Some experiments with the blue LD's

This is a forum to share experiences and ideas about holography.
Thieu

Some experiments with the blue LD's

Post by Thieu » Wed Sep 22, 2010 7:06 pm

Paulos, what kind of feedback do you use? Your reasoning with the interferometer and hologram results makes perfect sense. However it's quite different from what both wler and I see in the mode spectrum directly with a spectrograph at higher currents, for instance here:

http://piepklein.blogspot.com/2010/07/out-of-blue.html

While looking at the movie myself I noticed something though. Check out the modes at around the 6 second mark, this is at reasonably high current:
modes.JPG
modes.JPG (5.18 KiB) Viewed 3109 times
You see that the spectrum is not single mode. There's one intense mode at the left and a couple modes at the right. The notable thing is that the mode spacing is not equal for the modes at the right! The effect is even more dramatic later in the video, the mode structure becomes a fuzzy mess. The fuzzy mess is not bad though, because it won't add up to banding in the hologram (there is no periodicity in their sum), but to a general background. No problem at all for dcg. Maybe the fuzziness is also slightly there in the mode at the left, causing the decrease in coherence length you report, without any banding.

The impression I'm getting more and more is that the filamentary nature of the current in these diodes causes local refractive index changes inside the cavity. It not longer behaves as an ideal fabry perot cavity with a well behaved, equally spaced mode structure, but as a complex structure with complex mode patterns and frequencies. In a situation as the above, this might be an advantage.

Just an impression though... :-)

Paulos

Some experiments with the blue LD's

Post by Paulos » Thu Sep 23, 2010 7:06 am

Thieu wrote:Paulos, what kind of feedback do you use?
Instead of a glass plate, I put a 30%R beam splitter just at the front of the laser diode.

Image

Reflecting back a portion of the diverging beam may be advantageous.
Due to the divergence and the very small size of the emitting surface, only light coming from an extremely small area (may be 1/100 000) of the emitter returns to the cavity, participating to the feedback mechanism.
It is probably why the laser is forced to be single longitudinal mode relatively easily, and the spatial / transverse profile is also improved.

wler

Some experiments with the blue LD's

Post by wler » Thu Sep 23, 2010 12:42 pm

Thieu wrote:You see that the spectrum is not single mode. There's one intense mode at the left and a couple modes at the right. The notable thing is that the mode spacing is not equal for the modes at the right! The effect is even more dramatic later in the video, the mode structure becomes a fuzzy mess. The fuzzy mess is not bad though, because it won't add up to banding in the hologram (there is no periodicity in their sum), but to a general background. No problem at all for dcg. Maybe the fuzziness is also slightly there in the mode at the left, causing the decrease in coherence length you report, without any banding.
)
That an irregular fuzzy "mess" does not lead to banding is absolutely true. It will lead to a reduction of coherence length and a loss of contrast beyond that, but how much is a difficult question, as the answer depends on the details of the spectrum. The irregular chaotic mess typically spreads over a relatively wide range, therefore the useable coherence length would be very short if the spectrum would consist only of that. But if there is a strong peak and the fuzzy chaotic mess is strongly suppressed, the contrast would be just reduced a bit. Unfortunately all possible situations occur when scanning through the current beyond approx 250mA.

This is a general topic I always wondered about ... but hard to quantify without a detailed analysis: what are the actual limits on spectral purity for practical day-to-day holography? What percentage of a side mode would create a disturbing banding..0.1%... 1% .. 10% of the main peak ? And what amount of "side mess" can be tolerated for creating an unnoticeable reduction of contrast? It would be very interesting and important to have some guidelines... also for myself developing lasers, it would be important to know what to aim for, ie, what kind of spectra are good enough. Maybe I am putting more effort then necessary.

Obviously the easiest way would be to systematically investigate by experiment and shoot lots of test holograms. Unfortunately I don't have the space to do this concurrently with spectrum measurements.
Thieu wrote: Do you think it's worth the effort to test an ecdl setup with this improved beam?
Obviously yes. I got some paper where the authors propose to use a tilted cylindrical lens to improve the wavefront hitting the grating. One other possiblity would be to use a thin slit in order to cut down on transverse modes. A problem I encountered when playing with the cavity is that it tends to become mechanically unstable and long, while shorter ones seem preferable.

