DIY Silver Halide Film

From HoloWiki - A Holography FAQ
Jump to: navigation, search

Jeff Blyth has online instructions for making Silver Halide Plates.

Gelatin Film thickness

In the case of MBDCG a thinner film can give a bit more sensitivity because you can have more MB dye in. If x % of dye is optimum for a dried film of 10 microns then for a dried film of say 40 microns you need to cut it down to 0.25x% and that is usually bad for sensitivity. If you leave it at x% then you kill too much object beam light. However you can get away with x% if you are able to do split beam reflection work (unlikely you have enough laser power with MBDCG unless you are making say 1 sq cm size images). Single beam transmission H’s could be OK though.

Silver Halide

If the diffusion method is tried on coatings much thicker than 7 microns it is troublesome. I have tried it on ~ 100 micron gelatin and decided that it was not viable. The problem being that the rate of diffusion decreases exponentially with time and AgNO3 can carry on diffusing into the thick film and away from the incoming bromide ions . If you leave it long enough (several minutes) for the bromide ions to catch up and combine with the furthest Ag ions near the glass then the front end may have already been in the Br bath too long and you start to get grain growth .

If you don’t catch those unreacted Ag ions at the glass end then they quite rapidly develop up in the (ascorbic acid pH 6) sensitizer bath causing bad darkening or bad fog. You can stop this by giving the film a very prolonged soak in tap water (with its Cl- ions) after the Br bath before the sensitizer bath. What you get is then virtually all the silver bromide in the first few microns and none in the gelatin nearer to the glass , you might just as well have coated it thinner in the first place.

However using the conventional process with AgBr precipitated in molten gelatin solution you still get fundamental problems if you try to coat ultra thick layers. The processing chemicals take much longer to migrate in and grain growth and unevenness is inevitable. I needed to make some experimental coatings about 1 mm thick and although I got gratings they were sadly dim. I struggled to get any grating at all. Once upon a time, Agfa produced an experimental 8E75’B’ coating that was twice as thick as usual but still only ~ 15 microns . Nobody found they could get a good result from it compared to their standard 7 micron.


Hans' Diffusion Post

As promised, here is a revision of Jeff Blyth's diffusion method that will allow you to make very bright holograms, and some theory as to why I think it works so well.

When making your own holographic plates there are two requirements for bright results that work against one another: Lots of AgBr should be in the emulsion, and the AgBr crystals (grains) that are made up from all this AgBr should be as small as possible.

It normally is very difficult to make an emulsion that complies to both these requirements and lots of articles have been written up about solving this problem. Lots of methods have been invented, but few are as reliable as Jeff's diffusion method.

Why is it so difficult to make small grains: Imagine you have a gelatin solution into which you want to introduce AgBr grains. The traditional method would be to add to this gelatin solution a solution of AgNO3 and a solution of KBr. Both would be added at the same time at a certain rate. This method is called the double jet method. When the addition of the two solutions is started, at first nothing happens. Only the concentrations of both solutions slowly increase in the gelatin solution. Above a certain concentration, suddenly lots of Ag+ ions combine with lots of Br- ions to form very minute AgBr crystals. The formation of these crystals causes a fast decrease in concentration of both the AgNO3 and the KBr. As soon as the concentration drops below a certain value, any added AgNO3 + KBr will not create new grains (as we would like), but cause the grains that are already there to grow. This growing of the grains is not desirable.

So, what I mean to say is that for new grains to form, the concentrations of the AgNO3 + KBr need to be above a certain critical value. Concentrations below this value cause grain growth and even lower concentrations do nothing at all.

As you can see now the double jet method needs precisely controlled flows of liquids to allow for concentrations to be above this critical nucleation concentration to allow as many micrograins to form. Second problem is that if there are many small grains present, any newly added AgNO3+KBr prefers to settle onto those grains rather than forming new nuclei. And here lies the difficulty in making emulsions with both small grains and lots of grains at the same time. Most of the old recipes for making Lippmann emulsions are for making very fine grain emulsions, but with a very low amount of silver in them. Manufacturers of holographic film usually keep their methods for making their emulsions a secret just because of this reason.

Now if there were a method of instant mixing a gelatin emulsion of very high concentration AgNO3 with a liquid of very high concentration of KBr (and very quickly after mixing both the excess AgNO3 and KBr could be removed), very small grains in high quantity would be virtually guaranteed. And this is where Jeff's brilliant diffusion strategy comes to the rescue: Imagine you had a very thin gelatin layer that was soaked with AgNO3 and this would be suddenly dunked into a solution of KBr, the KBr would be introduced to the AgNO3 throughout the surface of the gelatin as it diffuses into the layer. So a great many small grains of AgBr would form instantly everywhere inside the very thin layer of gelatin. If the layer is then quickly washed after this step, all excess AgNO3 and KBr are removed and thus further growing of the grains is no longer possible.

