Talk:Laser

Archives

Old discussions with no (big) contributions after Jan 1, 2006 have been archived to Talk:Laser/Archive1. A couple of additions after January 2006 are also present.

Misconception?

Someone added the following to the "misconceptions" section:

Another common misconception involves the explanation of coherent light production. Many authors claim that laser light is coherent because stimulated emission is in phase with the stimulating wave. However, in-phase emission only produces gain (since superposed in-phase waves produce a wave of higher amplitude) and incoherent light is only amplified without losing its incoherence. In reality, both spatial and temporal coherence is due to the optical cavity. The cavity produces temporal coherence by excluding any light with wavelength which is not contained an integer numer of times in the cavity length. The cavity produces spatial coherence by acting as a spatial filter with gain; by reflecting only light with a spherical wavefront of a particular radius.

I don't think this is completely accurate. Coherence is more than simply cavity resonance. In particular, the linewidth of a laser above threshold is orders of magnitude narrower than the width of a cavity mode of the same cavity.--Srleffler 02:14, 8 February 2006 (UTC)

Right, it's not simply cavity resonance, yet the narrow linewidth still is caused by the cavity. We need to ask ourselves WHY the laser output linewidth is so much narrower than the cavity linewidth. As I understand it, the narrowness arises because the gain of the laser medium, minus the cavity-dependent losses through the partial mirrors, together puts the peak of the system gain vs. frequency right at 1.0. So, as light repeatedly passes through the system, the cavity's frequency curve is repeatedly multiplied by itself. Repeatedly multiplying a mathematical function by itself will produce a delta function, if the original function has a single maximum with a value of one. "The laser's coherence is caused by the cavity" is a compressed explanation. Also, my intent was to point out the common and widespread misconception: the wrong idea that stimulated emission explains laser coherence. Any explanation of the actual cause of coherence should be put elsewhere in the article. --Wjbeaty 04:31, 29 March 2006 (UTC)

The width of a cavity mode is usally smaller than the linewidth of the gain above threshold, but the original statement is very misleading. Let's take an example: in an Argon ion gain profile is about 4-12 GHz wide, while the linewidth of a one meter long cavity has a free spectral width of 150 MHz and thus a much smaller linewidth (1.5 MHz assuming mirrors of reflectivity 96.8%). But the cavity does not reduce the spectral width, it only clears some spectral regions. To reduce the laser linewidth to the linewidth of the cavity (monomode operation), an etalon is necessary, which is an additional very narrow cavity. And secondly, even without cavities the coherence would have been already excellent. Third, if the cavity becomes very narrow (as in VCSEL diode lasers or photonic crystals) the gain width perhaps becomes smaller than the width of the cavity mode. --danh 16:40, 8 February 2006 (UTC)

I guess I should have qualified what I was saying better. If you constrain a laser to operate in a single mode (longitudinal and transverse), it will have a linewidth much narrower than the cavity linewidth. For the example of your argon ion laser, the laser initially operates in multiple longitudinal modes. The role of the etalon is to force the cavity to operate in a single longitudinal mode. If you got the laser working with the etalon in, and then turned down the pump source so that the laser was a little below threshold, the on-axis stimulated emission would have linewidth equal to the cavity mode linewidth, as you would expect. If you then raise the current above threshold, you would find that the laser's linewidth gets orders of magnitude narrower. This is a well-known phenomenon that you can read about in any laser physics text. Look for the equation for the Schawlow-Townes limit. This gives the theoretical linewidth of the laser, above threshold. In practice, real lasers always have much larger linewidth than the S-T equation predicts, because vibrations and thermal fluctuations cause the effective cavity length to fluctuate on very short timescales. This introduces an extra source of broadening to the laser output. The laser will still typically be significantly narrower than the cavity modes, though. --Srleffler 00:24, 9 February 2006 (UTC)

I propose to just keep it out. Mode selection is done in several ways and we don't want to go in the details in the Misconception section. --danh 14:14, 9 February 2006 (UTC)

