Selenium Toning: Selenosulfate, Silver Selenide and the Protection Claim

Two prints on the same cold-tone paper, one soaked four minutes in a selenium bath and one not, can be laid side by side under a good light and told apart by nobody in the room. One of them is metallic silver. The other, in part, is not. Selenium toning is what forces this course’s distinction between what a print looks like and what a print is made of, because it is the one in common use that can be run entirely for what it does to the substance while leaving the appearance almost alone — and because one of the manufacturers who sells it will then tell you, in print, that there are more effective routes to permanence than his own product and that a fine printer uses selenium for other reasons entirely.

This page is four questions. What is in the bottle. What the bath does to the silver. Which part of the tonal scale it does it to first, and why that is less well established than it sounds. And what the hazard and the waste actually require, as opposed to what the reputation suggests. The permanence argument itself is quoted here and settled elsewhere: it is an evidence question, and it belongs on the archival-evidence page where the conservation measurements are.

!Why this page carries no safety level, and where the level is argued

The classification rubric assigns a level to an operation, and this page prescribes none: no bath to mix, no sequence to follow, no times of its own. Where the operation is, the level is argued — Level B for the selenium experiment, on the rubric’s own waste criterion rather than on custom, because Level A’s waste is “silver-bearing fixer or spent developer that can be collected for recovery, or dilute rinse water; nothing that requires specialist neutralisation”, and a bath whose own maker routes it to hazardous waste collection fails that test before anyone opens a toxicology sheet.

Two things below are nonetheless mechanism rather than housekeeping, and they appear where the chemistry produces them: the acid rule, because acid meeting a selenium salt is a reaction and not a rule of thumb, and the waste rule, because it follows from an aquatic classification the makers publish and not from a regulator’s preference. Elemental selenium itself is Level D and is not handled anywhere in this course; the reasoning is on the selenium page.

What is actually in the bottle

Start with what the maker who sells the best-known product will say. Kodak’s G-23 states that the active ingredient in its Rapid Selenium Toner is a sulfite salt at a concentration of less than 2 per cent, and that working solutions generally contain less than half a per cent “selenium sulfite”. Those are bare percentages with no basis stated — not per cent w/v, not w/w — and “selenium sulfite” is not a compound name that identifies anything. It is a trade description, and the course records it as one.

Two current European safety data sheets do better, and they disagree with each other about the one thing that matters most.

• HARMAN’s sheet for HARMAN SELENIUM TONER names three components and nothing else: ammonium thiosulphate at 10 to 30 per cent w/w, sodium sulphite at 10 to 30 per cent w/w, and sodium selenite — selenium(IV) — at 1 to 5 per cent w/w, with a CAS number.
• Moersch’s sheet for MT1 Selentoner names three components in the same three roles: ammonium sulfite at 10 per cent w/w, ammonium thiosulfate at 20 per cent w/w, and sodium selenate — selenium(VI) — at 2 to 3 per cent w/w.

Selenite and selenate are different compounds with different chemistry, sold under the same shelf label for the same job. Any account of selenium toning that speaks of “the selenium toner” as though it were one substance is wrong about at least one of the two. The full composition tables, with the classifications attached to each component, are on the product entries in the formulary — the HARMAN toner and the Moersch MT1 — and this page does not restate their dilution and time tables.

Where the selenosulfate in this page’s title comes from. It is the historical route and, as far as this course’s sources reach, the only selenium toner whose composition anybody has actually published. Kodak Limited’s T-55, printed in London in 1949, is sodium sulphite, selenium powder and ammonium chloride, and its mixing instruction is to dissolve the sulphite in hot water, add the selenium powder and boil until it dissolves. Elemental selenium is insoluble in water and soluble in sulfite solution, and the product of that dissolution is sodium selenosulfate, the selenium analogue of thiosulfate. That page carries the reaction and its sourcing; it is not repeated here. What must not be done is to read the selenosulfate back into a modern bottle. Neither current sheet names it, and Kodak’s “selenium sulfite” names nothing at all.

⚗Three slots, and why the thiosulfate is the one to worry about

Set the two composition tables beside each other and the same three functional slots appear in both: a thiosulfate, a sulfite and a soluble selenium salt. That reading is the course’s — neither sheet assigns a function to anything, two sheets are not a family, and Kodak’s product may or may not fit the pattern since it names no compound. It is recorded as an inference under Rule 7 and not as a manufacturer’s statement.

Moersch says the thiosulfate part outright in his own prose, in a different document: the selenium toners of all manufacturers he knows of contain thiosulphate. He says it as a warning to anyone about to bleach a selenium-toned print, because thiosulfate carried into a ferricyanide bleach turns the bleach into a reducer and the highlights vanish irretrievably.

The consequence for this page is different and simpler. At 10 to 30 per cent, the concentrate holds ammonium thiosulfate at a strength you would recognise in a fixer bottle, so the working bath has some silver-halide solvent action of its own. That is the course’s reading rather than a maker’s statement, and three things follow from it: the toner is not the place to finish an inadequate fix; the print must be properly fixed and washed before it goes in; and the spent bath carries thiosulfate and dissolved silver as well as selenium, which is why it belongs in the silver-bearing stream on a second count.

