Selenetype, a variant of the Van Dyke silver print

AbstractComplete guide to Selenetype, the stable variant of the Van Dyke silver print that I developed starting from the historical formulas of Nichol and Namias. You will find the chemistry explained by a professional, the two-solution recipe that keeps indefinitely, exposure times, development and fixing, and finally control of the dry-down effect with wax, linseed oil and fumed silica. No ready-made recipes: only the reason why each step works.

When I began studying the silver technique called “Van Dyke”, I immediately realised that the chemistry had serious problems and that recipes and clearly mistaken claims were circulating online.

First of all, let’s remember that the first to invent a very similar technique was, again, our chemist and photographer Herschel — the same one behind the cyanotype. The technique was later optimised and patented by W.W.J. Nichol. At first it was simply called Argentotype, but in the early twentieth century the makers of sensitised paper popularised it under the name Van Dyke purely for marketing purposes.

What are the problems with this formula?

The first is that the brown obtained is closer to a brown/violet than to a true Van Dyke brown; the second is that the formula is not stable. Even taking every precaution — amber bottle, cool dry place, etc. — after a day you’ll already see a heavy black precipitate. That precipitate is nothing but metallic silver. In practice the formula is stable only for a few hours after mixing the three parts; then the ammonium ferric citrate reduces the silver to metallic silver even without direct light. As you can imagine, obtaining a quality print becomes practically impossible without knowing the amount of silver in the solution.

That is why I wanted to experiment starting from this formula, trying to create my own personal recipe. I have called it Selenetype, after Selene, the name of the Greek goddess of the moon, depicted as a beautiful woman in long, flowing, silvery robes. As you will see, at the end of the process I developed the print looks more like a black-and-white / dark brown than the brown/violet of the original formula. But above all, Selenetype is a stable formula, storable indefinitely.

Nichol’s original formula

The original formula calls for making three bottles: one containing tartaric acid, one containing ammonium ferric citrate (FAC) and the other silver nitrate. These are then mixed together. There is no logic in separating the tartaric acid from the ammonium ferric citrate: they are not incompatible — quite the opposite. Adding tartaric acid lowers the pH of the solution and prevents the formation of mould (a classic cyanotype problem). Below, the historical formula.

Historical formula (Nichol) — three separate solutions

Reagent	Quantity	Solution volume	Final conc. % w/v
Ammonium ferric citrate	9 g	33.3 ml	9.0%
Silver nitrate	3.8 g	33.3 ml	3.8%
Tartaric acid	1.5 g	33.3 ml	1.5%
Final volume: 100 ml

For those who aren’t in the trade, remember that % weight/volume in chemistry indicates the amount of reagent in grams per 100 ml of solution. In other words, 9% means 9 grams in 100 ml of solution (generally water). Here we prepare three “stock” solutions of 33 ml at triple concentration; when we combine them, the final concentration will be the one given in Nichol’s patent. Compared with Herschel’s original formula, Nichol added tartaric acid in order to stabilise the chemistry. In a sense the chemistry is stable for a few hours — certainly better than adding nothing — but it certainly cannot be kept for more than a few days.

For analytical correctness it should be noted that Nichol used brown ammonium ferric citrate, an old form that became obsolete with the discovery of the green form, which is more efficient and faster. I have not tested the stability of Nichol’s formula with the brown version: if anyone has done this test, do get in touch. Certainly the brown version of FAC will produce a much less reactive, and therefore slower, solution.

Namias’s modified formula

Rodolfo Namias (in some English texts mangled as “Niamas”) is a giant of photography and chemistry, author of a boundless number of formulas, besides having taken part in the creation of many legendary films such as Kodachrome.

Searching in an old photography book (Cassell Photography, page 484) you’ll find a Van Dyke formula modified by Namias that calls for green ammonium ferric citrate, silver nitrate and citric acid. The formula is not easy to follow, since it uses both ounces and grams; for the calculations I used the gram version. Below, the final concentration of the three components:

Modified formula (Namias)

Reagent	Final conc. % w/v
Ammonium ferric citrate	22.0%
Silver nitrate	5.5%
Citric acid	5.0%

Looking at the original document, you can see that Namias uses only two solutions: one containing citric acid and ammonium ferric citrate, the other containing silver nitrate. This reiterates what was already said: there is no sense in separating the citric or tartaric acid from the ammonium ferric citrate. I have tested the formula personally: using citric acid at triple the concentration of Nichol’s formula strongly stabilises the solution. Namias was surely aware of the intrinsic instability of Nichol’s formula… Although it is more stable, the solution still does not last more than a few days. Moreover, as we’ll see, this change of chemistry affects the colour, which starts to shift toward a brown/black.

