Photographers' Formulary cyanotype and toning kits

This is a product page, and it is the odd one in the category. Every other bought product in this library is here because nobody outside the factory knows what is in the bottle. Photographers’ Formulary prints its cyanotype kit’s contents on page one of the instruction sheet, with a weight against every line, and then prints the two stock solutions those weights make and the ratio in which they are mixed. There is no secret recipe to be careful about.

So the honesty this page kind exists to enforce has to do its work somewhere else, and it turns out there is plenty for it to do. What the supplier does not publish is not the formula but the material: which grade of green ferric ammonium citrate is in the jar, and therefore how much iron is in it, which is the one number that decides how fast the paper prints and how blue it goes. The kit’s safety data sheets are four separate sheets for four separate solids, so nothing in the literature describes the liquid kits’ bottled solutions at all. Two printings of the same instruction sheet disagree by a factor of ten about the strength of the toning bath. And in the box, alongside the two chemicals that make a cyanotype, there is a gram of potassium dichromate that this course will not use at any level.

What follows is what the supplier publishes, which is a great deal; what its safety data sheets disclose, which is four substances at a concentration that cannot mean anything; where the boundary actually falls; and the open formula this kit is a printing of, which is the classic cyanotype sensitiser with a fifth less ferricyanide in it.

Photographers' Formulary, Inc. — sold as kit

Components the maker discloses — from the safety data sheet, which classifies hazards and does not state a formula

Component	Concentration as the sheet gives it	Hazard codes
Starch, from potatoCAS 9005-25-8The sizing starch. The instruction sheet in the same PDF calls it arrowroot; this sheet is for starch from potato, and no document in the supplier's literature reconciles the two	<= 100 %
Ammonium iron(III) citrateCAS 1185-57-5The light-sensitive half. The instruction sheet, not this sheet, states the function: the blue comes from ferrous ions produced by the photo reduction of the ferric ammonium citrate	<= 100 %	H315, H319, H335
Potassium ferricyanidenamed on the sheet as Potassium hexacyanoferrate(III)CAS 13746-66-2The precipitant. The instruction sheet states it: the blue is Prussian blue formed from the reaction of the ferrous ions with potassium ferricyanide. This sheet classifies it as not a hazardous substance and records separately that contact with acids liberates very toxic gas	<= 100 %
Potassium dichromateCAS 7778-50-9The supplier's optional contrast agent, added as a 1 per cent solution to the sensitiser. This course gives no procedure for it under the chromium ruling, so the gram in the box has no function here at all	<= 100 %	H272, H301, H312, H314, H317, H330, H334, H340, H350, H360, H372, H410

Not disclosed. Photographers' Formulary publishes the composition of these kits in full — four named chemicals with weights, two stock strengths and a mixing ratio — which makes this the only bought product in the formulary whose recipe is not the thing being withheld. What is withheld is the specification of the materials. The sheets say "ferric ammonium citrate (green)" and give no iron content, no supplier, no lot and no assay, although the iron content of the green salt is the single variable that sets the speed, the colour and the maximum density of every print the kit makes, and runs from 14 to 18 per cent by weight across present-day commercial material. They say "arrowroot starch" while the safety data sheet the same box supplies is for starch from potato, and no document reconciles the two. They give no grade or assay for the potassium ferricyanide. They give no pH for any solution, no keeping time for the kit, for the bottles or for the stock solutions beyond the two to four hours allowed to the mixed sensitiser, no coating volume per sheet against which the "approximately twenty-four 8 by 10" can be checked, no exposure scale, no maximum density and no sensitometry of any kind. And because the safety data sheets are four separate substance sheets each stating its own component at "<= 100 %", there is no document anywhere in this literature that describes the liquid kits' bottled solutions as the mixtures they actually are: nothing states their pH, nothing states whether they carry a preservative, a biocide or a wetting agent, and nothing states how long they keep.

Nearest open formula. Classic cyanotype sensitiser — They are the same two chemicals in the same two bottles at the same one-to-one mixing ratio, and the iron half is identical: Ware's 20 g of green ammonium iron(III) citrate made to 100 mL is exactly the kit's 100 g made to 500 mL. The difference is the ferricyanide. Ware's stock B is 10 per cent w/v and the kit's is 8 per cent, so the mixed sensitiser holds 4 per cent ferricyanide where the open formula holds 5. On this course's arithmetic from the same published figures, that moves the iron-to-ferricyanide molar ratio from about 1.7-2.1 to about 2.1-2.7 to one — further from equivalence, and further into the excess iron(II) that bleeds out of a shadow and blues the highlight beside it. What the kit adds beyond the open formula is a box: the arrowroot for sizing, the bottles, and a sheet carrying an exposure, a wash, a peroxide bath, a contrast bath, three toners and a troubleshooting table. What the open formula adds is a stated iron content for the salt, an account of why every number is where it is, and no dichromate.

Purpose

To sell a beginner every solid needed to make a cyanotype, pre-weighed, with a sheet that carries the whole workflow from sizing the paper to toning the finished print.

That is a narrower claim than it sounds, and worth stating precisely. The kit does not sell a formulation: the two active chemicals are the two that every classic cyanotype has used since 1897, in a printing of the traditional strengths that any reader could weigh out from bulk stock for less money. What it sells is the removal of four decisions and one weighing session — which chemicals, which grade, how much of each, what to dissolve them in, and the balance work — and a sheet of instructions written by somebody who has actually stood at the tray.

Three catalogue numbers cover the same chemistry at two scales and in two states:

	07-0090	07-0091	07-0092
Sold as	dry solids	solutions	solutions
Ferric ammonium citrate (green)	100 g	100 g in 500 mL	20 g in 100 mL
Potassium ferricyanide	40 g	40 g in 500 mL	8 g in 100 mL
Arrowroot starch	20 g	20 g	4 g
Potassium dichromate	1 g	1 g	0.2 g
Stated yield	approx. twenty-four 8 × 10	approx. twenty-four 8 × 10	approx. twenty 4 × 5
Ferricyanide first-wash contrast bath	on the sheet	not on the sheet	not on the sheet

The three sheets agree on the two strengths that matter — 20 per cent w/v citrate, 8 per cent w/v ferricyanide — and the dry kit simply asks the buyer to make them up. This page treats the three as one product because the chemistry is one product, and says which sheet a figure came from wherever they part company.

The second half of the purpose is in the entry’s own name. Each of these sheets also carries toning solutions, credited on the sheet itself to Jan Arnow’s handbook rather than claimed as the supplier’s own: an ammonia bleach followed by tannic acid for brown to black, dilute sulphuric acid for green, borax or lead acetate for violet. Those weights are the source of this course’s tannic acid toner for cyanotype, which is an open formula precisely because the supplier published it. The kits are therefore the reason a toner page exists, and the toner chemicals themselves are not in the box — the sheet prints formulas for baths you have to buy the ingredients for separately.

One point of housekeeping, because the register that commissioned this page calls it “cyanotype and toning kits”. No separate cyanotype toning kit with a catalogue number of its own was found in the supplier’s published literature. The toning this entry covers is the three baths printed on the cyanotype sheets themselves, whose reagents are sold individually. If a packaged toning kit exists it is not documented anywhere this course has read, and the page does not invent one.

Recommended uses

A first cyanotype, by somebody who has never weighed a chemical. This is the case the kit is genuinely good at. The sheet’s route — size the paper, mix two stocks, mix equal parts, float or brush, dry in the dark, expose under a sunlamp, wash — is complete, and nothing in it requires a balance once the two stocks exist.

Teaching, where the point is the process and not the chemistry. A workshop that wants twenty-four 8 × 10 prints out of one box, without a scale on the bench and without opening a jar of powder at all, gets exactly that from 07-0091 or 07-0092, because the two active solutions arrive already made. The dichromate packet arrives with them and is not opened; see Safety.

Printing on cloth. The sheet is explicit: soak material of at least 50 per cent cotton in the standard sensitiser, dry it in the dark, stretch it, hold the negative down with glass and expose as for paper. Masking is needed if the background is to stay white. The cyanotype process entry covers what changes on a textile support.

As the sizing supply for other iron processes. The 20 g of arrowroot starch is the same sizing the supplier ships with its Van Dyke and kallitype kits, and the boil-for-five-minutes-in-a-litre procedure is the same procedure. If you already size paper, this part of the kit is not a novelty; arrowroot has its own entry.

Where a sheet of instructions has value in itself. The 07-0090 sheet carries several things the open literature makes you assemble: the greenish-yellow test for correctly dried paper, the reversal end point for exposure, the two-sided warning about the wash, a hydrogen peroxide after-bath, an oxalic acid spot treatment for blue whites, and a ferricyanide first wash for contrast. It is a better document than its typography suggests.

Not for anything requiring a known iron content. If the work depends on the sensitiser being the same next year — a sensitometric study, a calibration, a print run that has to match — the kit’s unstated grade of citrate is the thing that will move underneath you. Buy the salt with an assay on the label instead, and mix the open formula.

Not as a source of potassium dichromate. The gram in the box is not a bonus reagent. It is a chromium(VI) compound in a domestic parcel, and what to do with it is under Safety and Waste below.

When another formula is preferable

• When you already own a balance, the classic cyanotype sensitiser is this kit with a stated iron content, a chemist’s reasons for every number, twenty-five per cent more ferricyanide and no dichromate. It is the same two bottles. The only thing the kit gives you that it does not is the pre-weighing.
• When the classic sensitiser’s short scale is the problem, the New Cyanotype (1995) replaces the citrate with ammonium iron(III) oxalate and buys a longer exposure scale, a higher maximum density and a printing time cut by roughly two thirds. Photographers’ Formulary sells it as kit 07-0095 and makes those claims itself, in its own words — “a much smoother texture, a more intense blue with Dmax near that of black, and a longer tonal scale”. That kit contains ammonium dichromate, so as an object it is further from this course’s rules than the classic kit rather than nearer.
• When ammonia is unwelcome and a balance is available, Mike’s Cyanotype reaches the Simple Cyanotype’s solution through triammonium citrate instead of a titration, and the Simple Cyanotype (2019) builds the iron complex in the beaker from nothing more exotic than iron(III) nitrate, citric acid and ammonia — with the volume of ammonia as the contrast control the classic formula never had.
• When you want to know what Herschel actually did, Herschel’s 1842 cyanotype is the original proportions from his own notebook, and they are not these.
• When more contrast is wanted from a classic sensitiser, the answer this course gives is a negative with a longer density range, a duller light for proportionally longer, or the supplier’s own 0.2 per cent ferricyanide first wash — which is on the dry-kit sheet, involves no chromium and costs nothing. It is not Ware’s improved classic cyanotype, which the course records at Level D for study only because it puts a dichromate in the sensitiser, and it is not this kit’s own dichromate route either.
• When a premixed two-bottle set is all that is wanted, other suppliers sell one; Bergger’s is in this course’s bibliography and is the ordinary case — it names its two chemicals and publishes no weights at all. Bostick & Sullivan’s cyanotype kit is a separate planned entry in this formulary and is not written yet.
• When the print has to survive being touched, nothing on this page helps and no cyanotype formula does. Prussian blue is destroyed by alkali and skin is alkaline; that is a property of the image substance, and the supplier says so on page one.

What the maker publishes

Six printed pages, undated, in a typeface that has been photocopied more than once, and they contain more usable procedure than most manufacturer sheets three times their length. Taken together with the two liquid-kit sheets, this is what is on the record.

The contents, with weights. Arrowroot starch 20 g, potassium ferricyanide 40 g, ferric ammonium citrate (green scale) 100 g, potassium dichromate 1 g. That table is the reason this page reads differently from every other product page in the formulary, and it is worth pausing on: a maker printing the mass of every solid it ships is publishing its formulation, and almost none of them do.

The two stocks, with volumes and a temperature. Stock A is 400 mL of water at 20 °C/68 °F plus 100 g of the citrate, water to make 500 mL. Stock B is the same with 40 g of the ferricyanide. Three dark brown storage containers are specified, two of 500 mL and one of 100 mL, and distilled water is specified throughout. That is a make-up volume rather than an addition, which means the strengths are exact: 20 per cent w/v and 8 per cent w/v.

The mixing ratio, and the life of the mixture. Equal parts, mixed in subdued light, used as soon as feasible; “the sensitizer solution is stable for about 2-4 hours after mixing”. The sheet gives two worked examples rather than a rule — 10 mL and 10 mL, or 25 mL of both — which is the right way round for a beginner.

Two contrast controls, and a third this course does not use. Dilute the mixed sensitiser with water for a softer print (“the greater the dilution the softer your print”). Use an initial wash bath of 0.2 per cent potassium ferricyanide, 2 g in a litre, for a harder one. Or add six drops of a 1 per cent potassium dichromate solution per 2 mL of sensitiser, at a stated cost of “about 2 steps using Kodak No.2 step tablet”. The older printing goes further and gives a table for the dichromate route: 11.5 mL of sensitiser from 5 mL of each stock and 30 drops of dichromate solution, or 28 mL from 12 and 12 and 4 mL, with the note that twenty drops is approximately 1 mL.

The negative it wants. “In general, if your negative will give a good print using grade 0 paper, it will give an excellent cyanotype print using the Standard cyanotype Sensitizer Solution.” A negative that prints on grade 1 or 2 is the case for which the higher-contrast sensitiser exists. That is a more useful statement than a density range, because it is in units the reader already owns.

Coating, two ways. Float the sized side down for three minutes, or brush top to bottom and then side to side. Dry in the dark; a hairdryer is allowed if it is not too hot; hang or lay flat, and rotate a very wet sheet end for end during drying because it will streak otherwise.

The one diagnostic on the sheet, and it is a good one. “The dry sensitized paper should appear greenish-yellow. If it is blue, ferrous salts are present either by exposure or through chemical contamination.” That single sentence catches fogged paper, contaminated brushes and a stock bottle that has gone over, before any of them costs a sheet of paper.