The bottle you buy in 2026 may not be the bottle the sheet describes. Moersch’s own shop page records a regulatory change: selenium toner based on sodium selenite may now only be marketed below 1 volume per cent, and MT1 was unaffected because it contains selenate rather than selenite. That is a maker telling you, without being asked, that the class of product has been re-formulated around him and that his dilution figures still hold. Read the sheet that came with the bottle in your hand, and check its revision date against the one the course cites.

The conversion, and the equation this course does not print

The end state is not in dispute and is not this course’s inference. Kodak states it in one sentence: Rapid Selenium Toner converts the silver image to silver selenide. Getty’s conservation atlas agrees from the other end of the object’s life, recording that selenium is detectable by X-ray fluorescence long afterwards — the selenium is still there, and still where the image is.

What the course cannot give is the arithmetic. No source it holds publishes a solubility product, a solubility figure or a formation constant for Ag₂Se, and no source it holds publishes a balanced equation for the conversion. That is a real difference from the sulfide route, where the exchange is driven by a solubility product of 1.6 × 10⁻⁴⁹ and the argument can be made numerically on the sulfide page. Here the mechanism is recorded as the makers and the conservators record it — an observed conversion and an observed increase in stability — and no equation appears on this page, because writing a plausible one is precisely the failure Rule 1 exists to prevent.

The conversion is partial, and a manufacturer says so with a number. Moersch toned a print in MT1 at 1+10 for six minutes, bleached it, and reports that after six minutes not all the silver has been transformed to silver selenide; the small remainder can be re-halogenated and toned in thiourea afterwards. So a selenium-toned print is normally a mixture of silver selenide and untoned silver, in proportions the printer chose by time and dilution without necessarily knowing it. Protection is proportional to conversion, and conversion is partial. Hold that: it is the load-bearing sentence for the permanence argument later in this part.

What a selenium bath reaches, and what it leaves

1. 1Shadow deposit, largest mass of silver — a converted shell over a substantial untoned core; the density change here is the largest and the easiest to measure
2. 2Mid-tone deposit — same shell, smaller core, so a larger fraction of this particle has been converted
3. 3Highlight deposit — the shell has almost closed over it; Moersch calls this membrane "hidden selenium toning" because it changes the colour barely at all
4. 4Gelatin layer — the toner has to diffuse through it to reach any of them, which is why agitation and paper weight change the time

Drawn to teach, not measured. Kodak states the end product; Moersch states that conversion is incomplete after six minutes and that a membrane forms around the highlight grains. The equal shell thickness is a drawing convention, because no source the course holds gives a rate law for the conversion.

On “covering power”. It is common to say that silver selenide has a higher covering power than the silver it replaces, and the observation underneath that claim is solid: Kodak states that Rapid Selenium Toner “tends to intensify the image”, Moersch reports a distinctive increase in maximum density, and Getty record a slight increase in contrast and brilliance. The step from those observations to a property of Ag₂Se is the course’s inference and is marked as one, for two reasons. No extinction coefficient for silver selenide appears in any source read — the comparison that can be made for silver sulfide, where Ware gives about 16,000 dm³ mol⁻¹ cm⁻¹ for nanoparticle silver against about 560 for the sulfide, has no selenide counterpart. And since conversion is partial, the density that is actually measured belongs to a mixture of selenide and silver as the printer happened to make it, not to the pure compound.

Why the shadows change first, and how much of that is established

Here is the observation, and it is as flatly stated as a manufacturer ever states anything. Moersch: irrespective of the dilution or the temperature, the toner always reaches the higher densities first. He repeats it in the brown-toning guide from the other direction — selenium is useful as a pre-tone precisely because it protects the shadows from a subsequent bleach, and how far up the scale that protection reaches is a question of time, whereas a sodium sulphide bath affects the complete range of tone values.

Kodak corroborates it commercially rather than mechanistically: the dilutions offered “to increase shadow contrast and maximum density with a minimum tone change” are dilutions that do something at the bottom of the scale while doing nothing visible anywhere else, and the paper/toner table’s entry for Rapid Selenium at the protective dilution reads, simply, No Tone Shift.

⁇The mechanism is not established, and the obvious explanation points the wrong way

It is often said — and this course’s own manifest said it before this page was written — that selenium attacks “the largest, most reactive silver deposits” first. No source this course has read states that. Moersch states the effect, not the cause, and neither Kodak nor Getty offers a cause at all.

The reason it is worth flagging rather than repeating is that the surface-area argument, which is the one piece of real theory the course has for how a toner reaches a particle, predicts the opposite. Ware’s account is that a printed-out image’s very small particles present “a very high surface area exposed to attack, proportional to its mass”, which is why print-out silver tones so readily, whereas a developed image is much larger filament bundles. On that reasoning the smallest deposits should convert most readily per unit of silver — which is, in fact, exactly what the diagram above shows happening, and exactly what Moersch’s “hidden selenium toning” describes: the highlight grains do get a membrane, and it changes almost nothing you can see.

