Protons or antiprotons ? April 26, 2007
Posted by dorigo in physics, science.trackback
I received a question which I think should be in the “FAQ” of CERN – I think the answer belongs to an independent post, for the benefit of those who never bothered to think the matter over.
So I am curious. Why does the tevatron collide protons-antiprotons, but the LHC goes for proton-proton? This has been puzzling me for some time, and even google has no anwsers
Running protons against antiprotons has the advantage of allowing one to use the same magnets to bend both beams, circulating in opposite directions, in the same way. The Lorentz force experienced by a moving charge in a magnetic field changes sign once for the opposite direction and once for the opposite charge, and the net effect is null, so that both protons and antiproton can travel in the same beam pipe – thus saving magnets, vacuum structures, and a lot of infrastructure.
So, what is the drawback ? It is that producing antiprotons is a maddeningly hard task. You produce antiprotons by smashing protons against a target, and sifting through the emerging bodies downstream, with magnet optics that select them and patiently direct and store them in an accumulation ring.
The art of producing antiproton beams has been perfected at the Tevatron in the last twenty years, but it is just impossible to reach the intensity required to achieve the rate of collisions that CMS and Atlas at LHC need in order to investigate very rare processes. Producing a proton beam is easy: you take hydrogen, strip electrons off, and there you go…
Finally, one must mention that colliding protons versus protons is not exactly the same thing as colliding protons versus antiprotons, as far as the physics output is concerned. Strong interactions do not care whether the projectiles are particle or antiparticle: what they care about is the color charge of quarks and gluons, which is the same in hadrons and anti-hadrons. But electroweak interactions do, because they are sensitive to the flavor of the quarks (electroweak processes do not “see” the gluons, by the way). So the relative rate, and kinematics, of electroweak processes is different at the two accelerators. In any case, these are details, and one can discover pretty much the same things one way or the other.
To summarize: deciding on protons versus antiprotons at the Tevatron at the end of the seventies was -as far as I understand it – a matter of cost versus effectiveness when a 900 GeV machine was designed. The Main Ring, a fixed-target (only protons circulating only in one verse) 400 GeV synchrotron which had helped discover the Upsilon mesons and the bottom quark at the end of the seventies, could be used to run hadron-hadron _collisions_, provided one injected antiprotons in one way and protons in the other. This yielded the possibility to try and discover the top quark without a complete redesign of the machine (although, well, major upgrades were needed anyways).
On the other hand the LHC, being a totally new machine, was instead built with two separate beampipes, and separate magnets in the bending tracts, thereby allowing proton-proton operation, with intensities that will exceed by two orders of magnitude those of the Tevatron – and allow to study 100-times rarer processes.
Comments
Sorry comments are closed for this entry
A Quantum Diaries Survivor 
I always answer to the same question in this way (more physics-oriented, while yours is more tools-oriented):
The proton momentum is distributed among its components, some of whom are the valence quarks and the others are virtual quarks and gluon.
Most of the momentum of a proton is carried by the valence quarks.
So if the threshold for creating some new particle or observing some new phenomenon is, let’s say, 200 GeV, and your protons have a momentum of 1 TeV, you need to pick up from the two protons (or from the proton and antiproton) two constituents of roughly 100 GeV each, i.e. with 10% of the momentum of a proton.
It turns out that most of the virtual particles carry much less than 10% of the momentum, while this is quite frequent for the valence quarks.
So, since Tevatron can afford momenta of roughly 1 TeV for the incoming particles, it’s quite convenient to collide protons against antiprotons: because the valence quark in the proton will easily find a valence anti-quark in the anti-proton.
In fact, the rate for the process “q anti-q -> W” is orders of magnitude higher in the Tevatron conditions than at an hypothetical proton-proton collider at the same energy.
Instead, LHC has 7 TeV per proton. To pass a threshold of 200 GeV, an elementary interaction between constituents needs only roughly 1 % of the momentum of the proton. This means that also the virtual quarks and gluons contribute easily. In fact, the cross section for “q anti-q -> W” at LHC energy is not so different between the proton-proton and the proton-antiproton case.
So, it would be silly to use antiprotons since it is so difficult to have them abundantly.
And in particular, for processes involving the strong force (like the top quark production), the possibility to involve the virtual gluons (which carry a small fraction of the total momentum each, so are unuseful at low-momentum collision, but are a lot) means that the rates can increase spectacularly.
In fact, a mere factor of 7 in the beam energy gives a factor of 120 increase for top quark production at LHC with respect to Tevatron.
Quite right, Andrea. Your answer explains well why the Tevatron went for antiprotons rather than a doubling of their magnets (they needed q-qbar interactions to discover the top at 1.8 TeV energy) while mine explains better why LHC chose the other route (they need as high intensity as they can to push the discovery limit).
Only one thing though, to clarify: even p-pbar collisions at 14 TeV would produce O(120x) more ttbar pairs. QCD is matter-antimatter blind when valence quarks are irrelevant.
Cheers,
T.
yes, sure, I could be misinterpreted on this last point:)
Anyway, also your explaination is a very good one: I found interesting to realize that while I always think first in terms of physics (“first we decide what is interesting to look for, then SOMEBODY ELSE will find a way to build a facility for looking at it” ;)), you (correctly) give a lot of weight to the engineering problems.
Yes, quite right. In the end, it all boils down to how much science you can do with the resources you have.
Cheers,
T.
Dorigo, it’s funny that I can actually see you talking with your nose in the air through your writings.
Cheers,
A.
Hi A,
hmmm, my English does not allow me to understand whether you are mocking me or what. I’ll take it as a compliment, because I am such an optimist…
Cheers,
T.