Lippmann Plate Prism Mounting Guide

The Lippmann Guide Part 4 - Mounting and Viewing

Once the plate has dried, you will be able to see the colors when angling the plate into a light source, the more diffuse the better. If you want to stop here, that's absolutely fine! But Lippmann plates, in their absolute top form, also have a "Wiener" prism attached to them. This solves a few issues - firstly, it separates the "surface reflection" (the glare of the light source off the plate) from the "metallic reflection" (the colors produced from the fringe pattern recording in the emulsion, aka the part that you want). Secondly, the prism stops the plate from redshifting or blueshifting in response to changes in ambient humidity. Although this is admittedly one of the coolest aspects of the process, the novelty has well worn off for me to the point where it's "kind of annoying to deal with". And lastly, it protects the image. The dry emulsion is incredibly fragile, and prone to scratches. Getting any oil on the surface (especially from fingers) will temporarily totally obscure the colors in that area, and it can be a bit of pain to clean off.

It is said that the prism should be about 10 degrees. I haven't played around with different angles all that much, but this seems "about right". Early on when I barely had the money for it, I accidentally commissioned a bunch of 4x5 prisms with a 6 degree angle instead of 10, and these still worked fine - they were just a bit more annoying to view, since the angle of the prism's surface reflection is a lot closer to the metallic reflection.

Sourcing The Prisms

Just Buying Them

I've been purchasing prisms from Nanyang Jingying Trade Co., Ltd. on and off for a few years through Alibaba, with generally positive results. After factoring in shipping, 6x6 prisms come out to about $5/piece, and quarter-plate about $6. Every once in a while there is one with a scratch or whatever, but it's pretty hard to argue with the price. You do have to order in bulk, though, which may not be ideal for someone who just kind of wants to play around with this process.

Making Them

You can actually pretty easily make your own prisms, by using two piece of glass and filling in the volume between them with a polyester resin. These tend to have a very slight amount of chromatic aberration compared to their glass counterparts, but it's very subtle. Since I usually coat the emulsion on 8"x10" plates and cut them down, I have a lot of extra 6cmx6cm plates that I don't really know what to do with, so I will clean off the failures and use them as glass stock for acrylic prisms.

Originally I used Eager EP4101, which was recommended in the original documentation by Hans and Darran. This appears to be no longer available. After some research, I picked out [INSERT BRAND HERE], which as far as I can tell, is exactly the same product, as the company seems to have changed hands.

You can use some 10 degree triangular spacers to tape up some glass roughly into prism-shape, and then pour in the resin to fill them up. Smaller prisms don't really have issues with bubbles, as they have plenty of time to float to the top before the resin starts to harden. You might need to look into a vacuum chamber or pressure pot if you have trouble with bubbles getting stuck in larger prisms.

Eventually I got sick of taping these things up all the time, and made a silicone mold of a few existing prisms. This worked extremely well, and is probably worth the investment if you plan on making a lot of them.

These prisms do seem to fail spontaneously sometimes. I do not know exactly what is the cause, but I believe it might be due to rapid changes in temperature. Prisms that have been perfectly fine hanging out in my house for years suddenly would experience a separation of the polyester from the glass if I would, say, mail the item to someone. Polyester might not be the most robust choice for this application, but it should work fine if you're just experimenting with the process.

Casting polyester prisms in a silicone mold.

Testing the Image

A lot of times, a plate that looks great without a prism, doesn't look so great with one on, particularly with underexposed plates. The issue is, when the surface reflection and the metallic reflection are combining, sometimes the surface reflection does a bit more heavy lifting than it would seem. There have been more than a few times where I cemented a prism on with Canada balsam, only to immediately regret this, as the image suddenly turned dim and weak. Additionally, particularly with mica plates, the image appears overly yellow when viewed out in the air, and much more balanced and colorful with a prism on -- this can make it difficult to discern when you've actually made a really nice plate. To avoid this, I like to test out the plate first before committing to Canada balsam.

Early on, I would just dump the plate in a tray of some sort of organic solvent - isopropyl alcohol, ethyl acetate, acetone, toluene, pretty much any of these work. However, having containers of this stuff just kind of hanging out in your house is a recipe for an accident. Additionally, the tray of solvent would eventually start to accumulate dust and whatnot.

Now, I prefer to just take some solvent and put a few drops of it on the plate, and then lay a prism down on top of it. I use lab-grade toluene for this, because it matches the final representation of the image versus Canada balsam pretty well, and it leaves a minimal amount of residue once it dries. Don't worry about the residue too much either - it goes away if you do decide to fully mount it.

If you have a lot of shadows that are totally clear, it might be useful to repeat the process, but use a black and solvent-resistant (not plastic) backing under the plate. I like to use an old piece of anodized aluminum leftover from my wet-plate days.

The Viewer

You can use any old diffuse light source to see these things, but a viewer makes it about 100 times easier to see with your eyes, and about a billion times easier to photograph them.

