"Gravitons"

[Sisyphus]

i.e., the hypothetical "particle responsible for gravity."

 

Beyond the fact that everything else seems to be particles/waves, is there any reason to expect that such a thing exists?

 

Given the ridiculously small effect/probability of such a thing, do we ever expect to actually detect such a thing? Wouldn't its effect fall within the range of quantum uncertainty?

 

If we can't ever detect one, is there any point in talking about them? Further, if it is impossible to detect one, shouldn't we be saying that it doesn't exist?

 

Talk down to me. I'm way out of my league with this stuff.

[ecoli]

I never had a problem with thinking of gravity as purely the way matter bends space-time around itself. Why does there 'need' to be a particle? I have no idea.

[swansont]

GR breaks down at the very small scales where quantum effects are seen.

 

See the "Quantum and Relativity" thread, in theaters now.

http://www.scienceforums.net/forum/showthread.php?t=26246

[ajb]

So what is a graviton...

 

In order to quantise general relativity one starts by power expanding around a classical around a classical solution to Einstein's field equations. Usually this solution is taken to be Minkowski space-time.

 

[math]g_{\mu \nu}(x) = g_{\mu \nu}^{0}(x) + \kappa h_{\mu \nu}(x)[/math]

 

 

where [math] h_{\mu \nu}(x)[/math] is the graviton field and [math]\kappa \approx \sqrt{G_{N}}[/math] (Newton's gravitational constant).

 

Now given this one can now expand Christoffel symbols, (and hence the curvature tensors etc..) so that we can write the Hilbert-Einstein action as a power series in the graviton field [math] h_{\mu \nu}(x)[/math] . In doing so, each term in the expansion contains two derivatives and an increasing number of [math] h_{\mu \nu}(x)[/math] . Hence the action is nonpolynomial and non-renormalisable (in the standard sence).

 

However, by treating it as an effective field theory you can still calculate the graviton propagator and extract Feynman diagrams. What you should note, is that we have an infinite number of interaction vertices. Usually, people only consider the three and four point interaction, but higher ones are present. So in principle we can calculate gravitational scattering processes to low order.

 

The graviton is a spin 2 boson (gravity is a tensor theory) and is thought to be massless based on astonomical observations.

 

So in words, the graviton is the quanta of a linear pertubation of the classical metric. They only make sence in a linearised approximation to gravity. Are they real will they be detected? If gravity follows similar rules to the other forces, then I would expect gravitons to be detected someday. If not directly, then maybe indirectly as loss of energy in some process. There are people on here who are much more awear of experimental issues, maybe they can help...

[Sayonara]

If we can't ever detect one, is there any point in talking about them?

I was rather under the impression that being unable to detect a particle does not necessarily preclude it from being useful in theoretical physics.

 

Further, if it is impossible to detect one, shouldn't we be saying that it doesn't exist?

Well... no. Impossible to detect is not the same as non-existent.

[YT2095]

actually IIRC, a phenomenon exists that goes a long way to showing that Gravitons Could very well exist.

 

it was something to do with a quartz sphere being spun at high rate and then super cooled to the point of making cooper pairs, the Gravitometers detected a 19 fold increase being exerted on them.

 

it was Originally part of a Gyroscopic mechanism for a satellite under test when it was observed, it`s also been repeated independantly about 30 times at last count in other Labs.

 

I wouldn`t be so eager to Diss the idea Just yet!

[Sisyphus]

Well... no. Impossible to detect is not the same as non-existent.

 

How is it different?

 

YT, I hadn't heard about that. If there's a phenomenon, then my whole post is moot.

[Sayonara]

How is it different?

By not having the virtue of being the same thing.

[Sisyphus]

How is it different in science, I mean.

[Sayonara]

Well let me pu it this way: assume we had never developed the ability to detect... say... X-rays. That would not make them non-existent.

[YT2095]

http://www.scienceforums.net/forum/showpost.php?p=310430&postcount=35

 

was the time I was aware of it (the date stamp on it).

I`ll try dig out some more data if I get time too.

[Sisyphus]

Well let me pu it this way: assume we had never developed the ability to detect... say... X-rays. That would not make them non-existent.

 

"We never developed a way to detect" and "it is impossible to detect" are two very different things, as I see it. My question is analogous to asking whether an electron "has a position" more precise than the threshold of quantum uncertainty.

[Sayonara]

If the graviton exists, then don't we detect its presence through the effects of gravity?

 

You seem to be asking the question "if we cannot interact with something in any way, then for all intents and purposes might we not say that it doesn't exist?", which is certainly a tricky one.

[Sisyphus]

We detect gravity, yes, but not quanta of gravity. As I understood, it becomes too weak to detect - perhaps even theoretically - long before the scale at which it would no longer be divisible.

 

And yes, that is basically the question I'm asking. It's an epistimology vs. ontology question along the lines of the tree falling in the forest, but taken even farther. And I know that physicists certainly did talk about it (Heisenberg said such things don't exist, Einstein said they did, etc.), I'm just not sure if they still do, because I know very little about theoretical physics from the last 60 years....

[Sayonara]

...

Deja vu. I think this has come up before on here.

[timo]

Beyond the fact that everything else seems to be particles/waves, is there any reason to expect that [gravitons] exists?

