Salted paper sensitising solution

A salt print is made from two bottles, and only one of them is dangerous, expensive and worth understanding twice. This is that one: a single solid dissolved in a litre of distilled water, at a strength chosen in the 1830s and still printed by suppliers today.

Ingredient	Quantity	Form the source specifies
Silver nitrate	120 g	Technical or ACS grade. Reilly states that the highly purified grades required for photographic emulsion making are unnecessary here
Water	to make 1000 mL	Distilled or at least de-ionised, and Reilly gives the reason rather than the rule: made with tap water, the silver nitrate reacts with the chloride and the hard-water ions already in the supply and throws a cloudy precipitate of silver chloride and silver carbonate, which robs the bath of strength and makes sensitising difficult. He states no temperature for the make-up. The same reaction, run deliberately, is what clouds the first wash water after exposure.

Purpose

To do two jobs in one three-minute float, and the second is the one nobody expects.

The first job is to make the light-sensitive substance. Silver chloride is insoluble in water and in most solvents, so it cannot be dissolved and coated the way a dye can. It has to be assembled inside the paper out of two solutions applied separately: a soluble chloride first, in the salting solution, and then this bath of soluble silver. Where they meet, silver chloride precipitates on the spot.

The second job is to leave far more silver behind than the first job needs. Reilly states the requirement plainly: there must be enough silver nitrate in the sensitising solution not only to react with all the chloride present on the paper, but also to leave a considerable excess of silver nitrate present in the light-sensitive layer. That surplus is not slack in the formula. It is the second reagent of the printing-out process, and without it a salt print stalls at a darkening you would struggle to see.

Everything else on this page — why 120 g and not 60, why the paper only keeps two days, why the bath goes black, why the first wash water is the one to save — follows from those two sentences.

Recommended uses

Plain salted paper, floated for three minutes. Paper salted to Reilly’s plain gelatin formula is sensitised on exactly this bath at exactly this strength, and the pair is the simplest complete silver process in the formulary.

Albumen paper, floated for two and a half to three minutes on a bath of 10 to 12 per cent. The albumen solution and the double-coated procedure describe the sheet; this is what makes it sensitive. Reilly’s dedicated treatment of albumen sensitising is a page of its own in the register, the albumen sensitiser, because albumen brings a complication no other salted paper has: the silver bath coagulates the binder as well as sensitising it.

Any hand-salted printing-out paper, which is a larger family than it sounds. Arrowroot, matte albumen, resin-arrowroot and the albumenised salting solution all take a silver bath built on 120 g/L; what changes between them is how much citric acid is in it and how long the sheet floats. Those are set out under Variants.

Brush or rod coating, for matte papers only. Reilly recommends brushing as an alternative to floating for matte salted papers but not for glossy albumen. It uses far less made-up solution — Bostick & Sullivan get about fifty 8 × 10 prints out of 100 mL, which is roughly 2 mL a sheet against a tray that has to be filled — and it removes the risk of getting silver on the back of the sheet altogether. What it introduces is streaking, and slightly stronger solutions are used to compensate.

In yellow safelight, at the sensitising bench, and nowhere else. The salting operation is a white-light job at the sink. This one is not, and the two should not share a tray, a bench or a pair of hands without a wash in between.

When another formula is preferable

• If the sheet has to keep more than a day or two, the citric acid version of this same bath. Five per cent citric acid gives what Reilly calls the maximum preservative effect; as little as one per cent extends the life noticeably. It is under Variants, and it changes the print colour.
• If the binder is starch — arrowroot, resin-arrowroot, matte albumen — the citric acid version again, because a starch paper carries no chemically active organic material of its own and prints grey and flat without one. See the arrowroot salting solution.
• If you want a developed image, a fixed speed and an enlarger, the silver chloride contact emulsion. It is a far harder thing to make, its silver is in gelatin rather than in the paper fibres, and every rule on this page reverses: a develop-out material wants a halide excess and no free silver at all.
• If what you want is a hand-coated print rather than specifically a silver one, the classic cyanotype sensitiser is one coating instead of two, needs no fixer, costs a small fraction as much and carries none of the hazards below.
• If you are printing thin negatives on albumen paper, Reilly records the historical answer — a 1 to 1.5 per cent chloride content with an 8 to 9 per cent bath, standard between 1880 and 1900 — and then advises against it, on the ground that better-looking and more permanent prints come from stronger salting and stronger silvering even at the cost of some negatives being unprintable.

Mixing

Distilled or de-ionised water, and this is the one requirement Reilly states as a requirement. Tap water carries chloride and hard-water ions, and silver nitrate finds them immediately: the bath goes cloudy with silver chloride and silver carbonate, loses strength by exactly the amount that precipitated, and sensitises badly thereafter. You can watch the same reaction on purpose in the first wash after exposure, where the milky cloud coming off the sheet is free silver nitrate meeting the ions in the tap supply.

Weigh 120 g of silver nitrate into about 800 mL of the distilled water, dissolve, then make up to 1000 mL. That order is what a make-up volume means: the litre is the finished solution, not the water you started with. Reilly prints no order for a solution with one solute; the Photographers’ Formulary sheet, for its own 13 per cent version, states it as placing the water in the vessel and adding the silver nitrate to it, which is the same instruction.

Eye protection goes on before the bottle is opened, not before the tray is filled. Reilly puts his caution at the head of the section that gives the formula, and it is about compounding as much as about coating: always wear eye protection when compounding or handling silver nitrate solutions. The silver nitrate handling SOP is the course’s procedure and the Safety section below says which criteria put this page at Level B.

Glass, and only glass, if you have it. Reilly wants the tray and every container used for the silver solution to be glass; a stainless steel tray is specifically not recommended, and a plastic tray must be new, because plastic holds on to whatever it has been used for and a trace of developer or fixer will stain. A Pyrex baking dish is his cheap answer.

