Ammonium persulfate

Kodak’s 1928 primer sorts every reducer into three classes and then makes a claim it makes nowhere else in the book: of the third class, the flattening reducers, only one is known, and it is ammonium persulphate. That sentence is why this page exists. There is no substitute for this salt, not even its own potassium twin, and a formula that says only “persulphate” has not told you which one it means.

In photography

The flattening reducer, R-1. A reducer removes silver; what distinguishes one from another is which silver. Kodak’s three classes are set out in the 1928 primer and they are a genuinely useful piece of teaching. A cutting reducer takes an equal quantity from every part of the image, which is a larger proportion of a thin shadow than of a dense highlight, so it clears fog and veils first. A proportional reducer takes silver in proportion to how much is there, so it exactly undoes development. A flattening reducer acts very much more on the heavy deposits than on the light ones, so it pulls a blocked highlight down and leaves the shadow detail where it was.

Ammonium persulphate is the third of those and, on Kodak’s account, the only member of its class. The formula is R-1, and the 1928 Eastman primer and Kodak Limited’s 1949 London handbook print it at the same quantities twenty-one years and an ocean apart:

• 60 g of ammonium persulphate and 3 mL of concentrated sulfuric acid, water to make 1 litre.
• For use, one part of stock to two parts of water.
• When reduction is complete, immerse the negative in an acid fixing bath for a few minutes, then wash.

The metric columns are identical. The avoirdupois columns are not quite — the 1928 primer prints 2 ounces to 32 fluid ounces, the 1949 handbook 4 ounces 350 grains to 80 fluid ounces, which work out to 5 and 4.8 ounces per 80 respectively — and the London handbook warns in its own opening pages that its two columns are not exact equivalents. The metric figures are the ones to work from.

⚗What the persulfate ion does to the image

Peroxodisulfate carries an oxygen-to-oxygen bond, and breaking it is worth a great deal of oxidising power. Each ion takes two electrons and falls apart into two ordinary sulfate ions.

2 Ag + S₂O₈²⁻ → 2 Ag⁺ + 2 SO₄²⁻

The step underneath the whole reducer

The silver ion meets sulfate and, in Kodak’s words, is transformed into silver sulphate, “which dissolves in the solution”.

2 Ag⁺ + SO₄²⁻ → Ag₂SO₄

Silver sulfate, the soluble form the image leaves in

The acid is not the oxidant. The primer states only that the reducer must be used in an acid solution, and gives no reason; the course therefore records the requirement and does not supply a mechanism it has not read.

What the sources do not explain, and it is the interesting half. Every Tier 1 source agrees on the behaviour: the reducer attacks the densest parts hardest, it is “somewhat uncertain”, “occasionally refusing to act”, and it “always acts more rapidly as the reduction progresses”. None of them explains why a reducer should prefer a heavy deposit, and none of them explains why it accelerates. The course has found no mechanism for the flattening action in any source it holds, so this page states the observation and prints no explanation. That is not an omission to be filled in later by guesswork; it is the state of the evidence.

R-5, the proportional reducer, is this salt plus a permanganate. The 1928 primer’s reasoning is worth following because it is an argument rather than a recipe. There is no single substance that reduces proportionally. But potassium permanganate is a slightly cutting reducer and persulphate is a flattening one, so a mixture of the two can be made to land in between. Kodak’s R-5 is Stock A of 0.3 g of permanganate with 16 mL of 10 per cent sulfuric acid per litre, Stock B of 30 g of ammonium persulphate per litre, used one part of A to three parts of B. The 1928 and 1949 printings give the same 30 g/L for stock B. Wall’s 1924 handbook gives the same pairing independently at 0.25 g of permanganate and 25 g of persulphate per litre, mixed in the same 1 + 3 ratio — two Tier 1 sources agreeing within about twenty per cent on a formula neither took from the other.

The hypo eliminator, where the two salts genuinely are alternatives. Wall’s 1924 formulary lists “potassium or ammonium persulphate in 1 per cent solution, made alkaline with ammonia” among the ways to destroy residual thiosulfate, and notes that the potassium salt was the one sold under a trade name. This is the one use where a formula legitimately says “persulphate” without specifying, and it is also a use the course does not teach: a modern sulfite-based washing aid works by ion exchange rather than by oxidation, does not attack the image, and carries no sensitisation statement. The potassium persulfate page carries that argument in full.

A contrast agent in the platinotype developer. Wall’s 1924 chapter on platinotype offers persulphate in place of the potassium bichromate normally used to shorten the scale of gradation. The course records the substitution and gives no quantity, for two reasons. The figure as printed in the copy read is ambiguous about what it is a percentage of, and the course does not complete an ambiguous line. And ruling 4.1 means platinum is taught and never handled in any case. It is worth knowing the option existed, because it means a dichromate-free contrast control for the siderotypes is a historical idea and not a modern invention.

