Which type of back end mirror for diode pumped Nd:YAG laser?

[Mich_K]

Good day,

I am busy designing a diode pumped solid state (DPSS) laser using a Nd:YAG rod, as shown in the attached figure.

I want to use a 808nm IR laser diode to pump the rod, but I am not sure what type of dielectric lens I need that would
allow the 808nm light into the cavity, but then reflect all of the 1064nm and 532nm light from the lasing process.

Any help or suggestions will be appreciated.

Thanks
Mike



Updated schematic:

 

[RedCowboy]

You can start here > https://www.rp-photonics.com/dichroic_mirrors.html

I think you are looking for an ( AR 808nm HR 1064nm ) but there's more to it than that, the link will help until someone with experience building these can hopefully help you.

 

[Jim H]

Good day,

I am busy designing a diode pumped solid state (DPSS) laser using a Nd:YAG rod, as shown in the attached figure.

I want to use a 808nm IR laser diode to pump the rod, but I am not sure what type of dielectric lens I need that would
allow the 808nm light into the cavity, but then reflect all of the 1064nm and 532nm light from the lasing process.

Any help or suggestions will be appreciated.

Thanks
Mike

View attachment 70044

You are missing your KTP in your diagram... No green without it

 

[Alaskan]

You asked for back lens, not mirrors, I just understood I headed the wrong direction, but had already written text below, I need to know what diode or 808 nm source you are using to pump the rod, I might be able to point you in the correct direction, especially if fiber output 808.

Edited:

Illustrations: Below you will find a diagram of the diode pumped ND:YAG, they illustrate the laser diode portion without showing it has a collimation lens inside that box, earlier I couldn't understand how this arrangement shown in the left most drawning could work without a collimation lens first, but I can now see there is a cylinder lens in a horizontal position in front of the laser diode. The only way a pair of cylinder lenses can correct a rectangle shaped laser diode beam into more of a square (approximating a dot), is if it is collimated first, they do not show the collimation lens, only the first cylinder lens as a small bump attached to the laser diode mount.

The second diagram to the right is without beam shaping. If you are using a diode array like are used in a FAP800, for myself, I wouldn't want to tackle that one, although I would not be afraid of a FAP800 with their nice small fiber output. Regarding the diagram on the left again, for other applications, a benefit of beam shaping a diode output is to increase the power density, and I am assuming this is what they are doing to pump the YAG too, but it may merely be to make the beam more uniform into the YAG. I need to study pumping YAG rods more to get a proper understanding of the benefit of beam shaping a pump diode output, I might be over speculating.

Fig. 1 Diode (end) pumped Nd:YAG laser schematic

I am working on the same thing, but my rod is already multicoated, if you can find a rod like that, maybe it would make things easier in some respects. According to the seller I bought my rod from, I am hoping they are coated to AR and pass 808 nm on the input end, they are also coated for 1319 nm, that is the wavelength I want to produce. I am building a database for this and other DPSS projects at laserhobbyist but not a lot there yet, better to go to http://donklipstein.com/lasercds.htm#cdstoc

This is what the seller told me about the coatings:

Definitely AR for 1319, possibly also HT 808 for end pumping (which would be unusual for a rod of this length). Also possibly HR for 1064nm since you need to suppress gain at 1064 to get lasing at 1319. That's usually done in the cavity mirrors though. Or maybe even HR 660 to keep any doubled light out of the gain stage.

You can probably extract some of the information you are wanting from his message, above. If I cannot end pump it, that's fine, I will side pump it. What I have found for a 1064 nm ND:YAG to SHG to 532 nm is the rod needs HR or high reflectivity for both 1064 nm and 532 nm on one end, external mirrors (or I suppose a mirror coated on the rod, but I am still learning about that) and on the other end an OC partial mirror for 1064 nm and high pass for 532 nm (intracavity SHG crystal) which allows only a percentage of the 1064 nm energy out, but all of the green without much loss.

Note: To produce green using a ND:YAG you should have the crystal inside the cavity, and even better also have a Q-switch inside. If the crystal is outside of the cavity of a CW laser the conversion to green is very weak, if you can see any at all which is more often the case. Although if the rod is Q-switched through some technique, you can have the crystal located outside, but it requires a lot of peak pulse power to have much efficiency when the crystal is on the outside. I have tried to get green out of a KTP crystal using a 2 watt 1064 nm CW pointer and only saw a faint spot of green on the surface of the input end, no apparent beam at all and the crystal itself did not light up either, no matter the angle which just now brings to mind the likely problem, the output of that pointer probably is not polarized, I need a type II crystal for unpolarized.

We have some experienced members who can likely clarify some of this and better advise, hoping they will chime in. There are different techniques, depending upon what your application is.

Note: First paragraph regarding the illustrations edited.

 

[Mich_K]

RedCowboy, thanks, the 'dichroic mirror' is the terminology I was missing, didn't know the correct terminology, but that article is very useful and now I at least know what I should be looking for (-:

Jim H, thanks, I updated the schematic since I was missing the KTP crystal.

Alaskan, thanks for the in depth explanation, much appreciated. Sorry for the confusion in the title, I changed it to 'back end mirror'. To answer your specific question on the diode, I am planning to use a DTR-G-8 glass lens to focus the beam of a simple 5.6mm 808nm diode somewhere in the YAG rod. So not fiber-coupled. Any idea if this will be sufficient to pump the YAG, or would the diode beam require more shaping? Unfortunately I already bought the uncoated rod, so I think I have to get a dichroic mirror at the back end, although I expect that to be a bit expensive so I'll have to weight up the costs of that lens vs coated rod.