Exposure. Sunlight or a General Electric 275 or 300 watt sunlamp at 12 to 18 inches; ten to twenty minutes; print without a glass plate where possible, “glass absorbs ultraviolet light”; the exposed print looks olive-green before washing. The end point is by inspection, and the instruction is explicit that the correct exposure looks wrong: expose until the high values carry considerably more tone than you want and the shadows have begun to reverse, “therefore an apparent overexposure is necessary”.

The wash, with the failure at both ends. Five minutes in softly running soft water. Too short and ferric salts stay in the paper and the print fades; too long and the image lightens, “particularly if the wash water is slightly alkaline”; and hard water can alter the print’s appearance because of the iron salts in it.

Two after-treatments. A hydrogen peroxide bath, 50 mL of 3 per cent peroxide in 500 mL of water, for a few seconds, to force in seconds the deep blue that air would take hours to produce — with the sheet noting that the peroxide is not supplied and can be bought at a chemist’s. And a 5 per cent oxalic acid spot application to clear whites that have gone blue, followed by a wash, with Prussian blue watercolour recommended for spotting the other way.

Sizing. Twenty grams of arrowroot starch creamed with a little hot water, made up to a litre, boiled five minutes, cooled, skimmed or decanted; brushed on across and then up and down and worked with a second dry brush until the surface loses its gloss. The older printing adds the reason: sizing “fills the pores of the paper with starch and provides a uniform surface for the image”, and the degree of sizing decides how far into the fibres the image sits. Both printings say it is optional.

Toning, in three directions, credited to somebody else. The sheet introduces the toners by naming Jan Arnow’s Handbook of Alternative Photographic Processes as where the ideas came from, which is an unusually scrupulous thing for a kit sheet to do. Brown to black: bleach in ammonia until the colour has gone, wash ten minutes in cool water, then tannic acid until the wanted colour appears, then fifteen minutes running water. Green: a 1 per cent sulphuric acid solution, made from the supplier’s own 48 per cent acid, two millilitres to a hundred, acid into water. Violet: a mild borax solution, or a warm 5 per cent lead acetate solution.

A troubleshooting table. Five rows, all of them about damp: pre-exposed coating from drying somewhere too humid or from chemicals that are too old; a negative whose emulsion was eaten away because the support was still damp; a print that started to develop in sunlight because the coating was damp; an overall blue from insufficient washing before drying; and skin stains that only strong soap will shift, with the observation that gloves prevent them.

A chemical safety section that argues rather than warns. It names potassium ferricyanide and potassium dichromate as the two needing special attention, and then explains the first one properly: the cyanide in potassium ferricyanide is bound to the iron and not free to act as a poison, “not the insidious poison that sodium cyanide is”, but the cyanide can be released as hydrogen cyanide gas if the compound meets a strong acid — “however, strong acid is not used in the cyanotype process”. That is correct, it is the right thing to tell a beginner, and it is more chemistry than most kit sheets attempt.

⁇Two printings of one sheet, and where they disagree

Two printings of the 07-0090 instruction sheet are in this course’s corpus: the one served from the supplier’s own store, and an older one mirrored by Freestyle. They are recognisably the same document and they differ in five places.

The toning bath, by a factor of ten. The older printing gives brown-to-black as ammonia 28 per cent 10 mL in 100 mL of distilled water; the current one, and both liquid-kit sheets, give 10 mL in 1000 mL. The tannic acid bath is 2 per cent in every printing — 10 g in 500 mL, or 20 g in 1000 mL, which are the same thing. So it is the bleach alone that moves, and it moves by a decade. The course’s tannic acid toner page carries both figures rather than choosing.

The headline claim. The older sheet says the kit contains chemicals to make approximately twenty-four 8 × 10 prints; the current one says it contains the chemicals to prepare 1000 mL of sensitizer solution to do approximately twenty-four 8 × 10s. The second is the more useful sentence and the first is the one most catalogues quote.

A footnote that was dropped. The older contents table carries an asterisk: “Ferric ammonium citrate (green) is somewhat light sensitive and should be stored in the dark.” The current printing prints the same table without it. Nothing else on either sheet tells the buyer to keep the jar dark, and the safety data sheet in the same PDF says “Light sensitive” and “Decomposes on exposure to light”. The dropped footnote is the only place the instruction sheet ever said so.

The peroxide bottle, reversed. One printing says the peroxide solution degasses slowly and “should be stored in a tightly capped container”; the other says it degasses and “should not be kept in a tightly capped container”. Both then say it loses potency and should be fresh. Neither reading is absurd — a degassing solution builds pressure, and a vented one loses strength — and this course does not choose between them; it uses fresh peroxide and does not store the bath at all.

The higher-contrast table. Only the older printing carries the worked table of sensitiser volumes and drop counts for the dichromate route. The current printing keeps the method and drops the table.

None of the five is a contradiction about the sensitiser itself, which is the same in every printing and in both liquid kits. But a reader who quotes “the Formulary sheet” as though there were one is quoting whichever copy they happened to download.

What is not disclosed

The composition is published, so the interesting question is what is left. It turns out to be most of what would let you predict a result.

The iron content of the citrate — the single most consequential omission on this page. The sheets say “Ferric Ammonium Citrate (green)” or “(green Scale)” and nothing else: no percentage of iron, no assay, no supplier, no lot number, no specification of any kind. Ammonium iron(III) citrate is not a defined compound. It is a family of complexes whose composition varies with how it was made, and Ware gives the green form’s iron content across present-day commercial material as 14 to 18 per cent by weight. Everything photographic follows from that number: how much iron(II) a given exposure makes, how much Prussian blue forms, how fast the paper prints, how much surplus iron is left over to bleed into the highlights. A twenty-eight per cent spread in the iron is a twenty-eight per cent spread in the sensitiser, and the buyer cannot find out which end of it they have got. The kit’s own 100 g is the same 100 g whichever it is.

Whether the starch is arrowroot. The instruction sheet says arrowroot starch, on every printing and every catalogue number. The safety data sheet the supplier ships in the same PDF is Sigma-Aldrich S4251, Starch, from potato, CAS 9005-25-8. Arrowroot is Maranta arundinacea; potato starch is not. Both are starches and both will size a sheet of paper; they gelatinise at different temperatures and give slightly different films, which is why papermakers distinguish them. No document in the supplier’s literature reconciles the two, and this page does not attempt to: it records that the instructions describe one material and the hazard documentation another.

The grade of the ferricyanide. No purity, no assay, no water content. The open formula’s source is specific where this is silent — general purpose reagent at 98 to 99 per cent is adequate — and potassium ferricyanide is at least a defined compound, so this omission has smaller consequences than the citrate’s. It is still an omission.

Any pH, anywhere. Not for stock A, not for stock B, not for the mixed sensitiser, not for the wash. The photochemistry of an iron citrate sensitiser is pH-dependent — the photoactive species is a dimer that forms above about pH 2 — and the wash instruction turns on alkalinity, since “prolonged washing will lighten the image, particularly if the wash water is slightly alkaline”. The one pH figure anywhere in the supplier’s documentation is on a Sigma sheet for a solid: potassium ferricyanide at 6.0 to 9 in a saturated solution.

