Can we produce a white laser from a visible light spectrum?

[athrun200]

I am doing a final year project, my Prof gave me 2 prism.
He wants me to do so:
A white laser is shot on a prism producing a visible light spectrum, use another prism to converge the spectrum back into a laser.

For me, it is theoretically impossible and I tried to convince him that it doesn't work as prism can only make the light more disperse. But he insisted that by principle of reversibility, there should be a way to convert the spectrum back to a laser.

Is he really correct? I want to disprove his statement but I cannot figure out a concrete proof.

I don't know what to do now.

 

[maajdl]

Newton did it.

http://www.thestargarden.co.uk/NewtonAndLight.html

 

[Drakkith]

You can easily produce white light by combining different colors via a prism. You can't make a laser out of it, as a laser uses stimulated emission to produce a coherent beam of light that consists of a small range of wavelengths, but you definitely make plain white light.

 

[athrun200]

You can easily produce white light by combining different colors via a prism. You can't make a laser out of it, as a laser uses stimulated emission to produce a coherent beam of light that consists of a small range of wavelengths, but you definitely make plain white light.

That means even if the original light source is a laser, it will become a plain white light after 2 prisms?

 

[UltrafastPED]

Do you have a white light laser? I think that they all use super-continuum white light generation techniques, like this one: http://www.nktphotonics.com/superk_compact

So let's assume you have a laser with significant bandwidth; the ultrafast lasers I work with have bandwidth of 20-50 nm. For a "white light" laser you would have about 350 nm of bandwidth: 400 nm (deep blue) to 750 nm (deep red). For any of these broad-bandwidth lasers you will also have temporally short pulses: typically less than one picosecond, down to 10 femtoseconds. This is the ultrafast pulsed laser temporal range.

Now send the pulse through a prism: the colors separate (disperse) due to the differing speeds of light through the glass. They will also exit the prisms with different angles, so your pulse will be spread spatially, and temporally: the colors are fanned out spatially, but the blues will have exited first, and the reds last-hence they are also fanned out temporally.

The usual way to use prisms is in pairs: you adjust both members of the pair so that the beams enters/exits at the "angle of minimum deviation". You can actually use your eyes to determine this physical alignment, and after the beam exits the second prism it will be "all travelling" in the same direction again: the dispersion is cancelled.

However, your temporal dispersion is still there.

The fix for this is to be found in the design of the ultrafast CPA laser: CPA=Chirped Pulse Amplification. My thesis advisor was one of the inventors of this method in the late 1980s; see http://www.cap.ca/fr/node/709

Here is a nice picture with explanations: http://en.wikipedia.org/wiki/Prism_compressor

But it is easier to work with gratings; then you don't get non-linear optical interactions within the prisms!
See http://en.wikipedia.org/wiki/Chirped_pulse_amplification

 

[athrun200]

Do you have a white light laser? I think that they all use super-continuum white light generation techniques, like this one: http://www.nktphotonics.com/superk_compact

So let's assume you have a laser with significant bandwidth; the ultrafast lasers I work with have bandwidth of 20-50 nm. For a "white light" laser you would have about 350 nm of bandwidth: 400 nm (deep blue) to 750 nm (deep red). For any of these broad-bandwidth lasers you will also have temporally short pulses: typically less than one picosecond, down to 10 femtoseconds. This is the ultrafast pulsed laser temporal range.

Now send the pulse through a prism: the colors separate (disperse) due to the differing speeds of light through the glass. They will also exit the prisms with different angles, so your pulse will be spread spatially, and temporally: the colors are fanned out spatially, but the blues will have exited first, and the reds last-hence they are also fanned out temporally.

The usual way to use prisms is in pairs: you adjust both members of the pair so that the beams enters/exits at the "angle of minimum deviation". You can actually use your eyes to determine this physical alignment, and after the beam exits the second prism it will be "all travelling" in the same direction again: the dispersion is cancelled.

However, your temporal dispersion is still there.

The fix for this is to be found in the design of the ultrafast CPA laser: CPA=Chirped Pulse Amplification. My thesis advisor was one of the inventors of this method in the late 1980s; see http://www.cap.ca/fr/node/709

Here is a nice picture with explanations: http://en.wikipedia.org/wiki/Prism_compressor

But it is easier to work with gratings; then you don't get non-linear optical interactions within the prisms!
See http://en.wikipedia.org/wiki/Chirped_pulse_amplification

Yes, I have a white light laser. Thanks so much, it's so informative.
In fact, my Prof restricts me to use 2 prisms only, no grating is allowed.

I will give it a try tomorrow in the lab!