Store it dark from the first minute. Amber glass, capped, labelled with the strength and the date, per the labelling SOP. Photographers’ Formulary put the reason in one line on their own sheet: silver nitrate itself is not light sensitive, but in strong light it decomposes to a black precipitate.

🧮One hundred and twenty grams, and what half of it is for

120 g ÷ 169.87 g/mol = 0.706 mol/L, a 12.0 % w/v solution

The bath, as made

20 g ÷ 58.44 g/mol = 0.342 mol/L of chloride, a 2.0 % w/v solution

Reilly's plain salting solution, for comparison

0.342 mol/L × 169.87 g/mol = 58 g/L, a 5.8 % bath

Silver that would exactly consume that chloride

So the bath carries about twice the silver the chloride can take, and something over half of it — call it 62 g in every litre — survives the double replacement as free silver nitrate sitting in the sheet. That surplus is what the section on the mechanism is about.

Reilly states the same relationship twice, and the two statements are one rule read at its two ends. He writes that the sensitising solution must be approximately four times as strong as the salting solution, worked as 3 per cent against 12. He also reports that Talbot found about six times more nitrate necessary, with a salting solution of 2 to 4 per cent against a silver bath of 12. Divide 12 by 4 and by 2 and you get the two multipliers. The rule is a band, not a number, and this page sits at its strong end.

One assumption, stated rather than buried. The step from bath concentrations to what is in the sheet assumes the paper carries away comparable volumes of the two baths. That is an assumption, not a measurement. What is not an assumption is that the silver is in large excess, because Reilly, Ware and Talbot all say so directly, and because the arithmetic above needs only the two published strengths.

🧮Why the replenisher is twice as strong, and what that number quietly admits

Reilly’s rule of thumb is to replenish the lost volume with a silver nitrate solution twice as strong as the original bath: a litre of 12 per cent that has come down to 850 mL is restored with 150 mL of 24 per cent.

Work it forwards. That 150 mL of 24 per cent puts back 36 g. For the finished litre to be at 120 g again, the 850 mL that remained must have held 84 g — which is 9.9 per cent, not 12. The bath did not merely get smaller; it got weaker at about the same time, and by about the same amount.

That is the whole content of the rule. Reilly names two mechanisms of loss — silver nitrate consumed by reaction with the chlorides in the paper, and silver solution carried off mechanically by the sheets — and says the amount removed by both causes stays fairly constant and predictable. Doubling the strength of the replenisher is the assertion that, over a printing session, roughly as much silver goes into the paper as chemistry as leaves the tray as liquid. Near enough half and half.

The arithmetic in that paragraph is the course’s, not Reilly’s; the rule and the worked example are his. He also says what to do about the assumption, which is not to trust it: the errors inherent in the method may accumulate with unpleasant and expensive results, so serious work titrates. See Experiments.

Behaviour

It is not light-sensitive in the bottle, and it is very light-sensitive on the paper. Ware gives the reason from first principles. Silver nitrate in isolation is not photosensitive at all, because the reduction of the silver ion has no available oxidation half-reaction to supply it an electron: nitrate is already fully oxidised, and water needs a larger redox potential than the silver couple provides.

Ag⁺ + e⁻ → Ag

The half-reaction that has nothing to pair with, until there is something oxidisable nearby

Put the same solution on paper and everything changes, because now there is something oxidisable — the cellulose, the sizing, the binder — and, once it has met the salting solution, a silver halide as well. Both suppliers say the first half of this in their own words: the silver nitrate is not light sensitive while it is kept separate, and the coated sheet is photosensitive as soon as it is dry.

Two mechanisms empty the tray and they are not the same mechanism. Silver leaves as chemistry, consumed by the chloride in the paper, and it leaves as liquid, carried off in the sheets. The first lowers the strength; the second lowers the volume; both lower the amount of free silver in the next sheet, which is what the print notices.

A weak bath is a false economy, and Reilly is unusually blunt about it. A 10 per cent bath is cheaper than a 12 per cent one and reaches a dysfunctional point faster, so it needs constant monitoring and replenishment. Start at 12, hold between 10 and 12, and you can print two or three batches — his example is 30 prints of 11 × 14 inches — before you need to measure anything.

It goes brown, then nearly black, and that is organic matter rather than dirt. Some of the sizing, the gelatin or the albumen always dissolves off the paper into the bath. It reacts with a little of the silver, that silver is spontaneously reduced to the metal, and the colloidal metal colours the solution. A slightly discoloured bath may be used; a nearly black one may not. The remedy is 15 g of powdered kaolin per litre, kept permanently in the storage bottle, shaken up with the bath after each use and left to settle overnight, after which the clear solution is decanted or siphoned off. If there is no time for it to settle, filter through medium filter paper.

A surface scum is a separate fault with a distinctive signature. It shows on the exposed print as a metallic marbled sheen. Filter the solution, and check that the surface is clear before every sheet goes on it.

An old bath and a fresh bath behave differently towards albumen. Sodium and ammonium nitrate accumulate in a working bath as the double replacement runs, and those salts coagulate albumen. A freshly made bath has none of them, which is why albumen paper is most likely to start dissolving off the sheet on a new solution. Reilly records that some old manuals therefore advised adding sodium or ammonium nitrate deliberately — and notes that the advice belonged to silver-poor baths of 5 to 8 per cent and does not apply at 10 per cent and above.

The sheet it makes has a two-day life, and the clock starts here. Salted paper keeps indefinitely; sensitised paper does not. See Storage.

Image characteristics

The print is colloidal silver, not filaments, and that decides its colour. Light dissociates silver chloride into silver and chlorine, and the metallic particles aggregate only up to a certain size — far smaller than the silver filaments a developer grows. Particles that size do not absorb all wavelengths, so they are not black; what they absorb depends partly on the refractive index of the medium around them. That is why the same negative on gelatin-salted, albumenised and starch-bound papers gives three different colours, and why the colour moves when the print is fixed and again when it dries.

