Friday, March 29, 2013

Antiprotons obey CPT within 5 ppm

Just a neutrino link: Two weeks ago, MiniBooNE reported some results that seem to conflict with the standard model of 3-flavor neutrino oscillations. Hat tip: Joseph S.
In January 2013, the ATRAP Collaboration that includes e.g. Gerry Gabrielse – an ex-colleague of mine who also led the most accurate measurement of the electron's magnetic moment, the most accurately verified prediction in all of science – published the preprint
One-Particle Measurement of the Antiproton Magnetic Moment
that finally made it to PRL this Monday. The article was accompanied by a popular review by Eric Hudson and David Saltzberg – who is also famous as the flawless science consultant behind The Big Bang Theory CBS sitcom; he's the man who makes sure that Sheldon Cooper in particular doesn't talk gibberish.

Saltzberg with Bill Prady. What part of 41 53 43 49 49 don't you understand? It does sound like a Sheldonite question...

The experimenters threw some of their Harvard devices to their luggage and pockets and they flew to CERN where the (\(5\MeV\)) antiprotons are cheap and abundant. A happy place, indeed. By measuring some frequencies of transitions in a magnetic field, they could quantify the magnetic moment of the antiproton – how strong a magnet each antiproton is. And yes, except for the sign, the result agreed with the figure for the proton within the antiproton 5 parts per million (0.0005%) error margin.

The proton's magnetic moment is measured with accuracy that is about 500 times better than for the antiproton. I think it's because you don't have to be afraid of the protons so much and they are cooler. When you are trying to determine how smooth the skin of various animals is, you may also get a more accurate measurement for a kitty than for a tiger.

I have written about CPT, CP, C, P, T, and all that many times, e.g. in November 2012, so I won't do it again. Here, let me just mention that while C,P,T,CP have been found to be broken in Nature, CPT has been proven to hold in a Lorentz-invariant Lagrangian quantum field theory since the mid 1950s when it was demonstrated by Wolfgang Pauli, John Bell, and Gerhart Lüders.

This symmetry – which replaces all particle configurations by their mirror images composed of antiparticles and that runs everything backward in time – has to hold because this CPT is the right interpretation of the rotation of the Euclideanized spacetime in the \(tz\) plane by \(\pi\) and this has to be symmetry because it belongs to the analytical continuation of the Lorentz symmetry.

The CPT-symmetry has many trivial consequences. For example, when you talk about the mass of the proton, there's just one number. We don't have a special mass for the "left proton" or the "right proton", a special mass for the "proton observed forward in time" and the "proton observed backward in time". It follows that as far as the value of the mass goes, the transformations P and T are irrelevant. CPT effectively reduces to C and it says that the mass of the particles and their antiparticles have to coincide.

(Most recently, this was verified at the LHC for tops and antitops, despite some previous implausible statements by CDF at the Tevatron.)

Similarly, there are just two possible values of the magnetic moment – one for the proton and one for the antiproton – and it's equally clear that the CPT-symmetry can't do anything else than to relate these two constants and demand that their magnitudes are equal. This is what the ATRAP Collaboration verified at a rather impressive accuracy.

I would think that if the top quark and the antitop antiquark had masses that would differ by \(3\GeV\) i.e. 2 percent or so, as previously claimed by the CDF, a comparable asymmetry would probably exist for other quarks as well and it would be more or less unimaginable that the antiproton – which is a rather complicated bound state of quarks and gluons that feels "everything" – would have the same properties such as the magnetic moment as the proton, within 5 ppm.

Your humble correspondent is a theoretically inclined person who takes all the proofs of the CPT-theorem very seriously and I think that the CPT-symmetry simply has to be exact. But even if you ignored all of theory, there are still experimental constraints. Measurements such as the measurement of the magnetic moment of the antiproton (and the proton) show that the typical differences between the properties of rather ordinary particles and their antiparticles are smaller than several parts per million (several times 0.0001%).

For the CDF to claim that they see evidence of a 2% difference between the top and the antitop was an immensely extraordinary statement that required extraordinary evidence and their 2-sigma deviations, probably caused by some human errors anyway, were simply not the adequate evidence to back similar statements.

At any rate, congratulations to ATRAP.


  1. Yes but took some while. Used IBM EBCDIC so I failed to "C1" even though the A was there "41" to see.

  2. Dear Lubos,

    I read a very disturbing article today claiming that some experiment has violated the complementarity principle, therefore the compahagen interpretation of QM.Now, I think there is something wrong with this experiment or they are misinterpretating the results. Can you please shed more light on this?

  3. I think he has: