Ammonium sulfite

Two ammonium ions and one sulfite ion. It is the ammonium member of the family this course already knows well — sodium sulfite is in nearly every developer, and sodium bisulfite and the metabisulfites are on the same shelf — and almost everything the sulfite ion does is already taught on those pages. What this entry has to add is what changes when the cation changes, and the answer is not cosmetic. A sodium sulfite can only give you sulfur dioxide. An ammonium sulfite can give you sulfur dioxide if you push it one way and ammonia if you push it the other, and the bath a photographer actually meets it in is held at a pH where the second of those is the live question.

In photography

The one current product that names it. Moersch Photochemie’s safety data sheet for MT1 Selentoner, version 3 of 16 February 2025, declares three components, and this is the first of them: ammonium sulfite, CAS 10196-04-0, EC 233-484-9, at 10 per cent w/w, classified on the sheet Eye Irrit. 2 and STOT SE 3, H319 and H335. The other two are ammonium thiosulfate at 20 per cent and sodium selenate at 2 to 3 per cent. The formulary entry for MT1 sets the whole sheet out line by line and this page does not repeat it.

What the sulfite is doing in that bottle is not known, and this page does not decide it. Moersch states no function for any of the three components. The formulary page sets out the two readings the chemistry allows — that a sulfite is the standard preservative of photographic solutions and would protect a thiosulfate-bearing bath from aerial oxidation, and that a sulfite is a medium in which selenium species are known to interconvert — and adopts neither, because choosing would be reconstructing a formulation by inference. That refusal holds here. What can be said without inferring anything is narrower and still worth having: every selenium toner in this course’s corpus contains a sulfite of some kind, and this is the only one of them that uses the ammonium salt. HARMAN declare sodium sulphite at 10 to 30 per cent in theirs, and leave it unclassified.

⚗Why the cation is not a detail in this particular bottle

Moersch measures his concentrate at pH 9.53 and records its odour as ammonia like. Those two entries on the same page are one fact stated twice.

Kodak’s J-215, writing about silver recovery from ammonium-based fixers and bleach-fixes, gives the threshold: in practice the pH should not be raised above 7.8 to 8.0, because of the evolution of ammonia that takes place in ammonium-based baths. The concentrate is more than a pH unit and a half above that.

NH₄⁺ + OH⁻ → NH₃ + H₂O

What an alkaline bath does to an ammonium salt

So the smell on the sheet is not an impurity or a perfume. It is the ammonium ion of two of the three declared components doing what an ammonium ion does when the solution around it is alkaline, and it is the reason the ventilation advice for this toner is not the ventilation advice for a sulfite developer. HSE’s EH40 sets ammonia at 25 ppm over eight hours and 35 ppm over fifteen minutes; those are far higher numbers than the sulfur dioxide limits below, but a smell that reaches you across a darkroom is telling you the bath is releasing something, and the answer to it is airflow rather than familiarity.

The documented photographic function, and it is not toning. The one place in this course’s corpus where somebody states what an ammonium sulfite does photographically is a Kodak patent. Richard Henn and John Crabtree’s Photographic washing accelerators, US 2,860,978, filed 1954 and granted 1958, is the disclosure behind the sulfite hypo clearing agent: a bath of soluble sulfite between fixing and washing, which the one per cent sodium sulfite washing aid page sets out. Example 6 of that specification is one sentence long and it is the whole of this substance’s photographic warrant:

In place of the sodium sulfite employed in Examples 1 to 5, we have repeated these examples employing lithium, potassium and ammonium sulfites and have found all of them are fully as effective as the sodium sulfite for hypo elimination from fixed photographic materials.

Read what that claims and what it does not. It claims that the sulfite ion does the work and that the cation is interchangeable across lithium, potassium, sodium and ammonium — which is consistent with the three-part mechanism the same specification proposes, where thiosulfate adsorbed on the gelatin is displaced by sulfite ions until equilibrium is reached. It does not claim any advantage for the ammonium salt, and no product this course knows of is built on it. So the washing role is recorded here as documented rather than as current practice: if a reader meets an ammonium sulfite in a washing aid, this is the paper it came from.