This is not the whole story by a long shot though. In normal kitchen gelatin there are often left-over chemicals from the fabrication process that actually encourage grain growth. Any chlorides present in the gelatin would hamper the formation of small grains because AgCl is a lot more soluble than AgBr and also because when the AgNO3 is added to the gelatin, the first nuclei that are formed are AgCl nuclei and that's not what we want because we want the sudden process of virgin AgBr nucleation as the gelatin is dunked into the KBr solution. Any chemicals with Sulfides in them also cause grain growth. Luckily there is a way to clean your gelatin. More about this later.

Just like there are chemicals that encourage grain growth, there are also chemicals that discourage grain growth. And that is where the second brilliant idea of Jeff comes to play. It just so happens that the dye used in the diffusion process (pinacyanol chloride) is one of those chemicals that help prevent grain growth. Adding this dye (that makes your emulsion sensitive to red laser light) to the KBr solution will help keep the grains small: as the gelatin is dunked into the KBr, the newly formed grains are quickly coated with dye molecules, preventing further settling of new AgBr onto them. One problem with the dye is that it does not like to be in water. That is why the KBr mixture is actually a mixture of water+methanol. The dye is very soluble of methanol and will stay even in solution if some water is present.

Ok, so much for theory. Here is the procedure. Rather than writing up the differences from Jeff's original procedure I will now proceed and type the whole recipe.

Washing the gelatin

For this you need a small glass jar. Fill jar with 20ml of de-ionized (DI) water and add about 2.2 gram of gelatin. Next slowly warm this mixture to about 45C until the gelatin is completely dissolved. Take a plastic tupperware and pour this liquid into it and allow it to gel. When the solution has gelled, cut this gel up into small cubes with a plastic knife. Pour about 100ml of cold DI water into the tupperware tray, rock it a little and let it sit for about 30 minutes. (this step allows any contaminants in the gelatin to diffuse into the DI water). Pour off the DI water and add fresh DI water, rock and let it stand 30 minutes again. Repeat this procedure about 4 times. When you're done washing the gelatin, put it back into the glass jar and put it in the fridge (not the freezer) for later use.

Preparing the glass

When put into a alkaline developer, gelatin does not want to stick to glass anymore. So the glass needs to be prepared for holographic use. Firstly wash a piece of glass (say 20x30cm) with vinegar (this will remove some of the grease that is on the glass). After that, vigorously rub the glass with household ammonia and be very careful not to get any of this into your eye because it will make you blind forever. Now the glass will be very clean. The next step will be to chemically treat the glass to make it sticky.

Making the glass sticky for gelatin

Add about 0.5ml of 3-amino-propyltriethoxysilane (less is better than more) to 100ml of Acetone and rub this solution onto your cleaned glass plate. Let the plate sit for about one hour and then clean it again. This time with a Ammonia based glass cleaner. Your glass has now been coated with a very thin layer of molecules that on one side stick to the glass. The exposed sides of these silane molecules have -NH3+ endings that bond well with the gelatin.


You will need a second glass plate to be able to make a nice gelatin coating. Throughly clean a glass plate of the same size as the plate that was prepared from step [2] and rub it with an automotive anti rain agent such as Rain-X. And then rub it with a clean dry towel. On two opposing edges of this plate stick a long piece of Scotch tape. (During the coating step, gelatin will be poured onto this plate and the silane treated plate will be put on top of this gelatin puddle. The Scotch tape acts as a spacer and allows a perfect gelatin coating with just the right thickness when dried.)

Preparing the chemicals

  • Mix 1g of pinacyanol chloride in 1000ml of methanol. This solution will last you a life time.
  • Mix 33ml of water with 66ml of Methanol. To this solution add 6 gram of LiBr and 2.5ml of the dye solution. Pour this liquid into a Tupperware tray that is big enough to hold your glass plate and close it.

Wear Safety Glasses

If you get AgNO3 in your eye you will be blind forever.:Take your washed gelatin from the fridge and warm it up to 45C again. When it has become completely liquid again, add 1.2 gram of AgNO3 to this solution. Often the solution becomes milky when you do this, but if you stir for about a minute, it will become transparent again.

Coating the Plate

Heat your Silane treated glass plate with a hair drier and while holding it level (USE KITCHEN GLOVES), pour a puddle of your gelatin on top of it. Quickly place the Rain-X treated glass plate on top of it and allow the gelatin to completely spread between the glass plates. After a few minutes the gelatin will gel and both plates will stick together. Now place this sandwich into the fridge and leave it there for a few hours.

