What is the difference between diffraction and refraction in waves?

[dervast]

Hi can u please explain me the difference between refraction and diffraction?

 

[mrjeffy321]

In refraction, the wave is being bent because the speed of propigation in the medium it is moving into is different than the medium it is moving out of. For instance, light moving from air to water at an angle is bent away from the axis that runs perpendicularly to the surface because light travels slower in water than in air.

In diffraction, the wave is being bent (often times through very small slits) when it hits somthing, causing a diffraction pattern to form.

Diffraction: http://en.wikipedia.org/wiki/Diffraction
Refraction: http://en.wikipedia.org/wiki/Refraction

 

[spacetime]

Refraction is the "bending" of a ray when it meets an interface between media of two different optical densities (two different speeds in the media).

Diffraction is the "spreading" of a wave, which may not be because of a change in the medium, it can occur even while the wave propagates in the same medium. It can be observed when one puts an aperture nearly of the same dimensions as the wavelength of the wave in it's path.

Another interesting fact is that though refraction can be explained if you take into account only the particle picture of light, diffraction cannot be explained.

spacetime
Physics

 

[ZapperZ]

Another interesting fact is that though refraction can be explained if you take into account only the particle picture of light, diffraction cannot be explained.

spacetime
Physics

This is not true.

Diffraction is a DIRECT manifestation of the Uncertainty principle based on the determination of a photon's transverse position by the slit. Furthermore, the Marcella paper that I've mentioned several times clearly derived purely using quantum mechanics of the diffraction pattern and the interference pattern from multiple slits.[1]

Zz.

[1] T.V. Marcella, Eur. J. Phys., v.23, p.615 (2002).

 

[vanesch]

This is not true.

Probably the poster meant the NEWTONIAN particle picture of light...

 

[ZapperZ]

Probably the poster meant the NEWTONIAN particle picture of light...

IF that is true (I doubt it), then someone is severely out-of-date.

Zz.

 

[lightgrav]

I'll bet most readers follow the context and read spacetime's comment as:
"... diffraction cannot be explained unless one includes wave aspects."
Quantum theory includes wave aspects.

Where spacetime's comment is misleading is in claiming that refraction "can be explained" using only the particle picture of light. I suppose he's referring to some sort of path-extremum principle (like Fermat's), but such a description is not an explanation unless you can come up with a plausible mechanism for a "particle" to obtain information about the variety of paths available, and to restrict its motion to be along such a path.
Huygens' construction (generallized by Feynman) uses wave properties to provide the mechanism.

 

[spacetime]

This is not true.

Diffraction is a DIRECT manifestation of the Uncertainty principle based on the determination of a photon's transverse position by the slit. Furthermore, the Marcella paper that I've mentioned several times clearly derived purely using quantum mechanics of the diffraction pattern and the interference pattern from multiple slits.[1]

Zz.

[1] T.V. Marcella, Eur. J. Phys., v.23, p.615 (2002).

Yes, I was wrong in what I said. Diffraction can indeed be explained using the photon picture, that is the particle nature of light. Actually, I was referring to the historical fact that diffracion could not be explained using the corpuscle picture, which could still have explained refraction by proper assumptions.

Thanks Zz.

 

[RAJOD]

Good point, I think he meant that the "particle" theory of light was not enough to describe what occurs with diffraction. Also the uncertainty principle is one of those "Its just that way it is" there is no explanation as the the "why" of the uncertainty principle. We really do not know what is going on, only that it seems to be consistent.

Yes, I was wrong in what I said. Diffraction can indeed be explained using the photon picture, that is the particle nature of light. Actually, I was referring to the historical fact that diffracion could not be explained using the corpuscle picture, which could still have explained refraction by proper assumptions.

Thanks Zz.

 

[Neo_Anderson]

IF that is true (I doubt it), then someone is severely out-of-date.

Zz.

Especially since Young showed a wave model of light like over 100 years after Sir Isaac Newton was around...

 

[Neo_Anderson]

Yes, I was wrong in what I said. Diffraction can indeed be explained using the photon picture, that is the particle nature of light. Actually, I was referring to the historical fact that diffracion could not be explained using the corpuscle picture, which could still have explained refraction by proper assumptions.

Thanks Zz.

But remember that diffraction is observed even after we send a single photon through the slit...

 

[Bob S]

Bragg diffraction (nλ=2d sin(θ)) on atomic crystal planes of thin crystals has been used for nearly a century for precision wavelength measurement of x-ray and nuclear gammas. There are two basic methods, transmission (meaning "refraction"** through the crystal) and reflection off the front surface. The two methods have similar resolution (≈1 second of arc). The reflection method is usually chosen at low x-ray energies when the attenuation coefficient of the x-ray through the crystal (transmission method) is large. Both methods have been used with very low photon counting rates.
Bob S

[added] ** Actually, diffraction of the incident beam in the transmission mode removes photons from the "refracted" beam. Both refracted and diffracted photons exit the rear of the crystal, but only the diffracted photons satisfy the relation nλ=2d sin(θ).

 

[armanzBSE]

thanks for the argument. i learned. :D

 

[sophiecentaur]

If the term 'diffraction' is taken to mean the way the propagation of a wave is affected by some sort of obstruction to its path then how is refraction anything other than subset of diffraction - just like interference is also a subset?
The results of all three can be calculated by integrating the contributions of the full width of incident wavefront to find out what happens beyond the structure of interest. (The Huygens idea)
The above is an essentially classical approach but I think it applies even if you want to include Quantum interactions.

 

[armanzBSE]

If the term 'diffraction' is taken to mean the way the propagation of a wave is affected by some sort of obstruction to its path then how is refraction anything other than subset of diffraction - just like interference is also a subset?
The results of all three can be calculated by integrating the contributions of the full width of incident wavefront to find out what happens beyond the structure of interest. (The Huygens idea)
The above is an essentially classical approach but I think it applies even if you want to include Quantum interactions.

oh yes, Huygens idea. thanks for this!

 

[sophiecentaur]

Mr H. rules if you want to understand anything to do with waves. I think it's the basis of lots of numerical wave modelling routines. It goes along with Fermat's Principle.