Holography Lab help

[christianpytel]

Hi, I'm a student that is attempting to print Computer-Generated Holograms onto Kodak film and holographic plates. My film printer is very good, it's a LaserGraphics LFR Mark 3, which has an 8K resolution. After printing a pattern onto my film, I am trying to basically "burn in" the image onto a holographic plate. I use Ultimate U08M plates and expose them for about 200microJ/cm^2 with a continuous 633nm HeNe. Here is a photo of my setup that I use to expose the holograms, it's just fiber with tape to diffuse it https://drive.google.com/file/d/1PeFg5A ... QhXJE/view. And here is how I place the film onto the plate https://drive.google.com/file/d/1l6vOQk ... sp=sharing. I clip the top glass and the film onto the plate, with the two emulsion sides touching each other (film emulsion side down, plate emulsion side up). My issue is that even when I expose these for the correct exposure time, accounting for the film's transparency, I still obtain horrible diffraction efficiency. This is a plate that I exposed and developed yesterday (https://drive.google.com/file/d/1hPKKHH ... sp=sharing ) and it's a simple diffraction grating. It looks good under a microscope, with clear fringes. The issue is that it has 1% diffraction efficiency despite looking so good. I want to make any pattern I have from film onto these holographic plates, so the grating ensures the correct phase difference from 0 to 2pi.

I have tried many different plates and exposures but cannot seem to get a good hologram. There was a student a few years ago who did the exact same method and theirs worked, so I am at a loss for words and ideas as to why mine have not worked. I appreciate any help, thank you!

[jrburns47]

You might have a polarization problem if the film substrate is polyester and not acetate. If your laser is linearly polarized, the beam might be able to be rotated to optimize film transmission brightness if that’s the issue.

[christianpytel]

The Ultimate U08M plates use a "triacetate" substrate (under "specifications" https://shop.ultimate-holography.com/in ... &id_lang=1). I also placed the grating before and after a linear polarizer and got the same results.

[Din]

If I understand you correctly, you are generating a fringe structure in software. You then print this fringe structure onto Kodak film, and then transfer the fringe structure onto Ultimate holographic film using laser light.

I don't think polarisation is an issue, since you're not exposing either the Kodak film or the Ultimate film with two beams. If I'm right, you're generating the fringe structure in a computer, either just as a set of lines or by using FT software. Polarisation only becomes an issue when recording with two beams, where the polarisation of one beam is π/2 rotated with respect to the other.

My thoughts are:
What is the spatial frequency of your computer-generated fringe structure? You say that your printer has an 8K resolution, but without the size of the print, it's not possible to determine spatial frequency in terms of lines/mm - the usual way of stating the holographic spatial frequency of a grating. The spatial frequency is given by sin(θ)/λ l/mm(assuming one beam is normal), or, in your case, sin(θ)/0.000633 l/mm. So, what is the sin(θ) you have simulated? As an example, if I recorded a grating with one normal beam and one beam at 30 deg with a HeNe laser, the spatial frequency would be ~790 l/mm.

The actual fringe structure, recorded by two beams in the conventional manner, is sinusoidal. If the printer cannot reproduce the sinusoidal nature of the fringes, and instead prints them as a square(-ish) variation, you'll lose efficiency. The reason is that the diffracted light is based on the sinusoidal components of the printed fringe structure, ie on the Fourier transform of the printed fringe structure.

What is the modulation, ie what is the intensity variation between fringe and not-fringe? The most important parameter in holographic efficiency is the modulation. If the modulation is low, the efficiency will be low. Generally, the modulation is given by:

M = {I(high) - I(low)}/{I(high) + I(low)}
where I(high) is the intensity of darkest part of the fringe, and I(low) is the intensity of the non-fringe.

I am curious, though, why you feel you have to transfer to a holographic film. If the print can take the spatial frequency, you can simply hit the transparency with laser light, you don't need to transfer to holographic film. In fact, it may be a cause of the low efficiency, since there is a loss of information due to the transfer gamma; you may be losing some of the higher frequencies.

[BobH]

I wouldn't use tape to diffuse the beam. That's probably why your DE is low. I'd recommend using a lens to spread the beam.

I'd also recommend using index-matching oil between the emulsions, and a smoked glass backer also index-matched to the back, all to prevent Fresnel reflections.

Polarization may be an issue as Jody mentioned. The tape messes with polarization. The Kodak base may as well, but you are emulsion-to-emulsion, so I don't think that's it.

I would not tape the Kodak film to the backer. I'd make a plateholder that clamps the plate uniformly over the film, possibly by using Kodak-film spacers around the edges where the clamping force is applied.

[christianpytel]

If I understand you correctly, you are generating a fringe structure in software. You then print this fringe structure onto Kodak film, and then transfer the fringe structure onto Ultimate holographic film using laser light.