Keeping, of everything except the mixed sensitiser. Two to four hours is published for the mixture and nothing at all for anything else. There is no shelf life for the unopened kit, none for the two stock solutions after they are made up, and — this is the one that matters commercially — none for the bottled solutions of the liquid kits, which are the entire product in 07-0091 and 07-0092. The sheet specifies dark brown bottles and then never says how long they last in them.

Coating volume per sheet. No millilitres-per-8 × 10 figure appears on any of the three sheets, in either coating method. The consequence is that the yield claim cannot be checked: a litre of sensitiser across twenty-four 8 × 10 sheets is about 42 mL a sheet, which is an absurd figure for a brush and a perfectly ordinary one for a tray you float paper in and then keep. That reading is this page’s inference and not the supplier’s statement — the supplier’s own New Cyanotype sheet gives 1.5 mL per 8 × 10 by rod for a different product, so the house does know how to publish the figure.

Any sensitometry at all. No exposure scale, no density range in numbers, no maximum density, no speed, no reciprocity note, no curve. The nearest thing to a sensitometric statement on the sheet is the grade 0 negative, and the nearest thing to a measurement is the two steps of a step tablet the dichromate is said to cost.

What the dichromate is doing. The sheet gives the dose and the cost in steps and no mechanism. That is a gap this course does not fill, because filling it would mean writing a procedure for a chromium(VI) contrast agent.

And, in the liquid kits, everything about the liquid. The bottles are the product. No document describes them: not their pH, not whether they contain a preservative or a biocide or a wetting agent, not their density, not their colour, not how long they keep once opened. The safety-data bundle for 07-0091 is the same four solid-substance sheets as the bundle for the dry kit.

!Why the four concentrations cannot be added, subtracted or read at all

Every other product page in this formulary has to warn a reader against totalling the bands on a mixture’s safety data sheet. Here the warning is stranger, because there is no mixture sheet to total.

The supplier’s kit “SDS” is a 35-page PDF that binds the instruction sheet to four Sigma-Aldrich sheets for four single substances. Each of them is a section 3.1 Substances sheet, and each states its one component’s concentration as <= 100 %. Four times “at most everything” is not a composition; it is four documents each describing its own jar, printed consecutively.

That is not a criticism of the supplier. A kit of four separately packaged solids is not a mixture, and supplying the substance sheet for each package is the correct thing to do. But it has two consequences a reader should hold on to.

First, the table on this page is doing a different job from the table on a developer’s product page. It is not a partial composition; it is a list of what is in the box with the hazard classification of each item, and the composition proper is on the instruction sheet, in grams.

Second, there is no hazard document for the working solutions. The sensitiser you actually handle is a mixture of two salts in water at 10 and 4 per cent, and no sheet anywhere classifies that. The Safety section below is therefore built from four substance sheets and the course’s own rubric, not from a manufacturer’s assessment of the thing in the tray — because no such assessment exists.

Disclosed components

Four, in the order the safety-data bundle prints them, each with the concentration its own sheet gives. The concentrations are all the same and all uninformative, for the reason set out above; the classifications are not, and the fourth line is why this kit needs a policy rather than a warning.

Starch, from potato, CAS 9005-25-8, EC 232-679-6, <= 100 %, no hazard statement. Sigma-Aldrich S4251, version 5.4, revision date 27 February 2015. The kit’s sizing material, and the one line where the hazard documentation and the instructions describe different substances: the instruction sheet says arrowroot starch and this sheet is for potato starch. The link above goes to this course’s arrowroot entry, because arrowroot is what the buyer is told is in the packet; the sheet is recorded here as what the supplier actually sends. The classification is the mildest in the box and is not really about photography at all: no pictogram, signal word Warning, and one hazard statement — “May form combustible dust concentrations in air” — with section 2.3 repeating combustible dust. That is a grain-elevator hazard, not a darkroom one; twenty grams creamed into hot water is not a dust cloud. The workplace limits, which the sheet gives as an ACGIH TLV of 10 mg/m³ with the remark dermatitis, not classifiable as a human carcinogen, and OSHA limits of 15 and 5 mg/m³, are nuisance-dust figures. Section 10.5 names strong oxidising agents as incompatible, which matters only because there is a strong oxidising agent in the same box. The sizing itself is boiled, and boiling starch is the one step in the whole kit that involves heat.

Ammonium iron(III) citrate, CAS 1185-57-5, EC 214-686-6, <= 100 %, H315, H319, H335. Fluka 09714, version 3.11, revision date 26 February 2015. The light-sensitive half of the sensitiser and the largest single mass in the box. Its function is the one thing the supplier does state, on the instruction sheet rather than here: the blue comes from ferrous ions produced by “photo reduction of the ferric ammonium citrate”. The classification is Skin Irrit. 2, Eye Irrit. 2A and STOT SE 3 for the respiratory system, signal word Warning — causes skin irritation, causes serious eye irritation, may cause respiratory irritation. Three things on this sheet are worth reading twice. Its formula line is printed as C₆H₈O₇ · x Fe³⁺ · y NH₃ with the molecular weight left blank, which is the manufacturer of the reagent saying in its own notation that the substance has no fixed composition — exactly the point the “not disclosed” section makes about the missing iron content. Section 7.2 says “Light sensitive. Hygroscopic.” and section 10.2 says “Decomposes on exposure to light”, which is the storage instruction the current instruction sheet dropped. And section 8 gives a glove specification most sheets omit: nitrile rubber, 0.11 mm minimum thickness, breakthrough time 480 minutes for full contact — see glove selection. Its only ecotoxicity figure is an LC0 of 200 mg/L over seven days to a killifish, which is a mild number by the standards of the fourth line on this table.

Potassium ferricyanide, named on the sheet as Potassium hexacyanoferrate(III), CAS 13746-66-2, EC 237-323-3, <= 100 %, no hazard statement. Aldrich 455946, version 4.12, revision date 23 May 2016. Section 2.1 reads “Not a hazardous substance or mixture” and the label section is empty — no pictogram, no signal word, no precautionary statement. Section 2.3 then says, on its own: “Contact with acids liberates very toxic gas.” Both are true and the pairing is the whole safety lesson of a ferricyanide bench. The compound is a stable low-spin iron(III) complex in which the cyanide is bound to the metal, and the supplier’s own instruction sheet explains it better than the hazard sheet does — the cyanide groups “are bound to the iron atom and are not free to act as a poison”, it “is not the insidious poison that sodium cyanide is”, and the gas comes only from strong acid, “however, strong acid is not used in the cyanotype process”. Section 10.5 lists the incompatibilities that follow: strong acids, strong oxidising agents, ammonia, hydrochloric acid, cyanides. Note ammonia on that list and then note that the same sheet of instructions prints an ammonia toning bath; they are used on different days and on a washed print, and the two must not meet in a bottle. The physical data are the fullest of the four: pH 6.0 to 9 at 329 g/L, relative density 1.890 g/cm³, water solubility 329 g/L and completely soluble, which is why an 8 per cent stock dissolves without argument. The limits are the tightest in the box apart from the dichromate’s — an ACGIH ceiling of 5 mg/m³ with a time-weighted average of 1 mg/m³, which is the same TWA as the citrate but with a ceiling on top of it, and an OSHA TWA of 5 mg/m³ carrying a skin designation — and the aquatic figures are unalarming, LC50 869 mg/L to rainbow trout at 96 hours and EC50 549 mg/L to Daphnia magna at 48 hours.

