Anti-matter = repulsive gravity?

[Frenetic Pony]

So for those following the latest scientific news they'll know there's a new astrophysical/physics hypothesis out there that states that anti-matter has both a negative gravitational charge, repulsing matter and attracting other anti-matter, and that like many theories state there is actually about an equal amount of it compared to normal matter in the universe.

You can read an overview here: http://www.physorg.com/news/2012-01-repulsive-gravity-alternative-dark-energy.html

Since this is the smartest forum I've found on the net, and I've only a general overview knowledge of physics, I'd thought I'd ask you guys what this hypothesis looked like. I mean, as an explanation for dark energy/no need for a cosmological constant, possible explanation for dark matter, etc. it all sounds very nice.

So, thoughts?

[SlyReaper]

It's not a new hypothesis, it's been around pretty much as long as we've known antimatter exists. It's certainly an elegant answer to questions like "why didn't the universe self-annihilate right after the big bang" and "why are distant objects accelerating away from us". But the real test is whether the theory works in the details, and as far as I'm aware, it encounters problems in the details. As a result, last time I looked into it, the consensus seemed to be that antimatter behaves reacts the same way as normal matter to gravitational fields. But of course it's well worth doing the experiment to be sure. We've only been able to observe antimatter for very fleeting moments up until very recently, and even then, gravitational forces have been overwhelmed by electromagnetic ones, so it has never been clear which direction it falls.

One of the stumbling points, if I remember correctly, is that "neutral" particles like photons which are their own antiparticle are attracted by gravity. If there's no difference between the positive case and the neutral case, why should the negative case be different?

[starslayer]

We see exactly zero evidence of this in the cosmos, so I'm going to go with "nice, but almost certainly not true." If antimatter really does exist in large quantities in voids, you would expect it to form antistars and antigalaxies just like regular matter does. But we don't see that; we would also expect to see gamma ray emission from the boundaries between matter and antimatter regions, since at least some of it would manage to overcome the gravitational repulsion. But we don't see that. You see where this is going.

Also, his assertion that "matter and antimatter having different gravitational charge is a prediction of GR" is news to me, and it appears that only this guy actually believes that. What matters in GR is the energy content of spacetime, and both matter and antimatter have positive rest energies and kinetic energies, so they should curve spacetime in the same way.

So anyway, what do we see? Flat galactic rotation curves indicating the presence of extra matter we can't see (oh hi, dark matter!), and a universe that is expanding, and that this expansion is accelerating. If matter and antimatter were gravitationally repulsive, you would expect to see filaments of antigalaxies, since there would be small primordial density fluctuations for antimatter as well as matter, and thus the voids would have these galaxy filaments running through them. But we don't see that (again); what we see is galaxy filaments that are all connected to each other skirting roughly spherical spaces where there is virtually nothing (the voids). All of which is easily explained by the dominant visible matter component of the universe being normal matter. There are no large-scale accumulations of antimatter in the universe. This guy's nutty.

[Charlie!]

The biggest conflict with what I know of is with the mass-energy equivalence in general relativity. Energy curves space just like mass does. And we know that antimatter has positive energy because if you collide an electron and a positron, you get two 511 keV photons.

[JWalker]

We see exactly zero evidence of this in the cosmos, so I'm going to go with "nice, but almost certainly not true." If antimatter really does exist in large quantities in voids, you would expect it to form antistars and antigalaxies just like regular matter does. But we don't see that; we would also expect to see gamma ray emission from the boundaries between matter and antimatter regions, since at least some of it would manage to overcome the gravitational repulsion. But we don't see that. You see where this is going.

How can you tell from billions of light years away if a star/galaxy is made of anti-matter or not?

[yurell]

Because if some parts of the universe are matter, and some are antimatter, there'd be an interface between the two, where we'd expect to see antimatter/matter annihilation, but we don't.

[starslayer]

You would tell from its interaction with normal matter at the boundary region; you would expect mixing of some sort there, as random matter particles find their way in/antimatter particles find their way out, resulting in gamma ray emission. But we see nothing like that anywhere. Remember also that its not like matter and antimatter would have formed in disparate parts of the universe separate from one another - they would have been mixed up and annihilated very quickly in the early universe. You can get baryogenesis asymmetries fairly easily from the Standard Model, but IIRC the quick estimates show that there would only be enough normal matter left to make one or a few large galaxies, not the tens of billions we see today. It remains an open question what exactly caused the asymmetry to go in favor of ordinary matter so heavily.