This misses my point. Many intro textbooks claim that the coherence of laser light arises only because the stimulated emission produces a special type of light called "in phase light." This is wrong, since the in-phase emission process only explains why the laser media is transparent and why it has gain. The light within the cavity is initially incoherent via spontaneous emission, and stimulated emission isn't a recipe for removing this incoherence. To explain the coherence properly, we have to view lasers as acting as amplified narrowband filters, with the light passing repeatedly through the "filter." The partially-mirrored cavity in a conventional laser supplies the "narrowband filter" effect. The Schawlow-Townes limit equation shows why the linewidth isn't zero, but doesn't clarify why the linewidth is so small. --Wjbeaty 04:31, 29 March 2006 (UTC)

I disagree

I take issue with Srleffler over the assertion that "Football field sized Nd:glass lasers have only two significant applications" (namely ICF and weapons design). The applications of giant Nd:glass lasers have expanded widely in the past couple decades. HEDP (high energy density physics) done at these facilities now includes sophisticated astrophysical jet modeling, supernova shockwave physics, liquid metallic hydrogen research, megagauss regime investigations relevant to neutron star environments, shock Hugoniot data for silica to better understand geophysics of conductivity in the earth's mantle and now with the ultra-high energy chirped pulse Nd beams: laser wakefield particle accelerators(!!), fast fusion ignition, visible wavelength photo-fission or photo-transmutation of radionuclei, attosecond regime x-ray spectroscopy. Really it goes on and on. I would even say that some of this research is MORE significant than the fusion stuff. There really should be a HEDP article.--Deglr6328 07:05, 21 February 2006 (UTC)

I'll adjust the caption. I haven't been following this field for a while and forgot about the other applications.--Srleffler 12:48, 21 February 2006 (UTC)

Error

"These lasers have particularly narrow oscillation linewidths of less than 0.01 cm-1 "

Shoudn't this be 0.01 nm?

Sorry if I misplaced this comment, but I'm in a hurry-

No error. Whoever wrote it was expressing the linewidth in terms of wavenumber instead of wavelength. -- The Photon 05:12, 3 March 2006 (UTC)

It's me again. The linewidth can of course be represented in both ways but 0.01 cm-1=1000000000 nm which doesn't make any scense for a linewidth. The laser would not have a narrow linewidth at all. I may be wrong, which is also the reason that I didn't correct it. (Hauberg 13:31, 3 March 2006 (UTC))

Yes, but the computation gives a funny result. He's talking about a center wavelength of, for example, 224 nm, which is 280,499 cm^-1. Similarly, 224.01 nm would be 280,486 cm^-1; so a laser line spanning these wavelengths would have a linewidth of about 12 cm^-1. So 0.01 cm^-1 linewidth is actually about 1200 times narrower than 0.01 nm in this wavelength range. -- The Photon 00:19, 5 March 2006 (UTC)

That makes sense for me. Thanks for the explanation :-) I am sorry for the inconvenience. Feel free to delete this section. Hauberg 17:08, 10 March 2006 (UTC)

Let's not use wavenumbers at all, I think they're confusing sometimes too and they're conventionally used in the IR range. changed to Hz. --Deglr6328 16:38, 11 March 2006 (UTC)

Pulsed laser plasma

Anything you laser experts could contribute to Volumetric display#Laser plasma would be appreciated. The press releases are rather short on details. In Chirped pulse amplification, it mentions "self-focusing". I assume this is something involving a change in the index of refraction from heating up the air, related to blooming? (needs a real article) It also says 700 gigawatts/cm² will ionize the air. Is that just an arbitrary example that happens to ionize air, or is it the actual breakdown threshold of air? (I'm more familiar with breakdown voltage, which happens at ~30 kV/cm, for instance.) If so, could we get a rough guess of the display's output power? (1 nanosecond pulses at 100 Hz.) — Omegatron 06:19, 14 March 2006 (UTC)

That 700 GW example is just an example. I'm not sure of the exact intensity level where the kerr effect becomes important in air. Anyway I doubt the laser used in that 3d thingy uses CPA at all. really all it needs to do is use a q-switched high power laser with a conventional lens and a couple mirrors to control the Z and X Y axes respectively. It's virtually certain that the technique used is essentially identical to that used to make those "laser crystal" things. Focus a 1 GW IR pulse down to a few micron spot (easy) and your talking intensities around the quadrillion W/cm^2 range (if I added right). Everything breaks down to plasma at those intensities! no cpa needed. --Deglr6328 07:30, 14 March 2006 (UTC)

Wow. Doesn't sound that difficult at all, now. This page has people creating the same effect with $200 Nd:YAG lasers off ebay, apparently. (Though without the mirrors or any long distance.)