Potassium dichromate, CAS 7778-50-9, EC 231-906-6, Index-No. 024-002-00-6, <= 100 %, H272, H301, H312, H314, H317, H330, H334, H340, H350, H360, H372, H410. Aldrich 483044, version 4.8, revision date 4 February 2016. One gram, optional, and it changes what this box is. The composition table records it as included in the REACH Candidate List of Substances of Very High Concern, and classifies it as Ox. Sol. 2, Acute Tox. 3, Acute Tox. 2, Acute Tox. 4, Skin Corr. 1B, Eye Dam. 1, Resp. Sens. 1, Skin Sens. 1, Muta. 1B, Carc. 1B, Repr. 1B, STOT RE 1, Aquatic Acute 1 and Aquatic Chronic 1 — signal word Danger. In words: may intensify fire, toxic if swallowed, harmful in contact with skin, causes severe skin burns and eye damage, may cause an allergic skin reaction, fatal if inhaled, may cause allergy or asthma symptoms if inhaled, may cause genetic defects, may cause cancer, may damage fertility or the unborn child, causes damage to the cardiovascular system through repeated inhalation, and very toxic to aquatic life with long-lasting effects. The exposure limits are the lowest in the box by a wide margin: an ACGIH TLV of 0.050 mg/m³, twenty times below the ferricyanide’s time-weighted average, and an OSHA PEL of 0.005 mg/m³ under the chromium(VI) standard 29 CFR 1910.1026 — the only limits in this parcel set for a carcinogen. The aquatic figures are worse than they look — EC50 0.035 mg/L to Daphnia magna at 48 hours, EC50 0.31 mg/L to a green alga at 72 hours, LC50 0.131 mg/L to bluegill at 96 hours — which is roughly four orders of magnitude more toxic to water life than the ferricyanide on the line above. Section 6.2 is unambiguous: “Do not let product enter drains. Discharge into the environment must be avoided.” The supplier’s own instruction sheet, bound into the same PDF, says the opposite. This course does not use this substance at any level; the chromium ruling is the policy and the chemical entry classifies it Level D. What to do with the gram is under Safety and Waste.

Behaviour

It prints out, and that changes how you work. The image forms during the exposure, so it can be judged by inspection if the frame allows a peek without losing registration. The exposed sheet is olive-green before it is washed, and it will go blue in the tray whatever you do.

The correct exposure looks over-exposed, and the sheet says so. Highlights lighten in the wash — that is the classic sensitiser’s known weakness, a loss of the top of the scale as pigment that never got inside the paper fibre is carried away — so the print has to come out of the frame with more tone in the high values than it should have, and with the shadows just beginning to reverse. A printer who stops when the print looks right in the frame will get a weak print every time.

The bath is not reused and there is no capacity to track. The sensitiser goes onto the paper and is gone. What ages instead is the mixture, which the supplier gives two to four hours, and the stocks, for which it gives nothing at all. This is a one-shot product in the strict sense: nothing here is replenished, exhausted or tested for capacity.

Speed is slow and it is set by the salt. “Cyanotype is extremely slow. Exposure will take 10-20 minutes” under a 275 or 300 watt sunlamp at 12 to 18 inches. That is a two-to-one range for the same kit under the same lamp, and it is honest; sunlight is more variable still. The part of that range you land in depends on the iron content of your particular jar, which nobody publishes.

It goes on darkening after the wash. Air oxidises the Prussian white back to Prussian blue over several hours of drying. The peroxide bath does the same thing in seconds. A print judged wet and fresh is not the print you will have tomorrow, and the difference is large enough to mislead a beginner into over-correcting the exposure.

Water quality is a variable and the supplier says which way it goes. Iron salts in hard water alter the print’s appearance; slightly alkaline wash water lightens the image; a short wash leaves ferric salts that cause fading later. The sheet asks for distilled water for the stocks and soft water for the wash, and both requests are load-bearing.

Alkali destroys the image, permanently. Not as a subtlety but as the central fact about the material: “Prussian blue will fade when alkaline. Since perspiration is alkaline a cyanotype print can be permanently damaged if it is touched.” That is on page one of the supplier’s sheet, and it is the reason the toning kit’s first bath is a bleach.

Light fading is partly reversible and the supplier explains it. “Cyanotype prints tend to fade in strong light. The colour will return if the print is stored in a dark damp area.” That is the photoreduction of Prussian blue to Prussian white running forwards under light and back again in the dark, and it is the same reaction the peroxide bath uses deliberately.

Image characteristics

Colour. Prussian blue, and the supplier names the substance rather than describing the hue. The kit’s own words are “a long full scale and distinctive blue color”, and the process it describes can be run either way round — a blue image on white from a negative, or a pale white image on blue from a positive transparency.

Tonal scale. Claimed long, and the exposure instruction quietly says otherwise. The classic sensitiser’s exposure scale is short at the top because the pigment that forms in the coarse pores between the fibres is washed away, taking proportionally more from the light tones than the dark ones; that is why the sheet asks for apparent over-exposure and why it recommends a negative that would print on grade 0. The classic sensitiser page sets out the argument and the numbers, which the kit sheet does not carry.

Maximum density. Not published for this kit, and the supplier’s own comparison is the most informative thing available: its New Cyanotype sheet claims for that product “a more intense blue with Dmax (maximum density) near that of black” over the classic formula. Read plainly, the supplier is saying that the classic kit’s maximum density is not near that of black.

Texture and wash-off. Same source, same comparison: the New Cyanotype sheet says the classic formula is the one with the rougher texture and the one that loses blue into the wash water. That is the peptisation of the pigment out of the paper — this course has a separate entry for the symptom.

Bronzing. Not mentioned on any Formulary sheet, and it is the commonest visual surprise in cyanotype: a heavily exposed area that goes metallic and coppery rather than deeper blue. See bronzing and reversal, and note that the sheet’s instruction to expose until the shadows reverse is deliberately taking the print into the region where it happens.

What toning does to all of the above. Nothing on the print’s blue, because the toner destroys the blue first. The ammonia bath bleaches the Prussian blue out entirely and the tannic acid bath puts something else in its place; the result is brown to black according to how long the second bath runs, and it is not reversible. The green and violet routes are colour changes to the Prussian blue itself rather than replacements of it. Toned cyanotype is the process entry.

The mechanism

The supplier states the mechanism in one sentence and it is correct: “The blue color of the print is due to the Prussian blue formed from the reaction of ferrous ions (from photo reduction of the ferric ammonium citrate) and potassium ferricyanide.” That is more chemistry than most kit sheets attempt, and everything below is either an expansion of it from the open literature or an inference this page marks as one.