The course’s own reading, offered as a reading. “Shadows first” may be a statement about where the effect is visible and measurable rather than about where the reaction is fastest. A highlight can be largely converted and still move its density very little, because there is very little silver there to move; a shadow can be converted only at its surface and still shift Dmax, because that is where the density lives. If that reading is right, then the two accounts do not conflict at all — one is about reaction rate per particle and the other is about density per unit area — and Moersch’s own diagnostic is what separates them, because it makes the invisible conversion visible.

The course has not found a source that settles it, and does not pretend to. The selenium experiment is where a reader can attack it with a step wedge and a densitometer, which is more use than a citation would be.

Moersch’s diagnostic deserves setting out on its own, because it is the only published method in this part for seeing something a print will not show you.

Moersch's bleach diagnostic: reading how far the toning actually got

1. 1Tone in seleniumthe shadows change; the highlights appear not to
2. 2Bleach in ferricyanide and bromideuntoned metallic silver goes to silver bromide; silver selenide does not
3. 3Read what survivesa red-brown image with tonality still in the highlights means the toning reached them; worn-out highlights mean it did not
4. 4Finish the print, or it is not a printthe silver bromide must be fixed out or converted by toning again, because it would tone uncontrolled over time by gas action

That last step is not optional and Moersch says so in terms: a print in which not-yet-toned metallic silver has been bleached to silver bromide is no longer archival. The diagnostic destroys the print unless it is finished, which is why it is run on a sacrificial second print and not on the one you care about.

What it does to the tonal scale

Kodak’s three statements about print density point in three directions, and the selenium one is the odd member: sepia and brown toners “will reduce print densities”, while Rapid Selenium Toner “tends to intensify the image”. The instruction attached to it is that print development time may be shortened slightly, with the qualification that toning a fully developed print in it “yields an increased tonal scale” — and the further qualification, which is the one that actually governs, that the modification depends on the paper emulsion type and grade.

Three independent accounts agree about the shape of the change and none of them gives you a number you can use.

• Kodak publish a characteristic curve for POLYMAX Fine-Art Paper, developed in DEKTOL at 1+2 for two minutes at 20 °C and toned four minutes in Rapid Selenium Toner at 1+40, showing an increase in upper-scale contrast and D-max.
• Moersch reports a distinctive increase in Dmax with the deep shadows more differentiated, a higher contrast range, and a shift of image tone towards cooler and less green tints.
• Getty describe it from the conservation side: properly done archival selenium toning caused no major colour change, “the only perceivable visual effect” being a slight increase in image contrast and brilliance.

The Kodak curve is not reproduced or redrawn in this course, and the reason is worth knowing. It is a copyrighted figure published as a picture and not as numbers, the course has no image pipeline, and the paper is discontinued and the developer is Kodak’s. The model curves that show the directions of the sepia and selenium changes on one pair of axes are on the mechanism page, drawn and labelled as drawn. Your own numbers come from the experiment.

The practical consequence runs backwards into the printing session. A print made to look right untoned and then selenium-toned comes out with shadows heavier than intended and a slightly longer scale. Kodak’s answer is to develop slightly less; the more useful answer, once you have the experiment’s numbers for the papers you actually use, is to decide before you make the print. What you cannot do is share a printing plan between sepia and selenium: one route loses density and the other gains it, so “print darker for the toner” is right for one and exactly wrong for the other.

Colour by paper, and why the same dilution does different things

The one thing every maker publishes about colour is a range, not a result, and all three ranges say the same thing in the same order.

Kodak: cool chocolate-brown hues with warm-tone papers, purplish brown with neutral-tone papers, and “very little or no change” with cold-tone papers. HARMAN: cool chocolate-brown, through purplish brown, to little discernible change, with warm-tone papers usually producing the most visible effects. Moersch: all papers are accessible to selenium toning, warmtone paper tones quickly with a clear change of image colour, and with coldtone paper the alteration is often barely visible even in a strong solution for a long time — and, from the same maker’s brown-toning guide, ILFORD Multigrade IV in a neutral-tone developer is a paper on which selenium toning is “hardly visible”.

ILFORD build that behaviour into their own recommendations from the opposite direction. For MULTIGRADE FB CLASSIC and for FB COOLTONE they recommend selenium toner for display prints because it has little effect on those papers’ image colour; for FB WARMTONE, where it would have a large effect, they recommend polysulphide and selenium toners for what toning does rather than for what it protects.

The reason sits in Part XVIII and is not re-taught here. A paper’s own image colour is a statement about the size and form of its developed silver, so three papers present three different deposits to the same solution — and the developer that made the deposit is part of the same statement, which is why Moersch’s rule is that the warmer the print was developed, the higher the toner should be diluted, or the image tone cools too far. The bath does not know what colour you wanted. It knows what it found.

📷What a claim of no tone shift is worth

Kodak’s table entry reading “No Tone Shift” for Rapid Selenium at the protective dilution is the strangest useful sentence in G-23, and it carries a footnote that matters: the papers in that table were developed in DEKTOL at 1+2 at 20 °C. It is a claim about a toner and a paper and a developer and a dilution together, on papers most of which no longer exist.