Joe Farina

Some experiments with the blue LD's

Post by Joe Farina » Thu Sep 23, 2010 5:46 pm

wler wrote:This is a general topic I always wondered about ... but hard to quantify without a detailed analysis: what are the actual limits on spectral purity for practical day-to-day holography? What percentage of a side mode would create a disturbing banding..0.1%... 1% .. 10% of the main peak ? And what amount of "side mess" can be tolerated for creating an unnoticeable reduction of contrast?
As for actual coherence length, I guess it would depend on the kind of holograms to be made. The 445nm wavelength happens to be very good for DCG, and most people seem to be content with fairly shallow and broadband images, so the coherence could be much less. Typically with silver, more depth is expected, but 445nm isn't especially good for commercial silver emulsions, considering that they are usually optimized for 633nm red, and a lot of fairly cheap power is available at 532nm. It would be good for full-color silver holograms if the coherence is long enough. But very few people are into that. It seems to me that the 445nm diode could really shine with respect to homemade DCG holograms.

Cristiano_Perrucci

Some experiments with the blue LD's

Post by Cristiano_Perrucci » Mon Sep 27, 2010 3:49 pm

Joe Farina wrote:As for actual coherence length, I guess it would depend on the kind of holograms to be made. The 445nm wavelength happens to be very good for DCG, and most people seem to be content with fairly shallow and broadband images, so the coherence could be much less. Typically with silver, more depth is expected, but 445nm isn't especially good for commercial silver emulsions, considering that they are usually optimized for 633nm red, and a lot of fairly cheap power is available at 532nm. It would be good for full-color silver holograms if the coherence is long enough. But very few people are into that. It seems to me that the 445nm diode could really shine with respect to homemade DCG holograms.
I don't know if spectral sensitivity of homemade silver emulsions is comparable to commercial products, but I have to say that pinacyanol red-sensitized emulsions work fine with the Nichia blue laser. In fact silver emulsions are intrinsically blue sensitive at short wavelenghts above 500nm, so in my opinion the same result I got with my experiments might be achived using plain emulsion without the add of any sensitizing dye.

A drawback when using blue laser with silver emulsion is the deep-blue/UV shift due to gelatin shrinkage.

To take a decent picture I had to breath the emulsion, note the fantastic nichia-blue tint where the emulsion wasn't wet!

Exposure energy : 1mJ/cm^2
Development : ascorbic acid-metol developer
Bleach : Fe-EDTA

Image

DJ Mathson

Some experiments with the blue LD's

Post by DJ Mathson » Mon Sep 27, 2010 7:43 pm

Cristiano_Perrucci wrote:A drawback when using blue laser with silver emulsion is the deep-blue/UV shift due to gelatin shrinkage.
Do you mean there is a general shrinkage when a blue wavelength is used with silver emulsions? Or, do you mean the general shrinkage due to some development processings?

Colin Kaminski

Some experiments with the blue LD's

Post by Colin Kaminski » Tue Sep 28, 2010 4:33 am

Another drawback is the strong scattering from silver crystals.

Cristiano_Perrucci

Some experiments with the blue LD's

Post by Cristiano_Perrucci » Tue Sep 28, 2010 6:23 am

DJ Mathson wrote:
Cristiano_Perrucci wrote:A drawback when using blue laser with silver emulsion is the deep-blue/UV shift due to gelatin shrinkage.
Do you mean there is a general shrinkage when a blue wavelength is used with silver emulsions? Or, do you mean the general shrinkage due to some development processings?
Yes. I mean the general shrinkage due to chemical process.
Colin Kaminski wrote:Another drawback is the strong scattering from silver crystals.
With my homemade emulsion the scattering seems not to be an issue, the light haze you see in the picture is due to overbleaching and not focused light coming from a white led during hologram reconstruction. Also the emulsion used for the blue laser experiment was carefully optimized for red HeNe. Anyway If I decide to go on with blue laser experiments, I'm decreasing silver grainsize in order to get more clearness at lower wavelenghts.

The actual problem I see is a simple, efficient and cost effective beam shaping for this laser.

Joe Farina

Some experiments with the blue LD's

Post by Joe Farina » Tue Sep 28, 2010 9:54 am

Cristiano_Perrucci wrote:In fact silver emulsions are intrinsically blue sensitive at short wavelenghts above 500nm, so in my opinion the same result I got with my experiments might be achived using plain emulsion without the add of any sensitizing dye.
Yes, I would think so.

Locked