There are many subsystems in ALICE, each of them with different purposes. The main tracking detector in ALICE is a Time Projection Chamber (TPC). The TPC is cylinder about 5m (16.4 ft) in diameter and 5 m long filled with gas – the largest time projection chamber in the world. You can see above that a person can actually fit in it. The basic principles behind a TPC are really simple:
- A charged particle in a magnetic field moves in a circle with a radius r = p/qB where p is the particle’s momentum, q is the charge of the particle, and B is the magnetic field. The magnetic field in the TPC is roughly constant, as shown below:
so that a charged particle will bend either clockwise or counterclockwise, depending on its charge:
- A charged particle moving through a gas will ionize the gas, knocking electrons (called secondary electrons) off the gas molecules
- The TPC has an electric field (400 V/cm), as shown below, that causes the secondary electrons to drift to the end of the TPC (the ends of the cylinder):
- We have pads on the ends of the TPC which collect the charge and each of these pads has to be read out for every event. The position on the end of the cylinder (usually called x and y) is determined by where the charge hits the end of the TPC. The position along the beam axis (usually called z) is determined by the time it took the charge to get there and the drift velocity of the TPC. (This conversion from time to distance is what gives the TPC its name – a time projection chamber.)
- This gives us a bunch of “hits” – the position of a charged particle in the TPC at different times so what we actually see are the red dots below:
We then have to figure out which hits belong to which track. Not so difficult if there’s only a few tracks but for heavy ion collisions we expect to create a few thousand particles in each collision.
- We have to collect data fast enough that the detector is ready for the next collision. Collisions occur several hundred times per second.
A lot of details have to be just right to get a TPC to work well. We have to know the electric field and the magnetic field very precisely. The amount of charge left by a particle is sensitive to the type of gas and to the temperature. We have to keep the temperature constant to within 0.1 degree Celcius. Because the TPC is so large, keeping everything constant and well calibrated is very difficult. But ALICE has done it.
And not only is it the biggest TPC in the world, but it’s also the best, in my humble opinion.
Here you can see some tracks reconstructed in the TPC from a 7 TeV proton-proton collision:
You can see some more event displays here. Some animations of event displays collisions at 7 TeV in ALICE are here, here, and here. (You can see some of the other detectors in these displays – I left them out of the diagrams above for simplicity.)
ps – Thanks to Jim Thomas, one of the many members of the TPC team, for helping me find event displays, technical details, and editing!
Tags: ALICE, detector, heavy ion physics
Hi Christine,
Stupid question time I am afraid. When the beams are circulating and upto speed there are 4 places (I believe) where they can be brought together for collisions. Are the collisions simultaneous in the 4 areas, or does each experiment get a collison, then the next one etc or even that on a given day it will be CMS only, then ATLAS etc? And secondly would there be any interaction from a collision at point A on the detectors at point B or are the particles that are produced so short lived that this can’t happen? Thank you again for giving an insight into this fantastic project.
Regards
Mark
Hi Mark
These are not stupid questions at all.
Now that everything is running well, there can be collisions at all four interaction regions at once, meaning that all of the experiments can be collecting data at the same time. That doesn’t mean that the collisions happen at exactly the same time in all interaction regions or that the rate of collisions is the same everywhere. The beam is not continuous – it’s actually in little bunches, like repeated shotgun blasts. When two bunches cross, the probability of a collision happening depends on how many protons are in the bunch, how big the bunch is (how spread out the protons are), whether or not the bunches hit head-on, and their cross section (roughly how big a proton is). The beam isn’t necessarily equally focused at each interaction region (the protons can be more spread out at one point) and the bunches may hit each other head-on at one interaction point but barely graze – or even miss – each other at another. Getting the beams to interact well at all four interaction points simultaneously is not easy. In fact, because it takes a while to get good at steering the beam, for the first collisions we only had interactions at one experiment at a time.
Collisions from one experiment affecting other experiments – The beams move in circles. Debris from collisions won’t move with exactly the same radius, so particles created in a collision in ALICE won’t affect ATLAS. But things that happen at one interaction point can affect where the beam moves and therefore what will happen at other interaction points. When I was on the STAR experiment at the Relativistic Heavy Ion Collider, sometimes our magnet would trip (meaning it rapidly changed from a 0.5 tesla magnetic field to 0 tesla). When this would happen, the beam would be deflected, we’d lose the beam, and sometimes the other experiments would get grazed by the beam. This can’t happen at the LHC. At full energy the beam can bore several feet into solid steel so if the beam suddenly changes course it can do a lot of damage – and if the beam hit one of the detectors, the detector would be destroyed. So the LHC has several safety mechanisms so this doesn’t happen. But even minor changes in the magnetic field move the beam. Moreover, just the fact that there are collisions at one interaction point makes the beam get broader and decreases the number of protons in each bunch. So debris from collisions at another interaction point are not an issue, but things that happen at one interaction point can affect what happens at another.
Just curious… In your pictures, the direction of magnetic field is perpendicular to electric field.
This is very strange. I can understand that the direction of electric field is along the pipe (that is how particles are propelled and that is how the charge is carried out), but how can this be aligned with magnetic field?
Also, protons in such magnetic field would run in circles even before the collisions, so how are they kept within the pipe?
sorry, not perpendicular, but parallel.
(my poor english)
Hi Marius – I’m not certain I understand exactly what you’re asking, so if this doesn’t make it clear, please ask a follow up question.
As for technical challenges, we can make constant magnetic field and the electric fields in any direction we want. (If we wanted very fancy patterns, this could get difficult.) The magnetic field in the TPC comes from a solenoidal magnet surrounding the detector. The electric field comes from applying a voltage between a membrane in the center of the TPC and the ends of the TPC.
Many of the particles created in collisions at the LHC move roughly perpendicular to the beam pipe (and the electric and magnetic fields) – not parallel. Imagine throwing two glass balls at each other head on (at equal speeds) so that they shatter. You will have some fragments moving in the direction of the original balls but shards of glass will fly everywhere, since the system has no net momentum. In ALICE, we’re most interested in the momentum of particles moving perpendicular to the beam pipe and most interested in their momentum perpendicular to the beam pipe. (There’s a lot of interesting physics to be learned from particles closer to parallel to the beam pipe, but that’s not the focus of the ALICE physics program.) A magnetic field parallel to the beam lets us measure this momentum most accurately.
The electric field isn’t actually what propels the particles out of the TPC. The particles we’re looking at are moving way too fast to care about the electric field. Think about it as like shooting a bullet (the fast particle) through Jello – the bullet is going to go through. As a (fast) particle moves through the TPC, it knocks electrons out of the gas molecules. These electrons are pulled by the electric field to the end of the TPC, where we collect them to see where the particle went.
Ideally, if the protons in the beam were going perfectly parallel to the magnetic field, they would not be deflected at all by the magnetic field in the TPC. The momentum (p) in my equation above is the momentum perpendicular to the magnetic field. (It comes from a cross product between the momentum and the magnetic field.) In practice, there are slight imperfections in both the magnetic field and the direction of the beam, and these imperfections can cause slight deflections of the beam. We just have to compensate for these imperfections by carefully tuned magnetic fields elsewhere.