Gravitons

[martu]

I need some help trying to understand why the graviton is postulated to exist in physics? I thought Special Relativity 'explains' gravity, that is Gravity is due to the shape and size of the universe?

I have tried Wikipedia and Google but can't find exactly what I am looking for, I have read enough to know that the graviton is theoretical and that's it.

 

[shadron]

Basically, all the forces in the universe (all three of the others, that is) use a particle at the quantum level to "mediate" the force; that is, to communicate it from point to point. The electromagnetic force, for example, uses photons; the weak force, "intermediate vector bosons" (W+, W-. Z and hypothetical X) and the strong force uses gluons. Gravity needs analogous gravitons, but gravity is so innately weak that gravitons haven't yet been detected. Newton held that these forces acted at a distance (that is, they projected a field that influenced the target object at a distance). This concept has been replaced in quantum mechanics by these mediation particles.

Special Relativity and quantum mechanics don't agree at the very small quantum level, and the definition of gravity as a coordinate distortion doesn't work there.

See http://people.cornell.edu/pages/jag8/sect4.html

 

[martu]

Thanks that link looks very interesting. So in simple terms we need the graviton to explain gravity at the quantum level.

 

[Ziggurat]

I thought Special Relativity 'explains' gravity, that is Gravity is due to the shape and size of the universe?

That's general relativity, not special relativity - special relativity ignores gravity. And it's reconciling GR (not SR) with quantum mechanics which we haven't been able to do yet.

 

[Ziggurat]

Thanks that link looks very interesting. So in simple terms we need the graviton to explain gravity at the quantum level.

More precisely, many physicists expect we need it. I think they're likely correct, but at this point we can't prove that they exist, so maybe they don't.

 

[sol invictus]

I need some help trying to understand why the graviton is postulated to exist in physics? I thought Special Relativity 'explains' gravity, that is Gravity is due to the shape and size of the universe?

In general relativity the force of gravity arises because the space is curved by the presence of matter and energy. Think of it as a rubber sheet stretched by heavy weights. Now take one of those weights and jiggle it a little - as you can imagine, ripples will propagate out across the sheet.

Those ripples are called gravity waves, and they've already been detected indirectly, and will probably be seen directly in the next few years. In every other type of physics, waves of that sort are quantized - that is, a wave of a given frequency carries an integer number of quanta of energy (for an elecromagnetic wave, those quanta are called photons). Gravitons are the putative quanta of the gravitational field.

 

[martu]

That's general relativity, not special relativity - special relativity ignores gravity. And it's reconciling GR (not SR) with quantum mechanics which we haven't been able to do yet.

Thanks.

 

[martu]

In general relativity the force of gravity arises because the space is curved by the presence of matter and energy. Think of it as a rubber sheet stretched by heavy weights. Now take one of those weights and jiggle it a little - as you can imagine, ripples will propagate out across the sheet.

Those ripples are called gravity waves, and they've already been detected indirectly, and will probably be seen directly in the next few years. In every other type of physics, waves of that sort are quantized - that is, a wave of a given frequency carries an integer number of quanta of energy (for an elecromagnetic wave, those quanta are called photons). Gravitons are the putative quanta of the gravitational field.

As I found out in another thread recently Gravity is different and unique in that like charges attract so how confident are we in this assumption? It seems a logical assumption that gravity is different.

 

[Perpetual Student]

In general relativity the force of gravity arises because the space is curved by the presence of matter and energy. Think of it as a rubber sheet stretched by heavy weights. Now take one of those weights and jiggle it a little - as you can imagine, ripples will propagate out across the sheet.

Those ripples are called gravity waves, and they've already been detected indirectly, and will probably be seen directly in the next few years. In every other type of physics, waves of that sort are quantized - that is, a wave of a given frequency carries an integer number of quanta of energy (for an elecromagnetic wave, those quanta are called photons). Gravitons are the putative quanta of the gravitational field.

Does that imply that the graviton exists when there is a relative motion that creates waves, but that if two objects are at relative rest no gravitons are present or created? Or is it an acceleration that creates the waves and the gravitons, but a uniformly moving object does not?

 

[martu]

Does that imply that the graviton exists when there is a relative motion that creates waves, but that if two objects are at relative rest no gravitons are present or created? Or is it an acceleration that creates the waves and the gravitons, but a uniformly moving object does not?

Good question, I think your thoughts are simialr to mine as this is one of the thoughts that led me to ask this question. I look forward to being educated!

 

[sol invictus]

As I found out in another thread recently Gravity is different and unique in that like charges attract so how confident are we in this assumption? It seems a logical assumption that gravity is different.

Well, the standard model of particle physics ceases to make sense when you couple it to classical (i.e. non-quantum) gravity. So something must modify gravity, and based on everything else we know about the world, it's almost certain to be quantum effects. However precisely what form those take is very uncertain.

