Iodide hypo check

A fixing bath gives almost no warning. It looks the same on the day it stops working as it did on the day it was mixed, and the prints it ruins do not look ruined for years. Kodak says exactly that: the appearance of a fixing bath changes very little during its useful life, so some means of determining when it is unfit for further use should be employed. This is that means — one salt, one bottle, five drops against five drops, and an answer in a second.

Ingredient	Quantity	Form the source specifies
Potassium iodide	190 g
Water	to make 1000 mL	at 26.5 °C; Kodak's first line is 750 mL of water at 26.5 °C (80 °F), which receives and dissolves the iodide, and its last is water to make 1.0 litre. Only the final volume fits this field. The stated temperature is Kodak's; 190 g/L is nowhere near the solubility limit of potassium iodide, so it is not a solubility requirement, and Kodak gives no reason for it.

Purpose

To find out how much silver a fixing bath is carrying, before the prints find out for you. Kodak’s own statement of what its two testing solutions are for is the sentence to keep: they permit a quick check on the acidity of the stop bath and the silver content of the fixing bath. That is the quantity this reagent responds to. Not the thiosulfate left, not the pH, not the age of the bottle — the silver.

Why silver is the thing to measure is stated on the next page of the same book. An overworked fixing bath contains complex silver thiosulfate compounds that are retained by the films or prints and cannot be removed completely by washing, and those salts lead to stains which may not become evident for a period of time. The failure is therefore invisible twice over: invisible in the tray, because the bath looks fine, and invisible in the print, because the stain arrives later.

The 1924 primer gives the chemistry underneath that sentence, and it is the whole reason a fixer has a life at all. Two compound sodium silver thiosulfates exist, one of them almost insoluble in water, while the other is very soluble, and as long as the fixing bath has any appreciable fixing power the soluble compound only is formed. A bath worked past that point starts leaving the insoluble one in the paper, and washing does not take it out.

!What this test is not

It is not a test of the print. A bath that passes says nothing about whether a particular sheet was fixed long enough or washed long enough. ST-1 tests the material for residual silver; HT-2 tests it for residual thiosulfate. This one tests the liquid.

It is not a test of fixing power. A bath can be weak because it was mixed short, or diluted by carry-over, or has thrown its thiosulfate as sulfur, and still hold very little silver. The clearing time catches those; this does not.

Recommended uses

On a print fixing bath, which is what Kodak publishes it for. The heading is How to Test a Print Fixing Solution and the surrounding text is about prints throughout. J-1 also recommends, on printed page 8, that one fixing bath be used for films and plates and another for papers, so a reader following Kodak has a print bath to test in the first place.

At the end of a printing session, before the bath goes back on the shelf. Five drops is nothing; the discipline is in doing it while the tray is still in front of you rather than at the start of the next session, when a bath that failed has already had a print in it.

On both baths of a two-bath sequence, because the answer decides which bath moves where. Kodak’s rotation is published with the test and is half of its value: if both baths give a precipitate, replace both; if only the first does, the first is replaced by the second and the second is replaced by a fresh bath. Without the test the rotation runs on counting prints, which J-1 gives as five changes before both baths are discarded, and ILFORD warns that throughput can only be a guide because it depends on the proportion of exposed to unexposed areas on the prints.

As a cross-check on a silver estimator strip, at the level where strips stop working. ILFORD is blunt about the limitation: silver estimator papers are usually not sensitive enough to test the very low silver levels suitable for optimum permanence. A drop test that gives a yes or no at some level is not a substitute for a measurement, but it is a second opinion that costs nothing.

On a film fixing bath, with a correction you have to make yourself. Nothing in the chemistry cares whether the silver in the bath came off a film or a print. What changes is the level that matters: ILFORD allows a film bath to reach 8 to 10 g/L of silver without serious effect, against below 2 g/L for fibre papers where commercial permanence is wanted and below 0.5 g/L for maximum stability. A film bath condemned by a test calibrated for prints may have most of its life left. Kodak publishes no silver figure for this test’s endpoint, so the course cannot tell you where it sits between those numbers, and says so rather than guessing.