More free silver means more density. This is the practical form of the whole mechanism. Reilly: when excess silver nitrate is available, more image silver will be formed and a greater maximum density attained. A bath that has fallen too far gives flat prints with weak shadows and, at worst, the blotchy light patches in the dense areas that the old manuals called the measles.

Bath strength also sets the size of the precipitated grains, which sets how deep the image sits. Reilly states it for starch papers, where it matters most: stronger solutions produce larger grains of silver chloride, and larger grains are less likely to be absorbed down into the paper fibres. An image that has sunk into the fibres is scattered by them, so its maximum density falls and the surface reads matte. Strong bath plus short float keeps the image up where you can see it.

Scale length is long and the paper is forgiving at the top end. A printing-out image is self-masking: silver already formed in the shadows shields the silver chloride under it, so density arrives more and more slowly there while the highlights are still building. That is why printing-out papers carry a far longer scale than develop-out papers and have much less tendency to the soot-and- chalk look, and it is why they need negatives of a density range no enlarging paper would tolerate.

Judge the wet print against what it will become, not against what you want. Bostick & Sullivan put the working rule on their sheet: pull the print when it is about one half to two thirds as dark as the intended final image, because it darkens through washing, fixing, toning and drying. Reilly explains the same movement chemically — the image loses density and yellows in the fixer as the residual silver chloride dissolves and the refractive index of the whole system changes, and darkens and cools again as the sheet dries and the particles pack closer.

The mechanism

Step one: double replacement, inside the paper. The two solutions never meet in a tray. They meet in the fibres, and because silver chloride is almost insoluble it precipitates immediately where they do.

NaCl + AgNO₃ → AgCl + NaNO₃

Reilly's own equation: silver chloride formed in situ, which is the only way it can be formed

The sodium nitrate is soluble, takes no part in the image and washes out — though not before it has done one small job, which is to help coagulate albumen in an older bath.

Step two: light, and a reaction that wants to run backwards. Ultraviolet and blue light reduce silver(I) to metal and oxidise chloride to chlorine.

2 AgCl + light ⇌ 2 Ag + Cl₂

Photolysis, and the back-reaction that limits it

Ware is emphatic that this is reversible, and quantifies how badly. In a pure silver halide crystal with nothing to take the halogen away, the chlorine simply re-forms the halide; the photolytic silver particles stop growing at about 10 nm, and the density reached is of the order of 0.02, which is a barely perceptible darkening. A salt print made of silver chloride and nothing else would be a blank sheet.

Step three: the free silver nitrate takes the chlorine away. This is the job the surplus computed in the maths callout exists to do, and Ware’s calculation is that water and free silver(I) ions together are the major halogen acceptor and contribute the greater part of the image silver.

Cl₂ + H₂O + Ag⁺ → AgCl + HOCl + H⁺

Ware: chlorine totally disproportionated by water in the presence of free silver(I)

The chloride made in that reaction meets more free silver and becomes fresh silver chloride, which light breaks down again, and the cycle repeats. Vogel reasoned his way to the same conclusion in the nineteenth century without any of the equilibrium constants, and Reilly quotes him: the liberated chlorine unites with the silver nitrate present to form new silver chloride, light breaks that down, and the cycle begins again to be repeated over and over.

Notice what the equation also produces: hydrogen ions. Ware points out that the acidity generated during exposure is itself significant, because it could assist re-oxidation and dissolution of the silver image by the nitrate as the pH falls, and is one of the diffusion-controlled back-reactions that degrade a print-out image over long periods.

Step four: why the ratio matters and not just the amount. Ware sets out the surface chemistry that turns a quantity into a property. A precipitated silver halide is not stoichiometric at its surface; whichever ion is in excess in the surrounding solution is adsorbed onto the crystal.

Where silver is in excess — Talbot’s case, Reilly’s case, this bath’s case — the crystal surface carries adsorbed Ag⁺. The excess positive charge draws photoelectrons to the surface, so silver specks form and grow there, free of the constraints of the lattice, and the positive holes reach the surface to release halogen into a medium that will accept it. Ware calls this “sensitized” silver halide.

Where halide is in excess, the surface carries adsorbed chloride instead, and the crystal behaves as “fixed” silver halide — stabilised, sluggish, hard to print. That is not a hypothetical: it is what Talbot’s earliest fixing method did on purpose, by flooding the print with strong salt solution.

So a bath at 6 per cent instead of 12 would not make a paper with less silver in it. It would make a paper whose silver chloride sits in the wrong ionic environment, and it would barely print. That is the answer to the question everyone asks when they see how much silver this formula uses.

And the manufacturers said the same thing from the other end. Kodak’s 1928 primer states the rule for developing materials as an absolute: they must contain no excess of soluble silver and the emulsion must always carry an excess of bromide or chloride, because any excess of soluble silver fogs the whole surface as soon as it meets a developer. Then it names the exception. Solio, the printing-out chloride paper of its day, was a chloride emulsion made with an excess of silver nitrate, and the primer gives the consequence in five words: this causes rapid darkening in the light. A factory emulsion and a hand-salted sheet are made in completely different ways and they agree on this one number’s sign.

⁇Whose 12 per cent is it? Reilly and Ware read Talbot differently

Reilly (1980) writes that Talbot found about six times more nitrate necessary and established that the salting solution should be 2–4 per cent in strength and the silver nitrate 12 per cent. That sentence is where this bath’s authority comes from in most modern accounts.

Ware (2019), working from Talbot’s own Notebook P and from Malone’s description of the Reading Establishment, gives ranges rather than settled figures: salting solutions implying 0.6 to 2.3 per cent w/v with a customary trough at 1.1 per cent, and silver solutions between 23 and 18 per cent w/v, with Notebook P recording values of 20 per cent and 11 per cent. On that reading Talbot’s usual silver bath was stronger than 12 per cent, sometimes twice as strong, and his salting weaker than Reilly’s band.

