I'm not sure if this is similar to what you are already doing, but you could use a spatial filter to generate a homogenous beam whereby you focus the beam through a lens onto a pin hole located at the focal length of the lens.

I would also check the alignment off all the elements in you beam expander as any slight misalignment can lead to some strange beam profiles. Is it possible to remove individual elements? This might help you to identify if a specific element is causing the apparent diffraction.

The best way to remove the diffraction rings from your beam is to use the spatial filter suggested by Alexander Malm. By passing only the zeroth order diffraction in the pinhole, you get rid of the rings to get a clean beam that you can collimate with a lens located one focal length down the path of the beam. You can do the expansion right after the filter by choosing a lens with the appropriate focal length. Also note that the shape of the filtered beam will be the same as that of the original laser beam.

Dear Mohamed and Alexander,

Thank you for your answers.

I'm already using a spatial filter (what I call an iris is a pin-hole sorry) to get rid of a first type of diffraction pattern that looks like vertical lines (certainly due to the diode laser cavity) by selecting only the zero order, but unfortunately I end up with a second diffraction pattern, this one being circular, in the zero order that I isolated. I checked the alignment several time and I think it is alright. I can't use a second spatial filter as I would loose to much power. You can have a look at the attached file the setup is well described.

Hi Elie,

What you show in your setup is not an spatial filter, you are just chopping the beam sith a circular aperture, that will always give you a circular ring pattern. A spatial filter needs a converging lens and at the focal spot you place your pinhole to only let the zeroth order through. Think of the lens as an element that gives you the Fourier transform of your beam at the focal spot and what you are doing is getting rid of all the frequencies in the beam except the zeroth frequency, this cleans the beam. In the link I send you there is a nice little explanation of what you need to know to implement your spatial filter.

Regards

Alex

http://www.newport.com/Spatial-Filters/144910/1033/content.aspx

If you set things up properly, i.e., you focus the beam tightly on an a pinhole with right size (adjustable one, for example) to block everything that focuses away from axis of the beam and pass only the portion of the beam focusing on the axis, then only the zeroth order diffraction will pass through and you should see no spatial structure beyond the pinhole. The zeroth order beam carries no information about the edges, the holes and other things the beam met before it left the laser. How much of your intensity survive the filtering? It depends on how much of the beam was undiffracted. If your laser operates in the zeroth spatial order, then the filtering process should not affect the intensity of the beam that much.

A bit more expensive, but better would be getting a single mode fiber. The output will be a Gaussian whatever you put in...

Also as the guys said, you need to focus the beam on the pinhole and make sure it's sufficiently small.

Thank you all for your answers! 

I will modify my setup to place the pinhole at the focal spot and use a converging lens. How small should the pin hole be in order to isolate only the zeroth spatial order?

Neven, could you tell me a bit more about the fiber, how should I couple my laser into it?

Cheers

Thorlabs sells fiber ports:

http://www.thorlabs.de/newgrouppage9.cfm?objectgroup_id=2940

A bit cheaper are couplers like these, they work well:

http://www.thorlabs.de/navigation.cfm?guide_id=2235

Also, be careful when buying, do you need PM or not and APC or just PC and so on... Depends on your laser and application

Also see:

https://www.youtube.com/watch?v=kQvhbJbDG0M

How small should the pin hole be in order to isolate only the zeroth spatial order?

Small enough that only the central spot passes through. Start with adjustable iris; it should do the job if it closes to sufficiently small size.

Elie, take a look at the link I shared, there you can find some formulae to help you decide what focal length and pinhole size you need to use.

Dear  Elie Nadal,

about spatial filter:

http://search.newport.com/i/1/q1/products/q2/Optical+Systems/q3/Spatial+Filters/q4/Three-Axis+Spatial+Filters/x1/pageType/x2/section/x3/chapter/x4/family/nav/1

Model M-900;

objective x20; pinhole 15 mkm and lens from 7cm  to 20 cm (depends on beam dimeter you need).

What is the laser model you use?

