Can Light Be Polarized Despite Its Three-Dimensional Wave Nature?

[minio]

I know those nice pictures showing light polarization, but the light wave should be three dimensional, so no up and down just more and less, like expanding spheres with different desities, right? So it would be more like soud wave, but then the I should be able to detect oscilations but should not be able to polarize it. What am I missing?

 

[DaveC426913]

Look here:
http://en.wikipedia.org/wiki/Electromagnetic_radiation

About a quarter of the way down the page in the right there are diagrams of how electromagnetic waves propagate.

 

[minio]

I can understand light polarization if light wave is represented beam like as on that wiki page. But how this could go together with diffraction? I can understand diffraction if the light waves are like spheres or planes (as on the blue pictures on this page http://en.wikipedia.org/wiki/Diffraction). But then where would the vector describing electric field would be pointing? It would be plane so only outside of plane. But then no polarization.

 

[netheril96]

I can understand light polarization if light wave is represented beam like as on that wiki page. But how this could go together with diffraction? I can understand diffraction if the light waves are like spheres or planes (as on the blue pictures on this page http://en.wikipedia.org/wiki/Diffraction). But then where would the vector describing electric field would be pointing? It would be plane so only outside of plane. But then no polarization.

Why does light waves have to be like spheres or planes in order to diffract? I don't understand your question.

 

[Philip Wood]

The direction of travel is at right angles to the wavefront at each point. Consider, then, the normal to the wavefront at any point, P. If the wave is (plane-)polarised then the electric field at P will be confined to one of the directions at right angles to this normal, that is to one of the directions in the plane tangential to the wavefront.

 

[minio]

So then the classical picture of electromagnetic radiation is just depiction of vector of electric/magnetic field along line going through wavefront, right?

But one more question in case of wavefront (lets say its plane) are vectors of electric field equal at all points of the wavefront at cetrain time or could they differ?

 

[Philip Wood]

Not quite sure I understand what you're getting at in your first paragraph.

Second paragraph. E-field vectors in the same wavefront have a common phase (by definition of a wavefront. Amplitude could vary, e.g. it would fall off towards the edge of a diffracted wavefront.

Directions of E field might or might not vary, depending on direction of polarisation. Consider cylindrical wavefronts diffracted from a narrow slit. If the direction of polarisation of the incident waves were parallel to the slit, then the direction of the polarised waves would be, as well, at all points on the curved diffracted wavefront. If the direction of polarisation were transverse to the slit, then those in the diffracted wavefront would be perpendicular to the normal at each point, and tangential to the circular section through the cylindrical wavefront.

 

[chrisbaird]

Light is an electromagnetic wave, so its essence is electric and magnetic fields throughout space oscillating. The electric field is a vector field, and as such must point somewhere. A vector has direction and magnitude. If there were no direction, the vector would be zero. The polarization of light just means the direction that the electric field vector is pointing.

Polarized em waves do not have to be plane waves or transverse waves. For instance, radio waves given off by a linear dipole antenna have a vertical polarization, but travel outwards spherically and have a non-constant field strength distribution on the wave front known as the antenna pattern. As another example, the laser light inside a quantum cascade laser are polarized normal to the semiconductor layers, but are not transverse. There is also a field component in the direction of propagation.

The electric field has x, y, and z components, but also x, y, z, and t dependencies, and these are all independent entities. As a result, you can get fairly complicated waves.

 

[minio]

Directions of E field might or might not vary, depending on direction of polarisation. Consider cylindrical wavefronts diffracted from a narrow slit. If the direction of polarisation of the incident waves were parallel to the slit, then the direction of the polarised waves would be, as well, at all points on the curved diffracted wavefront. If the direction of polarisation were transverse to the slit, then those in the diffracted wavefront would be perpendicular to the normal at each point, and tangential to the circular section through the cylindrical wavefront.

I think I am staring understand this. It does make sense. But if I am looking at planar wavefront frozen in time - all E field vectors would have same direction (magnitutde may vary) regardless if this is polarized light or not. I would be able to tell the difference only if I follow the changes in vectors direction over time. Am I right?

 

[Philip Wood]

Minio. I have to bow out at this stage. Not quite sure that we can even apply the notion of wavefronts to unpolarised waves. The issue seems to me to be bound up with that of coherence. Let wiser heads prevail.

 

[minio]

Ok. Thank you very much. You helped me to clear things anyway.

 

[chrisbaird]

I think I see your problem. There is no such thing as "unpolarized light" in the literal sense. Polarization is just where the electric field vector points and it always got to point somewhere. A better name for unpolarized light is randomly, non-coherently polarized light. So if you took a snapshot of unpolarized light, say the light coming off an incandescent bulb, the electric field vectors would be pointing in different directions.

 

[Drakkith]

The electric field is a vector field, and as such must point somewhere. A vector has direction and magnitude. If there were no direction, the vector would be zero. The polarization of light just means the direction that the electric field vector is pointing.

Hmmm. This confuses me. Probably because I don't understand vectors and EM theory well enough. How complicated is the math?