Washing Baths

When doing the diffusion method it is important to stop the grain growth as soon as the grains are formed. Also it is preferable to remove any excess silver nitrate from the coating as soon as possible. So, prepare two trays of DI water to remove most of the AgNO3 and LiBr that is left over in the gelatin and one tray with tap water (most tap water contains some chloride that will precipitate with whatever Ag+ ions that are left after washing). To the tap water bath you should add a few drops of liquid dishwasher fluid.

Diffusion Step

After a few hours remove your glass sandwich from the fridge into your safe lighted room. With a plastic knife remove the Rain-X treated plate from your Silane treated plate. If all went well, the gelatin coating should stick to the Silane treated plate in a perfect smooth coating. Without waiting for the plate to dry or become warm, immediately drop this plate into the LiBr bath and leave it there for about 45 seconds. Then quickly take the plate out and transfer it to the first DI water bath for about 1 minute. Then the second DI water bath, then the tap water bath. Let the plate drip dry by setting it almost vertically against an object on your table. After about 15 minutes when most of the water has dripped from the plate you can use a cool hair drier to finish drying.


Your freshly made plate will not be sensitive enough yet for practical use. Also the gamma of the emulsion will not be suitable yet for holography. Prepare a solution of 100ml water + 1.2 grams of Ascorbic Acid (=vitamin C) + 0.4 gram of NaCO3 + few drops of dishwasher liquid. Immerse your plate into this for about 2 minutes and dry again. When the plate is dry, it is ready for use and to be exposed for the brightest Denisyuk hologram you have ever made.

Ok, so that's about it. It looks like a complicated and long procedure. But after you have done it a few times, you will find it easy and simple to do and reasonably fast. It is possible to make a number of plates in one day and store them in the fridge for later use.

This procedure addresses a number of problems in the original procedure:

  • Lots more silver will be present in the gelatin. This will make your holograms a lot brighter.

The original recipe calls for first coating the gelatin and afterwards introducing the AgNO3. This can certainly be done, but the gelatin needs to be very very hard and squeegeed well after adding the AgNO3. Otherwise AgNO3 will crystallize on the surface of the gelatin layer and prevent diffusion from taking place.

  • This method will allow for very soft gelatin layers to be made. This is interesting if you want to experiment in doing SHSG.
  • In the original method there is also some Ascorbic Acid in the LiBr+dye bath. This certainly does work, but you run the risk of developing out any AgNO3 that has not precipitated out. This causes some darkening of the plate. It is better to do the sensitizing afterwards.


  • This adjustment will allow you to do the diffusion method very fast and still give the same brightness.
  • Skip steps [1] through [7]
  • Immerse a PFG-01 plate in a solution of 20% Sodium Thiosulfate (non hardening fixer) until it has become completely transparent. And rinse in DI water and dry.
  • Prepare a solution of 1 ml DI water + 0.18g AgNO3.
  • With a laser printer transparency spread a few drops of this solution over the surface of the fixed out plate and squeegie the plate very well.
  • Start from step 8 in the above procedure.
  • You have now upgraded your PFG-01 plate to a plate that competes well with the brightest plates in the world.

I have done the original method, my adjusted method and the quicker method many times over and they give predictable results but have now switched to a completely different method (using double jet) that I don't want to write about just yet.


  • Gelatin at low concentration is much easier to coat than the 10% that is required for the above procedure. It would be interesting to try to make your fresh dry gelatin into a very fine powder in a kitchen slurpy mixer. Then load this fine powder in a very cold solution of AgNO3. The solution needs to be cold because otherwise the powder will become a sticky mass. Next run this mix through a coffee filter to drain off excess water+AgNO3. Then pour your LiBr+dye mixture over the powder that is still in the coffee filter (do catch what drips out of the filter because it can be used again). Next pour large amounts of DI water through the filter. If all liquids are rather cold during this procedure the gelatin will not clump up and it just might work. After this procedure you would have holographic gelatin that can dissolved when needed at concentrations of about 2% and coated by just pouring on a horizontal glass plate. I don't know if this procedure would work. But if it does work it would be very nice. I tried it once but made a mistake in the dark and made a mess of it. So I was not able to conclude if it is possible or not.
  • To coat a layer of 2% gelatin on glass to which some dichromate is added. Then make this layer really hard in an oven and use the shortcut method I wrote about above. The layer should be very smooth and hard enough to allow a good squeegee. I have not tried this, but if it works it will be a lot faster.

Well, that's about all I know about the diffusion method. I very much enjoyed using it. Please understand that this is by no means the only way to do the diffusion method. If you want to have a go at it, try it this way first and then experiment with your own idea's. I am sure you will come up with idea's that will improve upon this method.