There's nothing super special or complicated about the viewer - it's a box with a cheap Fresnel lens on top, and a holder to angle the plate slightly. You can make it out of a cardboard box if you want to, there's nothing super specific about the dimensions I chose. I have a lamp positioned above the box with a bright, high CRI bulb. Positioning this close to the lens's inflection point (about 18 inches in my case) projects a very bright and even spotlight onto the plate.

Plans for a 1/8" (3mm) laser-cut version can be found here. By my own admission, after making a bunch of these with the intention to sell them, the 1/8" version is a little flimsy. I am currently redesigning it for 1/4" (6mm) wood, which will be available Soon™. Since I have also recently started making these in quarter-plate format, the old version was a bit too small to accomodate plates that size. I will also make the quarter-plate version available Soon™.

Cementing and Backing the Prism

If you're happy with how the plate looks, and you're ready to commit it to a prism, it's time to cement it with some Canada balsam. Beware, it's not difficult to do, but it is messy and lengthy.

Canada balsam is basically tree sap with a refractive index very close to that of glass. It's really sticky. Don't even think about opening that bottle unless you have some acetone and paper towels on hand. Washing your hands with soap will not work. Without acetone, the only way your fingers will stop being sticky, is when they've accumulated enough lint to fully cover all exposed sticky zones. You might get away with using gloves, but in my experience, the reduced dexterity involved with using them just causes me to accidentally touch the stuff more than I would have otherwise. Latex will straight up rip when trying to separate your hands. You have been warned.

I like to do all my work on an old piece of plywood, since it's flat, I can carry the workspace around, and I don't mind if it gets Canada goop all over it. Typically I will lay down a few layers of paper towels, with the plate flat on top. I've used a few different sources of Canada balsam, but Wood Finishing Enterprises seems to have the most competitively-priced product at the time of writing. I use the real shit, not stuff thinned with toluene or any other solvent.

For a 6x6 sized plate, I'll usually pour about a "quarter" sized blob in the middle of the plate. For larger plates, (especially if they are not square), start by pouring a little bit in the middle, and then towards a corner and back for each of the corners - this will draw a bit of an "X" shape. Slowly lay the prism down, first contacting the balsam blob at a bit of an angle, and then flatting it while applying firm pressure down. The balsam will spread across the plate and squish out the sides on to the paper towel that you hopefully laid down. While the downward pressure should be firm, don't "force" it too hard - applying too much pressure will overcompress everything, and when you finally do let off, air is liable to enter the sizes as everything relaxes a bit. Once the plate is fully covered, you can stop pressing it -- if more Canada balsam wants to flow out, it will do so on its own accord. Even if it doesn't seem like it, the plate is still pretty slipslidy, and the viscous nature of Canada balsam means the prism will be more than happy to start sliding off the plate at 1mm/minute the second you walk away. It's best to place a few objects against it all to gently brace it. After a few hours, wipe the excess goop off the sizes, and continue to brace it.

If you have a few bubbles between the plate and prism, ideally you want to squish them out when initially putting the prism on. If you still have some afterward, don't panic quite yet - very often, they will reduce in size over the next 24 hours until they are imperceptibly small. If you still have some after a day or so, and they really annoy you, you might need to remove the prism and start again.

The drying process is slow - typically after a couple days the plate/prism will be firm enough for general handling/admiration, but truly it will take a couple of weeks before the edges are truly dry. Any remaining prism cleanup can be achieved with acetone, Windex, paper towels, and a dream.

The final black backing can then be applied. I like to use black enamel for this purpose, and to put a good 3 coats of the stuff at least.

Development

We can't use just any old developer to develop these things. For the Lippmann colors to work at all, it's essential that the developed silver grain to end up in the exact same place that the exposed silver halide grain was. As it turns out, your bog standard black and white developer starts to do some really weird stuff on a micro-crystal level. They technically do work still, at least a little bit, but they will be incredibly dim with barely any trace of color. We'll be needing to use developers from the "physical" or "solution physical" classes to get good results.

Note that no colors will be visible at all until the plate has fully dried.

GP-2

Unless you're deliberately experimenting with other developers, GP-2 is hands down the winner, and the only one I explicitly recommend using. It has a pretty reasonable shelf-life, it doesn't have any explicitly toxic chemicals (don't drink it though), and it tends to produce the best "whites" out of any developer I've tried.

Stock Solution

Phenidone, 0.2g
Hydroquinone, 5g
Sodium Sulfite, 100g
Potassium Hydroxide, 5g
Ammonium Thiocyanate, 12g
Water, 1000mL

*The "official" version calls for "methylphenidone", aka "Phenidone B". Phenidone A is way easier to source. There isn't really much of a difference in terms of quality. I think Phenidone B just lasts a little bit longer.