⚗What the two bottles do, and which half of it the supplier claims

Light reduces the iron; the citrate pays. All the photochemistry belongs to the iron salt. The photoactive species is a dimeric citratoferrate(III) anion; light reduces iron(III) to iron(II) at the expense of the citrate, which is oxidised and loses carbon dioxide. The organic photoproduct has been identified as acetone dicarboxylic acid, and the iron(II) photoproduct has not been identified at all, which is why this course prints the change as a scheme rather than a balanced equation.

hν + [Fe₂(C₆H₄O₇)₂(OH₂)₂]²⁻ → 2 [Fe^II] + CO(CH₂COOH)₂ + CO₂

The photolysis, as a scheme

The iron(II) then precipitates the blue, in the dark. A plain electron transfer between two iron centres, needing no light at all:

Fe²⁺ + [Fe(CN)₆]³⁻ → Fe³⁺ + [Fe(CN)₆]⁴⁻

The oxidation states switch

and the iron(III) and hexacyanoferrate(II) precipitate together as Prussian blue. The supplier’s sentence names the reactants and the product and stops there; the two-step account, the identity of the photoactive dimer and the uncertainty about the iron(II) species come from the open literature and are set out in full on the classic sensitiser page.

Over-exposure reduces the blue in its turn, to colourless Prussian white, which is what “the shadows have begun to reverse” means on the exposure line of the sheet. Air puts it back over hours; the kit’s hydrogen peroxide bath puts it back in seconds.

Prussian white + hydrogen peroxide → Prussian blue + hydroxide ions

Reoxidation

The supplier describes the effect twice without naming the chemistry — a faded print recovers in a dark damp place, and a faded print can be restored with a peroxide bath — and both descriptions are this reaction.

🧮What the kit's own weights make, and how far it is from equivalence

This arithmetic is the course’s, worked from the supplier’s own published weights and make-up volumes. Nothing in it is read off a safety data sheet, and nothing in it is a number the supplier prints.

Stock A is 100 g of the citrate made to 500 mL: 20 per cent w/v. Stock B is 40 g of potassium ferricyanide made to 500 mL: 8 per cent w/v. Mixed in equal volumes, the sensitiser is 10 per cent citrate and 4 per cent ferricyanide — 100 g/L and 40 g/L.

Potassium ferricyanide is a defined compound of molar mass 329.24 g/mol:

40 g/L ÷ 329.24 g/mol = 0.121 mol/L

Ferricyanide in the mixed sensitiser

The iron salt is not a defined compound, so it has to be reckoned through its iron content, which for the green form runs from 14 to 18 per cent by weight:

100 g/L × 0.14 ÷ 55.85 g/mol = 0.251 mol/L

Iron, at 14 % Fe

100 g/L × 0.18 ÷ 55.85 g/mol = 0.322 mol/L

Iron, at 18 % Fe

So the kit’s sensitiser holds between about 2.1 and 2.7 moles of iron(III) for every mole of ferricyanide. The open formula, with the same citrate but a 10 per cent ferricyanide stock, holds about 1.7 to 2.1. The image-forming step is one iron(II) to one ferricyanide, so both are in iron excess and the kit is further into it — there is more iron(II) with nothing to react with, free to diffuse out of a shadow and blue the highlight beside it.

Two cautions on this arithmetic. The 14-to-18 spread is a property of commercial material and not of this kit, so the ratio a particular box delivers is uncertain by a factor of 1.3 before anything else goes wrong; and the difference between 2.1-2.7 and 1.7-2.1 is smaller than that uncertainty. What can be said without over-reading is the direction: the kit is a slightly weaker-in-ferricyanide printing of the classic formula, and both are well away from chemical equivalence. Whether the supplier chose 8 per cent for a reason is not published and is not guessed at here.

⚗The three toning routes, and how much of each is established

The sheet prints three toners and explains none of them. What the course can say, marked for how well it is established:

Brown to black — ammonia then tannic acid. The first bath is established: alkali hydrolyses Prussian blue irreversibly, and a buffer at pH 9.4 will decolourise it completely in one to ten minutes. That is the same reaction that makes a cyanotype vulnerable to a fingerprint, used deliberately. The second bath is not established: tannic acid puts a brown-to-black substance where the blue was, and no source read for this course identifies it. An iron-tannate is the obvious candidate and this page does not assert it. The open formula page is where the evidence and the gap are set out.

Green — dilute sulphuric acid. A colour change to the Prussian blue rather than a replacement of it, and unexplained by the supplier. Note the collision with the ferricyanide sheet’s section 2.3: acid and ferricyanide liberate very toxic gas. The print being toned has been washed, so the ferricyanide is gone; the acid must never meet the sensitiser, the stock bottle or an unwashed sheet, and this is the one place on the whole sheet where the supplier’s own instructions could be misread into danger.

Violet — borax or lead nitrate as the acetate. A mild alkali on one hand and a soluble lead salt on the other, offered as alternatives for the same result, which is an odd pairing and unexplained. The course does not give a lead procedure.

The nearest open formula

The classic cyanotype sensitiser — and for once the comparison is not between a published performance and a secret composition, because both compositions are published. It is a comparison between the same formula printed twice.

	Classic cyanotype sensitiser	Photographers’ Formulary kit
Stock A	20 g green ammonium iron(III) citrate to 100 mL	100 g to 500 mL — the same 20 % w/v
Stock B	10 g potassium ferricyanide to 100 mL	40 g to 500 mL — 8 % w/v
Mixed sensitiser	10 % citrate, 5 % ferricyanide	10 % citrate, 4 % ferricyanide
Iron : ferricyanide, molar	about 1.7-2.1 : 1	about 2.1-2.7 : 1
Iron content of the salt	stated: 14-18 % for the green form	not stated anywhere
Surfactant	Tween 20, named as not part of the traditional formula	none, and none mentioned
Life of the mixture	short; the source says so without a figure	2 to 4 hours, published
Sizing	not part of the formula	20 g arrowroot, in the box
Contrast control	negative, light, coating	dilution; a 0.2 % ferricyanide first wash; or a dichromate
Chromium in the package	none	1 g of potassium dichromate
What it costs	two chemicals, a balance and an evening	a box
Safety level	B	B, with a Level D solid in the parcel

Where the two are identical. The iron bottle, exactly. Twenty grams to a hundred millilitres and a hundred grams to five hundred are the same solution, and the coating, drying, exposure and washing are the same operations in the same order. If you buy the kit you are printing the classic cyanotype, not a proprietary variant of it.

Where they differ, and by how much. One number: the ferricyanide, 4 per cent against 5. The consequence is set out in the arithmetic above and it is a real difference in the same direction as the formula’s known weakness rather than against it. It is also smaller than the uncertainty in the iron, which is the honest way to hold it: you cannot tell the two sensitisers apart on a print unless you know the iron content of both salts, and the kit does not tell you its own.

What the kit adds. Pre-weighed solids, so no balance; three bottles’ worth of chemistry in one parcel; the arrowroot; and a sheet that is a better beginner’s document than the open formula’s provenance is, because Ware’s account is a chemist’s and the kit sheet is a printer’s. The exposure figure, the greenish-yellow test, the wash warnings and the peroxide bath are all on the kit sheet and none of them is a formula.