Take it as the existence proof it is — a print can be substantially converted and look untouched — and not as a prediction about your paper. The way to find out whether it holds for the paper in your hand is the reference print HARMAN builds into its own instructions: make two, tone one, keep the other in a holding dish of water, and compare them wet, side by side, under the light you will show them in.

Dilution as a rate control, not an endpoint control

Line up what the three makers publish and the range looks alarming. HARMAN gives 1+3 for normal toning and 1+20 where the primary reason is protection with minimal tone change. Kodak gives the equivalent of 1+20 for protection and 1+20 or 1+40 for shadow contrast and maximum density with minimum tone change, and puts complete toning at 2 to 8 minutes at 20 °C depending on paper type and weight. Moersch runs from 1+5 to 1+400, and his shop page states the working range of the concentrate as 1+10 to 1+400 — roughly a hundredfold between HARMAN’s normal working strength and the weakest bath Moersch publishes.

There is no contradiction, because the dilution sets the rate and the printer sets the endpoint. Moersch is explicit about how he uses the two ends: 1+5 to 1+20 for only 20 to 60 seconds intensifies the shadows alone, while reaching the highlights before shadow density starts to fall needs 1+100 to 1+400. Same bath, same chemistry, two different jobs, separated by how fast the reaction is allowed to run relative to how long you can watch it.

!A note on the notation, because two sheets use two conventions

This course writes a dilution as 1+20, meaning one part of concentrate plus twenty parts of water, which is how HARMAN and Moersch write theirs. Kodak’s G-23 prints 1:20 and 1:40 and nowhere defines the notation. The course renders Kodak’s figures in its own form, as the process atlas entry and the formulary do, and notes here that the identity of the two is an assumption rather than something Kodak states. Where a published figure matters — a capacity, a protective time — check which convention the sheet in your hand is using.

Three things make the clock the wrong instrument. The rate depends on the paper, on the developer that made the deposit, on temperature and on agitation. HARMAN states that the rate of tonal change becomes progressively slower as the solution is used and ages, so an identical bath on Sunday is not the bath it was on Friday. And Kodak state that toning continues to some extent in the wash, which means the endpoint you are aiming at is not the one in front of you: the print is pulled before it looks right.

So the endpoint is judged by eye against an untoned reference, and HARMAN publishes the most complete version of that method: three dishes, the toner in the centre and plain water either side, a second identical print kept in the holding dish as the reference, and — the instruction people skip — do not watch the print continuously in the toner, because a slow change under your eye is invisible while the same change against a reference is obvious. Lift it out and compare. Thirty to forty seconds of agitation in the second water dish stops the action, and if it needs more it goes back in.

Two cautions on stopping. Moersch’s is chemical: clearing agents are not stop baths for selenium toners, and where a short strong bath must be stopped abruptly the method is rapid dilution in running water with both surfaces gently wiped with cotton wool. The second is the acid rule below — whatever you stop it with, it is not an acid.

The one case where the clock is the right instrument is the protective dose, and it is the exception that proves the point. At the protective dilution there is nothing to see, so there is nothing to judge; HARMAN accordingly publishes a time for 1+20 and none at all for 1+3, and Kodak publishes about three minutes for protection in the combined bath below. When you cannot see the endpoint, the number is all you have — which is also why a protective bath is the one place a stated dilution has to be measured rather than estimated.

Combining the steps: toning inside the washing aid

Both major makers publish a way to fold the toning step into the archival wash sequence, and both are worth reading as engineering rather than as instructions to copy.

Kodak’s version: if Hypo Clearing Agent is being used after fixing fibre prints, a working solution of it can be used to dilute the toner instead of water, which eliminates the wash between fixing and toning. The conditions attached are specific and are the part that gets dropped. Do not rinse the prints after fixing — go straight in, or stains may follow. Tone about three minutes for protection, longer for a tone change. Do not reuse that bath for untoned prints, because it now contains traces of toner. And wash for at least thirty minutes afterwards in running water at 18 to 20 °C with frequent agitation.

ILFORD’s version is the same idea inside their published optimum permanence sequence: fix one minute, tone in selenium toner diluted with working-strength WASHAID instead of water, then WASHAID at 1+4 for ten minutes, then a final wash of thirty minutes. Set that beside their untoned sequence, whose final wash is five minutes, and the arithmetic is stark: adding a toning step multiplies the final wash by six. Whatever the combined bath saves, it does not save washing.

!Two Tier 1 makers give opposite instructions about the rinse, and the course reports it

Kodak: do not rinse the prints after fixing; immerse them directly into the combination bath, to avoid stains. Moersch, writing about a plain selenium toner rather than a combined bath: if you use an alkaline fixer you need not rinse between fixer and selenium toner, but after a sour fixer you should rinse for around ten minutes.

Both are Tier 1 manufacturer statements and neither addresses the other’s case. The difference the course can point at is that Kodak’s bath is diluted in a sulfite washing aid and Moersch’s is not, and that “sour fixer” is doing work in his sentence that Kodak’s paragraph never has to consider. What the course cannot do is decide between them from the documents it holds. Follow the sheet for the product in your hand, and if the two products in your hand come from different makers, follow the toner’s.