I personally have one modification to it - I omit the ammonium thiocyanate from the rest of the GP-2 stock, and instead keep it i a separate 3% solution. Being able to adjust the amount of silver solvent on the fly like this can be super useful, since there can be a little variance from batch to batch.

Typically I make the stock in a 500mL size rather than the full liter. It will begin to yellow after a month or so, which means it's beyond its expiration - even when shooting tons of Lippmann plates I never get through it all before I have to toss it.

Working Solution

Water, 50mL
GP-2 Stock, 10mL
3% ammonium thiocyanate, 5mL

Typically I would notate a working solution like this as "50+10+5"

The amount of silver solvent required will vary based on your crystal size - so unless you have an electron microscope, we'll only really be able to figure this out empirically. Too much solvent causes an overall muddy-gold tone to overpower everything. Too little solvent tends to produce dimmer, contrasty scenes with pure colors.

Development time can also vary. At 20C, development times can range from 90 seconds to 4 minutes. Usually 2m30s is a good starting point.

Left: 60 minute exposure, 50+10+5, 2m30s development

Right: 90 minute exposure, 50+10+3.5, 3 minute development

Reducing the solvent concentration produced more faithful colors, while taking a hit to brightness, thus requiring an increased exposure time to compensate.

Fixing

Fixing the plates has always been... weirdly contentious? Even going back to Gabriel's day. Apparently skipping fix and just letting the plate print out doesn't totally mess things up. But I grew up developing black and white film, so skipping the fixing step just feels wrong. Additionally, Darran Green has noted that his older mercurial Lippmanns that were unfixed ultimately experienced some serious degradation.

That all being said, the developed out silver particles are very small, meaning they are easily attacked by the fixer - so it's best to keep it to a minimum. Fix in a 5% solution of hypo for 60-90 seconds, and then give it a long wash.

(Mica only) Second Mica Removal

If you used the mica mirror, there is still probably a thin layer of particles that are stuck resolutely to the gelatin. These can be pretty easily removed. I wet a cotton ball with water, and then put a few drops of dish soap on it. Gently massage the plate - the mica should detach pretty easily. Just be gentle as to not scratch the gelatin.

Swelling

Swelling is ultra-important if you used the air-gelatin method, and is either less important or not even necessary with the mica method. Air-gelatin plates experience gelatin shrinkage during development, which will cause the colors to be heavily blue-shifted unless we soak some extra stuff to prevent the gelatin from collapsing all the way down. We have a few options:

Citric Acid
- This is my go-to these days. It's basically the exact same as glycerin in performance, but it's easier to measure out since it's a powder and not a super thick goop.
- Will grow stuff after a few weeks - dump it and remix if it starts to get cloudy.
- 0.5% is a good starting point
Glycerin
- This is the tried and true option that has been used for many years.
- Measuring it out is really annoying
- Toss when you start to see stuff growing
- Try 0.5% to start.
Triethanolamine (TEA)
- You might be temped to use this, since it is also used as a hypering agent. However, I'm not a huge fan - it seems to dry somewhat unevenly, with the colors blueshifting or redshifting. I think the holography guys get around this by squeegeeing the plate before setting it on the drying rack, but I've never tried this.
Sorbitol
- I have heard this can be used, however, I have not myself tried this.

Regardless of your choice in swelling agent, they all work more or less the same - the higher the % concentration, the more redshift you achieve. The goal here is to bridge the gap between the amount of shrinkage from development, and where the plate naturally rests in your local ambient humidity. In my experience there isn't really a great way to remove this stuff from the emulsion once its dried, so you might have to shoot a couple plates while tweaking the concentration.

In my opinion, this is the step that makes-or-breaks air-gelatin plates. It helps to keep a good log of additions/dilutions of the swelling solution. I tend to still be a bit conservative with the concentration. If the plate is just a bit too blue, you can always run the shower on hot for a few minutes to get the room nice and humid -- this lets the plate slowly redshift, until they look just about right. Then you can slap a prism on there and seal it in like that forever. If you overswell the plate, there's a little less wiggle-room to work with. You can always let it sit on a heater or something, I guess.

Getting the swelling exactly right is much less of a concern with mica. Depending on the hardness if the gelatin, it may not need it at all. Most of your color is coming from the contact area under the mica particles, and these are "metallic fringes" that do not require swelling. However, if there is a significant amount of gaps between the particles, air-gelatin fringes will form here - and those will throw off your colors unless properly swelled. It's a bit difficult to see because it's very subtle, but the two sets of fringes catch the light at ever so slightly different angles.

If your gelatin is a bit stickier, you might not even need to swell at all, if pretty much all the gaps are filled. If this is your first mica plate, I'd skip this step and just see what happens.

To swell the plate, just rock it in a tray of your swelling solution for about 2 minutes, and then transfer the plate directly onto a drying rack. If you have some gentle airflow, dry time is typically about 20 minutes.

The difference between a properly swelled plate, vs. an underswelled (blueshifted) plate