!What the databases say photography does with it, and why that is not enough

Both CAMEO and HSDB list photography among this substance’s uses. CAMEO’s general description says it is used in the manufacture of other chemicals, in medicine, and photography; HSDB’s use list reads “In photography; as reducing agent”; and Haz-Map names photographic processing among the industrial processes carrying a risk of exposure to it.

None of those says which photographic use, and a course that took “used in photography” as an explanation would be doing exactly what Rule 4 forbids. The two uses named above — a declared component of one selenium toner, and a washing accelerator in one 1958 patent — are what the evidence actually supports. The gap between that and a database’s single word is where a reader should expect to find work nobody has published.

Everywhere else it is not a photographic chemical at all. PubChem’s use listings for ammonium sulfite are the production of caramel colour, refractory bricks for blast-furnace linings, lubricants for metal cold-working, medicine, permanent-wave solutions and the manufacture of ammonium thiosulfate. The EPA’s Chemical Data Reporting system files its industry uses as intermediate, soil amendment, reducing agent, preservative and processing aid, and records an aggregated United States production volume of 100 million to under 250 million pounds in each year from 2020 to 2023. The FDA lists it as a substance added to food, 21 CFR 73.85, with the technical effect colour or colouring adjunct. This is a bulk industrial commodity that photography borrows a few kilograms of, and knowing that is useful when you go looking for a supplier: almost nobody is selling it to photographers.

Properties

Relative molecular mass 116.14 for the anhydrous salt, from the formula (NH₄)₂SO₃. It dissolves freely in water and barely at all in alcohol or acetone, and the aqueous solution is recorded by the Merck Index as alkaline to litmus. Density 1.41 at 25 degrees C. There is no clean melting point: the figures in the record are decomposition figures, and they disagree with one another because they belong to different hydration states, which is the subject of the next two sections.

The sulfite ion is a reducing agent, and CAMEO says so without hedging. Its reactivity profile for this substance is two sentences: AMMONIUM SULFITE is a reducing agent. Emits toxic sulfur dioxide and oxides of nitrogen when heated to decomposition. That first sentence is the property the whole page turns on. It is why the salt scavenges oxygen, why it is on an incompatibility list with every oxidiser, why it will not keep indefinitely in air, and why Kodak’s 1928 primer could build the entire preservative argument on the sulfite ion’s affinity for oxygen without caring which cation came with it.

2 SO₃²⁻ + O₂ → 2 SO₄²⁻

The reaction that limits the shelf life of any sulfite

The Merck Index states the same thing for this salt in words rather than symbols: under the influence of air and heat it loses all its water of crystallisation and is gradually oxidised to ammonium sulfate. A tub that has caked has been doing both.

✕Three substances, one name — and the databases do not agree on which is which

This is the single most important paragraph on the page for anyone about to buy a jar.

The anhydrous salt is (NH₄)₂SO₃, CAS 10196-04-0, EC 233-484-9, ECHA CHEM record 100.030.428, relative molecular mass 116.14. It is what Moersch’s safety data sheet names.

The monohydrate is (NH₄)₂SO₃·H₂O, CAS 7783-11-1, EC 684-263-2, ECHA CHEM record 100.209.730, molecular formula H₁₀N₂O₄S. CAMEO’s datasheet 2459 — filed under the anhydrous CAS number — prints the chemical formula as (NH4)2 SO3.H2O and the molecular weight as 134.2, which is PubChem’s 116.14 for the anhydrous salt plus the mass of one water. So the emergency-response datasheet for one substance is carrying the other substance’s formula and mass under the first one’s registry number. The same weight of the monohydrate carries about 87 per cent of the sulfite that the anhydrous salt does, which is the difference between a 10 per cent bath and an 8.7 per cent one.