⁇Wall calls it even; Kodak calls it flattening

Wall’s 1912 dictionary opens its account with “the reducer which is most even in its action on all parts of the negative is the persulphate of ammonium reducer”, and then immediately adds “indeed, this reducer rather attacks the dense than the light portions of the negative”. His own 1924 handbook drops the first half and keeps the second — “the particular value of this reducer is that it attacks the densest parts of a negative more than the shadows, thus considerably reducing contrasts” — which is Kodak’s flattening class in different words.

So the disagreement is Wall of 1912 against Wall of 1924 and Kodak of 1928, and it is a disagreement about a label rather than about what the bath does. All three describe the same behaviour. The course uses Kodak’s three-class scheme, because it is the scheme the formulas were printed under, and records the 1912 wording so that a reader who meets “the most even reducer” in an old text knows which bath is meant.

Properties

Colourless crystals or a white powder, odourless, density 1.9 g/cm³, relative molecular mass 228.21. Kodak’s 1928 primer adds the one word that matters for storage: stable when dry.

It decomposes at 120 °C and never melts. Chemical Safety Card 0632 gives the decomposition temperature and no melting point, and names the products — ammonia, nitrogen oxides and sulfur oxides. That last detail is a real difference from the potassium salt, which has no ammonium in it to give ammonia or nitrogen oxides at all, and it is a reason not to treat the two as interchangeable even in a fire.

It is extremely soluble, and this is the fact that forbids substitution. The safety card gives 58.2 g per 100 mL at 20 °C; HSDB gives 83.5 g per 100 g of water at 25 °C, on a different basis and at a warmer temperature, and the course quotes both rather than choosing. Against that, potassium persulfate’s own card gives 5.2 g per 100 mL at 20 °C.

🧮Why R-1 cannot be made with the potassium salt at all

R-1 asks for 60 g per litre, which is 6 g per 100 mL.

Potassium persulfate’s solubility at 20 °C is 5.2 g per 100 mL. The formula therefore asks for more of the potassium salt than a litre of water at room temperature will hold. The substitution does not produce a weaker bath; it produces an undissolved heap in the bottom of the bottle.

Even where the arithmetic does work — R-5’s stock B is 30 g/L, comfortably below saturation — the molar dose is wrong. 60 g of ammonium persulphate is 60 ÷ 228.21 = 0.263 mol of peroxodisulfate. 60 g of potassium persulfate is 60 ÷ 270.33 = 0.222 mol, which is 15.6 per cent less oxidant for the same weight. Matching the mole count would take 71.1 g, and it would also introduce potassium where the formula had ammonium.

This is what the course means when it says that substituting one salt for the other publishes a different formula. It is not pedantry about names; it is two numbers, 228.21 and 270.33, and one solubility that the formula overruns.

The solution in water is a medium strong acid, which Card 0632 states and which is not obvious from the formula. R-1 adds sulfuric acid on top of that, so the working bath is acid twice over.

⚗The iron problem, which two Tier 1 sources describe and do not agree about

Kodak’s 1928 primer records that the Eastman Research Laboratories found the action of persulphate to depend largely on its containing a very small quantity of iron salt as an impurity, and that its capricious behaviour is due to variations in how much iron is present. Kodak’s answer was to sell a tested grade; the R-5 directions in the 1949 handbook take the other route and recommend distilled water, or water free from iron, for making up the stocks.

Wall’s 1924 handbook describes the same dependence and prescribes the opposite. The action, he writes, “is much complicated by intentional or accidental additions, such as acid, chlorides, and iron salts, which may occur by keeping the solution or the use of ordinary tap water” — and then offers, as one of three ways to steady the bath, adding 1 mL of a 5 per cent solution of ammonia iron alum per litre. Where Kodak removes the iron, Wall doses it deliberately.

Both are Tier 1 and the course reconciles nothing. What they agree on is the fact that matters practically: a persulphate bath’s behaviour is set by a trace constituent nobody weighs, which is why the same formula behaves differently in two darkrooms. CAMEO supplies the reason iron is not an innocent bystander — acidic persulfate solutions dissolve iron violently — from which it follows, as the course’s own inference and not as any source’s statement, that an iron or steel utensil is both a contaminant and a hazard in this bath.

Handling

The harmonised classification under Regulation (EC) No 1272/2008 gives Danger, three pictograms and seven statements: oxidiser, harmful if swallowed, skin irritation, allergic skin reaction, serious eye irritation, allergy or asthma symptoms or breathing difficulties if inhaled, and respiratory irritation. Safe Work Australia files the same seven. Japan’s NITE-CMC added organ toxicity in 2006 and 2008, and its 2024 revision keeps H334 alone — the respiratory sensitisation — which is the statement this page is built around.

