Feasibility of diode-pumped dye laser (just a thought).

[DrMario]

Like it says in thread header; it's more of a thought. :thinking:

I was wondering whether if it's possible to use the green - blue Gallium-Nitride laser diodes which are nowadays powerful and cheap (true blacklight laser diode is of course prohibitively expensive so I won't bother covering it here, only those cheaply available to us as of now, including the exotic 510 - 490nm laser. And yes, Blu-ray Disc burner 405nm laser diodes still also qualify as the role for them here is to pump the dye to lasing threshold).

As for the 520 - 395nm pump laser range and availability, I would focus on both Rhodamine and Fluorescein dyes (even though dyes from green - red liquid glow in the dark stick and highlighter marker are a possible candidate), due to the absorption bands, you want the emission band wavelength to be two - three times longer than the pump band wavelength - eg. 605nm yellowish-orange lasing line via Rhodamine dye in the tunable cavity curvette pumped by 520 - 510nm laser diode (with optical correction, of course - if you have a green VCSEL, you may get away with using fewer optical lenses than with Fabry-Perot laser diodes due to the nature of the optical cavity therein). The bigger consideration of course is with the absorption wavelength of the dye (never hurt to check the datasheet), as there's no such thing as free lunch (ie. it's hard to make all-compassing multi-wavelength dye laser), so I would also recommend to make it so the laser diodes are easily removable and replaceable if you make that into the lab laser, that way you can handle multiple dyes.

Yes, I would like to build the diode-pumped dye lab laser, however metal is expensive so I will have to see what options I have in meanwhile. (Metal is preferable so I can be sure I get the optical configuration right or it would be a poor excuse of a laser.)

 

[Monado]

I too have been considering making a diode pumped dye laser. From what I know I think it will be difficult, just power wise. I know you can get some rhodamine for pretty cheap (30$) but thats in small quantities. You can also easily glue some microscope slides together and probably just use some isopropyl alcohol instead of the more commonly used methyl alcohol. Then I think you're just left with the issues of power and of beam specs. The low wavelength of those diodes you were talking about should do well to excite some rhodamine or whatever dye you use.

Maybe you would not be doing this on as much of a budget as I potentially would be, but if I decide to do this it would be pretty dang cheap.
I personally love the yellow laser beams, they're just so beautiful but difficult to produce. I made an rg laser projector which has a sort of yellow appearance but I prefer a real yellow 589nm light.

Im also researching a project to make a diode pumped solid state yellow laser using frequency doubling, but im having trouble finding the correct crystal, so in the mean time a dye laser might be a fun project.
Good luck with this and I hope I was helpful. :yh:

 

[DrMario]

I may think of machining a dye cell along with a holder for the horizontal cylinder lens so I can squeeze the laser light a bit after the light pass through vertical correction lenses close to the green or blue laser diode - that or I can use two laser diodes in tie ribbon cavity setup after the cylinder diode laser correction lenses - dye cell may be made narrow so that way entire pump laser light lits the dye, while side of the dye cell allow the laser light from the dye to leave. I will have to experiment with cheap highlight marker ink. I will have to also purify DCM (dichloromethane) if it so requires for certain dyes, or just go with isopropyl alcohol (100% pure spectrometer grade) - even though Everclear may be a potential buffer fluid, after all the dye laser actually exist for some experiment with the dye mixture.

Lastly, the diode-pumped dye laser may not be cheap to build at home but is actually cheaper than flashlamp-pumped variety due to the fact that up to 10 Watts blue lasers can be had for less than $200 each and easily outlast the Xenon flashlamp (up to 20,000 hours if properly cooled - I may water-cool the diodes, I have to think of how to make the laser diode sled easily removable so I can change out the diode for different wavelength to match the dye absorption while not affecting thermal performance). Let's say it's more of something to keep me busy.

 

[CynicalBrad]

Some searching on PL seems to suggest this may not be possible as the absorption band is not great, as well as Rhodamine 6G seeming to prefer high peak pulsed inputs to get the lasing going. I imagine about the cheapest way to make this work would be to pump it with a 532nm converted laserscope unit but those monsters are well beyond something I'd even want to play with. Even then, it would be pulsed, not CW, and would need a proper setup including dye circulating pumps. I think I recall a few TEA laser pumped Rhodamine 6G setups but i'm not digging that far right now.

