There exists a concept, by which a single photon is visualized as having an electrostatic dipole-moment, which does not lie in a plane, but which performs a corkscrew, either left-handedly, or right-handedly, to start the phenomenon of electromagnetic radiation as based on circularly-polarized light, as opposed to being based primarily on plane-polarization. A quantity of photons could then still form plane-polarized light, not because they interact with each other, but because they coincide with each other in such a way, that their electrostatic fields cancel along one axis, but reinforce perpendicularly to the axis along which they cancel.
In reality, it’s dangerous to make such statements, about what exactly one photon does, because nobody has ever ‘seen’ a photon. We’re mainly able to make more-coarse measurements of what light does, when composed of swarms of photons, and must then deduce what the properties of one photon could be.
(Edit 02/20/2018 :
According to This Experiment, this hypothesis is disproved.
In the macroscopic world, circularly polarized light seems to exist, just as plane-polarized light does, without shedding much light on the subject of how one photon behaves, unless the latter subject is studied in much greater depth. )
But there is the matter of how any of this agrees with the classical, electrodynamic explanation of ‘light’, which would say that it has a magnetic dipole-moment, that oscillates with the same set of frequencies, with which the electrostatic dipole-moment, oscillates, but perpendicularly to the electrostatic moment.
The question could be asked of, If the electrostatic moment was plotted against time, What its phase-position would be, relative to the magnetic moment. And what I claim to know, is that they’d be in-phase.
This subject has been confused at times, with the question of whether the electrostatic component along one plane of polarization, is in-phase or out-of-phase, with the electrostatic component, along the perpendicular plane of polarization. Those are out-of-phase, in the case of circularly-polarized light, as well as in the case of circularly-polarized photons.
(Edit 02/07/2018 : )
Now, the question about plotting this could get sidetracked, by the question of whether it’s more correct, if where the electrostatic dipole moment, which I’ll say is denoted by the Green line above, is pointing ‘upwards’, the magnetic dipole moment, which I’ll say is denoted by the Red line above, should be pointing ‘towards the viewer’, or ‘away from the viewer’. The way I presently have it, at the left end of the plot, the red line is towards the viewer at that instant. Because magnetic dipole moments differentiate between North and South, while electrostatic dipole moments differentiate between Positive and Negative, these signs of polarity are independent. By convention, the magnetic North pole is denoted by positive numbers. If it was assumed that the Red line corresponds to North, as shown above, then the photon would need to be traveling from the left, to the right, which also corresponds to an increasing parameter (t), just in case anybody is interested in actually analyzing the Math I entered.
(Edit 02/26/2018 :
This also implies that t does not correspond to Time. t us just a variable-name which often denotes a generic parameter. More-positive values of t will arrive on the right, before more-negative values do. )
In that case it should also be noted, that ‘Wolfram Mathematica’ switches the (Y) and (Z) plotting axes, so that (Z) actually faces upwards, but needs to be given as the 3rd input of a parametric 3D plot, while (Y) faces away from the viewer, which is different from how some other 3D plots work. The way I tend to visualize World Coordinates these days, (Y) should be facing Up, and (Z) should be facing Towards the Viewer.
(Updated 02/08/2018 … )
This type of 3D plot represents a real-world phenomenon. This means, that it can be rotated 180⁰ round one axis, and still represent the same phenomenon, but as if viewed from a different direction. When we rotate a 3D model around one axis this way, we are actually reversing it along two axes – I.e., by rotating such a model 180⁰ around (Y), we are reversing it simultaneously along (X) and (Z), thereby reversing it left / right as well as towards-viewer / away-from-viewer.
If we were to take such a model and reverse it linearly along only one axis, in fact we’d be creating a new object.
But what this means is that If the ‘light’ was traveling from right to left instead, then the North magnetic dipole moments should be facing away from the viewer, when the electrostatic dipole moments are facing up.
But then again, I did write at length on this subject, In this posting.
About the WiKiPedia treatment of this subject:
(Edit 02/20/2018 :
According to the experiment which I carried out and linked to a report of above, the WiKi article seems accurate, and does not need to be challenged by me.
— Much Text Removed — )