When another formula is preferable

• The clearing time, for a film bath, and for anything where you own no reagent at all. It is free, it uses a scrap of the film you are actually processing, and it measures fixing power directly rather than by proxy. J-1’s rule is twice the clearing time for films and plates; Troop and Anchell’s is to multiply the clearing time by two — conventional advice — or three, which is theirs, and to discard the fixer when the clearing time doubles from the fresh test. Anchell gives the same rule in the Darkroom Cookbook: when the initial clearing time has doubled, mix a fresh bath. That test answers a question this one cannot, and this one answers a question it cannot: a bath can clear briskly and still be loaded with silver.
• ST-1 or ILFORD’s sulphide test, when the question is whether this print was fixed. The bath is a proxy for the print. The print is not a proxy for anything.
• HT-2, when the question is the wash rather than the fix. Different failure, different reagent, different reading.
• Silver estimator strips, when you want a number rather than a verdict, remembering ILFORD’s caveat above and Kodak’s own about on-site methods: J-211 reserves the word compliance for laboratory analysis with USEPA protocol behind it, and lists three different silver analyses that a laboratory might perform, which is a good indication of how little a drop of anything can settle.
• A bought hypo-check product, if you have one. Its own instructions are the authority for it and its maker’s safety data sheet classifies it. The course does not know what is in it and does not guess.
• Two baths and a schedule, when you would rather not test at all. The two-bath method with a fixed print count is what Kodak recommends independently of any test, and it fails safe in a way a forgotten test does not.

Mixing

Water, then the iodide, then make up. There are two ingredients and one of them is water, so the order is not a chemical argument; it is Kodak’s, and the closing line is Kodak’s own convention, stated in the 1928 primer: as a general rule in published formulas the term “cold water to make” is always given at the end of the formula, which insures dilution to a definite volume and thus a known concentration each time the formula is mixed. That matters more here than on most pages, because the strength of the reagent is the only thing setting where the test’s endpoint falls.

Kodak’s water is at 26.5 °C (80 °F), and the reason is not solubility. Potassium iodide dissolves to about 148 g per 100 g of water at 25 °C, so 190 g in a litre is a small fraction of saturation and would dissolve in cold water without complaint. Kodak states the temperature and no reason for it; the course does not supply one.

Mix a fraction of the formula. This is the practical instruction on the page. Kodak’s own drop convention on the facing column — 20 drops to the US quart against 1 mL to the litre for its stop-bath test — puts a drop at about 0.05 mL, so the five drops a test consumes are about a quarter of a millilitre.

🧮What a litre of it would be, and why you should not make one

190.0 g ÷ 166.0028 g/mol = 1.14 mol/L of iodide, or 19 % w/v

Molar concentration

1000 mL ÷ (5 drops × ≈0.05 mL) ≈ 4,000 tests

Tests per litre

Four thousand tests is several darkroom lifetimes, and the bottle keeps for a year. Fifty millilitres is 9.5 g of potassium iodide and about two hundred tests, which is still more than a home darkroom will use before the keeping figure runs out.

Scaling a formula down is arithmetic and not a variant, and the concentration — the only thing that sets the endpoint — is unchanged by it. Weigh to the balance’s honest resolution: 9.5 g on a 0.1 g balance is a 1 per cent weighing, which is far better than this test needs.

Glass or plastic, and nothing that has held a fixer, a sulphide or a silver salt. A trace of thiosulfate in the vessel does not matter much in a reagent this concentrated, but a trace of silver nitrate turns the bottle into a suspension of silver iodide on the spot, and a trace of sulphide ruins both. Dedicate the glassware and label it.

Behaviour

The reading is instantaneous and it is a yes or a no. Kodak’s whole procedure for a single-bath fixer is one sentence: to 5 drops of the test solution, add 5 drops of the fixing bath to be tested and 5 drops of water; discard the fixer if a yellow-white precipitate forms instantly. There is no timing, no comparison card and no scale. Either the cloud appears at once or it does not.