I don't think polarisation is an issue, since you're not exposing either the Kodak film or the Ultimate film with two beams. If I'm right, you're generating the fringe structure in a computer, either just as a set of lines or by using FT software. Polarisation only becomes an issue when recording with two beams, where the polarisation of one beam is π/2 rotated with respect to the other.

My thoughts are:
What is the spatial frequency of your computer-generated fringe structure? You say that your printer has an 8K resolution, but without the size of the print, it's not possible to determine spatial frequency in terms of lines/mm - the usual way of stating the holographic spatial frequency of a grating. The spatial frequency is given by sin(θ)/λ l/mm(assuming one beam is normal), or, in your case, sin(θ)/0.000633 l/mm. So, what is the sin(θ) you have simulated? As an example, if I recorded a grating with one normal beam and one beam at 30 deg with a HeNe laser, the spatial frequency would be ~790 l/mm.

The actual fringe structure, recorded by two beams in the conventional manner, is sinusoidal. If the printer cannot reproduce the sinusoidal nature of the fringes, and instead prints them as a square(-ish) variation, you'll lose efficiency. The reason is that the diffracted light is based on the sinusoidal components of the printed fringe structure, ie on the Fourier transform of the printed fringe structure.

What is the modulation, ie what is the intensity variation between fringe and not-fringe? The most important parameter in holographic efficiency is the modulation. If the modulation is low, the efficiency will be low. Generally, the modulation is given by:

M = {I(high) - I(low)}/{I(high) + I(low)}
where I(high) is the intensity of darkest part of the fringe, and I(low) is the intensity of the non-fringe.

I am curious, though, why you feel you have to transfer to a holographic film. If the print can take the spatial frequency, you can simply hit the transparency with laser light, you don't need to transfer to holographic film. In fact, it may be a cause of the low efficiency, since there is a loss of information due to the transfer gamma; you may be losing some of the higher frequencies.

My spatial frequency is, I believe, 7.6 lines per mm. Shining the laser onto the Kodak film I used to originally expose the plate, I get extremely high efficiency, upwards of 80%, so I do not understand the drastic difference between the Ultimate plates and the Kodak film I used.

The modulation is good under a microscope, although I cannot calculate it currently. On my next set of gratings that I will print onto film, I can make a section to be able to measure that.

For my fringes, if I want a good sinusoidal structure, I need to make the distance between slits much higher due to my film's poor resolution. The distance between dots (or "pixels") is 4.27 micron. Would it be better to make the distance between slits higher, or use a binary grating with very small distance?

Finally, I am using the Kodak film and exposing them onto holographic plates to create a portable Spatial Light Modulator (SLM). I create the images on the spatial light modulator and then print them to the Kodak film; however the Kodak film has poor transparency, so I use the holographic plates instead. People in my lab group have done this before and I am trying to recreate it, however there was little documentation and am looking for answers.

[christianpytel]

I wouldn't use tape to diffuse the beam. That's probably why your DE is low. I'd recommend using a lens to spread the beam.

I'd also recommend using index-matching oil between the emulsions, and a smoked glass backer also index-matched to the back, all to prevent Fresnel reflections.

Polarization may be an issue as Jody mentioned. The tape messes with polarization. The Kodak base may as well, but you are emulsion-to-emulsion, so I don't think that's it.

I would not tape the Kodak film to the backer. I'd make a plateholder that clamps the plate uniformly over the film, possibly by using Kodak-film spacers around the edges where the clamping force is applied.

When I used a lens in the past, I would get Airy disks on my Ultimate plates due to the back reflections of the top glass and Kodak film. I have never heard of the index-matching oil, so I will look into that.

I clamp the top glass to the Ultimate plate as shown: https://drive.google.com/file/d/1Va0JEO ... sp=sharing
Top glass, then film (emulsion side down), then Ultimate plate (emulsion side up).

[Din]

My spatial frequency is, I believe, 7.6 lines per mm. Shining the laser onto the Kodak film I used to originally expose the plate, I get extremely high efficiency, upwards of 80%, so I do not understand the drastic difference between the Ultimate plates and the Kodak film I used.

Well, one thing that occurs to me is that when you copy a set of fringes from Kodak to Ultimate, you may be getting diffraction off the edges of the fringes. This may "fatten" the dark part of the fringe structure (make them larger than the original). This will change the diffraction profile.. Also, my recollection of using Ultimate is that it's a very, very fine-grained material. Therefore, it's also quite insensitive and may require more than 200 uJ/cm². Another possibility is that you're "contact copying" the hologram, as well as direct copying of the fringes. If you hit the Kodak with diffuse light, some component of the diffuse light is at the right angle to not only copy, but reconstruct the diffracted beam. In this case, the rest of the diffuse beam acts as a sort of reference beam to the reconstructed diffraction. If it is contact copying, the object/reference ratio is very low, giving you low efficiency.

For my fringes, if I want a good sinusoidal structure, I need to make the distance between slits much higher due to my film's poor resolution. The distance between dots (or "pixels") is 4.27 micron. Would it be better to make the distance between slits higher, or use a binary grating with very small distance?

