Can you suggest a commercial high damage threshold mirror or a beamsplitter for a high power CW laser beam? Looking to reflect a ≈3kW laser light.
1 Like
You can use a laser line HR dielectric coated mirror. Those usually have >99.9% reflectance. The rest of the light will be transmitted. Edmund, Thorlabs, Newport, Optosigma, EKSMA … offers these.
Regarding beam splitter you should take a dielectric plate beamsplitter or an optically contacted (not cemented) cube beamsplitter. 
It’s slightly connected: I’m also looking for a beam splitter which is non-polarizing and suitable for a 100uJ pulse energy 100fs laser with 2mm beam diameter. It must be a cube beamsplitter, plate is not good. Any idea where I can get this?
yes, but this is kW level CW beam. So the damage threshold has to be higher than normal.
For you, check MKS Ophir, they have cube beamsplitters that can handle 1MJ/cm2 pulses.
Thorlabs quotes 20kW/cm LPD damage thresholds for 1064nm mirrors. How large is your beam? You can always make it larger with a beam expander. To stay below the LIDT, I wouldn’t go below 5mm beam diameter. If it’s a Gaussian beam, the peak intensity will be double the average.
Laseroptik.de has usually somewhat higher damage threshold coatings, maybe it’s worth to browse their website.
If you don’t trust the dielectric mirrors, you can always get some oxygen free copper mirrors diamond turned. LT Ultra Precision would be a good place to ask. Prepare for 500-1000 USD per piece prices.
Another option could be molybdenum mirrors, those are also scratch resistant, but not sure about reflectance. Is this a fiber laser or CO2?
I can’t find any suitable 50/50 cubes on the MKS Ophir website, just beam samplers
It’s a fiber laser. Our beam has a ≈3.5mm diameter, and yes, it’s Gaussian.
Didn’t know you needed a 50/50 beamsplitter. Yes MKS ones reflecting a very low %.
Sorry I forgot to mention this detail 
Anyway, I can’t see anywhere non-polarizing ones with high damage threshold. I wonder what’s the reason for that. 
I will have to consider redesigning my setup for plate beamsplitters.
Your 3.5mm beam is indeed very close to the damage threshold but still below it. The alternative is really a bulk copper mirror with water cooling or a beam expander. At this power density you might want to check if thermal lensing will be an issue.
Hi
one of doctor in my team are interested in your needs, his major is the high threshold coating. in our lab, we have leybold syrus pro, intvalc ,and vecco spectra coating machine.
we want to know
1.pass band
2. stop band
3. angle of incidence
4. size
5.also other demands that you are caring .
Hi Omeda,
In the meanwhile, I redesigned my system to use a plate beam splitter, but it will still have to be custom. The following is my requirement. Maybe you can supply this coating too.
- The plate beam splitter should be 3mm thick, 25mmx25mm size, made of UV fused silica (Corning7980)
- On the back side it should be AR coated for 1050+/-10nm and 976+/-5nm.
- On the front side it should have high transmission (>97%) for 976+/-5nm and 50% transmission / 50% reflection for 1050+/-10nm, independently from the polarization.
- Surface quality: lambda/10 (or 20-10).
What is the expected laser induced damage threshold at 100fs?
Can you measure or estimate the GDD of such a coating?
Best regards,
Daniel
If i’m not mistaked your beam has a linear power density of abut 8.6kW/cm; being the beam gaussian, the peak linear power density doubles to 17.2kW/cm, as @Daniel_Csati pointed out.
Have you had a look at optoman.com?
They declare 30kW/cm for CW lasers, thanks to their IBS coating process. They can also provide custom solutions.
I’ve worked with them in the past: very devoted and precise.
1 Like
I agree, Optoman is superb. 
I wanted to avoid going custom and they did not have off the shelf what I needed. I will likely contact them too 
Hi sorry for my article about "Sapphire window ",i have delete that article .
Based on the requirements, we design a film structure and provide GDD simulation results and theoretical spectral curves. As a matter of fact, there may be some deviations, which need to be communicated and confirmed.
1.How much does the average transmission(976±5nm) need to reach after double-sided coatingth sides have been coated
2.I want to know what is the acceptable fluctuation range for 1050±10nm? 50%±1% or 50%±3%?
3. How large is the effective aperture of the substrate coating? Whether it is 95% effective light aperture;
4. Most critically, we have not yet tested our film for laser damage at the femtosecond level. However, we tested the damage threshold for nanosecond lasers. In the attached figure, you can see the results of 1-on-1 LIDT for anti-reflection film and reflective film.
1 Like
Thanks for the evaluation.
- If the average transmission is larger than 98%, I would be happy with that already.
- 50% +/-2% in the +/-10nm range is acceptable.
- Clear aperture of >90% will be needed
- The femtosecond LIDT should be roughly possible to calculate back from the nanosecond LIDT. At 100fs ca. >0.01J/cm2 would be OK.
The GDD is only important for 1050nm+/-10nm. Currently it’s very high and I can compensate for maximum 500fs^2. Is there a way to optimize this coating for low GDD too?
Thanks, Daniel
got it we will reply soon.
could we talk in email, because this website do not APP in phone . not so convenient.
i can reply at noon and night
my email :omedaoptics@gmail.com or do you have whatsapp or linkedin
your needs is not so easy , we will calculate something and reply tomorrow
Our first certainty is that the transmittance, spectrum, and effective light aperture are all met.
Secondly，after conversion, femtosecond laser damage threshold is met （0.01J/cm2 ）when nanosecond laser damage exceeds 3.5J/cm2.Our samples have shown that it can be satisfied based on the test results
Lastly, we optimize the design value of GDD so that it can satisfy both reflected and transmitted GDD requirements.
Thanks for this. I’m a bit worried about the reflected GDD chart. Even though the GDD is only 1 pixel high around 1050nm, that could mean 100 000 fs2 if I look an the Y axis scale. How much is it really? Can you rescale it?
The transmitted GDD seems to have bit of a break in the chart between 1035-1055nm. It might be nothing, but can you try to explanation?
What would it cost to do this beam splitter for 3 or 10 pieces?