Only thing I'm seeing as an option for pumping Rhodamine 6G CW is to use a dye jet couple with a high output 532nm system or an argon-ion setup along with the proper dye mix. (Copper vapor as a pump would be fantastic looking at the absorption spectra)

Absorption for Rhodamine 6G is rather poor in the 445-470nm range and peaks at ~560nm but is acceptable at 532nm.

 

[DrMario]

I am not restricted to just one dye. Some dyes favorably absorb 450 - 455nm blue laser light (which they were formerly pumped by blue Argon Ion lasers) while others towards 520nm laser light (Rhodamine reacts strongly towards 520nm laser from what I knew and yes I tried that), however I am open to experiment with the dyes and several laser diodes (from 530 to 405nm wavelengths, hopefully I can get most of them in optical output of a watt or above - while that failing, I can use two laser diodes to hammer the dyes). As for DPSS, I want to avoid that and challenge myself a bit (besides DPSS is kind of inefficient because it relies on nonlinear optical imprisonment which forces frequency doubling - ie. 5mW 532nm for 50 - 70mW 1,064nm pump laser - depending on setup, you can lose as much as 50 - 80% energy in frequency doubling, or worse, summing). Moreso, experiment curiosity thrives on "Why not try that?"

I know it's not trivial but that haven't stopped the others from using laser diodes in the compound laser systems like Titanium : Sapphire solid state laser (it is the most fussy laser that I know of asides the run of the mill DPSS, however it have been successfully pumped by the diode lasers themselves in certain optical setups).

Besides, I usually test the dyes with some lasers I already have in my disposal to see which responds the best to so I can swap in the diodes that I know will pump the dye I am using, just so I don't waste my time setting up the laser.

EDITED: As for diode pumped dye laser design, I am looking into pulsed laser diode drive along with recirculating dye pump (to keep the dyes moving if I use the pump laser in CW mode). I will be exploring the design a bit before I scrape together the laser (case and some custom diode driver electronics may be the first to be obtained).

 

[Jimmymcjimthejim]

Hi DrMario,

I researched this pretty extensively about two years ago. My conclusion was that it's definitely possible but would be really difficult to do. The big hurdle for me was putting together some kind of dye circulation system, and I couldn't commit to go through that work. Without circulating, the dye would degrade rather quickly (order of seconds). Here's a few takeaways from what I remember:

• Pulsing the laser is preferred for two reasons: higher peak power to overcome the lasing threshold and reduction of losses due to triplet states.

• To overcome a high lasing threshold, focus your pump laser as much as possible on the dye.

• Dye absorption can quench your output so you'll want to use a thin layer of dye to excite.

• The reflectivity of your output coupler is important. I recall lower is better for this application.

Something that I wanted to try was to sandwich the dye between a concave mirror and a dichroic reflector (pass the pump, reflect the fluorescence). This makes a little dye cell that might get you a second of lasing.

I found that academic papers were really handy for learning about this. One article that I found helpful is "High brightness diode-pumped organic solid-state laser" but I no longer have access to it.

This website may also be of use for selecting dyes:
http://www.photonicsolutions.co.uk/dyeinfo.php#

Just be careful because some organic dyes are toxic as well as the solvents required.

Good luck and I'd be happy to help where I can!

 

[DrMario]

As for thickness of dye, I can see why the thinner the better chance of lasing is important as some dyes are translucent. Also, I might have to experiment with the optics a bit with 520 nm green laser diodes so I can see which optical setup is best before swapping for blue laser diodes which some orange-yellow to green dyes may want (the laser diode setup will be made so it can be swapped).

Thanks also for the laser dye chart, it will be useful for selecting the laser diode to use with whatever dye I will be lighting up.

I am also hoping Gallium-Nitride VCSEL diodes will come soon enough so I can pump certain dyes which are fussy. The laser diode driver would be microcontroller based so I can change pump pulse width as required (either STMicroelectronics STM32F4 or Texas Instruments Hercules which both have built-in independent timers which I can use for optics like piezoelectric optical modulator and / or laser diode so pumping optics become a bit more favorable).