“Instantly” is doing real work in that sentence. A precipitate that forms after standing is not the reading, because several slower things can cloud a fixer sample on their own — see Troubleshooting. Kodak asks for the reaction that happens while you are still holding the dropper.

Slight milkiness is not a positive. Kodak says so explicitly: any slight milkiness should be disregarded. A faint general haze is the sample, not the silver. Anchell’s version of the reading draws the same line from the other side — if nothing happens, or if a clear cloudiness appears, the fixer is all right; a white, or yellow-white, precipitate means it should be thrown out.

The colour of a true positive is the colour of the compound. Silver iodide is a light yellow, odourless solid and it is that pale yellow, thickened by scattering, that makes Kodak’s “yellow-white”. A white precipitate that is really white is worth a second look, because sulfur is white.

The two-bath test differs only in water. For the second bath Kodak keeps the reagent and the sample at five drops each and raises the water from 5 drops to 15. What it does not do is change the proportion of reagent to fixer, which is 1 + 1 in both.

⚗Why the extra water does not move the threshold, and what it does move

Write the mixture out. Let the reagent be at concentration C, the volume of reagent be V_r, the volume of fixer be V_f with silver A and free thiosulfate T, and the water be V_w. Every concentration in the vessel is diluted by the same total volume V, so:

[S₂O₃²⁻]² ÷ ([Ag(S₂O₃)₂³⁻] × [I⁻]) = T² V_f ÷ (A × C × V_r)

The quotient that decides whether silver iodide can form

The total volume cancels. The quotient depends on the ratio of reagent to fixer and not at all on the water added, so on this idealised treatment Kodak’s second-bath test has exactly the same threshold as the first-bath test. What the extra water does change is the amount of silver iodide that can form in a millilitre of the mixture, and therefore how heavy the cloud looks at a given silver level.

This is the course’s arithmetic on Kodak’s published proportions, not Kodak’s explanation. Kodak publishes none.

And the threshold moves with the fixer’s own strength, which is the limitation to carry away. The same expression has T², the free thiosulfate squared, in the numerator: a bath at half the thiosulfate concentration precipitates at a quarter of the silver. A working dilution mixed weak, or one watered down by carry-over, will therefore fail this test earlier than a correctly mixed bath carrying the same silver — and arguably it should, since it is a worse bath. But it is not the same measurement, and a reader comparing two different fixers by this test is not comparing like with like.

Image characteristics

It has none, and if it acquires any something has gone badly wrong. Nothing in this formulary is further from the picture: the reagent never touches film or paper, and the volume it exists in is five drops in a vial.

Two consequences follow anyway, and both are about the darkroom rather than the print.

Iodide must not get into the working bath. Troop and Anchell quote Haist’s 1979 finding that sodium thiosulfate fixing “times increase when small amounts of dissolved iodide build up in solution” — the emphasis is theirs — and add that iodide levels have increased in both films and papers since Haist wrote it, because modern emulsions carry more of it. The tested sample is therefore a one-way trip. Never pour it back.

A silver iodide precipitate is not a stain you can wash off glass. It is a fine, light-scattering solid and it will haunt a vial that is only rinsed. Clean the vessel properly or use a disposable one.

The mechanism

Kodak publishes the formula, the procedure and the reading, and no chemistry at all. What follows separates what the course’s sources establish from what it infers, because Rule 7 asks for the separation and because the inference is the interesting part.

Established: fixing works by turning an insoluble halide into a soluble complex. The general chemistry text the course cites for this sets the photographic case out in full: silver bromide has a solubility product of 5.35 × 10⁻¹³ at 25 °C, so washing it out with water would take tens of thousands of litres; silver ion forms a two-coordinate complex with thiosulfate whose formation constant is 2.9 × 10¹³; and adding the two equilibria gives an overall constant of about 15 for the dissolution.