One further Kodak instruction belongs here because it is easy to get wrong. A print that has already been dried is not simply re-soaked before selenium toning: Kodak direct that it be rewetted by immersion in a fixing bath and then rewashed, or staining may occur. That is a different and stronger instruction than the two-to-three-minute plain water soak G-23 gives for toning in general, and it is specific to this toner.

The smell, the ventilation, and the rule about acid

⚠Selenium salts and acid: why the rule is absolute rather than proportionate

Princeton’s environmental health and safety guidance for photography states that treating selenium salts with acid may release highly toxic hydrogen selenide gas, and instructs that precautions be taken to ensure sulfide or selenium toners are not contaminated with acids. Moersch’s safety data sheet states the same thing about his own product from the manufacturer’s side: violent reaction with strong acids; release of toxic materials with acids.

Neither document names the species released, and neither gives the pH at which it starts. What HSE’s EH40 gives is the workplace exposure limit that makes the rule non-negotiable: dihydrogen selenide at 0.02 ppm over eight hours and 0.05 ppm over fifteen minutes, among the very lowest figures in the whole table and more than two hundred times lower than the 5 ppm it sets for hydrogen sulphide. Selenium and its compounds, hydrogen selenide excepted, carry 0.1 mg/m³ as Se.

The course has not verified which gas an acidified selenium bath actually evolves, and says so. It has verified that the family’s volatile hydride is controlled at a concentration a darkroom has no means of measuring. A rule that cannot be made proportionate is made absolute: no acid stop bath in the sequence without a full wash between, no acid poured into a toner or into its waste bottle, no acid tray cleaner on a toning dish, and selenium waste never combined with acid waste. The reasoning is the same one the incompatibilities page applies to sulfide, with a lower number.

That rule has to be stated carefully, because Kodak’s own combined method looks at first sight as though it breaks it: a print goes from a fixing bath straight into the toner with no rinse. The distinction the course draws — and it is a reading, not a measurement — is between a bath meeting an acid and a print carrying a film of one. ILFORD publish pH 5.0 to 5.5 for their non-hardening rapid fixer; the one selenium toner sheet in this corpus that publishes a pH at all gives 9.53 for the concentrate. A few millilitres of mildly acidic carry-over entering several litres of an alkaline bath is a different operation from decanting a stop bath into it, and the makers’ instructions are consistent with that reading. The rule governs baths and bottles. It is not a licence to be careless about the carry-over, which is why the plain method has a wash before the toner in the first place.

What it smells of, and one claim the course cannot reconcile. Moersch’s sheet records the odour as ammonia-like and the pH as 9.53, which is what an ammonium salt in an alkaline solution should smell of; HARMAN’s sheet records odour, pH, density and every other physical property as “Not known”, and Kodak’s sheet says nothing about smell at all. Against that, Princeton states that selenium toners “also give off large amounts of sulfur dioxide gas”. The course cannot reconcile those. Both of the sulfur species in these baths give sulfur dioxide when they are acidified — the bisulfite reaction is on the bisulfite page and the thiosulfate one on the sulfide page, each with its source — and the one measured pH in the corpus is firmly alkaline. No measurement of the headspace above a working selenium bath appears in any source read. The disagreement is recorded rather than resolved, and it happens not to change the control: both claims ask for the same ventilation.

What the sheets actually require. HARMAN calls the product a toxic chemical, strongly recommends working in a well-ventilated area “preferably with some form of air extraction system”, and requires rubber gloves in all use and disposal, with eye or face protection and an apron. Its safety data sheet records that no occupational exposure limit is assigned to the mixture, asks for ventilation, local exhaust ventilation or breathing protection, specifies eye protection to EN ISO 16321-1 and impervious gloves to EN 374, and notes that a filter type A mask may be appropriate. Kodak, for its own more dilute product, recommends print tongs and clean rubber gloves and immediate washing of splashes, and states that “these selenium concentrations are not considered hazardous when you handle the solutions under normal conditions”.

Those two manufacturers are describing the same class of product in very different registers, and the course does not average them. It notes that the ventilation check is run before the session rather than during it, that HARMAN’s is the more demanding specification and is therefore the one a page in this part designs to, and that Kodak’s separate instruction never to use metal trays or tanks for toning solutions applies here as it does to every other toner.

Selenium as a substance

Selenium’s reputation is that it sits a very short distance from a dietary trace to a poison. That is close enough to true to be worth stating properly, and the properly is where it gets interesting.

The WHO’s IPCS monograph is unambiguous on the first half. Its Task Group concluded that selenium meets the criteria of essentiality for man, and then makes the general point that carries the whole argument: “as is true for all essential elements, not only deficient but also excessive exposure results in adverse health effects”, with a range of safe and adequate intakes between the two. The monograph’s own summary of ordinary exposure is that dietary intakes usually fall in the range 20 to 300 micrograms a day, with extremes of calculated mean intake from 11 to 5000.

The extremes are not hypothetical, and they come from one country’s geology.