What is not in dispute is the thing that matters chemically: the halide was substantially short of chemical equivalence with the silver, deliberately, and Talbot discovered it in 1834. Both authors say so, and Ware calls it the key both to the sensitivity of the paper and to fixing it afterwards.

What the course uses. Reilly’s 12 per cent, because it is the strength this page’s formula is published at, because it is what two current suppliers and the Getty Conservation Institute describe, and because it is paired in Reilly’s book with a salting solution the course also publishes. What the course does not do is repeat “Talbot’s 12 per cent” as though it were a figure read off a manuscript. It is a modern summary of a range, and a reader following the attribution back should know that the range runs from 11 to 23.

Function of every ingredient

Silver nitrate, 120 g — a 12 per cent bath, and the only substance in the formula. It is the one soluble silver salt photography uses; everything silver does in this course starts with an Ag⁺ ion that came from a bottle of it. Reilly notes that it is made by the action of nitric acid on silver metal, that it is sold as colourless crystals, and that for this work a technical or ACS grade is sufficiently pure — the highly purified grades needed for emulsion making are an unnecessary expense here.

Its quantity does two separate things at once, and confusing them is the commonest way to misunderstand the process.

The first portion is stoichiometric. Whatever chloride is in the sheet gets converted, and the silver that does it — about 58 g in a litre, against Reilly’s 2 per cent salting solution — is consumed. More chloride in the salting solution consumes more of it.

The second portion is catalytic in effect if not in name. The remaining 62 g or so stays in the paper as free silver nitrate and acts as the halogen acceptor, taking up the chlorine that light liberates so that photolysis can continue instead of reversing. This is the portion that decides how dark the print can get.

More silver therefore means more free silver, more recycling of the liberated chlorine, more image silver and a greater maximum density — up to the point where the extra is simply being washed away, since only 6 to 8 per cent of the silver in a sensitised sheet ever becomes image. It also means a bath that lasts longer between replenishments, larger precipitated silver chloride grains that stay nearer the surface, and, on albumen, faster and firmer coagulation of the binder and therefore a longer float.

Less silver means the reverse, and it fails in a recognisable order: first prints that lack brilliance and density, then weak shadows, then the measles — blotches or light spots in the dense areas that Reilly diagnoses as insufficient sensitisation from a too-weak bath or too little residual sensitiser in the sheet. Below about 6 per cent for a 2 per cent salting solution you are approaching the point where the silver only just consumes the chloride and leaves nothing free, and the mechanism above says that paper should barely print at all. That prediction is cheap to test and Experiments sets out how.

What it interacts with is a long list and every item on it is on this page: the chloride in the salting solution, which sets how much of it is consumed; the binder, which it coagulates or does not; the nitrate that builds up in the bath as it works; the organic matter that dissolves off the paper and reduces some of it to metal; tap water, which precipitates it; ammonia, which is dealt with under Incompatibilities and is the only genuinely dangerous interaction in the process; and the fixer, which removes everything it made that light did not reduce.

Water, distilled or de-ionised, to make 1000 mL. Not a filler, and not an ordinary “add water to a litre” either — this is the one ingredient in the salted paper workflow whose purity Reilly makes a stated requirement, and he gives the mechanism rather than the rule. Tap water carries chloride and carbonate; silver nitrate finds both and throws a cloudy precipitate of silver chloride and silver carbonate; the bath loses exactly as much strength as it precipitated and sensitises badly afterwards. Distilled water costs almost nothing against the silver it protects.

The make-up volume is the formula. Everything above is a concentration, and 120 g brought to 900 mL is a 13.3 per cent bath, not a 12 per cent one. On a page whose whole subject is a ratio, that is not a rounding error.

Interactions

With the salting solution, which is the interaction the process is built on. The two never meet except inside the paper, and their ratio is the design. Change one without the other and the paper changes character: a weaker salting solution under the same bath gives a thinner, less dense print with more free silver than it needs; a stronger one eats the free silver the print-out mechanism depends on. Reilly’s paired recommendation for most salted papers is 2 to 2.5 per cent chloride with a 10 to 12 per cent bath, and the pairing is the recommendation.

With the binder, in three different ways. Albumen is coagulated by silver nitrate — rendered insoluble — so a strong bath sets the surface quickly and then permeates it slowly, which is why a strong bath needs a longer float and a weak one risks dissolving the albumen off the sheet before it has set. Starch is not coagulated, so a starch paper stays permeable throughout and a long float simply carries the silver chloride deeper into the fibres, giving grey, flat prints; hence strong baths and short floats for arrowroot. Gelatin sits between the two and is discussed at length on the salting solution page, where Reilly’s and Ware’s accounts of what a protein binder contributes are set against each other.

With its own reaction products. Sodium and ammonium nitrate accumulate in the bath as it works, and they are not inert: they coagulate albumen, which is why a used bath handles albumen paper better than a fresh one. This is the only respect in which an exhausted bath outperforms a new one.

With whatever the paper gives up. Sizing and binder dissolve into the solution, reduce some of its silver to colloidal metal and colour it brown to black. Kaolin adsorbs the organic matter and takes it out; nothing takes the lost silver back.

With tap water, twice, in opposite senses. In the mixing vessel it is a fault that costs you strength. In the first wash tray it is the process working as intended — the milky cloud is the free silver leaving the sheet, and it is why that water is a silver-bearing stream and not rinse water.

With metals. Silver nitrate attacks the metal ferrule of an ordinary brush and the resulting stains land on the print, which is why Reilly specifies a wide flat Japanese brush whose bristles are bound with thread. A stainless steel tray is not recommended for the same family of reasons. The silver nitrate page records that the classification carries H290, may be corrosive to metals.