AgBr + 2 S₂O₃²⁻ → [Ag(S₂O₃)₂]³⁻ + Br⁻

Fixing: the halide leaves as the soluble argentothiosulfate complex, K ≈ 15

Established: silver iodide is not in that league. OpenStax’s solubility-product table gives AgCl 1.6 × 10⁻¹⁰, AgBr 5.0 × 10⁻¹³ and AgI 1.5 × 10⁻¹⁶ at 25 °C. Silver iodide is about 3,300 times less soluble than the bromide and about a million times less soluble than the chloride. The silver iodide page gives the same fact as a mass: 0.0028 mg/L against 0.135 for the bromide and 1.93 for the chloride.

The course’s arithmetic, marked as arithmetic. Put the two published constants together — the solubility product from OpenStax’s table and the formation constant from the complex-ion treatment, which are two different compilations and are used together here because neither prints both:

K = K_sp × K_f = 1.5 × 10⁻¹⁶ × 2.9 × 10¹³ ≈ 4 × 10⁻³

Thiosulfate against silver iodide

AgI + 2 S₂O₃²⁻ → [Ag(S₂O₃)₂]³⁻ + I⁻

The same reaction with the iodide: K ≈ 0.004, about 3,400 times smaller

That factor of three thousand is the entire test. Thiosulfate dissolves silver bromide easily and silver iodide barely at all, so the reaction runs the other way when iodide is offered to a bath that is already carrying dissolved silver:

[Ag(S₂O₃)₂]³⁻ + I⁻ → AgI + 2 S₂O₃²⁻

The hypo check: iodide takes the silver back out of the complex

Which way it actually goes depends on how much silver the bath holds, and that is why the test works. A fresh bath is nearly all free thiosulfate and almost no silver, so the equilibrium sits on the left and nothing happens. As silver accumulates, two things move at once: the complex builds up and the free thiosulfate available to hold it falls, because every silver ion locks up two of them. The quotient in the callout above — thiosulfate squared over silver — collapses from both ends, and at some point the iodide wins and the vessel clouds.

!Why the course does not publish a silver figure for the endpoint

It would be easy to finish that calculation, put a number in grams per litre on this page, and be wrong.

Run it for Kodak’s proportions and for Anchell’s, and the two published tests come out about forty times apart — because Kodak offers roughly 1.1 millimoles of iodide per millilitre of fixer and Anchell about 0.027. Either the two tests genuinely have very different endpoints, which is possible and which neither author states, or the ideal treatment is missing something. It certainly is missing something: activity coefficients at an ionic strength near 1 mol/L are nowhere near unity, not all the thiosulfate in a fixer is free, higher silver–thiosulfate complexes exist, and a precipitate has to be big enough to scatter light before anyone calls it a precipitate.

So the course publishes the direction and the scaling — the threshold rises with the square of the thiosulfate concentration and falls as the iodide offered per unit of fixer rises — and refuses the absolute number, because Kodak publishes none and the arithmetic cannot honestly supply one.

One established thing that the mechanism explains, and that a reader will otherwise meet as a mystery. Iodide is also what modern emulsions shed into a fixer, and Haist’s finding that fixing times lengthen as dissolved iodide builds up is the same chemistry seen from the other side: iodide in the bath competes for the silver the bath is trying to dissolve. The reagent on this page does deliberately, in a vial, what a bromoiodide film does slowly and unhelpfully to the tray.

Function of every ingredient

Potassium iodide, 190.0 g in the litre — 1.14 mol/L, 19 per cent w/v, and the entire formula. It is here for the iodide ion and for nothing else; the potassium is a spectator that came with it. Its job is to be present in an excess so large that the silver in the sample has no choice, and the reason iodide can do that when bromide and chloride cannot is the solubility products above: silver iodide is thousands of times less soluble than the halide the fixer was designed to dissolve, so iodide is the only common halide that can pull silver back out of a thiosulfate complex under bench conditions.