Population (Yang et al., 1983, as tabulated by WHO)	Average daily intake
Low-selenium area with Keshan disease	11 µg
Moderate-selenium area (Beijing)	116 µg
High-selenium area reported as without selenosis	750 µg
High-selenium area with a history of chronic selenosis	4990 µg

WHO note that the intakes of the two high-selenium areas did not overlap, which is what makes the table evidence rather than anecdote.

What that does and does not establish. It establishes that the substance is essential, that both ends are harmful, and that a population living without signs of selenosis sat at 750 µg a day while one with a documented history of it sat at 4990 — a factor of about six and a half, which is the course’s own arithmetic on WHO’s figures and not WHO’s statement. It does not establish a tolerable upper intake level, because this monograph sets none and gives no recommended dietary allowance, and the course was unable to reach a current agency figure from an accessible document while this page was written. So the honest form of the claim is: the margin is narrow by the standards of a nutrient and not by the standards of a poison, and anyone wanting a number for a daily intake should get it from a current nutrition authority rather than from a photography course.

The darkroom’s exposure is a different exposure anyway, which is the point of stating the occupational numbers separately. NIOSH’s recommended limit and the OSHA permissible limit are both 0.2 mg/m³ as an eight-hour average, and NIOSH states that the recommended limit applies to selenium compounds as Se as well, excepting selenium hexafluoride; the concentration considered immediately dangerous to life or health is 1 mg/m³. HSE’s EH40 is stricter, at 0.1 mg/m³, and EH40’s own introduction adds the sentence every one of these pages needs: the absence of a substance from the list does not indicate that it is safe.

The GHS classifications of the two products are a lesson in reading the sheet rather than the brand. HARMAN’s mixture is Acute Tox. 4, Skin Sens. 1 and Aquatic Chronic 3 — Warning, with H302, H317 and H412 — while its sodium selenite component alone carries Acute Tox. 2 by two routes and Aquatic Acute 1 and Chronic 1. Moersch’s mixture is classified Acute Tox. 2, STOT RE 2, Eye Dam. 1 and Skin Sens. 1, signal word Danger, and section 11 records an oral LD50 of 7 mg/kg in the rat attributed to ECHA — the sheet does not say whether that figure belongs to the mixture or to its selenate component, and the course does not assume. Two products for the same job, two classifications a category apart.

That second sheet also contradicts itself, printing Danger and H300, fatal if swallowed, in one section and H302, harmful if swallowed, in the next; the formulary entry sets the conflict out line by line and takes the more severe reading, which is what a document that disagrees with itself leaves you. Read the sheet that came with the bottle you bought, and read all of it; the method is Part II’s.

Waste, and why the argument is environmental before it is legal

A selenium bath is collected. Not diluted, not neutralised, not poured away late at night.

Kodak’s environmental guidance for amateurs — not for commercial laboratories — is the most striking document in the corpus on this point, because it is generous about almost everything else. Its sewer page tells a domestic photographer that most photographic chemicals can go to the sewer and lists developers, stop baths, fixers after silver recovery and wash waters; then it lists what must not, and selenium toners are first on that list. In its table of routes for its own products, used Rapid Selenium Toner is the single entry with every route withheld except household hazardous waste collection: no sewer, no treatment works, no household bin. HARMAN says the same in one line on the label — do not empty into drains; dispose of the material and its container at a hazardous or special waste collection point. Moersch’s sheet classifies the waste under the European hazardous properties for target organ toxicity, acute toxicity and ecotoxicity, and directs it to hazardous waste.

The reason underneath all three is on the composition tables rather than in a regulation. Sodium selenite carries Aquatic Acute 1 and Aquatic Chronic 1 — very toxic to aquatic life, with long-lasting effects — and sodium selenate carries the same pair. Dilution does not make an aquatic hazard go away; it moves it. And the British reading, which is the course’s and not legal advice, does not turn on concentration either: the disposal ruling reads a home darkroom’s separately collected photochemistry as an absolute hazardous entry in the List of Waste, and WM3 is explicit that an absolute hazardous entry must be used and applies even where the waste displays no hazardous property. There is no threshold for a reader to compute. The containers and the labelling are in the general chemical waste and silver-bearing waste procedures.

Check your local regulations. They govern, they differ between authorities inside one country, and they change. What does not change is the chemistry: this bath contains a compound classified very toxic to aquatic life, plus dissolved silver, plus thiosulfate.

The protection claim, introduced and deferred

Here is the claim, in Kodak’s own words, because the wording is what will be examined later. Toning “converts the black-and-white silver image to an inert compound, which reduces the harmful effects of intense light, ultraviolet radiation, oxidizing gases, extremes of temperature and humidity, and fumes”, and — the sentence that does the most work — “All Kodak toners will protect the image whether or not they produce a color shift.”

Three other parties have said something about it, and they do not all say the same thing.

• ILFORD recommend that prints made for display are given protective toning against the oxidising gases found in many environments, and name selenium first because it has little effect on their papers’ image colour, with sulphide toning and silver image stabilisers as alternatives.
• Getty’s conservation atlas records that selenium was recognised for its ability to increase the chemical and environmental stability of silver particles, after which it entered archival processing for the art and museum market.
• Moersch, who sells one, opens his own instructions by saying that recent discussions about image silver stabilisation by means of selenium toning appear pointless to him, that gold, sulphur and Sistan are more effective routes to archival permanence, and that a fine printer has other reasons for using selenium — while adding that even weak selenium toning is better than none.

