# Any photon is a spin 1 particle

member 11137
Any photon is a spin 1 "particle"

Once more time an easy question for the specialists. Any photon is a spin 1 "particle". In which manner is this spin physically connected with the magnetic field (rot A = H) associated with a photon? Is there any physical reality in which the spin (if // to H) can also be // (parallel) to the speed of the photon? I think that the answer is no but I cannot really explain why. The un-said question is the following: let us consider that the speed of a photon only is an average one around c (i.e. c + or - delta v). For the case where v < c: no special problem; for the short duration where v should be > c because the photon would have encounter any exterior physical influence (e.g. gravitational field): then we are in a forbidden region of the theory. But let us think that statistical arguments, at least concerning the precision of the measurement of the speed, allow this case. The speed of the photon under consideration cannot increase over c; then it has only one possibility: to deviate. Then it has a new speed direction and the later could be decompose into two components: one // to the initial direction, the other // to the initial plane wave, thus to H. Is it an acceptable representation of the things? Thanks for explainations.

The spin of a photon is always parallel/antiparallel to it's direction of motion. The fact that there are only 2 and not 3 spin degrees of freedom is explained in QFT because one degree of freedom leads to non physical entities like negative probabilities. We have treated this subject on pf many times, so i suggest you perform a general search with the key words spin of photon, polarization

http://www.mathpages.com/rr/s9-04/9-04.htm

marlon

member 11137
marlon said:
OK. I have red these threads and things are a little bit clearer in my head. As I discovered, it was and it is not so easy to explain why and how one of the 3 degrees of freedom is lost. But it's really interesting (the theories involved in). Next question: what is mean in the litterature with the words:" spin fluctuation"? The change of direction of these spins in a set of electrons, molecules, ...?

Blackforest said:
:" spin fluctuation"? The change of direction of these spins in a set of electrons, molecules, ...?

There are different definitions but in general a spin fluctuation is indeed 'a spin that flips from one positon to another'. Thus a fluctuation corresponds to a variation in energy due to the spin flip. By energy i mean the potential energy of the spin in an 'extern' magnetic field

marlon

reilly
In free space. em is governed by Maxwell's Eq,s -- both classically and quantum theoretically (cf. use the Heisenberg eq.of motion. The fields can have both a transverse and longitudinal structure, and the radiation fields -- in empty space and inertial frames -- are transverse, hence perpendicular to the direction of wave propagation.

Also, E&M fields, both transverse and logitudinal, transform like 3-vectors under rotations, hence they carry spin 1. In the momentum rep, the transverse nature can be expressed as, say, E(k)*k =0, where E and k are vectors, with k pointing in the direction of propagation. Classically, we describe radiation fields in terms of two polarization vectors, say like in the x and y direction if k is in the z direction. Quantum mechanically, we do the same thing, and call it spin.

(But, a thorough treatment of the vector and rotational aspects of basic QED is very technical, tedious, and tricky involving, as it does, vector spherical harmonics, and other joys of angular momentum theory. Edmund's book on angular momentum, and an old chestnut, Blatt and Weiskopf's Nuclear Physics do a great job. more modern treatments involve the helicity formalism of Jacob and Wick. And, there's always Google.)

If you send light into a material medium, then the actual speed of light, v, obeys v<c, and it's a brand new ballgame. For example, material-filled waveguides can have dispersive modes resulting in a loss of power, and, longitudinal waves in addition to transverse ones. (Jackson's E&M book does a good job on wave guides.) Note that you can have periodic structure in longitudinal fields, in empy space, but as their field strength goes as 1/r**2, these 'waves' do not contribute to the radiation field at appropriately large distances.
Regards,
Reilly Atkinson

Last edited:
member 11137
reilly said:
In free space. em is governed by Maxwell's Eq,s -- both classically and quantum theoretically (cf. use the Heisenberg eq.of motion. The fields can have both a transverse and longitudinal structure, and the radiation fields -- in empty space and inertial frames -- are transverse, hence perpendicular to the direction of wave propagation.

Also, E&M fields, both transverse and logitudinal, transform like 3-vectors under rotations, hence they carry spin 1. In the momentum rep, the transverse nature can be expressed as, say, E(k)*k =0, where E and k are vectors, with k pointing in the direction of propagation. Classically, we describe radiation fields in terms of two polarization vectors, say like in the x and y direction if k is in the z direction. Quantum mechanically, we do the same thing, and call it spin.

(But, a thorough treatment of the vector and rotational aspects of basic QED is very technical, tedious, and tricky involving, as it does, vector spherical harmonics, and other joys of angular momentum theory. Edmund's book on angular momentum, and an old chestnut, Blatt and Weiskopf's Nuclear Physics do a great job. more modern treatments involve the helicity formalism of Jacob and Wick. And, there's always Google.)

If you send light into a material medium, then the actual speed of light, v, obeys v<c, and it's a brand new ballgame. For example, material-filled waveguides can have dispersive modes resulting in a loss of power, and, longitudinal waves in addition to transverse ones. (Jackson's E&M book does a good job on wave guides.) Note that you can have periodic structure in longitudinal fields, in empy space, but as their field strength goes as 1/r**2, these 'waves' do not contribute to the radiation field at appropriately large distances.
Regards,
Reilly Atkinson
Thanks.

As I now understand, to my initial question:"Is there any physical reality in which the spin (if // to H) can also be // (parallel) to the speed of the photon? " the answer is NO because E is in the x direction, H (magnetic field) in the y direction and v (speed) in the z direction; as said in free, empty space refered to inertial frame; and the polarization's vector as the spin are in the plane (x,O,y) where O is the position of the photon.

If I come back to the small scenario I have made concerning a deviation of a photon in free space under the influence of gravitation: this seems to be a realistic situation where the acceleration due to the sun (for exemple) could be in the (x,O,y) plane, thus if one makes the choice of a convenient frame where H could be // to a lateral acceleration (pointing to the sun) and consequently to a lateral component of the speed of the photon... ? But the presence of gravitation complicates everything and we are no more in a totally free space... Thus the initial triedre is perhaps no more perfect (orthogonal). The absolutely necessary condition is that |v|² = c² Correct?

member 11137
Blackforest said:
... and the polarization's vector as the spin are in the plane (x,O,y) where O is the position of the photon.
Not correct for the spin which is the angular momentum of the photon and which is supposed to be in the direction or in the opposite direction of the propagation.

An other question, always concerning the spin (which is the item here): is there any good theory (book, chapter) explaining the contribution of a spin-spin interaction to the total stress energy-momentum of an atom?