What changes with more or less. More iodide moves the endpoint to a lower silver level: the test condemns baths earlier, which is safer for fibre prints and wasteful of fixer. Less moves it the other way, which is exactly what Anchell’s 4 to 5 per cent version does — see Variants — and it is why his instructions insist on exact volumes while Kodak’s are content with drop counts. Neither change is available to anyone who wants their result to mean the same thing as Kodak’s, because the reagent’s strength is the calibration. There is no scale to read against, only a threshold, and the threshold is this concentration.

What it does not do. It does not report thiosulfate, pH, sulfite, hardener or age. A bath that fails this test is carrying silver; a bath that passes may still be unfit for half a dozen other reasons, and the clearing time is what catches most of them.

Water, 750 mL first and then to make 1000 mL, at 26.5 °C. The split is procedural — the salt goes into the 750 mL and the volume is closed afterwards — and closing to a stated volume is what makes the reagent a calibration rather than a guess, per the 1928 primer’s rule quoted under Mixing. The temperature is Kodak’s and unexplained; the solubility figures show it is not needed to get the salt in. A bottle made up short reports every bath as worse than it is.

Not an ingredient, but part of the published method: the water in the test itself. Five drops, or fifteen for the second bath. It is not part of the formula and it is part of the measurement, and the callout under Behaviour is the course’s account of what it does and does not change.

Interactions

With the silver–thiosulfate complex, which is the whole point. Everything else on this list is interference.

With free thiosulfate, competitively, and this is the reason the test reads a ratio rather than a level. The thiosulfate the bath still holds is what keeps the silver in solution, and it appears squared in the quotient. Two baths carrying the same silver but mixed at different strengths do not read the same.

With sulfur, confusingly. Kodak’s 1928 primer records that an acid hypo solution gradually becomes milky on keeping, and that an excess of bisulphite turns hypo milky in warm weather through liberation of sulphur; it lists frothing, milkiness and sludging as the signs of a bath at the end of its life. Anchell says the same in modern words — a fixer that turns yellow or throws a white precipitate on its own should be discarded, because sulfur is precipitating out. A sample that was already cloudy before the reagent went in is telling you something real and is not telling you about silver, which is the course’s reading of why Kodak asks for a precipitate that forms instantly and tells you to disregard slight milkiness.

With the alum of a hardening fixer. F-5 and its relatives carry potassium alum, and an aluminium salt in a bath that has drifted towards neutral throws its own sludge. The course has no source describing what alum does inside this particular test and does not invent one; treat a haze in a sample from a hardening bath as unresolved rather than as a reading.

With the reagent’s own bottle, over a year. Potassium iodide solutions liberate free iodine in light and moisture. The chemical page puts the consequence sharply: a yellowed stock is no longer only potassium iodide — it is iodine dissolved in iodide, and that mixture attacks a silver image. On this page that matters twice: a yellowed reagent is out of specification, and a yellowed reagent near a print is a reducer.

With everything that fixing is for, if it gets back into the tray. See Image characteristics, and Haist’s finding there.

Variants

Kodak prints no variant of FT-1. What the literature has instead is a second, independently published iodide fixer test at a quite different strength, and the difference between them is worth more than a variant would be.

Anchell’s Fixer Test Solution, Darkroom Cookbook, formula 188. Water 80.0 mL, potassium iodide 4.0 to 5.0 g, water to make 100.0 mL — a 4 to 5 per cent solution against Kodak’s 19. Anchell credits it to Manuel A. Garcia Maceda. The procedure is different too, and not only in scale: add 10.0 mL of the test solution to 100.0 mL of used fixer and shake; if nothing happens, or if a clear cloudiness appears, the fixer is all right, and if a white or yellow-white precipitate forms it should be thrown out.

Four differences, and each of them is a decision:

• Strength. 4 to 5 per cent against 19.
• Proportion. One part of reagent to ten of fixer, against one to one. Put those two together and Kodak offers about forty times as much iodide per unit of fixer, which cannot leave the two tests with the same endpoint.
• Sample size. 110 mL against about 0.75 mL. A hundred millilitres of fixer poured out for a test is a hundred millilitres that must not go back.
• Agitation. Anchell requires the mixture to be shaken to create a precipitate; Kodak asks for one that forms instantly and says nothing about shaking.