And ninja'd.

[doogly]

The current article makes it clear this guy does not understand GR.
The linked article to past work, from last April, makes it clear he also does not understand QFT.
C symmetry does not get you antimatter - CP does. CPT does not get you antimatter - that gets you back to normal matter. This is why people will sometimes say antimatter is normal matter moving backwards through time. It's cute and maybe a little misleading but not nearly so awful as this guy's straight up nonsense.

You can propose new theories, go to town, but his work is based on flawed understandings of basic subjects.

[JWalker]

Because if some parts of the universe are matter, and some are antimatter, there'd be an interface between the two, where we'd expect to see antimatter/matter annihilation, but we don't.

I don't quite understand why this should be the case. Isn't intergalactic space largely devoid of matter? Is there still enough matter in intergalactic space that an anti-galaxy should radiate strongly?

[yurell]

Intergalactic space is largely empty, but there is still stuff there, and there is a lot of said space. For example, the forbidden transition in hydrogen (the 21-cm line) is incredibly unlikely: 2.9E-15 s^-1. The density of space is a few atoms per cubic centimetre, but nevertheless the entire sky glows at this frequency.
The Intergalactic Medium is 10 to 100 hydrogen atoms per cubic metre, so we would expect a lot of interactions at the boundary.

[PM 2Ring]

FWIW, here's some stuff I wrote yesterday on another forum in response to this article. The guy who posted the article was particularly puzzled by the remark in the article that "the gravitational repulsion between matter and antimatter is a prediction of general relativity".

> In this case, the repulsive gravity could stem from antimatter hiding in voids.

Well, it's *possible* that antimatter has the opposite gravitational charge to normal matter, but this is *extremely* hard to test without access to macroscopic quantities of antimatter. The consensus in the field is that antimatter doesn't have the opposite gravitational charge to normal matter, but at this stage it hasn't been ruled out by theory or experiment. If antimatter turns out to behave like this it would be fairly easy to incorporate it into GR, but to say that "the gravitational repulsion between matter and antimatter is a prediction of general relativity" is somewhat misleading IMHO.

...............................................................

> When scientists discovered in 1998 that the Universe is expanding at an accelerating rate, the possibility that dark energy could explain the observation was intriguing. But because there has been little progress in figuring out exactly what dark energy is, the idea has since become more of a problem than a solution for some scientists.

True.

> One physicist, Massimo Villata of the National Institute for Astrophysics (INAF) in Pino Torinese, Italy, describes dark energy as “embarrassing,” saying that the concept is an ad hoc element to standard cosmology and is devoid of any physical meaning.

Not really - assuming that the cosmological constant is exactly equal to zero is just as ad hoc as assigning a non-zero value to it. It's not totally devoid of any physical meaning - it represents a global tension acting on the whole of spacetime. It would be nice to be able to predict the value of the CC from first principles, but that may turn out to be impossible, and the best we'll get is a range of "legal" CC values that lead to a reasonably stable universe.

> “Cosmic voids (and in particular the nearby Local Void) are observationally very well known and constitute the largest structures of which our Universe is composed,” Villata told PhysOrg.com.

True.

> “The problem is whether they are really empty or contain the repulsive antimatter.”
> In Villata’s paper, which will soon be published in Astrophysics and Space Science, he suggests that antimatter could be hiding in these large voids, separated from matter by mutual gravitational repulsion.

Why is it hiding? If antimatter is just like matter, then in large enough concentrations it should form stars.

> As he explained previously, the gravitational repulsion between matter and antimatter is a prediction of general relativity.

No. The gravitational repulsion between matter and antimatter is consistent with GR, it's certainly not a prediction of GR, which has little to say about the nature of matter. All GR cares about in this regard is that matter has mass, and it postulates that the gravitational mass of a body is the same magnitude as its inertial mass.

> The gravitational repulsion between matter and antimatter could be so powerful, in fact, that Villata has calculated that it could be responsible for the accelerated expansion of the Universe, eliminating the need for dark energy and possibly dark matter.

It would be nice for a single theory to simultaneously account for both dark energy and dark matter. I'm glad he's done some calculations using this theory, but I'd be interested to know if his reasoning and calculations have been independently verified.