Lastly, would it be a good idea to cool the dye with active water cooler and heat exchanger combo if I am going over a watt?

 

[Emialias]

Definitely following this thread because this is probably something super cool to do with the new 505-495nm diodes becoming so easily available. kinda want to try this aswell

 

[Jimmymcjimthejim]

Lastly, would it be a good idea to cool the dye with active water cooler and heat exchanger combo if I am going over a watt?

I don't think this is necessary as long as you are circulating the dye and stay under 10W. Just as a point of reference, lots of electronic components dissipate a few watts without a problem, and that's without active cooling. I would start off without active cooling, and if heat is a problem, then add a few heatsinks to the system.

 

[Singlemode Laser]

One article that I found helpful is "High brightness diode-pumped organic solid-state laser" but I no longer have access to it.

The preprint is for free:


https://www.google.com/url?sa=t&sou...FjACegQICRAB&usg=AOvVaw1JR3YlQjou9MV_o-_rleNG

Singlemode

 

[paul1598419]

Great find, Singlemode. And a very interesting read, too. I love getting little jewels of knowledge like this. Makes my whole day. :yh:

 

[DrMario]

Jimmymcjimthejim, I think I may leave the radiator in as a backup, that if the dye decides to boil, I can turn on the second water pump and fan to take care of this problem. As for electronics, I would rather to deal with thermal management since how hot the electronic components run at affect the overall lifespan of the laser driver - if I pick a microcontroller with CPU speed of 200 - 400 MHz, I'd rather hearsink it to be on safe side (both Cortex R4 and M7 tend to run a bit warmer because they are superscalar so they have extra more transistors compared to both M3 and M4, hence increased performance at the cost of increased heat output).

Singlemode, thank you very much for the link, it makes for an interesting read. LED-pumped dye laser is one interesting subject as it tend to be difficult to make it work (I remember the work on LED-pumped Nd : YAG laser more for weapon research, and it was apparently finicky that all works on it eventually stopped).

 

[Jimmymcjimthejim]

The preprint is for free:


https://www.google.com/url?sa=t&sou...FjACegQICRAB&usg=AOvVaw1JR3YlQjou9MV_o-_rleNG

Singlemode

Awesome, thanks! The article is just as helpful as I remembered!

Jimmymcjimthejim, I think I may leave the radiator in as a backup, that if the dye decides to boil, I can turn on the second water pump and fan to take care of this problem. As for electronics, I would rather to deal with thermal management since how hot the electronic components run at affect the overall lifespan of the laser driver - if I pick a microcontroller with CPU speed of 200 - 400 MHz, I'd rather hearsink it to be on safe side (both Cortex R4 and M7 tend to run a bit warmer because they are superscalar so they have extra more transistors compared to both M3 and M4, hence increased performance at the cost of increased heat output).

Oh, of course you'll have to heatsink the electronics. What kind of dye circulation system do you have, or are you building it from scratch?

 

[DrMario]

I will be doing dye circulation system from scratch, that way I can fit all the optics into the metal box (I hope metal tariff doesn't kill me). Why metal box? that's so the optics can be permanently held down and steadied so the pump laser light doesn't walk off to the point dye stop lasing. Mode hopping and walk-off is very challenging to troubleshoot especially when you have the platform that can move or be bent.

 

[Singlemode Laser]

No need to thank me here. The preprint was just a Google second away!

Singlemode

 

[Jimmymcjimthejim]

So I have been trying to convince myself that making a diode-pumped dye laser is possible using a small dye cell without circulation. With a big enough dye cell, there should be enough liquid to maintain lasing for a decent amount of time.

If the dye degrades after 1000 pulses, and there are 8000 "fresh spots" on the cell, that means I can get 8,000,000 pulses. Using a single-mode 100mW laser, 100ns pulses, 100kHz pulse rate, and 10% efficiency, I should be able to get 0.1mW out for 80 seconds. Of course this is really optimistic, but I'd be thrilled to get anything lasing at all. Any thoughts?