Anchell is also explicit about the tolerance, in the chapter rather than the formula: the hypo check is a reliable means to test fixer exhaustion but only if you are accurate in your measurements and testing procedure — the iodide must be 10.0 mL of a 4 to 5 per cent solution and the hypo tested exactly 100.0 mL. That warning makes sense of the strength: a test running much closer to its threshold is a test that exact volumes matter to. The snippets the course was able to read stop mid-sentence after “otherwise the test is prone to”, and the course does not complete it.

⁇Which of the two is right, and what the course does about it

Nothing in the corpus reconciles them. Kodak publishes a manufacturer’s reagent for print fixing baths with no stated endpoint in grams per litre; Anchell publishes a second reagent, credited to a contributor, with a different strength, a different proportion and a different reading, and no endpoint either. Neither cites the other.

The course publishes Kodak’s as the formula on this page, because it is tier 1, because it carries a complete published procedure including the two-bath rotation, and because the drop protocol needs a quarter of a millilitre of fixer instead of a hundred. Anchell’s is recorded here in full because it is a real published alternative and because a reader who owns the Cookbook will meet it.

What the course does not do is average them, split the difference, or present either endpoint as calibrated. Where they disagree, they disagree.

The bought products are a third thing again. Reagents sold as “hypo check” are what gave this test its bench name. The course does not know their composition, does not assume it is potassium iodide, and records only that their own instructions and their maker’s safety data sheet are the authority for them.

Safety

Level B, on the rubric, and potassium iodide is what earns it — not the volume, which is trivial, and not the finished reagent in use, which is five drops.

The classification. PubChem’s aggregation of 721 reports across 53 notifications to the ECHA C&L Inventory gives potassium iodide Danger, with H302 (harmful if swallowed), H315 (causes skin irritation), H317 (may cause an allergic skin reaction), H319 (causes serious eye irritation), H334 (may cause allergy or asthma symptoms or breathing difficulties if inhaled), H372 and H373 (organ damage through prolonged or repeated exposure) and H411 (toxic to aquatic life with long lasting effects). Read the spread as well as the codes: 13 per cent of the reports say the substance meets no GHS criteria at all, and the individual statements are given by between 11 and 44 per cent of those that classify. That is a range of opinion, not a single ruling — but H317 and H334 are what set the level, because the course’s rubric puts a sensitiser at Level B and sensitisation is not reversible.

The step that carries the hazard is weighing the solid. Nine and a half grams of a fine salt on a balance is where the dust is, and dust is where the respiratory statement bites. Everything after that is a clear solution being counted out five drops at a time.

Controls. Nitrile gloves, changed rather than rinsed, because with a sensitiser repeated small contact matters more than one splash; chemical splash goggles rather than spectacles while the solid is open; an apron; and a mask rated for fine particulates while weighing, where your supplier’s safety data sheet calls for one. Weigh over a tray, not over the bench.

Kodak prints no caution for FT-1, and that is worth stating plainly rather than reading as an endorsement. The only Caution on that page belongs to the stop bath test solution beside it, which contains sodium hydroxide and phosphoric acid. A 1970s data book not printing a warning is not evidence that a substance is safe; the classification above was published decades later, and it is the current document that governs.

What is not a hazard here, and why. There is no vapour to control: a neutral aqueous solution of an alkali-metal iodide produces none at room temperature, and the test is run in a vial. J-1’s own safe-handling page names formaldehyde and acetic acid vapours, and the sulfur dioxide a fixing bath may liberate — that last one belongs to the tray you are sampling, not to this bottle, and is a reason for the room to be ventilated rather than a control specific to this reagent. Nothing is heated. There is no ultraviolet and no mains. And the finished reagent is not corrosive: it is the sensitisation, and the weighing that makes it airborne, that put this page at Level B.

