# Can you combine laser beams?

1. Feb 1, 2005

### parlous

Suppose I have an arbitrary amount of diode lasers, all of identical construction and abilities. Assume I can determine/control the polarization of each diode beam. Does anyone know how I can increase the amount of photons in a single laser stream via combining all the diode beams without cancellations or energy losses? Is there an optical device/arrangement that acheives this? Also, assume the polarization of the result beam doesn't matter.

Thanks,
Parlous

2. Feb 1, 2005

### Integral

Staff Emeritus
Yes, multiple beams can be combined. One way would be to use beam splitters. A beam splitter is simply a partially reflecting mirror, if you were orient the mirror so a laser reflected an individual beam 45 deg into the main beam path, you could combine the beam of different lasers since the main beam would then be passing through the mirrors in the transmission direction. How good of a beam would you get? The trouble with this arrangement is that you would be very likely to have significant amounts of interference between the beams unless you had some way to adjust the optical path length of each laser so they entered the main beam with the correct phase relationship. How do you do this? Could be very hard.

One thought that might work, depending on your application. By timing pulsed lasers the individual lasers pulses could be combined into a single beam by interlacing the pulse trains. The resultant beam would be at twice the rep rate of the individual lasers. This is something I just dreamed up sitting here, so if you use it in a commercial application I will expect royalties.

3. Feb 1, 2005

### krab

You can combine them, but you will not end up with a single beam of the same dimensions as one of the original beams. This is because the phase space volume occupied by beams is an invariant.

4. Feb 1, 2005

### Integral

Staff Emeritus
Krab,
Could you go into more detail on your response. I am not sure that I have my head wrapped around the phase space volume of a laser beam!

In the next few months I may learn much more about this. It is rumored that we will be getting 'stacked" laser heads, essentially 2 heads working together to form a single beam. This laser will be made by one of the well known manufactures of lasers.

5. Feb 1, 2005

### Moonbear

Staff Emeritus
Integral, are you suggesting something that can really be done, or just coming up with hypotheticals with regard to timed pulses of two lasers? Why would there be an advantage to pulsing two lasers alternately rather than just one in higher pulses? My reason for asking is I'm just working on wrapping my mind around the idea of how pulsing a laser increases power (not even sure power is the right word).

I was just trained last week on using a 2-photon microscope (very cool toy! ). Of course the people doing the training are the ones who know how to work the software and know about microscopy in general, but don't really know the physics of how the thing works. The explanation was rather minimal. Anyway, what I got out of it is that it differs from a traditional confocal microscope because the lasers are pulsed, so somehow they can utilize a longer wavelength, which does less damage to the part of the tissue you're not focusing on, but increases the chances of 2 photons colliding in the area you're focusing on. Somehow 2 photons of a long wavelength colliding leads to a single shorter wavelength hitting the fluorophore and causing the emission that we then detect.

Anyway, if that sounds really screwy, it probably is because I didn't understand what the rep was saying completely, and I'm positive he didn't understand it any better. I just thought it was cool when I saw the software and realized I hadn't been mishearing the rep all through his presentation...the laser really is called a Mai Tai laser (if that means anything special to you)! And if you're not careful with the power settings on this one, you get to watch a hole burn through the middle of your sample.

This is all a round about way of saying, if I understood just a smidge about how pulsing a laser affects its power (or anything else about it), I'd like that. Now that I was among the first group of users trained on this new toy, I'll be among those who will help do some of the training in the future. That means someone else is bound to ask me too, and I'd like to give them a better answer than I got.

6. Feb 2, 2005

### Integral

Staff Emeritus
Moonbear,
I hope I can help a bit. If I can't then I'll bet that someone will be along shortly who can!

First caveat... The exact answer to your question may well depend strongly on your system. My understanding of my system may or may not apply!

I work with http://www.coherentinc.com/Lasers/index.cfm?fuseaction=show.page&id=868&loc=834 [Broken] lasers these are Solid state diode pumped Q switch pulsed lasers. These lasers can generate roughly 1 microsecond pulses high power pulses at a 60kHz rep rate. The reason for the pulsing is that it requires a bit of time to get enough atoms in the excited state required for a high power pulse. So essentially you fill up the container, dump it out, then wait till it fills before you dump it again. You can run them in continuous mode for beam alignment purposes but the beam power is not high enough to even warm your hand up. ( you do not want to stick your hand in the full power beam, it instantly ignites paper!) In this mode you are drippling out the photons at something similar to the rate that they build up in the lasing media. Does that help?