> Unlike the first two components that are attractive, the third component could be repulsive, according to Villata. In support of this possibility, he notes that the Leo Spur galaxies, which would be located in between the Local Sheet and the attractive area, appear to be at rest with respect to this motion. Villata suggests that the origins of the third component may be on the opposite side, repelling the Local Sheet instead of attracting it. He calculates that a reasonable antimatter mass, located in a particular void, could account for the local velocity anomaly by the mechanism of repulsive gravity.
> In this way, the antimatter would act like dark energy in our local neighborhood.

Ok.

> On a large scale, numerous antimatter voids could drive the expansion of the Universe without the need for dark energy, and possibly even without the need for an explosive Big Bang (perhaps implying a cyclic Universe). The theory also implies that we live in a Universe with equal amounts of matter and antimatter, as expected by standard theories. To Villata, these results make repulsive gravity an alluring alternative to dark energy.

Grrr. I wish people would stop saying the BB was some kind of explosion - it leads to misconceptions about the nature of the BB. Yes, there was expansion during the BB, especially during the (postulated) inflation phase, but that expansion was very smooth and symmetrical, not chaotic, like a typical explosion. If you blow a soap bubble with a bubble-pipe, the bubble certainly expands, but you'd hardly call that an explosion. The other thing with typical explosions is that they involve a rapid release of energy, which causes the exploding material to heat up. But the BB was at its hottest at its onset and thenceforth proceeded to cool down, with various types of force and matter "freezing out" as the temperature became cold enough for them to persist.

[dockaon]

The main problem I see with this is most of the conventional mass (non-dark matter) in the universe isn't the mass of the fundamental particles but rather nuclear binding energy. The quarks that make up a proton or neutron are ~1% of the mass of the proton or neutron. The rest of the mass comes from the interchange of gluons that are binding the proton together.

Gluons don't have a well defined status as matter or anti-matter since they carry both color and anticolor charges, so there's no reason to believe that the binding energy from the strong interaction in an anti-proton is any different from the binding energy in a proton. So even if the anti-quarks had negative gravitational mass, anti-protons would still have a positive gravitational mass although slightly less than that off a proton. Since protons and neutrons make up the overwhelming majority of the mass of conventional matter, anti-matter would still have an overall positive gravitational mass. Therefore, there would be no separation of matter and anti-matter.

[doogly]

We actually have on great authority that antimatter has positive inertial mass. If you want to say gravitational mass and inertial mass are different, you are violating the equivalence principle, and are not GR's friend.

[PM 2Ring]

Nice work, dockaon.

We actually have on great authority that antimatter has positive inertial mass. If you want to say gravitational mass and inertial mass are different, you are violating the equivalence principle, and are not GR's friend.

Well, I did say they do have the same magnitude. I'd be very surprised if antimatter has the opposite gravitational charge to regular matter, but I don't see why it'd be a major theoretical calamity. OTOH, what you said earlier about CP / CPT does make a lot of sense.

Still, the fact remains that we haven't yet directly measured the gravitational mass of antimatter, so I'm prepared to allow weird theories a bit of slack if they want to entertain anti-gravitational notions, even if it does offend my physical intuitions. No matter how good are intuitions are, or how sophisticated are theories are, the universe always has the final word - what ultimately counts is how stuff actually behaves.

[doogly]

It wouldn't fuck with Newtonian gravity at all, but it ruins the equivalence principal. If you can tell whether you are undergoing constant acceleration or in a uniform gravitational field, you have just made Einstein cry. Are you happy with yourself?

[legend]

Another argument against this sort of behavior is that you could violate the conservation of energy; you could take a particle and corresponding antiparticle, move them from a gravitational well (this requires no energy as the total gravitational mass of the system would be zero), let them annihilate there, let the resulting photons fall back in the well (thus increasing their frequency/energy) and use them again for pair production. This closes the cycle, expect that you just generated a little bit of energy out of nothing.

[PM 2Ring]

It wouldn't fuck with Newtonian gravity at all, but it ruins the equivalence principal. If you can tell whether you are undergoing constant acceleration or in a uniform gravitational field, you have just made Einstein cry. Are you happy with yourself?

Maybe I'm just being dense, but I don't understand how it'd break the equivalence principal for antimatter to have negative gravitational mass and positive inertial mass. I cerytainly agree that negative inertial mass would be evil.

[doogly]

OK, you're in an elevator. Einstein says, "you can't tell if you're in a constant gravitational field, or uniformly accelerating!" You produce a proton in your left hand and a positron in your right, and you let go.