Keep it away from food, from drink and from any kitchen container, jug or utensil, and out of any bottle a household might mistake for something else. Potassium iodide is a pharmaceutical substance as well as a photographic one, which makes an unlabelled bottle of a 19 per cent solution a genuinely bad object to leave lying about.

Storage

Kodak’s instruction: a brown, stoppered glass bottle, and one year. Mixed solutions can be stored in brown, stoppered glass bottles for one year — the sentence covers this reagent and the stop-bath test solution printed with it.

Brown glass is not decoration. Light and moisture accelerate the decomposition of potassium iodide and liberate free iodine, and PubChem’s storage advice is a tight container below 40 °C, preferably between 15 and 30 °C. A clear bottle on a shelf near a window is the fastest way to end up with a reagent that has become a mild oxidising bath.

Date the bottle. With a one-year figure published and no visible change until the yellowing starts, the label is the only record there is. Follow the labelling procedure and write the strength on it as well as the date: potassium iodide 19 % w/v, fixer test, so that nobody has to reconstruct it.

Keep it away from the silver bench. A bottle of this next to a bottle of silver nitrate is an accident waiting for a careless hand; see Incompatibilities.

Keep a control. A drop of the reagent into a millilitre of spent fixer you have kept for the purpose should cloud immediately; a drop into fresh fixer should not. Two vials, thirty seconds, and you have tested the tester — which is the one check available to a reader who has no calibrated endpoint to appeal to.

Incompatibilities

Silver salts of any kind. Silver nitrate and this reagent meet irreversibly and instantly, and the product is the precipitate this test exists to see. Cross-contamination destroys both bottles and, in a shared dropper, produces a false positive that will condemn a perfectly good fixing bath.

Oxidising agents, strong or weak. Iodide is a reducing agent, and PubChem and CAMEO both class it as one. An oxidiser liberates iodine, which is the yellowing described under Storage in its fast form: hypochlorite bleach, persulfate, dichromate, ferricyanide, peroxide. Keep it off the bleaching shelf entirely, and note that a ferricyanide reducer and this bottle share a darkroom in most people’s cupboards.

Acids, in combination with the above. Iodide plus an oxidiser plus acid is the standard way to make iodine deliberately. None of the three alone does much; the combination is why the bottle lives on its own.

Sodium sulfide and every sulphide toner. The general darkroom rule applies — sulphides and silver-testing reagents do not share a bench — and the incompatibility matrix carries the pairs.

The working fixing bath itself, which is the interaction nobody expects to be told about. Iodide is not a contaminant that dilutes harmlessly; it lengthens fixing times in a sodium thiosulfate bath at concentrations described as small amounts. The tested sample and the rinse water from the vial are both waste.

Waste

Three streams meet in a very small volume.

The reagent itself carries H411 — toxic to aquatic life with long lasting effects — given by 14 per cent of the notifiers who classify potassium iodide. A 50 mL bottle holds about 9.5 g of the salt, which is not much and is not nothing.

The tested samples are silver-bearing waste, and that is the stream that governs. Silver is among the parameters most often limited in a discharge consent, which is where Kodak’s J-52 guidance on small volumes puts it. Every sample you test came out of a fixing bath, and the ones that gave a positive are by definition the most loaded liquid in the darkroom. They go in the silver-bearing waste container with the spent fixer, following the silver-bearing waste procedure.

But the reagent’s own residue does not go in the spent-fixer bottle, and the distinction is the same one the HT-2 page draws for silver nitrate. Spent fixer is held for recovery; adding a concentrated iodide to it precipitates silver as a solid that an electrolytic or metallic- replacement recovery process is not designed to deal with. Bottle unused or out-of-date reagent separately, label it, and follow the general chemical waste procedure.

None of that is a disposal instruction. It is the chemistry, and the ordinary practice that follows from it; what you may lawfully pour, bottle or carry is decided where you live, and the disposal ruling explains why this course will not pretend otherwise. Find your own regulations and follow those.