With the toner and the fixer, downstream. Everything this bath deposits that light does not reduce has to be dissolved out by the fixing bath, and how much there is to dissolve was decided here and in the salting tray. A heavily silvered sheet is a longer fix and a longer wash; the sulfite washing aid exists because of it. Toning in gold or platinum happens before fixing in this workflow precisely because it acts on the printed-out silver that this bath’s free silver made.

Variants

Reilly prints five further silver baths in the same book, and every one of them is this formula with citric acid in it. They are recorded here rather than as separate entries because the formulary register holds no ids for them; that is bookkeeping, not a judgement about the formulas.

The preservative addition — 5 per cent citric acid. Reilly’s general recommendation for extending the life of a sensitised sheet, and the one to reach for first:

Ingredient	Quantity
Silver nitrate	120 g
Citric acid	50 g
Distilled water	to make 1000 mL

Five per cent gives the maximum preservative effect; one per cent extends the usable life noticeably. It is not free: on glossy albumen paper the print colour turns reddish brown and stays reddish brown even after prolonged gold toning, and Reilly notes that purple tones were difficult to produce in the presence of citric acid — one of the two reasons professionals went on sensitising their own paper after ready-sensitised paper reached the market in 1872. Replenishment differs too: citric acid is replenished strictly on volume lost, not doubled, because unlike the silver it is not consumed by any reaction.

Arrowroot papers — 12 per cent silver with 4 to 5 per cent citric acid, floated for half a minute on a light coating and up to a minute and a half on a heavy one. Here the acid is not primarily a preservative: a starch binder is chemically inactive, and on Reilly’s account an active organic substance has to be brought in separately or the prints are grey and flat. It also gives arrowroot paper a keeping time of several weeks after sensitising.

Hübl’s matte albumen paper — 120 g silver nitrate with 15 g citric acid in a made-up litre, a 1 to 2 minute float, and again several weeks of keeping. Brush application takes two coats with drying between if the shadows come out weak.

Hübl’s resin-arrowroot paper — 120 g silver nitrate with 80 g citric acid in a litre, floated 4 to 5 minutes. Eight per cent citric acid is the strongest of the set and goes with the most inert binder in it.

The bichromated salting solution — a 10 per cent bath with 5 per cent citric acid, floated 3 minutes. Reilly gives it as the sensitising half of his contrast-raising salted paper. The course does not publish that salting solution: it contains a chromium(VI) salt, and under the chromium policy any formula containing one is a Level D historical-study entry with no working procedure. The salting solution page explains what is worth knowing about the technique without the recipe.

The modern kits, for comparison. Bostick & Sullivan supply a ready-made 12 per cent solution. Photographers’ Formulary make theirs at 13 g in 100 mL — 13 per cent — against a salting solution of sodium citrate and ammonium chloride rather than plain salt. The Getty’s description of the process as practised today gives around 12 per cent by weight, sometimes with a small amount of citric acid. Three independent sources, one band.

⁇What is the citric acid actually doing? The two accounts do not fully agree

Reilly’s account (1980). Citric acid belongs to a class of “active” organic substances — albumen, gelatin, and the organic acids citric, tartaric and oxalic — which facilitate the more complete reduction of silver chloride and additionally form light-sensitive silver salts of their own, such as silver citrate. On this reading the acid participates in the photochemistry, which is why an arrowroot paper needs it and a gelatin-salted one does not. He separately observes that it preserves the sensitised sheet, reddens the image, and increases both sensitivity and maximum density on albumen.

Ware’s account (2019). The dominant halogen acceptor in a Talbot-type sensitiser is the combined action of water and free silver(I) ions, and the organic component of the paper is not essential to the photochemistry of print-out, which he says can be demonstrated experimentally. He makes that argument about the sizing agent rather than about added organic acids, but its logic does not obviously stop at the sizing.

What can be said with confidence is the empirical half, which nobody disputes and which is directly observable: a citric acid bath yields a sheet that keeps for weeks instead of days, and a print that is redder. Reilly reports both from practice, and both are reproducible in an afternoon.

What the course does. It reports the effects as Reilly reports them, marks his mechanism as his interpretation, and does not present “citric acid facilitates the reduction of silver chloride” as established chemistry. A reader who wants to know which account is right has an experiment available to them, in the last section of this page.

☠The ammonia-nitrate bath, which this page describes and does not give as a procedure

For roughly twenty years the standard strong sensitiser for salted paper was ammonio-nitrate of silver: ammonia added to a silver nitrate solution of about 20 per cent until the brown precipitate of silver oxide that forms at first just redissolves, leaving a colourless solution of the diammine complex. The two accounts of when it arrived differ slightly and neither is disputed by the other: Ware credits the formulation to Alfred Swaine Taylor early in 1839, taken up at once by Talbot; Reilly writes that the effect was first noticed in 1840 and dates the process’s wide use to 1840–1860.

2 Ag⁺ + 2 OH⁻ → Ag₂O + H₂O

Ware: the oxide precipitates first

Ag₂O + 4 NH₄OH → 2 [Ag(NH₃)₂]⁺ + 2 OH⁻ + 3 H₂O

Ware: and redissolves as the diammine complex

It worked. Talbot recorded that the pictures were more resistant to fading, and Ware notes a colder image colour consistent with a larger colloidal particle — plausibly because the ammonia buffers the acid that exposure generates. It was particularly useful for printing from weak negatives.

The course gives no procedure for it, at any level. Ware’s cautionary note is that solutions of diammine silver can, with time, deposit a black precipitate of silver nitride — fulminating silver — which is a highly sensitive contact explosive that detonates at a touch and sometimes even when wet, and that storing such solutions is not recommended. Reilly’s independent warning is about the same substance from the other direction: some old manuals recommend “boiling down” an overloaded ammoniacal bath to rejuvenate it, he has never had occasion to do it, and it is not recommended for anyone to undertake that dangerous operation.