A manufacturer arguing against the main selling point of his own product is not marketing, and the course treats it as evidence accordingly. Put it beside his measurement that six minutes at 1+10 leaves silver unconverted, and the shape of the eventual argument is already visible: protection is proportional to conversion, conversion is partial and unmeasured in ordinary practice, and “protects the image” is a claim about degree that has been written as a claim about kind.

That is as far as this page goes. Whether the conservation literature has measured any of it, and what it measured, is the archival-evidence page, where what has been measured is separated from what practitioners believe.

Film, and why this part does not follow it

Kodak record that Rapid Selenium Toner can be used on black-and-white films: diluted 1+29, four minutes at 20 °C with occasional agitation, then a wash of 20 to 30 minutes, with a slight increase in contrast at that dilution and more at 1+3 or 1+9.

The figure is recorded here and the operation is not taught in this part, for a reason that is about scope rather than hazard. Everything above is built on a print — on reflection density, on a paper’s image colour, on judging an endpoint against a reference print in a dish. Toning a negative is an intensification, its consequences are read on a characteristic curve and land in the next print rather than in this one, and this part has no lab in which a negative is toned. A reader who wants to try it has Kodak’s figures above, the same acid rule, the same waste route, and one extra caution the print pages do not need: a negative that has been intensified is a negative whose printing exposure and contrast grade have both moved, so it needs a fresh test strip and not the old one.

Alternative route

This page prescribes nothing and needs no darkroom to read or to understand. The part it introduces does, and the honest position for a reader who cannot meet HARMAN’s ventilation specification is that selenium toning indoors is not the thing to compromise on, because the control the makers ask for is extraction and the hazard it addresses cannot be smelled reliably.

Three routes remain, in descending order of how well they substitute.

Tone outdoors. This is the route the sepia session takes, on the course owner’s ruling of 5 September 2026 and on Moersch’s own recommendation for sulfur baths, and extending it to selenium is the course’s extension and not any maker’s instruction — no selenium sheet in the corpus mentions working outdoors. The mechanics are the same: the toner tray goes outside, the print is rinsed before it comes back in, and the waste goes into the collected bottle rather than onto the ground or down a drain, which matters more here than it does for sulfide because the aquatic classification is the whole reason the bottle exists. It keeps the published chemistry intact and changes nothing about the dilution or the time.

Protect the print by another route. ILFORD name two alternatives themselves — sulphide toning, and silver image stabilisers — and Moersch names gold, sulphur and Sistan as more effective than selenium anyway. Gold is the one that survives in this course as an open, published formula rather than as a bought bottle: GP-1 is Level B and is covered in the gold and other toners lesson. The course has not researched proprietary silver image stabilisers and names them only because ILFORD does.

Read the part and skip the tray. The experiment states its own alternative route for a reader who cannot run the bath; this lesson, the mechanism page and the permanence page between them carry the whole argument, and a reader who never tones a print can still tell a toned one from an untoned one and can still read a permanence claim critically. That is most of what this part is for.

Recap

• Three makers, three different selenium species, and one of them undisclosed. HARMAN names sodium selenite, Moersch names sodium selenate, Kodak names only “a sulfite salt” at under 2 per cent and “selenium sulfite” at under half a per cent in the working bath. The selenosulfate of the title is the historical route, published as a formula in Kodak Limited’s T-55 in 1949, and must not be read back into any modern bottle.
• Both disclosed sheets show the same three functional slots — a thiosulfate, a sulfite, a soluble selenium salt — which is the course’s inference, not a maker’s statement. The thiosulfate is the one with practical consequences: the bath has some solvent action of its own, so the print is fixed and washed before it goes in, and the waste is silver-bearing as well as selenium-bearing.
• The end state is silver selenide and the course prints no equation for it. No solubility product, no formation constant, no balanced equation exists in any source it holds. The sulfide argument can be made numerically; this one cannot.
• Conversion is partial. Moersch bleached a print toned six minutes at 1+10 and found silver still unconverted. A selenium-toned print is a mixture, and the “higher covering power” of the selenide is an inference from an observed rise in Dmax rather than a measured property.
• Shadows first is an observation, not a mechanism. Moersch states it flatly and irrespective of dilution or temperature; nobody read states why, and the surface-area argument predicts the opposite. The course offers a reading — that the change is most visible where there is most silver — and labels it as one.
• Dilution sets the rate; the printer sets the endpoint. 1+3 to 1+400 across three makers with no contradiction. The endpoint is judged against an untoned reference print, not by the clock, because toning continues in the wash and the bath slows as it ages. The protective dose is the exception: no visible endpoint, so the number is all there is.
• The combined bath is real engineering with real conditions. Kodak dilute the toner in working-strength hypo clearing agent and skip the wash between fixing and toning; ILFORD do the same with WASHAID. Do not rinse first, do not reuse the bath for untoned prints, and expect the final wash to be thirty minutes rather than five.
• No acid, ever, in the bath or the bottle. EH40 sets dihydrogen selenide at 0.02 ppm, more than two hundred times below hydrogen sulphide. The course has not verified which gas an acidified selenium bath releases, which is exactly why the rule is absolute.
• The waste is collected. Kodak’s amateur guidance withholds every route from used selenium toner except hazardous waste collection, and it is the only product in that table so treated. The reason is the aquatic classification, not the paperwork. Check your local regulations.
• The protection claim is a claim about degree written as a claim about kind, and one of the people selling the toner says so. It is examined, not settled, here.

