This question has its origin to the reference on the Aegis experiment at CERN where they aim to produce super cooled antihydrogen and detect whether its reaction to gravity is negative.

It set me thinking that the beams in the Tevatron circulate for more than a second and everything falls about 4.9 meters in a second, so the bunches must be falling too. This of course will be compensated by the fields that keep the bunches in track among all the other corrections necessary. If though the antiprotons have a different behavior under gravity, this difference would appear in the orbits of protons and antiprotons.

The question has two points: a) since the beams are travelling equal and opposite paths through the magnetic circuit, a negative gravity effect on antiprotons would disperse the antiproton beam up with respect to the path of the proton one. Could one get a limit on the magnitude of the gravitational effect difference between protons and antiprotons from this?

I found one reference where the antiproton beam has a different behavior in chromaticity than the proton one, and it is explained away.

Now I am completely vague about beam dynamics which I have filed under "art" rather than "physics" but b) am wondering whether this observed difference could be interpreted as a gravitational field difference in a dedicated experiment.

Maybe there are beam engineers reading this list. My feeling is that if antiparticles had negative gravity interactions , beam engineers would have detected it since the first e+e- machine, but feelings can be wrong.


The vertical position of the beams is maintained by quadra-pole (and possibly other high n-pole configurations, I'm not actually familiar with the details of the Tevatron) magnets in the ring as part of the general positioning and focusing mechanism.

Because of the difference in coupling strength between gravity and magnetism, any difference due to different behavior in the beams would be too small to measure with the available equipment (or indeed anything you could install in the ring).

There was a proposal to measure $g$ for anti-matter at Fermilab that's been floating around for some time, but the PAC has turned it down repeatedly and now made the decision to alter some of the hardware it would have relied upon. So, the idea is dead for the time being.

I believe the reasoning went like this

  • It ought to work, but no one has used all these principle in one experiment at one time before.
  • It would be a moderately expensive, single measurement experiment
  • It's a good idea, but it doesn't really fit in with our future plans. I mean, it not really "intensity frontier" physics, is it?
  • $\begingroup$ @dmckee Thanks. What caught my attention is that all bodies fall in vacuum 4.8 meters a second, that is a large number, and the bunches circulate much longer than a second. If the antiprotons have negative gravitational attraction, should they not go up by 4.8 meters in a second? True, the orbits are adjusted as I said by "art". An orbit adjusted for protons though should behave drastically differently for antiprotons in the case the antiproton is repulsed by gravity and the parameters are kept the same. $\endgroup$ – anna v Feb 20 '11 at 13:53
  • $\begingroup$ The protons pass focusing magnets every with a frequency on order of $10^8$ times a second (in the lab frame), during the intervals they fall $~10^{-15}\text{m}$ a distance comparable to their own diameter and much shorter than the distance scales on which the magnets can be tuned. Which is to say that they are not falling freely on time scales of 1 second. In the mean time they are also subjected to the electo-static repulsion due to the other (anti-)protons in their bunch which overwhelm the puny gravitation attraction of the Earth. $\endgroup$ – dmckee Feb 20 '11 at 14:16
  • $\begingroup$ @dmckee It is evident that the magnets are tuned to keep a beam in a circe. I was thinking on the lines: tune the proton beam alone, dump it, bring the antiproton beam on with the same parameters.I would expect a different vertical behavior if the antiprotons had negative attraction. Now also , even though the individual turn has a very small drop, if this drop is not compensated in time it will describe a helix and not a circle, so there must be some time compensation programmed in the magnets on the order of microseconds, otherwise the fall would be there. $\endgroup$ – anna v Feb 21 '11 at 6:20
  • $\begingroup$ I would point out that the electrostatic force is not a problem while the beams are circulating because they are kept separate until the beam collision points. Beams may circulate without colliding. $\endgroup$ – anna v Feb 21 '11 at 6:59
  • $\begingroup$ sorry, I just saw that you talk of the repulsion within the bunch. That should not affect the motion of the center of mass of the bunch. $\endgroup$ – anna v Feb 21 '11 at 10:43

I am not good at all in all that but as soon as the particles in accelerators are charged and relativistic, the gravity force is always negligible with respect to other forces present. I think it is like trying to measure the gravity effect on a beam of light.

  • $\begingroup$ The high velocity makes measurements difficult in the cm and meter range. Beam bunches circulate over circles that have kilometer periphery and last a long time. In order not to lose the beams, by hand compensations are decided in the quadrupoles end dipoles ( corrections to calculations) so the fall must be within that compensation. I am arguing that if antiprotons lifted instead of falling the difference should be measurable, because of the value of g. $\endgroup$ – anna v Feb 20 '11 at 14:02
  • $\begingroup$ I understand but apart from systematic gravity action there are other systematic forces and one cannot distinguish what is responsible for what. $\endgroup$ – Vladimir Kalitvianski Feb 20 '11 at 14:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.