Chemical Safety Card 0632 turns the codes into a description of what actually happens. Inhalation of the dust may cause asthma-like reactions. Repeated or prolonged inhalation may cause asthma; repeated or prolonged skin contact may cause dermatitis and sensitisation; and it may cause a general allergic reaction such as urticaria or shock. It gives a TLV of 0.1 mg/m³ as an eight-hour average and a MAK notation of sensitisation of the respiratory tract and skin. Then it prints the notes that decide the classification: the symptoms of asthma often do not appear for some hours and are made worse by physical effort, so rest and medical observation are essential; anyone who has shown symptoms of asthma due to this substance should avoid all further contact; and, in capitals, do not take working clothes home.

HSE’s EH40 sets no British workplace exposure limit for persulphates or peroxodisulphates, and paragraph 6 of its introduction states plainly that the absence of a substance from the list does not indicate that it is safe — exposure is to be controlled to a level at which nearly all of the working population could be exposed day after day without adverse effect. The NIOSH Pocket Guide has no entry either; its CAS index carries neither 7727-54-0 nor the potassium salt’s 7727-21-1. So the only numerical limit anywhere in this record is the ILO-WHO card’s 0.1 mg/m³, and a British reader working under COSHH has no figure of their own to work to.

Why Level C. Against the course rubric, Level C’s first criterion names reagents classified as sensitisers at the concentrations used, where a fume cupboard or specialist disposal is the recognised control — and for a respiratory-sensitiser powder with a 0.1 mg/m³ limit, engineered extraction is the recognised control, which Card 0632 asks for in terms. The oxidiser statement is a second and independent reason: a Level A darkroom has nowhere to keep a class 5.1 solid away from combustibles. This is the same reasoning, on the same evidence, that put potassium persulfate at Level C, and the course reached it once more for glutaraldehyde on sensitiser status alone. The level does not soften because the salt is cheap, old, or printed in a manufacturer’s handbook.

⚠The operation that fixes this at Level C is opening the jar

Everything else about a persulphate reducer could be managed at Level B. The working bath is one part of stock in two of water, it is used cold, and the failure mode in the tray is an over-reduced negative. What cannot be managed in a home darkroom is putting a respiratory sensitiser on a balance, because the control the safety card asks for is extraction and the consequence of getting it wrong is permanent. A sensitised person reacts at doses far below anything they can perceive, and the only treatment is never to meet the substance again.

The operation-sets-the-level ruling does not rescue this one. That ruling permits a Level C weighing to be moved outside the session, or avoided by buying the stock ready made; neither route exists here, because there is no ready-made R-1 stock on sale and the weighing is the whole of the preparation.

So the course gives no home procedure using ammonium persulphate. Where a highlight must come down, the practical answers are a shorter development next time and a different paper grade this time. Where a negative genuinely needs reducing, the course teaches Farmer’s reducer and its two-bath variant, which are cutting and almost proportional respectively. What is not on offer is a home procedure in the flattening class, and that limit now follows from this page — from a respiratory sensitiser with a 0.1 mg/m³ limit — rather than from any gap in the formulary.

Waste

Because the course gives no procedure, the best waste advice is not to acquire the substance. Where it exists, it is an oxidiser and it goes nowhere near organic waste: not into a bottle with spent developer, not absorbed into sawdust or paper, and — Card 0632 is explicit — not into the environment, the substance being harmful to aquatic organisms.

A spent reducer is also a silver stream, because dissolving image silver was the point of it, and Kodak’s silver-recovery literature treats any solution that has taken up image silver as a recovery stream rather than an effluent. Kodak’s J-52 pH window of 5.6 to 9.4 does not settle the question, since the hazard is the oxidising power and not the pH. Bottle it separately, label it with the substance and the date, keep it away from anything combustible, and route it as hazardous waste; ILFORD’s guidance for domestic users in the United Kingdom points at a household waste and recycling centre. Check your local regulations, under the disposal caveat — they govern, and they differ.

History

The persulphate reducer arrived with a reputation for temperament that it never lost. Wall in 1912 already describes a bath that needs a drop of sulfuric acid per two ounces “to make the action more regular”, and prints H. W. Bennett’s stock solution — persulphate with sodium sulfite and sulfuric acid — specifically because the plain solution would not keep. By 1924 the same handbook is offering three separate ways to steady it: stand a 5 per cent solution in distilled water for three or four hours, or acidify it, or add a measured trace of an iron salt. Kodak’s Research Laboratories then found what all three were compensating for, and Kodak’s answer was commercial rather than chemical — buy the tested grade.

That sequence is worth noticing, because it is the same shape as several other stories in this encyclopaedia. A reaction that depends on an impurity is a reaction nobody can standardise from a formula alone, and the industry’s solution was to standardise the jar instead. The formula outlived the jar. R-1 was still in Kodak Limited’s London handbook in 1949, printed at the quantities the Rochester primer had given in 1928, and it is still the only flattening reducer anyone has named.

What ended it was not the chemistry. It was the occupational-medicine literature on persulphate asthma that accumulated across the twentieth century — hairdressers, not photographers, supplied most of it — and turned a convenient oxidiser into a substance with a sensitisation statement, a 0.1 mg/m³ ceiling and a safety card that tells you not to take your overall home.