I am searching for alternatives to 3rd harmonic generation out of Nd:YAG lasers (fundamental mode at 1064nm) to emit pulsed laser radiation somewhere in the range 360nm to 375nm.
If I may further ask, I would like to get pulse durations somewhere between 10ps and 30ps, repetition rates somewhere between 100KHz and 1MHz, and pulse energies of at least 1.5µJ.
That means the laser either emitting at 360nm to 375nm, or at 720nm to 750nm, or at 1080nm to 1125nm. (I think 4th harmonic out of a 1440nm to 1500nm laser would be too much).
Mode locked alexandrite would tune between 720 - 800 nm and has enough bandwidth to support 10 ps pulses. Ti:Sapphire may also go this low. Search the literature from the Fujimoto group (MIT) in the 1990's I think and you'll find a design for a mode locked laser cavity with an intra-cavity Herriott cell that will give you a huge path length, so lower repetition rate than is common for most mode locked lasers (in the single digit MHz range). You get a decent increase in per-pulse power with such a long cavity also, so you might be able to do your power requirements with oscillator-only designs.
I agree with the previous answers except for the fact that you will not get low repetition rates that way. You could add a pulse-picker but that would throuw away a lot of energy. I think I would suggest looking into gain switched diodes as a low power way of generating pulses between 10 and 30ps at low to medium repetition rates. This would then be followed by a solid state (Alexandrite) power amplifier to get the desired pulse energies and a doubling stage to get to the desired wavelength.
Hi Samuel: Here's a couple of references to mode locked laser cavities with intracavity Herriott Cells that give you low double digit and single digit MHz pulse repetition rates (down to 3.8 MHz I think). One of the papers gives some hints on how to go lower in rep rate. It might be worth getting in contact with the group at MIT to see where they got to on this? John
Optics Letters September 1, 2003 / Vol. 28, No. 17 / OPTICS LETTERS 1597
Optics Letters March 15, 1999 / Vol. 24, No. 6 / OPTICS LETTERS 417
OPTICS EXPRESS 5 May 2003 / Vol. 11, No. 9 / 1106
Thank you very much to all of you for your excellent inputs.
I will start "right away" contacting different people.
To be honest, my question was eventually wrong formulated, but your answers are very valuable.
I am actually just searching for alternative ways of generating laser radiation somewhere in the range 360nm to 375nm (so the "unconventional" is not a requirement).
We do not need and do not want to enter the fs regime, for that reason the pulse durations somewhere between 10ps and 30ps.
The system will scan a sample so it will be limited in performance by the scanning system (likely a pivot galvo scanning system), thus repetition rates must be kept "low" somewhere between 100KHz and 1MHz.
The process need to reach a threshold pulse energy to be started, so considering optical losses in the beam delivery optics, pulse energies of at least 1.5µJ are required.
Further ideas are that the temporal and spatial profiles of the beam are smooth (ie close to gaussian, without pedestal or carrier pulses, if possible).
Finally, if all can be packed compact and manufactured at low cost (say below 50K€ per unit). Then my dreams may become true ;-)
Now you suggest me several ways:
1) Mode locked alexandrite supporting 10 ps pulses
2) Ti:Sapphire
3) Contact Fujimoto group (MIT)
4) gain switched diodes as a low power way of generating pulses between 10 and 30ps at low to medium repetition rates followed by a solid state (Alexandrite) power amplifier to get the desired pulse energies and a doubling stage to get to the desired wavelength.
Thank you again, I will explore the 4 possibilites (as well as read the papers, thanks to John), and let you know.
If you have further ideas, your post them here.
Well, you could still take tripled YAG and pump the appropriate dye laser with that... Dye lasers are messy, but they reach everywhere. I am not sure, though, if they support picosecond pulses, I worked in nanoseconds.
Hi Samuel: This just showed up in the optics literature. It describes an architecture that will give uJ level 1 MHz rep rate UV pulses. This is Nd at 355 nm, but that fact that an architecture exists is promising. Best, John
39.1 μJ picosecond ultraviolet pulses at 355 nm with 1 MHz repeat rate
Peng Zhu, Daijun Li, Qingyong Liu, Jun Chen, Shaojun Fu, Peng Shi, Keming Du, and Peter Loosen »View Author Affiliations
Optics Letters, Vol. 38, Issue 22, pp. 4716-4718 (2013)
http://dx.doi.org/10.1364/OL.38.004716
Second harmonic of a mode-locked dye laser can meet these requirements. The dye laser is of course tunable.