[JWalker]

Intergalactic space is largely empty, but there is still stuff there, and there is a lot of said space. For example, the forbidden transition in hydrogen (the 21-cm line) is incredibly unlikely: 2.9E-15 s^-1. The density of space is a few atoms per cubic centimetre, but nevertheless the entire sky glows at this frequency.
The Intergalactic Medium is 10 to 100 hydrogen atoms per cubic metre, so we would expect a lot of interactions at the boundary.

Thats a very low density, are there any actual calculations as to what kind of luminosity we would expect from this with normal gravitational attraction? For that matter, are there any calculations on what kind of luminosity we would expect assuming a repulsive gravitational interaction? I sort of doubt anyone has seriously bothered to look into the negative gravitational interaction case, but I don't know for sure. Even the normal case would be interesting to see, and I'm sure thats out there, but I have no idea where to look for it.

[yurell]

Hmmm, I wouldn't know. I can try to create ballpark estimates, and I'm sure someone here will tell me I'm wrong at every step, but here goes:

How fast is the intergalactic medium travel? Well, it's at about E7K, so that puts the protons travelling at about one thousandth the speed of light. Let's assume that our interaction zone has hit equilibrium (i.e. the number of particles entering are the number of particles being annihilated). At ten atoms per cubic metre, we have 10⁶ interactions per square metre per second along the interface.
Just ftr, I have made a LOT of simplifying assumptions, so I hope someone better at this than I can come up with a way to do it using fewer assumptions.

At 10^-10 Joules per interaction, the interface will be glowing at 10^{-4[/sp] W m[sup]2}.

Where would this interface be? Well, our own halo is about 200 000 light years in diameter, so let's call the interface boundary a sphere 100 000 light years in radius. This is 10^{21[/i] metres, giving us 10[sup]43} m².

This gives a grand total of 10³⁹ watts.

A hundred billion stars each as bright as the sun would give us c. 10³⁷ watts, so the interface is about as bright as a galaxy, shining at whatever frequency gamma rays come out of proton-proton annihilation. I've no idea how far off I've become with my estimations, but even if I'm off by five or six orders of magnitude, it's still really bright and obvious.

[PM 2Ring]

OK, you're in an elevator. Einstein says, "you can't tell if you're in a constant gravitational field, or uniformly accelerating!" You produce a proton in your left hand and a positron in your right, and you let go.

Yeah, ok.

If I think the elevator's accelerating, then both particles should move in the same direction, but if we're really in a constant gravitational field, then the two particles will move in opposite directions if the antimatter has negative gravitational mass theory is correct.

Thanks, Doogly.

[Anaximander]

I don't think it is a currently accepted theory, but, once upon a time, one of the theories for anti-matter was that it was matter traveling "backwards" in time. If that were true then perhaps gravity would work "backwards" as well within that...I don't know what you'd call that...bizarro world.

I don't have any other details and I don't get to fool around with physical theory anymore as much as I'd like to , but I'd be interested to hear what the conventional wisdom is on something like that.

[legend]

The interpretation of antimatter as matter moving backwards through time might be somewhat confusing and misguiding , but it's not that far from the truth. The problem is rather that inverting your time axis doesn't change the sign of your energy density (you can easily see this by a simple dimensional analysis). And the only thing GR actually cares about is the energy density (we're assuming an isotropic, pressure less gas cloud) and not what this energy is made of and what direction in time it is moving.

[doogly]

Yeah, antimatter is defined to be a CP (charge and parity) transform of regular matter. If CPT is an exact symmetry, then antimatter is also a T (time) transform of regular matter. And as said, this does not change gravity.

[Minerva]

My understanding is that since gravity is a second-rank tensor field (hence why gravitons are theoretically spin-2) a consequence of that is that gravity can only be attractive, not repulsive.

And anyway, haven't we already done plenty of experimental confirmation that antiparticles (including neutral antihydrogen atoms, in the ATHENA/ATRAP etc. Antiproton Decelerator experiments) do fall down in a gravitational field as you'd expect?

[SlyReaper]

Nope. All the antimatter observed so far has either been to short lived or held in magnetic fields that completely overwhelm the effect of gravity, so there's nothing conclusive.

[doogly]

My understanding is that since gravity is a second-rank tensor field (hence why gravitons are theoretically spin-2) a consequence of that is that gravity can only be attractive, not repulsive.

If something had negative rest mass, it would curve things the other way. That is the sort of thing folks in the gravity business like to call "nonsense," though.
Sometimes with negative energy quantum states, like Casimir stuff, you can also get similar effects. But they might not be possible to build up enough. Current research (my research!) will tell more.

