plasma was formed. Some of them optimised their detectors for one rare signal, while others developed multipurpose detectors which were sensitive to multiple signals.
The direct approach
The only way to "see" the quarks directly is to detect the electromagnetic radiation which they emit. This radiation is in the form of photons. Unfortunately there are also many other processes which can produce photons, so there is a huge amount of background noise. One experiment was looking for these photons, but the overwhelming background made it hard to detect a clear signal. An enhancement of the number of photons was seen at high energies, but we cannot yet be absolutely certain that it originated from free quarks. But what the experiment did do was establish an upper limit on the photons that could be coming from the quarks. This means that they could explain at least 95% of the photons they saw by background processes alone, but up to 5% could have been coming from free quarks.
However 99.9% of the particles reaching the detectors from the lead ion collisions are hadrons, particles made of quarks. By definition these particles must have been formed after the quarks had already condensed, and the quark-gluon plasma no longer existed. But although these particles don't allow us to see the quark-gluon plasma directly, perhaps they can allow us to work backwards to deduce what must have been happening beforehand.
Full of charm - the J-psi
An important predicted signature of free quarks involves a particle called the J-psi. J-psis are made of a charm quark and a charm anti-quark. They are rare because charm quarks are very heavy, and can only be produced at the very first stages immediately after the collision, while the constituents of the nuclei still have their full energy. However the formation of J-psis would be suppressed by the presence of quark gluon matter,so a strong reduction in the number of J-psis leaving the fireball suggests that hot quark-gluon plasma was made in the initial stages of the collision. Sure enough, this is exactly what the physicists saw.
There are other kinds of so-called "vector mesons" called phi, rho zero and omega, which can also tell us some intriguing things. Like the J-psi, they can decay into lepton pairs (for example an electron and a positron, or a muon and an antimuon). If we draw a graph of the number of lepton pairs we see at different lepton pair masses, we end up with a peak for each different kind of vector meson. Usually the rho zeros form a wide peak, with a sharp peak from the omega meson on top. But in these high energy collisions we see no rho zero peak at all, only a broad smear. What the physicists think is happening here is that the rho zeros are being formed and decay while the hadrons are still interacting with each other. The short time between collisions does not give them a chance to develop into a state with a well-defined mass, hence the smearing. In other words, the graph acts as a snapshot of the collision stage directly after the liberated quarks have condensed, but well before the hadrons have stopped interacting.
When analysing the abundancies of the hadrons emerging from the nucleus-nucleus collision, a peculiar feature was observed which strongly distinguishes such collisions from collisions of light particles such as protons or electrons, both at the same and at much higher energies. The original lead ions only contain up and down quarks, there are no strange quarks. We can look at the particles coming out of the collisions and ask how many strange quarks and antiquarks are formed relative to the newly produced up and down quarks and antiquarks. For proton-proton or electron-positron collisions, the fraction of extra strange quarks made is 0.2. This stays the same however much you increase the energy. But for the nucleus-nucleus collisions, the fraction is twice as high, at 0.4. This is what is meant by the fantastic phrase "global strangeness enhancement by a factor of two"! The point is that we know from detailed calculations that once hadrons have been formed, essentially no more strange quarks can be made. So most of these extra strange quarks must have been created before the hadrons were made, ie in a quark-gluon plasma. These extra strange quarks are mostly reflected by an increased production of the very rare particles that contain two or three strange quarks (for example the Omega baryon, whose numbers are increased by up to 15 times), rather than in the generally more common particles that only contain one. This again hints that the "extra strangeness" was formed before the hadrons themselves, and is the other central prediction for the formation of a quark-gluon plasma.
Signatures of chaos
If we look at the particles leaving the fireball, they retain signatures of their recent past. We can use these signatures to work backwards in time, to discover clues about the transition from quarks into ordinary matter. The hadrons that escape and reach the detectors are fixed, in terms of their identity and their momentum. But we can run the film back to the point where the system was dense enough for the particles to be colliding, and bouncing off each other. These are the type of collisions which billiard balls undergo, the identity of the particles does not change, but their momentum does. As the fireball expands, the energy density decreases until the hadrons no longer interact. Their momenta "freeze out", and are fixed from that moment on. So when we look at the momenta of the different particles leaving the fireball, their distribution acts as a memory or snapshot of the conditions when this freeze out occured. We can use the momentum distrubution of the emerging particles to calculate the temperature at which this occurred, it works out to about 100 MeV. We can also calculate that at this point the fireball was expanding at an enormous rate, at over half the speed of light!
But we are interested in what happened even earlier, when the fireball was much hotter and denser, and the quarks were condensing into hadrons. After the hadrons have emerged from coalescing quarks, the system is simply no longer dense enough for the hadrons to interact in this way. The finally measured particle composition is therefore fixed, right at the point of hadronization! We can then use the distribution of the numbers of different types of particles to calculate the temperature at which the transition of quarks to hadrons occurred. It works out to around 180 MeV, which agrees with the critical temperature as predicted by theory.
Particles in pairs
Another technique called Bose-Einstein Interferometry is used to look at pairs of particles. It allows us to measure the final size of the system. We already know how rapidly the system expands, so once we know how big it is, we can use the information to extrapolate backwards and work out the original energy density. It works out at two to four times the theoretically predicted value of Eo, the critical energy density for quark liberation. The initial energy density in the collisions was thus plenty for the formation of quark-gluon plasma, according to theory.
Our picture of quark-gluon plasma therefore resembles a jigsaw puzzle, with many pieces provided by the different experiments. We need all the experiments working in concert to have a complete picture, and to check that all the pieces agree and fit together. There is one important piece still missing, the photon signal that would allow us to "see" the free quarks directly. But we have enough other pieces to be fairly certain of what is happening. It is possible to think up alternative explanations for the individual results, but there is no known theory apart from quark-gluon plasma that can explain all of the results together. Therefore although the evidence is circumstantial, we are confident that it is compelling enough to say that we have formed a new state of matter, in which quarks and gluons are "deconfined".
We have recreated matter in a state we have never seen before, at energy densities twenty times higher than that inside the atomic nucleus. The same conditions have only existed in the first few microseconds after the Big Bang (comparable energy densities, although at much lower temperatures, may also exist in the centre of neutron stars). Our understanding about how the universe was created, which was previously unverified theory for any time point before the formation of ordinary atomic nuclei (about three minutes after the Big Bang), has now been experimentally tested back to a point only a few microseconds after the Big Bang, the time at which we believe the hadrons making up those nuclei were created. Going back further will involve new challenges, which future experiments at CERN's new LHC and also at RHIC at Brookhaven, will address.
|[ Home ] [ Press release ] [ Story ] [ Science ] [ Animations ] [ Experiments ]|