Check your understanding

Question 1. HARMAN’s safety data sheet names sodium selenite in its toner and Moersch’s names sodium selenate in his. What does this course conclude from that?

• That one of the two sheets must be wrong, since both products do the same job
• That the toning reaction is the same either way, so the difference does not matter
• That "the selenium toner" is not one substance, and a statement about one product does not transfer to another
• That sodium selenosulfate is the true active species in both, under two trade names

Show the answer and why

Answer: That "the selenium toner" is not one substance, and a statement about one product does not transfer to another

Selenite is selenium(IV) and selenate is selenium(VI): different compounds sold for the same job. Neither sheet is wrong about its own product, and neither licenses a claim about the other. The selenosulfate is a third thing again - the historical route in Kodak Limited’s T-55, where selenium powder is boiled into a sulphite solution - and no modern sheet names it.

Question 2. A printer tones a fibre print for exactly the four minutes a book recommends, pulls it when it looks right, and washes it. The dried print is cooler and heavier in the shadows than intended. What went wrong?

• The bath was too dilute, so toning was uneven across the scale
• Toning continues in the wash, so the print had to be pulled before it looked right
• The print was fixed in a hardening fixer, which made the emulsion less receptive
• The wash was too short to remove the toner from the paper base

Show the answer and why

Answer: Toning continues in the wash, so the print had to be pulled before it looked right

Kodak state it directly: remove the print when it has almost the required tone, because toning will continue to some extent in the wash. The endpoint you are aiming at is not the one in the tray. The clock is also the wrong instrument for a second reason - HARMAN state that the rate slows as the bath is used and ages, so four minutes in a fresh bath and four minutes in a tired one are different doses.

Question 3. Which of these statements about selenium toning are made by a source this course holds, rather than being the course’s own inference? Select all that apply.

• The toner always reaches the higher densities first, irrespective of dilution or temperature
• After six minutes at 1+10, not all the silver has been transformed to silver selenide
• Silver selenide has a higher covering power than the metallic silver it replaces
• Selenium toner converts the silver image to silver selenide
• The reaction begins at the largest and most reactive silver deposits

Show the answer and why

Answer: The toner always reaches the higher densities first, irrespective of dilution or temperature, After six minutes at 1+10, not all the silver has been transformed to silver selenide, Selenium toner converts the silver image to silver selenide

The first two are Moersch, writing from the bench; the fourth is Kodak’s own sentence about its own product. The third is the course’s inference from an observed rise in maximum density, and it is weakened by the fact that conversion is partial, so the density measured belongs to a mixture. The fifth is stated by nobody the course has read, and the surface-area argument predicts the opposite.

Question 4. Why does this course refuse to publish a balanced equation for the conversion of image silver to silver selenide, when it publishes one for the conversion to silver sulfide?

• Because the selenium reaction is too complex to balance
• Because no source it holds gives an equation, a solubility product or a formation constant for Ag₂Se
• Because the reaction depends on which selenium salt the maker used, and that varies
• Because Kodak forbids reproduction of its chemistry

Show the answer and why

Answer: Because no source it holds gives an equation, a solubility product or a formation constant for Ag₂Se

The sulfide exchange can be argued numerically because the solubility products are published - 1.6 x 10 to the minus 49 for silver sulfide against 5.0 x 10 to the minus 13 for silver bromide. For silver selenide the course has no solubility product, no solubility figure and no formation constant from any source, so it states the conversion in words on the makers’ authority and writes no equation. The variability of the selenium salt is a real complication, but the reason given here is the absence of a source, which is Rule 1.

Question 5. A darkroom has an acid stop bath in use and a selenium toning session planned for the same evening. Which precaution follows from the chemistry rather than from housekeeping?

• Use fresh stop bath, because an overconcentrated one causes mottle in toned prints
• Keep the two waste streams in separate containers and never acidify the toner or its waste
• Tone before printing, so the toner is fresh
• Use plastic tongs, because the toner is alkaline

Show the answer and why

Answer: Keep the two waste streams in separate containers and never acidify the toner or its waste

Princeton state that treating selenium salts with acid may release highly toxic hydrogen selenide, and Moersch’s own sheet records a violent reaction with strong acids and the release of toxic materials with acids. EH40 sets dihydrogen selenide at 0.02 ppm, more than two hundred times below hydrogen sulphide, which is why the separation is absolute rather than proportionate. The mottle answer is true and is a fixing-and-stop fault Kodak describe, but it is a print-quality consequence, not the hazard.

Check my answersClear