What the open formula adds. A stated iron content, which is the variable the kit hides; a reason for every number; a surfactant recommendation with an explanation of what it is for; and no dichromate. The last one is not a small difference. Buying the open formula’s two chemicals from the same supplier gets you the same sensitiser without a chromium(VI) compound arriving in the post.

So which. If you own a balance, mix the open formula: it is cheaper per print, it is fully specified, and it does not oblige you to dispose of a gram of a REACH Substance of Very High Concern you never wanted. If you do not own a balance, or you are teaching, or you want the process before the chemistry, the kit is a reasonable purchase and 07-0091 or 07-0092 — where the solutions arrive already made and no jar of solid is ever opened — is the better one of the two forms. In either case the dichromate is not part of the procedure this course gives.

Safety

Level B, and the reasoning has two halves because the box has two halves.

Why the working chemistry is Level B and not A. The mixed sensitiser is a solution of two salts in water at room temperature, poured into a tray or brushed onto paper. The most severe classification on either of the two active substances is Skin Irrit. 2 / Eye Irrit. 2A / STOT SE 3 for the citrate; the ferricyanide is not classified as hazardous at all. That is within Level A’s ceiling on hazard. What lifts it is the rest of the operation: the dry kit requires weighing 100 g and 40 g of fine powders, one of which carries an ACGIH ceiling value of 5 mg/m³, and the process requires a UV exposure source — a 275 or 300 watt sunlamp at 12 to 18 inches, which is a skin and eye exposure that does not feel like one. Fine powders that must not be inhaled and UV units are both named in the Level B criteria. Follow weighing a solid and the UV unit SOP.

Why B and not C, when there is a chromium(VI) compound in the parcel. The Level C criteria name “chromium(VI) compounds, formaldehyde and similar substances in any quantity”, and this box contains one, so the question has to be answered rather than stepped around. The level is assigned to the operation the page gives, not to the contents of the packaging, and the operation this page gives uses two salts and water. The dichromate is not a reagent here: no procedure on this page opens it, no solution is made from it, and its own entry is Level D, historical study only. It is handled as a hazardous item — labelled, isolated, and routed to licensed disposal — which is a storage-and-waste problem rather than a laboratory operation. The same reasoning, in the same words, is what puts the Van Dyke Brown sensitiser at Level B with five grams of the same substance in its kit. A reader who does open the packet and follow the supplier’s contrast method has left this course’s procedure and is at Level C at best, needing engineered extraction and a licensed waste route, and this page gives them nothing to work from.

Why the liquid kits are the same level and a smaller risk. 07-0091 and 07-0092 arrive with the two active chemicals already dissolved, so no jar of powder is opened in the procedure and nothing is weighed — the dichromate packet is the only solid in those boxes and it stays shut. That is the same trade the whole product-page category exists to describe: you can buy your way out of a weighing step, and here you can do it without buying your way out of knowing the formula, because the formula is on the sheet either way.

⚠The gram of dichromate: what it is, and what to do with it

Every one of these kits ships potassium dichromate — 1 g in the litre-scale kits, 0.2 g in the small one — as an optional contrast agent, with a published method for using it.

Potassium dichromate is a chromium(VI) compound. The supplier’s own safety data sheet for it, bound into the kit’s own PDF, records it as a REACH Substance of Very High Concern and classifies it with twelve hazard statements including H340 may cause genetic defects, H350 may cause cancer, H360 may damage fertility or the unborn child, H330 fatal if inhaled, H334 asthma symptoms if inhaled and H410 very toxic to aquatic life with long-lasting effects. Its occupational limits are an ACGIH TLV of 0.050 mg/m³ and an OSHA PEL of 0.005 mg/m³ under the chromium(VI) standard, the lowest in the box by a wide margin and the only ones in it set for a carcinogen.

The course’s chromium ruling is that chromium(VI) is never used, at any level, anywhere. So this page gives no dichromate procedure, and the higher-contrast sensitiser is recorded above as something the supplier publishes rather than as something to do. The chemical’s own entry classifies it Level D, historical study only.

What to do with the packet, in order. Do not open it. Do not tip it down a sink, whatever the instruction sheet says. Label it as what it is, following labelling a container, including the CAS number 7778-50-9 and the words chromium(VI) — hazardous waste. Store it away from paper, wood, cardboard and anything else reducing or combustible, because it is also an oxidiser (H272). Then put it into a licensed hazardous-waste route as set out under Waste, and record it in your stock list so that the next person to open the cupboard knows what is in there.

If you have already used it, everything it touched — tray, brush, measuring cylinder, the wash water — is chromium waste, and the same route applies. Do not decide this in the darkroom on the night; the disposal page has the standing arrangement.

☠Two disposal instructions in one box, and the course follows neither

The kit’s instruction sheet says: “To dispose of excess solid potassium dichromates always wash the solid down a drain with copious amounts of water.”

The Sigma-Aldrich safety data sheet for potassium dichromate, bound into the same PDF, says at section 6.2: “Do not let product enter drains. Discharge into the environment must be avoided.”

They cannot both be followed, and the drain instruction is the one that is wrong. Chromium(VI) is classified H410, very toxic to aquatic life with long-lasting effects, with a 48-hour EC50 to Daphnia of 0.035 mg/L; a gram of dichromate is enough to exceed that concentration in ten thousand litres of water. Domestic drainage is not a treatment process. In many jurisdictions the instruction is also unlawful.

The safety data sheet’s own section 13 is not much better. It directs the reader to “burn in a chemical incinerator equipped with an afterburner and scrubber but exert extra care in igniting as this material is highly flammable” — boilerplate that contradicts the same sheet’s classification of the substance as an oxidising solid three pages earlier. Potassium dichromate is not flammable; it makes other things burn.

What the course does instead is on the disposal page: a labelled container, a licensed hazardous-waste route, and nothing chromium-bearing in a drain, ever. The lesson generalises beyond this box — a supplier sheet and its own safety data sheets can contradict each other, and the reader has to be the one who notices.

Gloves, and a specification for once. The citrate’s sheet gives nitrile rubber at a minimum thickness of 0.11 mm with a breakthrough time of 480 minutes for full contact, which is a real number rather than the usual “wear suitable gloves”. Glove selection is where the choice goes. Gloves also solve a cosmetic problem the sheet mentions: iron stains on skin come off only with strong soap, and “rubber gloves prevent skin stains”.

Eye protection. Safety glasses with side shields to EN 166 on all four sheets. The citrate is Eye Irrit. 2A — causes serious eye irritation — and it is the substance you will be handling in the largest quantity.

Dust, if you bought the dry kit. Both the citrate and the starch sheets warn about dust formation and ask for exhaust ventilation where dust is formed; the citrate is STOT SE 3, may cause respiratory irritation, and the ferricyanide carries an ACGIH ceiling value of 5 mg/m³. Weigh into a container rather than onto a paper, do not brush spilled powder about, and follow weighing a solid and PPE donning and removal.