Best regards, Pavel

Actually, your situation may be simpler than needing a spatial filter. After the light passes through the iris, arrange your lenses so that the iris is imaged onto your sample.

In your current setup, the collimating lens and the beam expander effectively project the iris into infinity (assuming you've aligned everything correctly, which I'm sure you have).

I would recommend starting with the following arrangement: negative / positive lens pair to expand and approximately collimate the beam out of the laser onto the iris, picking out a portion from the central fringe. Then a single positive short-focal length lens for projecting the iris onto your sample with some magnification (if the lens is closer to the iris, its image will be magnified).

If you want to get more advanced, you can try replacing the imaging lens with two or three lenses to have better control over magnification and image plane location, and to make sure the beams going through the beam splitter and the quarter-wave plates are reasonably collimated.

I would also urge you to figure out why there are diffraction fringes coming out of the laser itself. This suggests to me that something is wrong inside - either a damaged optic or a mis-aligned cavity. If you fix those issues, you'll have a better beam and considerably more power.

Looks like your laser is oscillating in multimode. You either have a misaligned optic within it resulting in such a beam profile or one of the optics is damaged. I think that you need to work on the alignment of the laser cavity so that it operates in single mode.

Making conjugated/deconjugated images, as suggested by Boris, is very intelligent in such a complex case. I guess you need to realign your laser cavity, if possible, or buyaa better laser. Maybe the manufacturer can realign it. I also believe it is working as multimode.

Other classicla option for interference rings, silly as it sounds, is dust. Just confirm the laser optics (adnd the rest are clean enough. But multimode emission will still be there. 

Thank you all again for all these interesting propositions and analysis.

I closely look at the spatial filter option and indeed it seems to be the perfect way to improve the beam profile, I also saw that it can be use to increase the beam diameter by choosing the right lenses. So I might try to use this solution.

Concerning the possibility to use additional lenses in order to image the iris on my sample, I'm not sure it would be compatible with the final beam spot size that I would need, approximately 5mm in diameter or more and a pretty hight power. As you said, the beam as to go trough several optics (beam splitter, quarter wave plate...) and so needs  to be collimated, such that the image of the iris would be at infinity again no? I have a 25 mm positive lens and several lenses with larger focal length. Boris, could you please be more precise about the configuration you have in mind?

Finally, I clearly would like to open the laser and look what is wrong but I loose any warranty if I open it. This is the laser I'm using: http://www.cnilaser.com/PDF/MLL-F-447.pdf

I will contact them to ask what I can do.

All the best to you all,

Elie

Another suggestion: Couple your laser into a single-mode fiber to filter out the fundamental, Gaussian mode. Then out-couple and collimate the laser light again and use the resulting Gaussian beam for your experiment will this be homogeneous enough?). Of course, if the laser is oscillating between different modes, this approach will lead to a fluctuating laser-intensity on your sample.

As a side-note: Be cautious with those lasers from CNI. I know they are relatively inexpensive and we used to use them quite a bit in our experiments. However, they tend to die after a few months/years of operation. You might want to organise a backup laser if you want to avoid longer downtimes of your experiment in the future.

Dear Elie,

I think the best way is to clean up the beam by using pinhole as some guys suggested. Some DPSS laser have really bad beams, you should first pay attention to M2 factor. As I see from your laser datasheet M2 factor is about 5, it's really bad.

Good luck

if you use a pinhole filter be very careful and wear goggles when aligning the beam to pass thru the hole.  The focus of the laser on the pinhole substrate is as intense as that laser will ever get!   That scatter can burn a hole in your retina (or draw a line on your retina when you move)  in an instant!!

Thank again to you all for all these precious advices.

Indeed you should try to use a spacial filter (pinhole). It causes losses, maybe you have to play with the diameter o find a good compromise between losses and beam-profile. Put them as last optical element before your sample holder in your setup. There will be also an interference pattern, but in the center there is a nice circular beam shape.

I did this with 330nm UV for interference lihography, same as you try I think. In worked.

Just follow the recomendations above from mall the other authors and you will be successful. ;-)