[DrZiro]

OK, you're in an elevator. Einstein says, "you can't tell if you're in a constant gravitational field, or uniformly accelerating!" You produce a proton in your left hand and a positron in your right, and you let go.

Interesting example. I'll have to think more about that.

I think I have another argument. We assume that matter/antimatter is symmetric, right? That is, the only way you can tell that something is antimatter is by letting it interact with matter (or something else established to be antimatter). For example, if it annihilates in contact with matter, it's antimatter. But you can't just take any particle and antiparticle and expect them to annihilate; they have to match. So we can decide for each particle/antiparticle pair which is matter and which is antimatter. Am I right?

If so, then if we decide for example that the proton is matter by definition, how should we define the other particles? The most logical choice surely would be that those with negative charge are antiparticles. The commonly accepted idea that the electron is also matter, is just for convenience since they happen to more common.

Obviously my point being that the electron doesn't have negative gravitational mass.

Actually, I don't know if we've ever reliably measured the (gravitational) mass of electrons anyway. Have we?

[mfb]

Particles do not work that way DrZiro. While the definition of "matter" and "antimatter" is arbitrary (and can be problematic sometimes, as shown below), there are groups which you can identify:

The total number of quarks minus the total number of antiquarks is conserved within the Standard Model (of particle physics). Naming one quark "matter" or "antimatter" determines the labels for all other quarks.
Protons and neutrons consist of 3 (valence) quarks. Therefore, protons and neutrons have to be in the same group. In addition, all other baryons are fixed, as they consist of quarks.

In a similar way, the number of leptons minus the number of antileptons is conserved. I think that there is no fundamental reason why electrons have to be in the same "matter" group. It is interesting that some theories predict a conservation of a difference between leptons and baryons, so the convenient way would be to call electrons "antimatter". Well, of course the current definition has historical reasons.

Some other interesting things:
Mesons consist of a quark and an anti-quark. They have their own antiparticles as well - but a definition of one as "matter" and the other one as "antimatter" would be arbitrary. If they are charged, this charge is used to name them: \pi^+ is a positive pion, its antiparticle is \pi^- and the antiparticle of this is again a \pi^+. For uncharged mesons, the names have just historical reasons.

>> We assume that matter/antimatter is symmetric, right?
This is not true, the weak interaction behaves different if you exchange particles and antiparticles (charge conjugation, "C"). You can make them more similar by additionally mirroring the space ("parity transformation", P), but even then there are small differences, this is called CP-violation.
There are good reasons that you might get the same result if you additionally reverse the time direction, too.

[DrZiro]

Well, I'm no expert. But suppose we were to name the down quark an antiparticle. Wouldn't the quark conservation rule still hold? Can you give me an example when it doesn't?

[doogly]

C doesn't exchange particles and antiparticles, only CP does that.

Neutron decay is one where up and down transition, so if you replaced down with anti down and did nothing to up, you would have a reaction that violates conservation for quarks.

[thoughtfully]

Actually, the Standard Model can violate baryon number, but only nonperturbatively, via the sphaleron process. B-L (baryons/quarks minus leptons) is conserved.

I don't know about you, but I'm glad there's some baryon nonconservation going around, even if it was only in the very earliest instants of the Big Bang

[doogly]

Oh yes, it's not an exact symmetry. But if you could violate it in beta decay we'd be in deep trouble.

[DrZiro]

I see. Well, my point still stands about the electrons.

[doogly]

No it doesn't. Electrons interact with things beside their own antiparticles. What if you have an electron and an antimuon interact? They can interact electromagnetically and just produce a pair of photons, conserving leptop number. If you wanted to say the positron is "regular" and the electron is now called an "antipositron," you'd run into trouble here, because now you have an interaction that does not conserve lepton number.

This keeps happening. You have to change everything if you want to change things.

[DrZiro]

Yes, but protons being particles still wouldn't mean that electrons have to be particles. Right? That's how I interpreted mfb:s post.

[doogly]

Weak interactions conect everything, you have no freedom piece by piece.

[thoughtfully]

I think you can flip all leptons and manage, right? Other than it being silly to have a bass-ackwards convention, totally defeating the purpose of conventions

[doogly]

Oh, hmm, yeah this might be fine.
But clearly you want to define the stuff that exists in your average atom as "regular," no?

[yurell]

I think they wanted to define it so all positive charge is 'regular' matter.