The practical consequences are simple and they belong in a working darkroom whether or not anyone intends to make this bath. Never add ammonia to a silver solution. Never store one that has had ammonia in it. Never heat one down. If ammonia and silver have met by accident, the ammoniacal silver quench SOP is the course’s procedure. Reilly’s own summary of the nineteenth-century additive literature — sugar, camphor, alum, alcohol, the nitrates, ammonia — is that most of these are superfluous, if not potentially harmful, and that for modern practice only the preservative additions are recommended.

Safety

Level B. The criteria that applied: a corrosive, oxidising solid weighed out in quantity; a concentrated solution of it handled in an open tray with wet sheets lifted, drained and hung; a substance whose harmonised classification carries H314 and whose aggregated classification adds H318; a cumulative systemic effect (argyria) from repeated skin absorption; and a waste stream that is hazardous to aquatic life at low concentrations and must be collected rather than drained. The classification page sets out what Level B assumes: splash goggles, eyewash within reach, experience with concentrated reagents, and controlled waste.

Eye protection is the control, and it is first for a reason. Reilly gives his caution twice in one chapter, once for compounding and once for coating, and states plainly that silver nitrate can cause permanent, irreversible eye damage and that sheets dripping silver solution pose a special hazard to eyes. That is the specific geometry of this operation: you are lifting a wet sheet above head height of anything you bend over. Goggles that seal, not glasses.

Gloves, and the reason is not the stain. Reilly is careful here in a way that is worth repeating accurately: infrequent contact and the resulting staining of the skin does not appear to be dangerous to health, but prolonged exposure leads to permanent staining, and silver absorbed through the skin is ultimately deposited around the body — conjunctivae, blood vessel walls, gums, mucous membranes — as argyria, which is irreversible. He records that consequences more serious than staining of the tissues have also been reported. Tightly fitting gloves, and hands washed immediately after any contact.

The solid, before it is a solution. Weighing 120 g of an oxidiser is the moment for the weighing SOP: no dust raised, no paper scoop left lying, nothing combustible on the balance. Silver nitrate is itself non-combustible but accelerates the burning of anything that is, and paper loaded with it and drying is exactly such a material.

What is not a hazard here, and why. Nothing in this bath produces a vapour or a gas, so ventilation is not among its controls — a general room airing is enough, and the ventilation check is for the pages that need it. Nothing is heated, so there is no hot-work or scald risk; Reilly gives no mixing temperature because none is needed. Nothing here is explosive as published: the explosion hazards recorded on the silver nitrate page — silver nitride, silver acetylide, silver fulminate — all require a second reagent that this formula does not contain and that the Incompatibilities section says to keep away. There is no acid in the bath and no alkali, so no neutralisation hazard and no route to a gas. And the ultraviolet exposure that follows is a real hazard with real controls, but it belongs to the UV unit SOP and the outdoor exposure session, not to this tray.

Naming that list is part of the assessment rather than a reassurance. The hazards of a salt print are concentrated almost entirely in this one solution and in the light that follows it.

✕How much light can you sensitise in? Two sources disagree, and one of them is talking about the bottle

Reilly: the operation of sensitising and drying must always be conducted under proper safelight illumination — for which he says yellow 60 W “bug-lite” incandescent bulbs work perfectly, bright enough to see and work comfortably, with no need for the dim safelights modern papers require.

Bostick & Sullivan: the silver nitrate in the kit is not light sensitive as long as it is kept separate, a safelight is not necessary, and normal incandescent lighting may be used during the coating process in a windowless room.

They are not describing the same moment. The supplier’s claim is true of the solution in the bottle, and their own sheet says the rest of it a paragraph later: once the silver coat has dried over the salting coat, the paper is photosensitive, and the negative is to be loaded in the darkened room.

What the course does: Reilly’s practice, from the moment the silver meets the salted sheet. Silver chloride begins forming as soon as the two coatings touch, and from then on every minute of white light is fog. Yellow light for coating, drying, loading and unloading; white light only for the brief inspections during exposure, which Reilly notes is unavoidable because the progress of printing-out cannot be judged under yellow. Incandescent rather than fluorescent for those inspections, because the chance of fogging is lower.

Storage

The made-up bath: dark, capped, labelled, with the kaolin left in it. Amber or brown glass, per the storage rotation SOP and the labelling SOP, with the strength and the date on the bottle and the date of the last titration if you have one. About 15 g of kaolin per litre lives permanently in the bottle: shake the bath and the clay together after every session and leave it overnight, and the clay settles carrying the dissolved organic matter with it. Reilly’s own arrangement is a two-hole stopper and a siphon so that the clear solution can be drawn off without disturbing the layer at the bottom.

Reilly publishes no keeping time for the solution, and the course will not supply one. A silver bath is not a dated consumable; it is maintained. What ends it is the point where it can no longer be cleared, or where replenishment has stopped keeping up with what the titration says.

The bottled product, as bought, is a different question. Bostick & Sullivan state a shelf life of at least five years from purchase for their solutions, stable at room temperature. That is a claim about a sealed bottle that has never touched paper, and the difference between it and a working tray is precisely the organic matter the paper gives up.

The solid. Amber glass, closed, cool, dry and dark, away from every combustible material and from ammonia and every reducing agent, in a container that has never held food. Kodak’s 1928 primer names silver nitrate as the outstanding photographic example of a chemical decomposed by light and notes that a solution darkens faster than the solid; the chemical page carries the detail.

The sensitised paper: one or two days, and less than that in a warm damp summer. Reilly is specific about the failure mode. The sheet yellows first — that yellow is spontaneously reduced metallic silver in a very finely divided state — then goes deep reddish brown, and finally black with a shiny greenish bronzed surface, which may take several months in dark storage. In extremely humid and warm conditions the first stage can arrive in 8 to 12 hours or less. Slightly yellowed sheets may still be usable, because the fixer removes some of the finely divided silver from the highlights. Store what you must in a tightly closed container, cool and dry, since air and moisture accelerate the decomposition; and see storing unfixed prints for the same problem one step further on.