"I saw a bright pinpoint flash of light at 1.5 inches from the lens in mid air! Very very cool (first time I have seen this phenomenon, though I have heard of it). I guess this gives another data point to the output power level... Air sparks at 200 to 400 mJ? :)"

"I have also sparked the air using a short focal length (about 1.5 cm) lens."

Hmmm... — Omegatron 01:42, 15 March 2006 (UTC)

Metal ion lasers

It states the the metal ion lasers have a prticularly narrow linewidth of 3GHz. Maybe for that UV wavelength region. but not generally is it? Most gas and fiber lasers have kHz linewidths.

Military Uses

Has the use of lasers to burn the enemy troops' retinas at a distance been outlawed? Anon user 12:30, 20 March 2006 (UTC)

Yes. One of the later Hague Conventions, I believe. --Carnildo 05:20, 21 March 2006 (UTC)

Although tank laser rangefinders might accidentally have that effect when simply ranging...Midgley 15:46, 26 March 2006 (UTC)

Why...

"Some types of lasers, such as dye lasers and vibronic solid-state lasers can produce light over a broad range of wavelengths; this property makes them suitable for the generation of extremely short pulses of light, on the order of a femtosecond (10-15 s)." It is not apparent to me (I know little of lasers) why this property makes them ... Midgley 15:45, 26 March 2006 (UTC)

Vibronic lasers are often used in a technique called modelocking to produce ultra short light pulses. A large gain bandwidth is needed in order to lock an high number of longitudinal modes. Lasah 15:55, 26 March 2006 (UTC)

The "why" is the Heisenberg uncertainty principle, which says that the product of our uncertainty in knowing the momentum and the position of a particle is bounded by a constant, e.g. $\Delta x\Delta p>\hbar /2$. A short pulse requires the position be very accurately known, since the duration of the pulse is related to the position of the constituent photons by the speed of light. But the wavelength of the photons is closely related to the momentum, so for the pulse to be very narrow, it must contain a wide range of wavelength components. -- The Photon 06:05, 28 March 2006 (UTC)

Thank you. Makes perfect sense. Is it gettable intot he article without moving from encyclopaedic to physics textbook I wonder? I suppose the best way might be as a footnote to the para at which I went "Why?". Midgley 18:45, 29 March 2006 (UTC)

Hmm, it happens that the Heisenberg uncertainty applies to light as well, but I wouldn't add that as an explanation. It is just a property of the Fourier transformation (from which you can derive the Heisenberg relation): $\Delta t\Delta \omega >1/2$ without any fundamental constants. The minimum for a time-bandwidth product also applies to e.g. sound waves, where Planck's constant doesn't play a role. Han-Kwang 12:49, 31 March 2006 (UTC)

Why are lasers so cool?

Lasers categorized

FT2, I'm not sure what is added to the article by changing from "Types of lasers" to "Lasers categorized" and adding the "by output power" section. Do you plan to add any further means of categorizing lasers, for example "by wavelength", "by average power", "by efficiency", "by pulsed or cw", "by cost", "by weight", ... The article is already near the maximum recommended article length, and its difficult to imagine adding all these different ways of sorting out lasers without going far beyond the recommendation.

The information added seems to belong more in the "Uses of lasers" section. For example, the list of important laser characteristics currently in the history section might be moved in following some phrase like "Laser characteristics that determine the best laser for a particular application include ..." And the CD, DVD, and DVD-R articles might be updated to include the information on the power of the laser involved in those systems. The information on the most powerful laser claim from Livermore could be moved under "Recent innovations" in the history section (with a citation).

--The Photon 03:51, 18 April 2006 (UTC)