Troubleshooting

Every bath fails, including one you have just mixed. Suspect contamination of the reagent by a silver salt, or a dropper that has been in both bottles. Run the fresh-fixer control from the Storage section: a drop into freshly mixed fixer must stay clear. If it does not, the reagent is finished.

Nothing ever precipitates, including in a bath you know is spent. Suspect the reagent again, from the other side — a bottle made up long, or a sample of fixer diluted before testing. Keep a small bottle of deliberately exhausted fixer as the positive control; a reagent that cannot cloud that is not reporting anything.

A haze that appears slowly rather than instantly. This is the case Kodak legislates for by asking for an instant precipitate and telling you to disregard slight milkiness. The likeliest cause is the bath rather than the silver: an acid hypo solution goes milky on keeping through liberated sulfur, and Anchell treats a fixer that has thrown a white precipitate on its own as one to discard anyway. Look at the neat sample in the vial before adding the reagent — if it is already hazy, the test cannot tell you anything about silver and the bath is probably finished for a different reason.

The precipitate is white rather than yellow-white. Silver iodide is pale yellow. A pure white solid is more likely to be sulfur, or alum sludge from a hardening bath. Neither is a silver reading.

The reagent has turned yellow or brown in the bottle. Free iodine. Light, warmth, an oxidiser or age; the one-year figure has probably run out. Discard it, and check that the replacement is in brown glass with a stopper rather than a cork.

A film bath fails and the films look perfectly fixed. Very likely both are true. Kodak publishes this test for print baths; ILFORD’s own tolerances allow a film bath 8 to 10 g/L of silver against below 2 g/L for fibre paper. Test that bath by clearing time instead, and keep separate baths for film and paper as J-1 recommends.

The test says the bath is fine and the prints stain anyway. The bath was never the whole question. Run ST-1 or ILFORD’s sulphide test on a print for residual silver, and HT-2 for residual thiosulfate. Under-fixing a print in a good bath and fixing it properly in a bad one produce different faults, and only the tests on the material tell them apart.

You cannot see anything in a vial that small. Use a clear glass vial against a dark background with a light from the side, which is how any turbidity is judged. Kodak’s 30 mL vial for the stop-bath test is the right sort of vessel; a white tile behind it is the wrong one.

Experiments

Calibrate the endpoint you were not given. This is the experiment this page most wants somebody to run. Take a litre of freshly mixed F-24, fix sheets of scrap paper through it in a counted sequence, and after every fifth sheet draw a sample and run the drop test, recording sheet number rather than a silver figure. The point at which the precipitate first appears is your bath’s endpoint in the only currency you can actually measure at home. Repeat it with a second bath at half strength and the threshold should arrive much earlier, because the quotient in Behaviour carries the thiosulfate squared.

Test the water claim directly. Two vials. In both, five drops of reagent and five drops of the same sample of part-used fixer; in one, five drops of water, in the other fifteen. On the course’s arithmetic both should precipitate or neither should, and the heavier-looking cloud should be the one with less water. If the more dilute vial stays clear while the other clouds, the ideal treatment above is wrong and the course would like to know.

Put the two published tests side by side. Kodak’s five-drop protocol and Anchell’s 10 mL into 100 mL, run on the same bath at the same moment, through a sequence of increasingly loaded samples from the experiment above. The prediction from the arithmetic is that Kodak’s fires a long way before Anchell’s. Any result at all is worth recording, because nothing in the course’s corpus compares them.

Watch the iodide poisoning that Haist described. Two identical small baths of plain hypo; into one, put a few millilitres of the reagent. Then clear matched strips of the same film in each and time them. The treated bath should clear more slowly. It is the cleanest possible demonstration of why the tested sample never goes back in the tray, and it takes ten minutes.

Compare a verdict with a measurement. Run the drop test and a silver estimator strip on the same series of samples. Where the strip reads below its floor and the drop test still says nothing, you have found the gap ILFORD describes — the region where the levels that matter most for permanence are the levels neither bench method sees. Record where each of them stopped being useful; that boundary, established once with your own baths, is worth more than either method’s marketing.