I was also wondering: Can the KTP crystal be placed anywhere in the optical cavity, or its exact location or orientation important? And what would be the effect of having a Q-switch inside the cavity as well, higher output power? I have heard of Q-switches, but not sure what the physical components are, are they also some form of crystal?

Thanks in advance
Mike

 

[Alaskan]

When using an 808 nm laser diode into a YVO4 crystal to produce 1064 nm, you would place the KTP at a location where the beam can be focused inside of the crystal, I mean the waist of the beam as it is focused down to a point needs to be inside the crystal, not on the surface of the crystal. Having brought up YVO4 crystals, you may want to take a look at them to produce 1064 nm, instead of a ND:YAG. All of these materials have their advantages and disadvantages.

If using the output of a CW YAG to SHG a nonlinear crystal, I expect the same would be required to have the power density needed for efficient SHG to 532 nm, or what is known as doubling. If you have a Q-Switch in the cavity too, you may not need to focus the beam down to a point to achieve the needed power density inside the crystal. I have not built a DPSS yet, only working on it, so you can research this further from what my understanding is and if you find I am incorrect, please let me know.

Also, I finally understand what was being illustrated in one of the drawings I attached in my last post, please see edits.

This document has some good info, in general. https://www.osapublishing.org/Direc...1-25-30532.pdf?da=1&id=275155&seq=0&mobile=no

....and this article on YVO4: https://www.osapublishing.org/Direc...7-25-22761.pdf?da=1&id=190775&seq=0&mobile=no

ND:YAG pumped at 938 nm: https://photos.imageevent.com/qdf_files/laserlinks/ND_YAG_938nm_pump.pdf

Note, 808 nm is not the only wavelength you can pump these materials with, 808 is a wavelength low and high power diodes are made for and have gained popularity as a pump diode due to being a very efficient pump wavelength, however that area of Near-IR is rather narrow for use as a pump and when building a stable laser where the output power remains consistent, requires temperature stabilization to keep the wavelength output inside a narrow range of only a very few nanometers, otherwise the efficiency drops off fast.

Unless you buy a temperature stabilized pointer like the Hercules pointer Laserglow produced (illegal to sell in the USA now), the cheap DPSS lasers outputs vary all over the place, when getting too warm, or cold for that matter when used in low ambient temperatures, dropping off. Temperature affects the SHG crystal too, with both the diode drifting in wavelength and the crystal temperatures shifting in our cheap 532 nm pointers, I am impressed they do as well as they do.

 

[Mich_K]

Hi Alaskan,

Thank you very much for all the very useful information. I have some reading to do (-:

Kind regards
Mike

 

[Alaskan]

There isn't a soul on the earth who has mastery of all aspects of this technology and never will there be one.

 

[Mich_K]

I think I might have found an end-mirror that will work for me, but I was curious about the option of side-pumping.

Is this as simple as the updated schematic shown below, by focusing the 808nm beam somewhere in the YAG rod from the side? Or is there more to it that that?

Some Google searching has shown some much more complicated side-pumping setups, but I think the setup often doesn't need to be
as complex as what is shown in existing setups to work.

PS: I am not building this laser in a pointer, it will be built as a bench test setup, so I am not dealing with many constraints
that I think will be relevant for a laser pointer build.

Thanks!
Mike

 

[Jim H]

I think I might have found an end-mirror that will work for me, but I was curious about the option of side-pumping.

Is this as simple as the updated schematic shown below, by focusing the 808nm beam somewhere in the YAG rod from the side? Or is there more to it that that?

Some Google searching has shown some much more complicated side-pumping setups, but I think the setup often doesn't need to be
as complex as what is shown in existing setups to work.

PS: I am not building this laser in a pointer, it will be built as a bench test setup, so I am not dealing with many constraints
that I think will be relevant for a laser pointer build.

Thanks!
Mike

View attachment 70065

Not an expert, but I think you would need a focused line along it's length, not just a point.

 

[Alaskan]

I’ve owned some small side pumped ND:YAG’s and mine did not have any kind of focused beam into the rod at all, just a array of diodes butted up against the rod, although I have seen one on ebay which did have a long PCX cylinder lens in front of the array. Everything was tightly together with that one, no apparent focal distance.

Also, Coherents “donut” diode pumped YAG’s do not have a lens between the rod and the diodes either. Flashlamp pumping does not use a lens, but they do have reflectors behind them to get back the energy from the flashlamp going the wrong direction away from the rod.

 

[Mich_K]

Thanks all!

Mike

 

[Alaskan]

Welcome, from what I have seen when pumping a rod, if side pumped, you want to fill the whole rod with pump energy as evenly as you can, if you focused the energy into a small area that just causes heat stresses between areas and also does not allow the whole medium to be pumped.

I suppose you would waste pump energy if you overshot the rod too much when using diodes, and not pump a large enough area if the diodes were too close, probably a happy medium needs to be achieved between too close or too far from the rod when directly side pumping with diodes. If the rod is too long, end pumping doesn't work well either unless the rod is a gradient rod.

 

[whiterod14]

There isn't a soul on the earth who has mastery of all aspects of this technology and never will there be one.