Uncovering the Mystery of Photons, Mass, and Gravity: A Guide for Beginners

[Dave0101]

Sorry folks, I couldn't find the "newb" section for this undoubtedly stupid question.

I started reading the wikipedia article on photons, and it states that a photon is massless. My understanding was that all light is photons, but also that light is affected by gravity. It was also my understanding that gravity only affects things with mass.. so I'm confused about how, say, a black hole, would bend light. Could someone lay this bag of snakes out straight for a physics dummy?

Thanks!

 

[MeJennifer]

It was also my understanding that gravity only affects things with mass.. so I'm confused about how, say, a black hole, would bend light.

Both mass and energy influence the curvature space-time.
The path that all (free falling) particles follow in space-time is completely determined by the curvature of space-time.

So note that it is not only mass that influences the curvature of space-time and that all (free falling) particles adhere to this curvature.

 

[rbj]

and there is also a never ending debate here (and in other places, it is not just a Physics Forum noob vs. veteran debate) about whether the term "massless" without qualification is the most appropriate for describing the mass of photons. i always say (and this rankles some folk here) that photons have inertial mass of $h \nu / c$ but, because their velocity is c for every inertial observer, they must have zero rest mass .

 

[anantchowdhary]

Sorry folks, I couldn't find the "newb" section for this undoubtedly stupid question.

I started reading the wikipedia article on photons, and it states that a photon is massless. My understanding was that all light is photons, but also that light is affected by gravity. It was also my understanding that gravity only affects things with mass.. so I'm confused about how, say, a black hole, would bend light. Could someone lay this bag of snakes out straight for a physics dummy?

Thanks!

i think if you go through Newtonian gravity,it also predicts bending of light by masses.But,GR predicts twice the measure

 

[pervect]

Sorry folks, I couldn't find the "newb" section for this undoubtedly stupid question.

I started reading the wikipedia article on photons, and it states that a photon is massless. My understanding was that all light is photons, but also that light is affected by gravity. It was also my understanding that gravity only affects things with mass.. so I'm confused about how, say, a black hole, would bend light. Could someone lay this bag of snakes out straight for a physics dummy?

Thanks!

Just remember that while photons do not have mass, they do have energy and momentum.

The idea that "mass" causes gravity is essentially a carryover from Newtonian theory. In General Relativity, it's better to think of energy (and not mass) as causing gravity. To be really precise, it is the stress-energy tensor that can be thought of as the "source" of gravity in GR. Energy, momentum, and pressure can all contribute terms to the stress-energy tensor, so they can all contribute to gravity.

As we have already seen that photons have energy, the problem is resolved. Energy is one of the things that causes gravity, and photons have it.

It's also worth noting that in GR, gravity isn't really a force, but a curvature of space-time.

 

[pervect]

and there is also a never ending debate here (and in other places, it is not just a Physics Forum noob vs. veteran debate) about whether the term "massless" without qualification is the most appropriate for describing the mass of photons. i always say (and this rankles some folk here) that photons have inertial mass of $h \nu / c$ but, because their velocity is c for every inertial observer, they must have zero rest mass .

What problem do you have with simply explaining that energy causes gravity?

Especially since the "relativistic mass" you are usually fond of is just another name for energy?

I note that you're calling it "inertial mass" now, rather than "relativistic mass". I'm not sure if the change in name has any significance. If you're talking about the relationship between momentum and velocity, though, that would be a second rank tensor, not a single number.

 

[MeJennifer]

One short note about these recurring debates about mass:

It is far more important, IMHO, to distinquish between Lorentz invariant and Lorent variant properties.

Energy, momentum and relativistic or inertial mass are Lorentz variant properties. You would have a hard time arguing, at least with me, that those properties could be properties of a particle. Mentioning those "properties" in isolation is doomed to cause confusion.

But rest mass and energy-momentum, but not energy or momentum in isolation, are Lorentz invariant properties and thus much more suitable descriptions for a particle's properties.

With regards to a tensor being the source of curvature, I understand what is implied here, but tensors are obviously unphysical entities.

 

[rbj]

What problem do you have with simply explaining that energy causes gravity?

i don't have a problem with it. but i can't "simply" explain it, rather i grope around for explanations like: http://math.ucr.edu/home/baez/einstein/einstein.html

nonetheless, there are so many contexts where photons seem to have properties akin to particles with mass of some sort, i'll still toss it back to you guys and ask: what problem do you have with the qualification of "massless" w.r.t. photons that it is rest mass or invariant mass (whatever term works for you) that photons lack?