We have pretty clear evidence that gravity exists, but that´s of course not what you were asking. Since gravity affects mass in large quantities, it seem natural that it should affect mass in smaller quantities to a proportionally lesser extent. This would then of course include scales at which we currently use quantum mechanics to describe the behaviour of particles. I can imagine that all arguments for the gravitational field to be quantized indeed relate back to all the other interaction-carrying fields also being quantized, but that already seems like a good reason to investigate that possibility (relating different effects back to common mechanisms and stuff).

[Severian]

Let me put it another way. Look at the equations which ajb wrote down. There is no quantum mechanics there - it is simply expanding the metric in terms of small perturbations, and that is well defined and reasonable to do anywhere that gravity is weak (i.e. everywhere except right next to a black hole).

 

So the graviton, as a fluctuation of the metric, most definitely does exist.

 

The question is, is this fluctuation quantized? Can we rewrite the field [math]h_{\mu \nu}[/math] in terms of an operator and apply it to quantum mechanics (the process of second quantization)?

 

Since (as atheist implicity said above) we have been able to do this for every other field we have ever come across, it does not seem much of a stretch that we should be able to do it here too. The difficulty is that this field is intimately linked with the space-time symmetry, rather than an internal symmetry, so things may be a little different...

[YT2095]

Bingo, I found the article I read, If I`de have know New Scientist printed their stuff on the net I wouldn`t have bought the Magazines

anyway here it is: http://www.newscientist.com/article/mg19225771.800

[ajb]

I agree with Severian, all the equations I wrote down were classical. In fact, (classical) gravitation waves are usually considered as linear pertubations of a metric.

 

Applying the standard methods of quantum field theory via path integrals to these pertubations we should arrive at the notion of a graviton as a "quantised gravitational wave". However, as I said this does not work as the theory is non-renormalisable.

 

But if we ignore these complications for the mean while (i.e. treat gravity as an effective theory), then assuming that gravity is no different to the other forces then, as Atheist and Severian say we have no reason to question the existence of gravitons.

 

The whole point of my posts was to give people an idea of what gravitons are.

[fredrik]

So the graviton, as a fluctuation of the metric, most definitely does exist.

 

I think part of the logical problem is howto achieve a unique and nonambigous decomposition of the spacetime geometry into a background metric and a fluctuation because you can't measure them separately.

 

The problem is that there is no obvious static background metric. Thus the mathematical perturbation expansion has no full fundamental justification I can see at least. (Computational problems aside)

 

The way i see it, this is somewhat similar to the idea that given an event space, the most natural prior would be that all events are equally likely - this is IMO an illusion. The only real justification I see for a prior is experience, defining the prior. If there is no prior, one need to build/create one from history. This is how I think is the implications of a true relational model should work. I think we need a fundamental review of this.

 

/Fredrik

[fredrik]

Applying the standard methods of quantum field theory via path integrals to these pertubations we should arrive at the notion of a graviton as a "quantised gravitational wave". However, as I said this does not work as the theory is non-renormalisable.

 

I apologize for the sidetrack, I'm aware that this thread was about gravitons not QG in general but to add a personal reflection here to keep he discussion going...

 

I've always had a clear feeling something is incomplete with the path integral formulation. The basic idea is beautiful, but I can't help feeling something is missing. I've taken as a task to find out what it is.

 

I personally suspect the proper formulation should be where the action to be minimized is simply related to a generalized total relative physical information entropy of the states (information divergence). The transition which is most likely is the one with minimal transitional entropy. But from my point of reasoning, this relative entropy is only well define in the differential sense. This is also what requires the integral summing of small steps.

 

I've been working on some ideas on this and I think it should be possible to evaluate the exact transition amplitude, instead of leaving the normalization constant for later. I think this is part of the problem and why it gets screwed up in QG. I thikn leaving the normalization outside of the integral is a mistake because gravity hides in the constant. I'm making some progress and eventually I hope to come up with an expression that I can actually test.

 

IMO, there are many things we take for granted that may or may not need to be revised.

 

/Fredrik

[ajb]

So, back to gravitons...

 

Any sensible questions about the subject that we can discuss or calculate?

[ether]

so it is not sensible to say - gravitons do not exist - but it is sensible to say they might, despite not having any empirical evidence to prove this... ok, well you go and theorise then, but i do not see how you are different from any religious group... is atheism not a belief, founded without fact?

 

 

 

 

 

[YT2095]

I thought as much

 

you`re a Religious Nut-case!

I KNEW there was something I didn`t like about you

[ajb]

so it is not sensible to say - gravitons do not exist - but it is sensible to say they might, despite not having any empirical evidence to prove this... ok, well you go and theorise then, but i do not see how you are different from any religious group... is atheism not a belief, founded without fact?

 

 

 

 

 

 

All I mean is that thsi thread seems to be going off track and becoming a bit off a personal attact on "beleivers" and "non-beleivers". I don't really wish to fight about the existance or not of gravitons, as gravitons at the moment are a pure theoretical objects, lets ask "sensiable" theoretical questions.

 

Things you might like to think about is why gravitons are thought to be massless and why spin-2 and not spin-0 or higher spin?