This is just my version of the need for pulsing, I hope that it is close enough that some of the young hot shots around here do not feel the need to tear me a new... to correct my misconceptions. (They may well exist, but please be kind!)

Last edited by a moderator: May 1, 2017
7. Feb 2, 2005

### Laser Jock

Another way of saying this is the most important aspect of a pulsed laser is that they can produce high peak powers. There are two "powers" associated with pulsed lasers: the peak power and the average power. For a CW laser (non-pulsed) these are the same. The average power is the amount of energy in a single pulse multiplied by the number of pulses produced per second. The average power does not depend on the pulse width, it just tells you the average rate that energy is coming out of the laser. As an example, a laser that produces pulses that contain a Joule of energy at a rate of 5/second will have an average power of 5 Watts.

Peak power depends on pulse width and is independent of the pulse rate. The peak power is the pulse energy divided by the pulse width. In my example above, if the pulse width is 1 millisecond (10^-3 s), the peak power is 1000 Watts. If the pulse width is 10 femtoseconds (10^-14 s), the peak power is 10^14 Watts .

Obviously when you shine light with that high of power on something, unusual things can happen. The average power of a laser will usually tell you how likely you are to burn things (like paper, your hand, your shirt, etc.). A laser with a high peak power will ionize things, tear molecules apart, and/or produce unusual nonlinear effects. One of these nonlinear effects is 2-photon absorption. Normally an atom or molecule will only absorb one photon, but if the intensity (power/unit area) of the light is high enough, it will absorb two photons at the same time. Now, I'm not real familiar with 2-photon microscopy, but I would guess that it uses the same effect. As you said Moonbear, the longer wavelengths (and lower average power) would be less likely to damage the tissue while the higher peak powers available in a pulsed laser would induce 2-photon absorption in the fluorophores. The fluorophores would then decay to a lower energy state by emitting another photon that you can detect.

Sounds like a pretty cool toy to me.

Last edited by a moderator: May 1, 2017
8. Feb 2, 2005

### Gonzolo

Here are 3 essential ways to make laser pulses :

1. Mechanically chopping a continous beam.

2. Q-switching : Intermittently changing the beam's polarization so it gets out only once its ready. (Integral's laser)

3. Modelocking : Having the cavity length so carefully ajusted that destructive and constructive interference of beam components cause a train of pulses. Some nonlinear tricks-of-the-trade are involved of course. (Moonbear's laser).

Two-photon absorption can takes place with femtosecond pulses because so many photons essentially arrive at the same time to excite energy levels. Energy levels absorb 2 photons, and then emit only 1 like ususal, so that it has twice the energy (half the wavelenth) of the incoming ones. (your "guess" is correct Laser Jock).

As for the original question, beams adding their energies together is roughly how laser cavities work. Outside the cavity though, beamsplitters will necessarily cause energy losses as each time you use one, a beam will be "split", so you can't end up with a single beam containing all the energy. Fiber optics might allow part of the effect Parlous is looking for. (shooting many diode in a single aperture?).

9. Feb 2, 2005

### Moonbear

Staff Emeritus
Thanks Integral, Laser Jock and Gonzolo! So the idea with pulsing is you are sort of storing up the excited atoms and letting them (or I guess you mean the energy they release?) all go at once instead of allowing them to steadily running out in a stream, so you pack more wallop (there's a technical term for you) into each pulse than would be in a steady stream over the same time interval? Okay, and the 2-photon absorption happens in the tissue, not because 2 photons are somehow striking each other before penetrating my cell to excite the fluorophore, but because the tissue manages to absorb two at a time because they are coming at it rather fast and furious (don't you love how I can mangle technical terms?). This is making more sense.

Yes, Laser Dude, it is a cool toy, a very very cool toy! This one is even improved over previous versions because you don't have to tune the lasers by hand anymore (I think that might have been a bit much for me), and someone thought to put tubes over most of the beam paths so you don't accidentally lean into one (though I'm sure you'd realize it pretty quickly). There's only one place where the laser runs through an open path, and they've put that way at the back of the instrument table, so there's no reason to get close to it.