Especially since the "relativistic mass" you are usually fond of is just another name for energy?

it's not just "relativistic mass". matter is converted to energy in nuclear reactions. maybe someday some Nobel prize will get awarded to some physicist who shows that all matter is just another encapsulated form of energy. maybe not. doesn't matter because there is a conversion factor between the two. a quantity of mass can always be referred to as a quantity of energy (as is commonly done) and, just the same, the other way around.

I note that you're calling it "inertial mass" now, rather than "relativistic mass". I'm not sure if the change in name has any significance. If you're talking about the relationship between momentum and velocity, though, that would be a second rank tensor, not a single number.

that i dunno. I've always thunk that the momentum vector of a particle (relative to some observer) was in the same direction as the velocity vector (relative to the same observer) so the two vectors can be related to each other with a scaler which would be the inertial mass.

 

[Dave0101]

So let's see if I can manage a simpleton's summary...

Matter ~= Energy ~= Matter

Photons aren't what you'd call matter, but they are energy, and so they're affected by large sources of gravity. Is that about right?

Also, trying to wrap my mind around the variant and invariant properties part. Basically you're saying that a photon's momentum or relativistic mass is not a property of the photon as much as it is a state, and as such, can't be formally factored into whether or not it has mass, per se. Except that it doesn't matter, because it has energy, and as such is subject to gravity?

 

[MeJennifer]

Also, trying to wrap my mind around the variant and invariant properties part. Basically you're saying that a photon's momentum or relativistic mass is not a property of the photon as much as it is a state, and as such, can't be formally factored into whether or not it has mass, per se. Except that it doesn't matter, because it has energy, and as such is subject to gravity?

A Lorentz variant property is a property whose value is determined by the (kinematic) relationship between the measured particle and the observer. Due to this, the value of such a property can vary from one observer to another. And thus it follows, that such properties cannot be proper, i.e. they cannot pertain to the intrinsic properties of the measured particle.

Relativistic mass, energy and momentum are all Lorentz variant properties.

A Lorentz invariant property is a property whose value is proper, i.e., it is an intrinsic property of the particle. Each observer will obtain an identical value if measured or calculated.

Rest mass, energy-momentum and proper time are all Lorentz invariant properties.

So, clearly, when we limit ourself to Lorentz invariant properties in measurements we avoid the kinematical relationship between particle and observer and, above all, confusion.

With those notions in mind it should not be hard to understand the fruitlessness of questions like "is energy conserved in relativity?"

 

[fireball3004]

"it’s better to think of energy (and not mass) as causing gravity."
-pervect
I have a question now. If photons give off the theoretical graviton particle (assuming it's real, as we have never found one) then doesn't that imply that photons are continually giving off energy that will become a graviton? And therefore does this not also imply that it must be continually degrading without actually degrading to less and less charged electromagnetic radiation?

Now I'm confused.

 

[pervect]

Photons don't "give off" real gravitons. (Personally, I think you'd be better off avoiding gravitons, BTW - gravitons are a quantum concept, and GR is basically a classical theory).

GR in its pure form explains gravity not as an exchange of virtual particles, but as a warping of space-time.

Consider the analogous question of the force between two charges. Your argument would be "the force between two charges can be regarded as an exchange of photons". But note that charges are NOT continually giving off energy, as your argument would (by analogy) state. The correct statement would be "the force between two charges can be regarded as an exchange of virtual photons". Virtual photons have many non-intuitive aspects, which a lot of people misunderstand by treating virtual photons as if they were real. Virtual particles don't have positive energies (one of the problems with your argument - it doesn't take any energy to create virtual photons), they don't follow definite paths, and have many other strange characteristics.

It's possible to use "virtual photons" correctly to explain forces, but even the job of explaing how a + charge attracts a - charge via an exchange of virtual photons is quite difficult, especially if one wants to get the direction of the force correct (the force points towards the instantaneous position of the charge, not the retarded position).

If you don't want to take my advice and avoid virtual particles altogether, you probably ought to read up on how to use them correctly. See for instance

http://math.ucr.edu/home/baez/physics/Quantum/virtual_particles.html

also couple of related FAQ's which cover some other aspects that people who are fond of virtual particles frequently get wrong are

http://math.ucr.edu/home/baez/physics/Relativity/GR/grav_speed.html (the correct way to measure the speeds of gravity and light)

http://math.ucr.edu/home/baez/physics/Relativity/BlackHoles/black_gravity.html
(how gravity and electromagnetic forces can get out of a black hole)