Does that imply that the graviton exists when there is a relative motion that creates waves, but that if two objects are at relative rest no gravitons are present or created? Or is it an acceleration that creates the waves and the gravitons, but a uniformly moving object does not?

You can ask the same questions about the electromagnetic field, where I can answer with certainty. There, static configurations create static electric and magnetic fields. But any static field can be decomposed into waves using Fourier transforms - a static field is the superposition of a stream of waves going in opposite directions. In quantum theories, that means static classical fields are composed of photons - but enormous numbers of them.

 

[Ziggurat]

In quantum theories, that means static classical fields are composed of photons - but enormous numbers of them.

To add to this, although talking about the number of photons is a useful handwaving argument, one shouldn't try to take it too literally: in some sense it's an infinite number of photons (since you get contributions from a continuous spectra, which means an infinite number of modes), but of course, the energy is not infinite. Edit to add: the Fourier decomposition is not handwaving (it's mathematically rigorous), but it's not quite counting photons either.

 

[martu]

Edit to add: the Fourier decomposition is not handwaving (it's mathematically rigorous), but it's not quite counting photons either.

Why not?

 

[Ziggurat]

Why not?

First, a bit of math.

If you've got a (well-behaved) function bounded over the interval [0,pi], you can break it down as a sum of functions of the form sin(nx) and cos(nx), where n is an integer. It makes sense to say that your function then has a certain contribution from, say, sin(5x). And that's just the prefactor in our sum, which we get by doing a discrete Fourier transform. There may be an infinite number of terms in your series, but it's a countable infinity.

But if you've got an unbounded function, then you cannot (in general) express it as a sum of sine and cosine terms, because you've got an uncountable infinity of possible sine and cosine terms. So instead of a discrete sum, you have to express it as an integral over a continuous distribution (which we get from the continuous Fourier transform). The Fourier transform doesn't give you prefactors in a sum, but something more akin to a probability density for the integral. An integral has a lot of similarities to a sum, but it's not the same thing. In particular, if you pick a particular wavelength and ask, what would the prefactor need to be if I were to try to build this function as a sum, you'd typically get an answer of zero, even if the Fourier transform has a nonzero value for that wavelength. This is analogous to having the probability density for a continuous variable be nonzero, but the probability to find that value within a certain range shrink to zero as that range shrinks to a single point.

 

[martu]

Thanks

 

[sol invictus]

To add to this, although talking about the number of photons is a useful handwaving argument, one shouldn't try to take it too literally: in some sense it's an infinite number of photons (since you get contributions from a continuous spectra, which means an infinite number of modes), but of course, the energy is not infinite. Edit to add: the Fourier decomposition is not handwaving (it's mathematically rigorous), but it's not quite counting photons either.

That's true, but divergences of that sort are pretty easy to handle. For example, just put the whole configuration in a huge conducting box and you're done. Or just remember that the apparatus you'll use to count the number of photons will not be infinitely big.

There's a related fact here, though: the photon number operator doesn't have a definite value. In the non-relativistic limit, the number of particles is a well-defined quantity. But relativity lets you create and destroy particles, quantum mechanics guarantees that whatever can happen will, and there's no non-relativistic limit for photons, so... every time you measure the number you're likely to get a different answer, even if nothing has changed in the field configuration.

 

[Ziggurat]

There's a related fact here, though: the photon number operator doesn't have a definite value.

Indeed. If you've got the quantum mechanics background to understand that statement, then that statement is all you need to understand what I meant about not quite "counting" photons, and one can save the trouble of bothering with my posts at all. But since many here don't have any formal exposure to the mechanics of quantum mechanics, I thought it useful to try to provide some more details with less prerequisites, even though it's longer.

 

[martu]

What is the SR equation that breaks at the quantum level? Or if this is a silly question I'll learn either way.

 

[sol invictus]

What is the SR equation that breaks at the quantum level? Or if this is a silly question I'll learn either way.

Presumably you mean GR? If so, it's Einstein's equations - the source term on the right hand side is infinite once one takes quantum effects into account. That's possible to deal with, but a more severe problem is that the equation itself will be strongly modified at sufficiently high densities or short distances... so if you apply it to the region near a black hole singularity or to a time near the big bang, it will give wrong answers.

 

[Delvo]

As I found out in another thread recently Gravity is different and unique in that like charges attract so how confident are we in this assumption? It seems a logical assumption that gravity is different.

It's not that gravity has like or unlike charges attracting or repelling; it's that it doesn't even appear to have two separate charges at all. But that doesn't really matter. Another of the fundamental forces (which keeps quarks together) has three kinds of "charge" instead of just two, metaphorically referred to as colors, with yet another completely unrelated set of rules for when they attract or repel. But that doesn't affect that force's dual descriptions as both a particle and a wave/field. The basic idea is that there should be both a wave/field and a particle for all of them, without their numbers or attraction/repulsion of "charges" having anything to do with that duality. It's like comparing the mechanical similarities and differences between two cars, and getting a question thrown in there about their seats' upholstery.