The bisulfite, the mono-ammonium salt, is NH₄HSO₃, CAS 10192-30-0, EC 233-469-7, ECHA CHEM record 100.030.414. It is a different substance with a different anion, and it is the one industry actually makes in quantity: its ECHA tonnage band is 10,000 to 100,000 tonnes against the sulfite’s 10 to 100. They are made in the same pot — HSDB describes the ATS process, in which sulfur dioxide from incinerated Claus tail gas is absorbed into aqueous ammonia to give ammonium sulfite and ammonium bisulfite in solution — which is why a supplier’s “ammonium sulfite solution” needs reading carefully.

And the confusion is not only the reader’s. MeSH files “ammonium hydrogen sulfite” as an entry term of the concept “ammonium sulfite”, so the two salts are not distinguished there at all. One CLP notifier has filed the name Ammonium hydrogen sulfate — sulfate, not sulfite — against the bisulfite’s ECHA record. Two other notifiers have filed Ammonium Sulfite Monohydrate and Diammoniumsulfit Monohydrat against the anhydrous record. The names on all three records overlap in ways the CAS numbers do not, so read the CAS number and not the label.

⁇Acidic salt or alkaline solution?

CAMEO assigns this substance to two reactive groups: Sulfite and Thiosulfate Salts, which is uncontroversial, and Salts, Acidic.

Set that beside the Merck Index, which records that the aqueous solution is alkaline to litmus, and beside Moersch’s measurement of pH 9.53 for a concentrate in which this salt is the second-largest declared component. The two do not sit together comfortably.

The course does not resolve it, and states what each source is for rather than picking. A CAMEO reactive group is a family label used to predict what will happen when two materials meet in a spill; it is not a measurement of the pH of a solution. The Merck Index entry and Moersch’s section 9 are statements about a solution. Nothing in this corpus measures the pH of a plain aqueous ammonium sulfite at a stated concentration and temperature, so no pH value for the salt on its own appears on this page. If you need one, measure it, and record the concentration and temperature with it.

The HSDB annotations, and why several of them say the wrong thing. Read the PubChem record for this compound and a number of the experimental properties carry the slash annotation /AMMONIUM SULFATE MONOHYDRATE/ — sulfate — including the solubility figures, the melting range, the storage instruction and the statement about water of crystallisation. This course reads that annotation as a transcription slip for the sulfite monohydrate, and the reason is internal to the sentence itself: one of the annotated statements is that the substance is gradually oxidised to ammonium sulfate, and nothing is gradually oxidised into itself. That is an interpretation and it is marked as one under Rule 7; the figures are attributed above exactly as HSDB labels them, and they are given as the monohydrate’s rather than the anhydrous salt’s.

Handling

Why Level A. Argued against the rubric rather than asserted, and the argument is short because the classification is mild and mostly absent.

Level A’s first criterion admits substances whose GHS classification is “at most irritant, harmful if swallowed, or corrosive at the concentrations actually handled”. Every statement any authority has put on this substance falls inside that ceiling: the European notified codes top out at H314, skin corrosion category 1B, carried by 13.4 per cent of the classifying reports, with the rest irritant; Safe Work Australia adds H302, harmful if swallowed, and H318, serious eye damage. Corrosive and harmful-if-swallowed are the two things the criterion names by name. Nothing here is acutely toxic, a sensitiser, a mutagen or a reproductive toxin on any record this course has read — which is the test that put ammonium bromide into Level B despite its milder-sounding name, because that salt carries a harmonised reproductive-toxicity entry and this one carries no harmonised entry at all. Level A’s other criteria are met without argument: nothing is heated, and the quantities are darkroom quantities.

This is the same level and the same reasoning as sodium sulfite and sodium bisulfite, and it should be: the notified classification for this salt is milder than sodium sulfite’s on the identical evidence basis. Sodium sulfite’s ECHA aggregation has 51.8 per cent of classifying reports giving H314 and only 24.9 per cent saying the substance meets no criteria. This one has 13.4 per cent giving H314 and 71.1 per cent saying it meets no criteria.

⚠The two steps that leave Level A, and they go in opposite directions

A sulfite is a salt of a weak acid, and an ammonium salt is a salt of a weak base. This one is both, so there are two ways to make it give up a gas and they are at opposite ends of the pH scale.

