Atoms, the Equivalence Principle, and Dueling Laureates

Good to know that our Secretary of Energy, Steve Chu, is still able to unwind from a long day of bureaucracy by thinking about atom interferometry and the Principle of Equivalence.

Equivalence Principle and Gravitational Redshift

Michael A. Hohensee, Steven Chu, Achim Peters, Holger Mueller

We investigate leading order deviations from general relativity that violate the Einstein equivalence principle (EEP) in the gravitational standard model extension (SME). We show that redshift experiments based on matter waves and clock comparisons are equivalent to one another. Consideration of torsion balance tests, along with matter wave, microwave, optical, and M”ossbauer clock tests yields comprehensive limits on spin-independent EEP-violating SME terms at the $10^{-6}$ level.

The Principle of Equivalence says that, if you’re in free fall, there’s no way of detecting the gravitational field around you in a local region of spacetime. (You’ve seen Inception, right?) Unlike electromagnetism, with gravity there’s no local “force” that can be detected by comparing what happens to particles of different charges. In other words, all particles feel the same “charge” as far as gravity is concerned; they all fall in the same way.

So to look for violations of the EP (which are certainly conceivable, even if it sometimes just sounds like technobabble), you do experiments that look for particles doing different things in different kinds of gravitational fields. For example, you can use the EP to predict the gravitational redshift, which can be thought of as “time running more slowly when you are deep in a gravitational potential.” (Not the most precise formulation, but it will do.) And therefore you can test the EP by measuring the different amount of time elapsed by sending clocks on different trajectories.

A relatively new technique for performing such tests is atom interferometry. Rather than literally sending clocks along two different trajectories — which has also been done, of course — you take advantage of the quantum-mechanical wave nature of matter, and simply send two “wave packets of atoms” along different trajectories. Waves oscillate, and the number of oscillations they experience serves as a clock. The benefit is that the waves interfere when they recombine, and we’re very good at measuring tiny deviations in the predicted amount of interference.

But apparently there’s been some controversy over whether the phase of the atom wave really counts as a “clock,” for purposes of this kind of experiment. This new paper by Hohensee and collaborators is saying “yes, it does.” A previous paper by Wolf et al. said “no, it doesn’t.”

At a glance, I tend to agree with Hohensee et al., but I haven’t gone through the arguments very carefully. I’ll just note that one of the authors of the Wolf et al. paper is Claude Cohen-Tannoudji, who shared the Nobel Prize in 1997 with Bill Phillips and … Steve Chu! Always good to see a Nobel-level tussle. I think I’ll let them figure it out.

This entry was posted in arxiv, Science. Bookmark the permalink.

13 Responses to Atoms, the Equivalence Principle, and Dueling Laureates

  1. Jasper says:

    … not that it actually IS technobabble, of course. 🙂

  2. Sean says:

    Of course! (Fooled my dissertation committee, anyway.)

  3. Carl Brannen says:

    I looked briefly at both papers and I think they’re both right. Let me give a different way of looking at the Schwarzschild solution. Instead of Schwarzschild coordinates, use Gullstrand-Painleve (GP). See the wikipedia article for details, or better, see the generalization to Kerr metric (rotating black hole) coordinates in the “river model of black holes”:

    Anyway, in GP coordinates (which of course give the same physical results as any other GR coordinate choice), the gravitational potential does not change the time rate of clocks. Instead, objects closer to the gravitating body are modeled as moving. So the effect on clocks becomes a pure Doppler effect (the test body closer to the gravitating body is the traveling twin). This is basically a question of the universality of free fall.

    I think that the real point of the Wolf et al paper is not that the experiment doesn’t measure anything, but instead that we’re lacking a theory that would predict a difference. If there’s any problem, it’s with the imagination of the gravitation theorists.

  4. Pingback: 23 February 2011 « blueollie

  5. Baby Bones says:

    Actually, I read a criticism of atomic clocks not being good clocks for relativistic experiments back in high school in the 1970’s. The book was Relativity Reexamined by Leon Brillouin, a treatise basically contentious of all things relativistic, but probably worth the notes on good clocks and bad clocks.

    Also, Julian Schwinger had an idea about the Mossbauer effect that went against his contemporaries, whereby there is no initial impulse on one atom from the decay that is transferred to other atoms of the lattice but rather the impulse is on the whole ionic lattice through their quantum superposition. That idea now sounds sounder, at least to me, since I heard that the protein folding problem was recently ‘solved’ by two Chinese physicists. If large protein molecules can select a shape via a large superposition, such effects should also be present in metal lattices. So maybe the Mossbauer effect has a similar explanation, and that might affect the definition of a Mossbauer clock.

  6. Shantanu says:

    Sean, there is one more paper on this
    Also this is another example of why interpretation of results in experimental gravity is hard.
    another example been the measurement of speed of gravity in jovian deflection experiment.
    Sean, now that you finally blogged about this , maybe you could also do the same on that experiment(although its now 9 years old now)

  7. Jolyon says:

    There’s a second paper which disputes whether or not the phase of an atom is a clock.
    I find this reference much easier to read than the Wolf et al paper, and I think that the argument presented here is more compelling.

  8. Paul says:

    Why is the secretary listed as a coauthor anyways? Was this work performed before he took the post, or is he truly engaged in this work in his ‘free time’?

  9. ian says:

    The guy is a Nobel laureate. I’m sure he can contribute to a paper or two without neglecting his job. Plus, it’s in the national interest that he keep doing some science on the side.

  10. wds says:

    @Paul #8: One can only guess how much time the man still has for research, but likely this is something he was already working on before he switched jobs.

  11. Anonymous_Snowboarder says:

    Regardless of the merits of the work (if any), this is highly objectionable if it was done while Energy Secretary. That is his full time job and frankly I don’t think it allows for any free time for moon lighting. Do you think the President should be writing for law journals in his spare time? Or the Sec. Defense perhaps indulging in writing a spy novel during his spare time? Any prior obligations are just that – prior – once you accept a position such as his.

  12. Dave says:

    Consider two hypotheticals:

    If Chu watches a soccer game every week, should citizens complain?

    If Chu spends every Saturday afternoon doing physics research, should citizens complain?

    Science can be done as a hobby.

  13. Matt B. says:

    Anonymous_Snowboarder, if it actually is spare time, then it’s okay.

    As for the technobabble, a better phrasing is “Time runs more slowly for an object when it is in in a deeper gravitational potential than the observer.” Because when you are deep in a gravitational potential, everything runs more quickly than if you weren’t as deep.