The Namias formula greatly increases the amount of ammonium ferric citrate. In both Nichol’s original formula and Namias’s modified one we are well beyond the minimum amount of FAC needed for the reaction to occur completely. We must, however, consider that the reaction takes place in the paper fibres — not an ideal environment — so it becomes necessary to work in a large excess of FAC relative to the silver present. Increasing the FAC also improves the print’s contrast.

The Selenetype formula

Hoping to obtain an even more stable formula, I tried to optimise Namias’s formula. First, I slightly reduced the amount of ammonium ferric citrate: beyond a certain threshold (around 20–25%), FAC becomes “self-screening” (see Mike Ware’s studies). Being a very coloured salt, if the layer is too thick UV light cannot penetrate all the way down into the fibres to activate the silver beneath. Consequently I also slightly reduced the amount of silver.

I then greatly increased the amount of citric acid. The formula, as expected, becomes even more stable and — contrary to what I thought — keeps a high reactivity (very short UV exposure times are needed). Despite everything, a small amount of metallic silver or silver citrate precipitates in the solution. So I abandoned the idea of a single stable formula and opted for two solutions, stable indefinitely. Increasing the citric acid was not in vain: it shifts the colour definitively toward a near black-and-white and allows a very respectable tonal scale.

The Selenetype formula

Reagent	Final conc. % w/v
Ammonium ferric citrate	20.0%
Silver nitrate	5.0%
Citric acid	11.0%

Below, the formula split into two separate solutions to be combined at the moment of coating. Unlike cyanotype’s unified solution, stable for only a few minutes, here you can take your time: the solution is very stable.

Two-solution recipe — for 10 ml

Solution A — Ammonium ferric citrate + Citric acid	3.0 g ammonium ferric citrate + 1.65 g citric acid in 10 ml of distilled water
Solution B — Silver nitrate	1.5 g in 10 ml of distilled water
Mix 2 ml of A + 1 ml of B, or 1 ml of A + 0.5 ml of B.

Exposure

Exposure time, as we saw in the cyanotype article, depends on many factors. In any case the Selenetype formula is very reactive: a few minutes will be enough. To give you an idea, in my case Selenetype needs an exposure time about 1/3 of that required for cyanotype.

Coating the emulsion and using a wetting agent

With the Selenetype formula, using a wetting agent (Tween 20) produces very different results. If you don’t use a wetting agent, the silver penetrates the paper less: the result is an image with almost no grain and very soft. Using a wetting agent lets the emulsion go deeper into the paper; the image will be sharper because the silver is locked between the fibres, and a little more grain appears (though still very low).

I have run various tests. If you don’t want to go crazy coating the emulsion, I recommend brushing on the wetting agent first (at a very low concentration, between 0.05 and 0.1%) with a dedicated brush. Coat your emulsion straight after. The exposure time will also change slightly between the two cases.

One clarification: what I’ve said about the wetting agent applies only if you do not use fumed silica (which I discuss later). The two additives have opposite effects — the wetting agent carries the silver deep, the silica holds it at the surface — and they should never be used together.

Development

Normally, for Van Dyke, plain water is used to wash the print. Some recommend citric acid at 1–2%. I have run several tests to optimise this phase.

Using citric acid at 1–2% can make chemical sense, since as a mild chelator it helps remove traces of iron from the print. The big problem is that at this step you still have traces of unreacted silver nitrate; lowering the pH heavily means producing nitric acid, which will dissolve the silver in your print. Within just a few minutes you’ll see your image lose optical density, a sign of silver loss.

I also tried sodium EDTA, a far more effective chelator than citric acid, with an acidic pH between 5 and 6 that ensures the silver is not dissolved. Unfortunately, although chemically it is the best product, sodium EDTA visibly reduces the print’s tonal scale. So I preferred to drop it.

I also tested distilled water. With the classic or Namias formula it works perfectly; on the contrary, with the Selenetype formula (very acidic) random silver spots will form on the print. This likely happens due to a pH “shock”: the emulsion is very acidic (it contains 11% citric acid) and we are washing with distilled water at pH 7, hence an unwanted silver precipitation.

I solved the problem by using citric acid at a concentration between 0.05 and 0.1%. This way the print washes easily and you won’t lose silver, thanks to the low acid concentration.

Fixing

This is a very standardised step: there’s not much to invent, just a few precautions. The classic fixer is sodium thiosulfate at 2%. This compound reacts with the unreacted Ag⁺ silver and turns it into Na₃[Ag(S₂O₃)₂], a silver salt very soluble in water that is washed away. It ensures the silver nitrate won’t react with light over time, staining the print.

You must, however, be careful: it’s true that at first sodium thiosulfate will remove only the unreacted silver, but in the long run it will also attack the metallic silver in the print. So never go beyond 1–2 minutes for this wash.