And I bet you physicists didn't know we got to play with such cool toys in biology (we just don't have to know how to build them ourselves). Of course now I need to put some thought into designing experiments that will give me an excuse to play with it more. It's one of those things where until you have the toy, you can't really plan on things to do with it.

10. Feb 2, 2005

### Gonzolo

That's the idea for Q-switching (my 2.), Moonbear : "store and dump".

The laser you are using probably has a few boxes. The first (the Mai Tai) shoots out a modelocked train of pulses. These are not "stored and dumped" pulses. They are best view as superposed sine waves of different wavelengths, such that they destructively interfere everywhere except for one point where their crests coincide to give out an extremely intense maximum. This point is a (femtosecond) pulse.

This train of pulses then goes into another box or two where they get shaped and processed for your specific application (2-photon microscopy). Q-switching ( = store-and-dump) tech may be used in one of these extra boxes.

The tubes between boxes are actually to prevent dust particles from burning on the lenses and mirrors ... though I guess it also prevents biologists from cooking themselves :surprised:rofl:

11. Feb 2, 2005

### Moonbear

Staff Emeritus
Oh, now it really is making so much more sense. Yep, during the training session, they pointed out that we need to make sure the box in the software that says "mode locked" is checked to use the 2-photon laser.

Cool, yep, there are two boxes. One with the laser that's always on (the one they put in the back so we can't toast ourselves by accident), and the one that sends out the laser to the microscope when it's in use. So that must be the one that does the Q-switching. I'll check my notes on that at some point and see what they called the second box (of course it might not mean anything, might be just another proprietary German word...the instrument we got is from Zeiss, so I got a German lesson too; finally understand why all the filters, plates and mirrors in the microscope have these weird sounding names - they make perfect sense when you translate them to English).

Oooh, good to know they're supposed to be dust covers! That wasn't explained to us; we were left thinking it was to prevent us from toasting ourselves. That's pretty important to know so the lenses and mirrors don't get ruined.

12. Feb 2, 2005

### krab

I work with particle beams, but I think the physics is analogous. Imagine you have an axially symmetric laser beam. Then in the x direction, it could be said to occupy a finite area in x-p_x phase space. The point is that you cannot put another laser beam smack into the same area. You can put it beside (same p_x, different x), you can coincide x momentarily by crossing them (same x, different p_x), you can interleave pulses if it is pulsed, but you cannot put two separate beams into the same phase space. That's why it is called stacking instead of merging. Let me know whether this agrees with your understanding..

13. Feb 3, 2005

### Laser Jock

I'm not sure if the physics is analogous. Photons are bosons so they can occupy the same state. As long as the phase of both beams are carefully controlled, I don't believe that there is anything that prevents the merging of two beams at a beam splitter.

Beam tubes also serve the purpose of keeping the whole laser system stable. Air currents can cause the beam going from one box to another to shift slightly and disturb the sensitive alignment. Another aspect of the safety reason for the beam tubes is eye safety. If your hand gets in the beam, and part of the beam hits your watch or ring, you could blind yourself or someone in the lab at power levels far below the pain threshold of your hand. Mode-locked lasers are especially dangerous in this regard.

14. Feb 3, 2005

### parlous

Thanks a bunch! You're ideas and suggestions have cleared up a lot for me and my project. I appreciate it very much. :)

15. Feb 3, 2005

### Gonzolo

In agreement with Laser Jock on both statements.

16. Feb 3, 2005

### Moonbear

Staff Emeritus
Yep, I did my laser safety training today and learned all the scary things they can do. I think I'll leave my watch off when I go work there, just in case, since the laser comes in on the left side and my watch is on my left hand -- especially while I'm getting used to the layout of things in the room.

17. Feb 4, 2005

### Gonzolo

I have made it a habit to always remove my watch in any laser lab, and to ask anyone else who enters to do the same.

It has also become a reflex to close my eyes if I have to bend over below table height (for switches or to pick something up). I imagine the entire plane of the room, at laser height, to be illiminated. Especially with IR beams that you don't see anyway.

Eyes are that important.

18. Feb 4, 2005

### Laser Jock

I do the same. I'm already legally blind in my right eye (non-correctable, born that way). If I lose my left to a laser... :surprised