The best storage is not to need any. Reilly’s general rule is that albumen and salted papers deliver their best results used as soon after sensitising as possible, and that for best results it is necessary to sensitise, print and process all on the same day. The nineteenth century solved the keeping problem commercially, with citric acid and with ammonia fuming, and the papers that resulted were often so silver-poor that they needed reviving with ammonia fumes before they would print at all.

Incompatibilities

Ammonia, absolutely, and this is the one that can hurt somebody. Adding ammonia to a silver solution makes the diammine complex, and diammine silver solutions deposit silver nitride — fulminating silver — a contact explosive that can detonate when touched and sometimes when wet. The course’s silver nitrate page carries the corresponding entry from CAMEO’s datasheet, a case in which ammonium hydroxide and silver nitrate treated with sodium hydroxide gave a black precipitate that exploded on stirring. Do not add ammonia, do not store a solution that has had it, and never boil one down. See the ammoniacal silver quench SOP and incompatibilities.

Every soluble chloride, bromide and iodide — including the ones in tap water. This is the reaction the whole process is built on and the reaction that ruins the bath, and which of the two you get depends only on where it happens. Separate trays, separate graduates, separate stirring rods, separate tongs, and the salting bench physically away from the sensitising bench.

Fixer, in either direction. A trace of thiosulfate carried into the silver bath dissolves silver and contaminates the tray permanently; a trace of silver carried into the fixer is not a disaster but is a stain waiting to happen on the next print. Reilly’s answer is one tray per operation, bought new and labelled, and it is the cheapest insurance in the process.

Developers and every reducing agent. A silver bath and a bottle of metol or hydroquinone should not share a shelf, let alone a splash. This is one of the reasons Reilly rejects trays that have been used for ordinary photographic work.

Alcohols. The silver nitrate page records, from CAMEO’s datasheet, that mixtures with alcohols may produce silver fulminate, which can explode when disturbed. Do not clean the tray or the coating rod with methylated spirit.

Metals. H290: may be corrosive to metals. In practice this shows up as a brush ferrule staining prints and as the recommendation against stainless steel trays, but the underlying reason is that many metals reduce silver out of solution and are themselves attacked.

Combustible material, because it is an oxidiser. Paper, cloth and wood carrying dried silver nitrate are a fire risk, which includes the blotting paper under the drying line and the newspaper under the coating table.

Waste

Every drop of this is a silver stream, and so is most of what happens after it. Reilly’s figure is the one to keep in mind: only about 6 to 8 per cent of the silver present in a sensitised sheet forms the image. The rest is washed or fixed away, and it goes somewhere.

Where it actually is. Most of the recoverable silver is in the first and second changes of wash water — the milky cloud that comes off the sheet in the first five to seven minutes is unexposed sensitiser leaving. A great deal more is in the fixing solution. Trimmings and spoiled prints carry silver too. The exhausted bath itself is the most concentrated stream of all.

What Reilly does with it, as a workshop practice rather than a regulation: save the first and second wash waters, precipitate the silver by adding sodium chloride or sodium carbonate, let the insoluble silver chloride or carbonate settle, decant the water off and send the solid to a refiner. Fixer is evaporated to a sludge or to dryness for the same journey. He notes that a typical nineteenth-century gallery recovered about 60 per cent of the silver it consumed, and that for a modern printer using small amounts, refining charges and minimums usually make recovery uneconomic — which is an argument about economics and not about whether the silver may be put down the drain.

Why it may not be. Silver nitrate carries H400 and H410: very toxic to aquatic life, with long lasting effects. Kodak’s J-214 records that under United States federal rules a liquid waste containing 5 ppm or more of silver is a characteristic hazardous waste, EPA Hazardous Waste Number D011. Collect it: one labelled container for silver-bearing liquids, per the silver-bearing waste SOP, and the general chemical waste SOP for the rest.

Local regulation decides what may go to drain and this course cannot tell you what it says where you are. See disposal.

Troubleshooting

Blotches or light spots in the dense areas — the “measles”. Reilly’s diagnosis: insufficient sensitisation, from a bath that is too weak or from too little residual sensitiser left in the paper. Nothing can be done for prints already exposed. Unexposed sheets from the same batch can be re-floated or re-brushed on a stronger solution. If it is happening at all, titrate the bath.

Paper-white round spots with hard, definite edges and no stain. Air bubbles trapped under the sheet during the float on the silver bath: those areas were never sensitised. Lift the sheet from one corner as you lower it, burst any bubble with a glass rod or a toothpick, blow the rest to the side of the tray before the sheet goes down, and start timing only when the sheet lies flat and no bubbles remain.

Paper-white spots with a light brown stain around them. The same fault one step earlier, in the salting tray: that area got no chloride, so the silver that arrived later had nothing to react with and stained the paper on its own. The two faults look almost identical and the stain is what tells them apart.

Patches of uneven density on the face of the print. Silver reached the back of the sheet — from immersion instead of floating, from a sheet lifted too fast, or from drips running round an edge — and locally sensitised it. Lift slowly from one corner so the run-off drains evenly; a properly lifted sheet hardly drops anything while it drains over the tray. Hang from two corners of the long edge, with the line at 5 to 7 degrees so the run-off collects at one corner, and blot the drop that forms there. Blotting is not tidiness: it is what makes drying even.

Streaks, in a brushed or rodded coating. The constant hazard of brush sensitising. Reload and work in alternating horizontal and vertical strokes until the solution has gone in; two coats with drying between will fix weak shadows but will not fix streaks, which have to be avoided rather than corrected.