Acid. HSDB’s entry for sulfites as a class states that if swallowed they may cause irritation of the stomach by liberating sulfurous acid, and sulfurous acid in water is sulfur dioxide leaving it.

SO₃²⁻ + 2 H⁺ → SO₂ + H₂O

What an acid does to a sulfite

EH40 controls sulphur dioxide at 0.5 ppm, 1.3 mg/m³, over eight hours and 1 ppm, 2.7 mg/m³, over fifteen minutes. Moersch records a violent reaction with strong acids for the diluted toner and, separately, the release of toxic materials with acids. Deliberately acidifying a sulfite is a procedure generating a vapour, which is a Level B criterion — the same conclusion the sodium bisulfite page reaches from the same chemistry.

Alkali. Push the other way and it is the ammonium ion that goes. Kodak’s J-215 threshold of 7.8 to 8.0 is the practical line, and EH40 sets ammonia at 25 ppm over eight hours.

So the rule is not “keep it away from acid”, which is what a sodium sulfite would need. It is keep it away from both ends: no acid stop bath carried into a bath containing it, no alkali added to raise the pH of a solution containing it, and no acid in the bottle its waste goes into. And it has the same corollary as the rest of the sulfite family: never mix a spent bath containing it with anything until you know what the other thing is. The general rule is on the incompatibilities page.

The dust is the exposure route that matters, and there is no limit for it. The Coast Guard’s health-hazard entry is three short sentences: inhalation of dust causes irritation of nose and throat; ingestion causes irritation of mouth and stomach; dust irritates eyes. Its protective-clothing line is dust mask; goggles or face shield; rubber gloves, and a dust mask appearing before gloves on a list of that kind is a signal about which route the writer thought was first.

What does not exist is a number to work to. HSE’s EH40 has no entry for ammonium sulphite, and neither the NIOSH Pocket Guide nor the ILO-WHO International Chemical Safety Card scheme covers it — CAMEO’s identifier block records both as “none”. EH40 does list two neighbouring sulfite salts, sodium hydrogen sulphite and disodium disulphite, at 5 mg/m³ over eight hours each, and EH40’s own introduction supplies the sentence that governs the gap: the absence of a substance from the list does not indicate that it is safe. The only air figures attached to this compound by name are the U.S. Department of Energy’s Protective Action Criteria — 30, 330 and 2000 mg/m³ for PAC-1, PAC-2 and PAC-3 — and those are emergency-planning thresholds for a release, not workplace limits, and must not be used as though they were.

🧮Reading the PAC numbers against the other salt in the same bottle

Those three PAC figures are worth one comparison, because the same database publishes them for sodium selenate, the third component of the same toner: 1.4, 1.6 and 2 mg/m³.

PAC-1 for the sulfite is 30 mg/m³; PAC-1 for the selenate is 1.4. The sulfite’s figure is more than twenty times the selenate’s, and its PAC-3 — the level at which life-threatening effects are anticipated — is a thousand times the selenate’s PAC-3. Two components of one bottle, on one scale, three orders of magnitude apart.

The lesson is not that the sulfite can be treated casually. It is that the level a mixture needs is set by its worst component, not by its largest one. The sulfite is declared at 10 per cent and the selenate at 2 to 3, and the selenate decides the handling of the bottle. A reader who reasons from the percentages rather than from the hazards will get this exactly backwards.

And the sulfite question that has nothing to do with the darkroom. HSDB records, under populations at special risk, that some asthmatics are said to be dangerously sensitive to minute amounts of sulfites in foods. That entry is about ingestion of sulfited food, not about handling a photographic bath, and this course does not extend it into a claim about darkroom exposure that no source supports. It is recorded because a reader with asthma who finds “sulfite” on a label is entitled to know that the word carries a specific medical history, and because the answer to an unquantified sensitivity is the same as the answer to the unmeasured dust: keep it out of the air, and out of you.

Waste

Almost everything about this substance’s waste in this course is decided by something else in the bottle.