A sinusoidal structure involves continuous change in the intensity (following a sin wave). Clearly any printing process cannot follow a continuous sin wave intensity variation, so the printed structure is not sinusoidal. In order to answer your question, I'd suggest a binary grating with small separation. This way, the primary Fourier component will be the spatial frequency of the binary grating.

Finally, I am using the Kodak film and exposing them onto holographic plates to create a portable Spatial Light Modulator (SLM). I create the images on the spatial light modulator and then print them to the Kodak film; however the Kodak film has poor transparency, so I use the holographic plates instead. People in my lab group have done this before and I am trying to recreate it, however there was little documentation and am looking for answers.

I see (I think). Have you considered ordinary silver halide from, say, Integraf? Their kits have all you need to copy a grating (https://www.integraf.com/shop/hologram-kits ). Ultimate is not easy to work with, it's insensitive and takes several minutes to develop, if memory serves; having said this, I have made gratings on Ultimate, but several years ago (2016), but I don't have my notes with me.

These are first thoughts. If anything else occurs to me, I'll let you know.

By the way, I read your other reply. I believe what you're seeing is Newton's rings, not Airy rings.

[BobH]

When I used a lens in the past, I would get Airy disks on my Ultimate plates due to the back reflections of the top glass and Kodak film. I have never heard of the index-matching oil, so I will look into that.

I clamp the top glass to the Ultimate plate as shown: https://drive.google.com/file/d/1Va0JEO ... sp=sharing
Top glass, then film (emulsion side down), then Ultimate plate (emulsion side up).

I misunderstood. Index-matching oil should prevent the rings you saw. An index-matched piece of dark smoked glass underneath the recording plate will help too as it prevents Fresnel reflection off the bottom and scatter off the table top or fabric.

I've used Cargille oil at index of 1.64 for emulsion-emulsion contact and 1.52 for glass-glass.

[christianpytel]

My spatial frequency is, I believe, 7.6 lines per mm. Shining the laser onto the Kodak film I used to originally expose the plate, I get extremely high efficiency, upwards of 80%, so I do not understand the drastic difference between the Ultimate plates and the Kodak film I used.

Well, one thing that occurs to me is that when you copy a set of fringes from Kodak to Ultimate, you may be getting diffraction off the edges of the fringes. This may "fatten" the dark part of the fringe structure (make them larger than the original). This will change the diffraction profile.. Also, my recollection of using Ultimate is that it's a very, very fine-grained material. Therefore, it's also quite insensitive and may require more than 200 uJ/cm². Another possibility is that you're "contact copying" the hologram, as well as direct copying of the fringes. If you hit the Kodak with diffuse light, some component of the diffuse light is at the right angle to not only copy, but reconstruct the diffracted beam. In this case, the rest of the diffuse beam acts as a sort of reference beam to the reconstructed diffraction. If it is contact copying, the object/reference ratio is very low, giving you low efficiency.

For my fringes, if I want a good sinusoidal structure, I need to make the distance between slits much higher due to my film's poor resolution. The distance between dots (or "pixels") is 4.27 micron. Would it be better to make the distance between slits higher, or use a binary grating with very small distance?

A sinusoidal structure involves continuous change in the intensity (following a sin wave). Clearly any printing process cannot follow a continuous sin wave intensity variation, so the printed structure is not sinusoidal. In order to answer your question, I'd suggest a binary grating with small separation. This way, the primary Fourier component will be the spatial frequency of the binary grating.

Finally, I am using the Kodak film and exposing them onto holographic plates to create a portable Spatial Light Modulator (SLM). I create the images on the spatial light modulator and then print them to the Kodak film; however the Kodak film has poor transparency, so I use the holographic plates instead. People in my lab group have done this before and I am trying to recreate it, however there was little documentation and am looking for answers.

I see (I think). Have you considered ordinary silver halide from, say, Integraf? Their kits have all you need to copy a grating (https://www.integraf.com/shop/hologram-kits ). Ultimate is not easy to work with, it's insensitive and takes several minutes to develop, if memory serves; having said this, I have made gratings on Ultimate, but several years ago (2016), but I don't have my notes with me.

These are first thoughts. If anything else occurs to me, I'll let you know.

By the way, I read your other reply. I believe what you're seeing is Newton's rings, not Airy rings.

You called it, the diffraction grating on the plate has fatter fringes compared to the original film. I will try overexposing, but I believe you're correct that I cannot get great results due to the small-scale diffraction as I expose them. We used to have some Integraf plates until we used those up, then purchased ones from Ultimate instead due to Integraf not having stock for months. I have still made successful holograms with Ultimate, but they were very basic reflection holograms and did not use the process this post is about. My only thing left to do is overexpose them a bit and instead of clipping the glass and plate together, instead just have a large glass sheet to cover the film.