Can anyone explain the unusual energy losses in my UV-laser mirror setup?

[S.Hollas]

I have a problem with which I would appreciate any help that is offered:

I have a 266nm laser (frequency quadrupled 1064nm Nd:YAG), and the energy losses along the route are a little strange. The beam is reflected by two UV mirrors; however, the first mirror loses ~7% of the in-going energy, but the second loses ~12% of the in-going energy. I have swapped the mirrors, and the effect remains.

The beam that is emmitted by the laser contains some 1064nm and 532nm fractions, and I feel that the losses of these frequencies (they should be minimally reflected by the UV mirrors) at the first mirror must relate to the problem, but I cannot figure out why.

Can anybody understand this effect?

 

[JeffKoch]

Over what bandwidths are the 7% and 12% numbers measured? Energy meters tend to integrate over broad ranges, and your quadrupled beam will contain significant amounts of 1w, 2w and 3w as well as 4w.

 

[mgb_phys]

What surface figure are the mirrors and how big is your power meter?
It could be that they are each scattering some light out of the beam - the effect of this is cumulative.

 

[Claude Bile]

Some questions...

1. How are you measuring the power loss? What detector type are you using? Are you using a filter of some kind? Does the detector have a window?
2. Is the Frequency doubling crystal intra-cavity? Is it in a separate cavity from the YAG crystal? Is the entire lasing and doubling process done in a black-box (i.e. you only get to play with the output)?
3. Were the mirrors hand-me-downs (for want of a better term) or were they chosen specifically for this type of application? Are the mirrors coated?

Claude.

 

[S.Hollas]

Dear Friends,

Thankyou for your responses, and I shall try and answer your questions:

The power meter is a Gentec Duo, measuring from 100µJ to 100mJ. The beam strength that I am measuring is ~30mJ; the measuring head is a Gentec QE25 series.

The mirrors are at 45º; I have both changed the position of each mirror in the beam line, and replaced one of the mirrors over the time this effect has been seen; they were new when I began, though the first mirror does suffer coating damage over time from the non-266nm sections of the primary beam.

The frequency quadrupled output is in a separate box immediately after the laser cavity.
There is no filter in the beam line.

I am measuring the energy by removing the beam cover immediately before and after each mirror; the gap between the mirrors is ~2m.

I hope this helps,

Simon.

 

[mgb_phys]

What bandwidth does the power meter measure over - it is probably sensitive to 532/1064 as well. Unless the doubler has blocking filters it will leak some of the original wavelength and your mirrors are probably reflecting some of this.

 

[S.Hollas]

Yes, it is sensitive to the other frequencies, and I am sure I am measuring them. If the situation were reversed, and the first mirror lost more than the second I would understand.
I am investigating the scattering effect, but initial results show nothing significant.

 

[Claude Bile]

I agree with mgb_phys. As it stands, the data you are getting is too unreliable as it is almost certainly "contaminated" with other wavelengths. You need some filters to get a reliable measurement.

Claude.

 

[JeffKoch]

You can probably estimate the amount of 1w, 2w and 3w coming out of your box, and from the mirror properties you can calculate the reflectivity at those wavelengths, so given the energy meter sensitivity curve (probably provided with the meter) you ought to be able to make a reasonable estimate for how the total energy measurements would differ after the first and second reflections. I'll bet it's roughly consistent with your numbers, I've seen these sorts of discrepancies myself. Experimental physics is harder than many theorists think it is.