Stains that look like the brush drew them. Silver nitrate reacting with the metal ferrule. Use a thread-bound brush, or a Blanchard’s brush — flannel folded over a strip of wood or acrylic — and keep it exclusively for sensitising, washing it in distilled water after every use. It will discolour and that does not matter.

The bath is brown, or nearly black. Dissolved organic matter reduced to colloidal silver. A slightly discoloured bath is usable; a nearly black one is not. Kaolin, shaken and left overnight, or filter it.

A metallic marbled sheen on the exposed print. Surface scum on the bath. Filter the solution, and inspect the surface before every sheet.

Albumen dissolving off the sheet during the float. A weak or freshly made bath, which lacks the accumulated nitrate that helps coagulate the layer. Shorten the float or use a stronger bath, but do not add nitrates to a modern 10 to 12 per cent solution: Reilly is explicit that the old advice belonged to baths of 5 to 8 per cent.

Flat, grey, weak prints with no black. Work through the causes in Reilly’s own order of likelihood: the rawstock is too porous, the negative has too short a density range, the salting is too weak, the bath has fallen below strength, or — on a starch paper — the float was too long and the silver chloride has gone down into the fibres.

The sensitised paper has yellowed. Spontaneous reduction, from time, warmth and damp. Slightly yellowed sheets often still print, because the fixer clears some of the finely divided silver from the highlights. Badly yellowed sheets are lost. Sensitise less and print sooner, or move to the citric acid variant.

The bath went cloudy the moment you made it. Tap water. Make it again with distilled.

Experiments

The bath-strength series, which is the experiment this page exists for. Salt one batch of identical sheets at 2 per cent, then sensitise four of them on baths of 6, 9, 12 and 15 per cent — the same three-minute float, the same negative, the same light, printed to the same visual endpoint. The mechanism section makes a specific prediction about the 6 per cent sheet: it sits close to the point where the silver only just consumes the chloride, so it should print out visibly more slowly and to a lower maximum density than the others, while the difference between 12 and 15 should be small. Record the exposures, because they will differ, and record the maximum density if you have a densitometer. This is the cheapest available test of a hundred-and-ninety-year-old piece of chemistry.

The float-time series, run separately on two binders. Half a minute, three minutes and ten minutes, on a gelatin-salted sheet and on an arrowroot sheet from the same session. Reilly’s claim is that the starch paper degrades badly with a long float — grey, flat, the image gone down into the fibres — and that the gelatin paper is much less sensitive to it. Look at the prints in raking light as well as head on: the fault you are looking for is depth, not density.

The citric acid keeping test, which is where the contested callout gets settled empirically. Sensitise six sheets from one salting batch: three on the plain 12 per cent bath and three on the same bath with 5 per cent citric acid. Print one of each on day one, day three and day ten, from the same negative, and keep the unprinted sheets together in the same closed box. Reilly predicts one or two days for the plain sheets and several weeks for the acid ones, and a redder image from the acid bath even on day one. The day-one pair is the important one: it separates keeping from colour, and it tells you whether the acid is doing something to the sensitiser or only to its stability.

Measure the bath instead of guessing at it. Reilly publishes a Volhard titration for silver-bath strength, worked out for this purpose by Irving Pobboraysky: sodium or potassium thiocyanate is run in from a burette against a measured sample of the bath, with saturated ferric ammonium sulfate as the indicator and nitric acid to keep the other reactions out of the way, until the faint brown colour will not stir away.

Ag⁺ + SCN⁻ → AgSCN

The silver is titrated out as insoluble silver thiocyanate

Fe³⁺ + SCN⁻ → [Fe(SCN)]²⁺

And the first excess of thiocyanate colours the iron(III) indicator

Multiply the normality found by 170 — the molar mass of silver nitrate to three figures — and the answer is grams per litre. Reilly’s own worked example: 0.65 N × 170 = 110.5 g/L, an 11 per cent bath. His standard is 0.1 N silver nitrate made as 4.25 g in 250 mL, which is 17.0 g/L; check it against 169.87 × 0.1 = 17.0 g/L and the arithmetic holds.

⚠The titration is a Level C operation as published, and it is the acid, not the silver

The procedure calls for 6 M nitric acid, and Reilly gives the dilution — 100 mL of the concentrated acid into 163 mL of water. Nitric acid is a Level C substance in this course: diluting the concentrated acid is exothermic, produces fume, and needs a fume cupboard and supervision the home laboratory does not have. Reilly’s own instruction with the procedure is to wear approved laboratory protection and work in a well ventilated area.

The course therefore records the method and does not present it as a home procedure. Two honest routes remain open. Buy the acid ready-diluted, or buy standardised thiocyanate, from a laboratory supplier — the reagent then arrives at a strength somebody else measured, which is also better analysis. Or use Reilly’s volume rule of thumb, understand from the maths callout above exactly what it assumes, and accept that its errors accumulate. What is not open is estimating the strength of a silver bath by eye, which is how the measles arrive.

Prove the halogen-acceptor argument on a sheet of plain paper. Coat one sheet with the 12 per cent bath alone, with no salting at all, and expose it beside a properly salted and sensitised sheet. Ware notes that paper impregnated with silver nitrate alone has been used to make prints — Herschel did — with the oxidisable component supplied by the cellulose or the sizing, and that such papers have never been popular because of their low sensitivity compared with silver halide papers. You are measuring how much of the process is the silver halide and how much is the free silver, and the answer should be visible in minutes.

Weigh what you throw away. Collect the first two wash changes from a single 8 × 10 print in a clean jar, precipitate with a little sodium chloride, let it settle, and look at it. Reilly’s claim is that 92 to 94 per cent of the silver you bought never becomes image, and that most of the recoverable part of it is in the first two wash changes. This is not a quantitative experiment unless you filter, dry and weigh the precipitate — which is worth doing once — but as a qualitative one it changes how people handle their wash water permanently.