Where a reader meets it, it is at 10 per cent w/w in a selenium toner whose 2 to 3 per cent of selenate makes the whole bath a hazardous-waste stream. Moersch directs the material and its container to hazardous waste and instructs that it not be emptied into drains; the argument for that is on the sodium selenate page and it is an argument about selenium. The sulfite comes along with it.

On its own account, the sulfite’s environmental problem is oxygen rather than poison, and it is the same property that makes it useful. A reducing agent in water reduces things; what it mostly reduces in a river is the dissolved oxygen the river’s inhabitants were using. HSDB’s aquatic figures are consistent with that reading — median tolerance limits in the hundreds of milligrams per litre, 240 ppm over 48 hours for the mosquito fish and 203 to 299 mg/L for Daphnia magna depending on exposure time, all with the conditions of bioassay unspecified. Those are not the numbers of an acutely poisonous substance and they are not a licence either: an oxygen-demanding load does its damage by consumption rather than by toxicity, and a fish kill from deoxygenation is still a fish kill.

The United States regulates it on exactly that footing. Ammonium sulfite is designated a hazardous substance under section 311(b)(2)(A) of the Federal Water Pollution Control Act, and 40 CFR 302.4 gives it a CERCLA reportable quantity of 5000 lb, or 2270 kg. No darkroom is within four orders of magnitude of that quantity. The listing is quoted here not as a rule a photographer can breach but as a statement, by a regulator, about which environmental compartment this substance belongs to.

So: collect it, label it with the substance named and the date, keep every acid out of that bottle, keep it with the selenium-bearing stream if that is where it came from, and route it as the toner’s own sheet directs. The course’s reading of the British position is set out under disposal, and the containers and labelling under general chemical waste. Check your local regulations; they govern, they differ between authorities inside one country, and they change.

History

There is no period-manual history to give, and saying so is more useful than implying one. A search of this course’s historical corpus — Hardwich, Hunt, Towler, Abney, Vogel, Wall’s 1912 dictionary and his 1924 Photographic Facts and Formulas, Cassell’s 1911 cyclopaedia, the Kodak primers of 1924 and 1928 and the Kodak Limited formularies — finds no photographic use of an ammonium sulphite under any of the spellings those books would have used. The nineteenth-century literature is full of hyposulphite of ammonia, which is ammonium thiosulfate and a different substance that Herschel himself prepared and described in 1819. The sulphite is not there.

The earliest photographic naming this course can trace is therefore 1958: Example 6 of Henn and Crabtree’s washing-accelerator patent, filed for Eastman Kodak in April 1954, where the ammonium salt appears not as a discovery but as one of four cations the inventors checked and found equivalent. That is a characteristic way for a substance to enter photographic literature — as a control in somebody else’s experiment.

Between that and Moersch’s 2025 safety data sheet this corpus holds nothing. Whether the ammonium sulfite in MT1 was chosen for a reason, or simply arrived with the ammonium thiosulfate it is blended with, is not stated anywhere the course can read, and the formulary page declines to guess.

This page was written late, and the reason is the same as the last one’s. The chemical encyclopaedia planned three sulfite entries — sodium sulfite, sodium bisulfite and sodium metabisulfite — plus the potassium metabisulfite, because the inventory was built by asking which sulfites a developer is made of, and every developer in the course uses the sodium salt. The ammonium salt does no developer chemistry at all. It arrived from the toning shelf, in a bottle made in Hürth, and the register learned of it on 6 September 2026 from two written Part XX lessons, selenium toning chemistry and selenium toning and the print curve, both of which named ammonium-sulfite in their dependencies and neither of which could link it — the same two lessons, on the same day, that had already produced sodium selenate. Two of the three components of one bought bottle were missing from an encyclopaedia of 137 substances, and it took a written lesson rather than a plan to find either. The planned count moved from 138 to 139. A substance two written lessons depend on is not optional, and a reader who meets “ammonium sulfite” in the composition table of the toner in their hand should not find the encyclopaedia silent.