Also, during this phase you’ll see the print darken immediately. This is due to the removal of the iron traces not yet eliminated in the previous step: ammonium ferric citrate has a “masking” effect, and its complete removal leads to denser silver deposits and a darkening of the image.

Final wash

As you can guess, leaving traces of sodium thiosulfate in the paper could fade the print over time. Very long washes in tap water are therefore needed (at least 30 minutes). If you want to speed up the process, you can treat the print with a slightly alkaline solution: it will widen the paper’s pores and help remove the sodium thiosulfate faster.

The dry-down effect and the use of wax and linseed oil

A problem common to all silver-based historical prints is the dry-down effect. Van Dyke, salt print, kallitype, Selenetype: it makes no difference. In practice, when the print is wet it looks beautiful, but as it dries the blacks or browns lose depth.

This happens because the paper’s pore, previously full of water, empties as it dries, giving this effect of reduced colour depth.

There are two solutions: Renaissance Wax, or a mixture of turpentine and linseed oil.

In the first case you must spread the wax on the print with a microfibre cloth when the print is completely dry. The colour will immediately come back to life. Besides benefiting the colour, the wax will have a protective effect on the print, thus avoiding any complicated protective toning.

In the second case, a little harder, you must mix turpentine and linseed oil in a 1:1 ratio. Again with a microfibre cloth, apply the solution to the print, leave it for 5–10 minutes and then remove the excess with a cloth.

At this point the linseed oil must polymerise in contact with oxygen and will need 5–10 days to complete the effect. The oil will penetrate deeper than the wax, but it may give the print a slight yellowish tone. There is a pre-polymerised linseed oil on the market, clearly much denser: if anyone has tried it, do get in touch.

Fumed silica: deep blacks without dry-down

There is a third approach, more radical than wax and linseed oil, to counter dry-down: fumed silica. While wax and oil act on the finished print, the silica intervenes upstream, changing the way the silver deposits in the paper.

Fumed silica is amorphous silica in nanometric particles, with an enormous specific surface area. Distributed over the paper, it forms a thin porous layer that holds the emulsion at the surface, preventing the silver from migrating deep into the fibres. The result is that the silver stays where the light formed it: the blacks (and browns) keep their depth even when the print is dry, because the pore that, by emptying, generates dry-down is gone.

Unlike wax and oil, the silica must be applied beforehand, in solid form, on the paper before the emulsion. The dry powder is distributed with a high-density closed-cell foam paint roller: the closed-cell structure does not absorb the powder and deposits it uniformly on the surface. Only afterwards do you coat the emulsion on top.

For this very reason, with fumed silica you must not use a wetting agent: Tween 20 has the opposite effect, carrying the emulsion deep into the fibres and nullifying the silica’s action. The two additives are alternatives, never together.

Notes and curiosities: colloidal silver and reactivity

Remember that Selenetype, like Van Dyke and kallitype, are techniques in which the silver particle is nanometric in size. This gives the particle a much larger contact surface than the classic silver gelatin, where particles are on the order of micrometres. The silver inside the paper will react with atmospheric pollutants and with the most common toning reagents far more easily. Classic toning techniques will have to be adapted to these peculiarities: for example, while a classic selenium toning requires the toner to be diluted from 1:20 to 1:40, in Van Dyke the ratio will need to be between 1:100 and 1:500.

For this reason, protection with wax or linseed oil can be a valid alternative.

Notes and curiosities: colloidal silver and colour

Why does Van Dyke have a colour different from the classic black-and-white we all know? And why does Selenetype shift toward a “near black-and-white”, losing the classic brown tones of Van Dyke?

Particle size also has an impact on colour. Technically we speak of Surface Plasmon Resonance: because of their nanometric size, the silver particles “choose” to absorb certain wavelengths of visible light and reflect others.

The very small particles of Van Dyke tend to strongly absorb blue and violet light. Since blue is removed from the white light hitting the print, the human eye perceives the complementary colour: brown/yellow/red.

In the case of Selenetype, the change of chemistry has altered the process of aggregation and formation of the silver particles, most likely producing particles that are still nanometric but larger. This is what plausibly makes it tend toward a dark brown.

Note — Using fumed silica also shifts the colour: the print tends toward a warm brown, close to sepia, while without silica the result veers toward a brownish-black. The explanation lies, once again, in particle size. The silica offers a kind of scaffold on which the silver aggregates: the particles form in the confined spaces between one silica particle and the next, ending up smaller and more uniform in size. Without silica, aggregation happens chaotically, generating particles that are on average larger and less uniform — hence the shift toward brownish-black.

This method is just one of many historical procedures that ennoble the image; you can explore the full picture in my section on Alternative Processes.