The acid rule of a ferricyanide bench. The ferricyanide sheet’s section 10.1 and section 2.3 say the same thing twice: contact with acids liberates very toxic gas. Nothing in the standard cyanotype procedure is acidic, and the supplier says so. But three things on the same sheet of instructions are: the 5 per cent oxalic acid spot treatment, the 1 per cent sulphuric acid green toner made from a 48 per cent stock, and any citric-acid clearing bath a reader might bring from another process. Those are used on washed prints, never near the stock bottles, and never in the same tray. Keep the acid on a different shelf and read incompatibilities.

Ammonia, if you tone. The brown-to-black bath is 28 per cent ammonia solution diluted into water, and the ferricyanide sheet lists ammonia among its incompatible materials. Dilute it in a ventilated space, add ammonia to water and not the other way round, and keep it away from the sensitiser entirely. Ammonia and dichromate must never meet.

Ultraviolet. Ten to twenty minutes under a 275 or 300 watt sunlamp at 12 to 18 inches is an exposure to your skin and eyes as well as to the paper, and the lamp is hot as well as bright — the sheet’s own warning is that considerable heat is generated and the lamp must not be brought close to the negative. Enclose the source, use a timer, and do not watch it work.

What is not a hazard here, and it is worth naming. There is no free cyanide. The supplier’s own explanation is correct and is the right thing to tell a nervous beginner: the cyanide in potassium ferricyanide is bound to iron and is not available to poison anybody, and the compound is classified by its own manufacturer as not a hazardous substance or mixture. Nothing in the standard procedure is heated except the sizing water, nothing is volatile, and nothing is flammable — except the dichromate, which is an oxidiser and therefore makes other things flammable instead.

Storage

The kit as sold: no published shelf life at all. Not for the box, not for the sachets, not for the bottled solutions of the liquid kits. That is the largest single gap in the supplier’s literature for a product whose selling point is that the solutions come ready-made.

The green citrate wants to be in the dark, and only one printing says so. The older 07-0090 sheet footnotes its contents table: “Ferric ammonium citrate (green) is somewhat light sensitive and should be stored in the dark.” The current printing prints the table without the footnote. The Fluka sheet independently says “Light sensitive. Hygroscopic.” in section 7.2 and “Decomposes on exposure to light” in section 10.2. Take the footnote and the safety data sheet together as the instruction: dark, dry, tightly closed, and a hygroscopic solid weighed promptly rather than left open on a bench.

The stocks: dark brown bottles, and then silence. Three containers are specified — two of 500 mL and one of 100 mL — and no keeping time is given for what goes in them. Practically, treat stock A as the one that ages: it is the light-sensitive, hygroscopic, indefinite-composition half, and a stock that has started to reduce will announce itself the first time a sheet dries blue instead of greenish-yellow. That test is the supplier’s own and it is the most useful storage instrument on this page.

The mixture: two to four hours, and it is not a suggestion. Mix what the session needs and no more. This is the one keeping figure the supplier publishes and the one most often ignored.

The peroxide bath: make it fresh and do not store it. The supplier says it loses potency, and its two printings disagree about whether to cap it tightly. The course’s practice is to mix the 50 mL in 500 mL on the day and pour it away afterwards, which makes the disagreement moot.

The dichromate: locked, labelled, and separated from everything reducing. Its own sheet asks for a tightly closed container in a dry, well-ventilated place and files it under strongly oxidising hazardous materials. It must not sit next to paper, wood, cardboard, powdered metals or organics, and it must not sit next to acid. If your darkroom does not have a place that satisfies all of that, that is a reason to get the packet out of the building rather than a reason to improvise.

Labelling. Every bottle you fill from this kit gets a label with the contents, the strength, the date and your initials — labelling a container — because stock A and stock B are both pale yellow-brown liquids in brown bottles and the sheet gives you no other way to tell them apart.

Incompatibilities

• Ferricyanide and acid. Sections 2.3 and 10.1 of the Aldrich sheet: contact with acids liberates very toxic gas. This governs the whole bench. No acid stop bath, no citric acid clearing bath, no oxalic acid, no sulphuric acid anywhere near stock B or the mixed sensitiser.
• Ferricyanide and ammonia. Also on the section 10.5 list, alongside strong oxidising agents, hydrochloric acid and cyanides. The kit’s own toning bath is ammonia; it belongs in a different part of the room and a different part of the evening.
• Dichromate and organics. Its section 10.5 names organic materials, acids, powdered metals and hydrazine. In a darkroom that means paper, cardboard, wooden shelving, cotton rag and every one of the other three chemicals in the box.
• Dichromate and the drain. Not a chemical incompatibility but the operative one. See Waste.
• The citrate and strong oxidising agents. Section 10.5 of the Fluka sheet, and the strong oxidising agent in question is in the same parcel.
• The starch and strong oxidising agents. Section 10.5 again, and the same remark applies. The supplier ships an oxidiser in a box with three things it should not be stored against.
• Alkali and the finished print. Prussian blue is destroyed by alkali. That includes soap, hard-water scale, alkaline buffered mount board, most household cleaners, and skin. The supplier’s page-one warning about perspiration is the version of this that matters most often.
• Iron in the wash water. Hard water can alter the print’s appearance because of the iron salts in it, which is the supplier’s own statement and the reason it asks for soft water.
• Cross-contamination with silver processes. A ferricyanide-contaminated brush or tray taken to a Van Dyke or a kallitype bench will bleach silver. See cross-contamination between alt processes and keep dedicated, labelled utensils.

Waste

The wash water from a normal cyanotype is the bulk of it, and it is not innocuous. It carries unexposed sensitiser: iron citrate and ferricyanide, at whatever the coating left behind. The ferricyanide’s own aquatic figures are mild — LC50 869 mg/L to rainbow trout — and the volumes in a home darkroom are small, but “mild” is not “nothing”, and the standing arrangement is on the disposal page and in general chemical waste. Check what your local authority accepts before the first session rather than during it; the supplier’s own advice is to do exactly that.

Never pour a ferricyanide solution into anything acidic, including a drain that has just taken an acid stop bath from another process. This is the same rule as the incompatibility, restated where it is most likely to be broken.

Spent stock solutions. Stock A and stock B, if they have to be discarded, are the same materials at twenty and eight per cent and are the concentrated case of the paragraph above. Do not mix them before disposal; keep them separate and labelled, exactly as they were stored.

The dichromate is hazardous waste and there is no domestic route for it. Not the drain, not the bin, not diluted, not “a gram is nothing”. It is chromium(VI), classified H410, and its own safety data sheet forbids discharge to the environment. Bag it, label it with the substance name and CAS 7778-50-9, and take it to a licensed hazardous-waste facility or a household hazardous-waste collection that accepts chromium. Follow solid spill if the packet has burst, and do not sweep it dry.

The supplier’s own drain instruction is not followed here, and the reason is set out in the callout under Safety. It is worth saying once more plainly, because the sentence is printed in a box that beginners buy: washing dichromate down a drain with copious water is bad advice, and the safety data sheet in the same envelope contradicts it.

Toning waste, if you go there. The ammonia bleach is alkaline and dilute; the tannic acid bath is organic and dilute; the sulphuric acid green toner is acid and must not meet ferricyanide waste; the lead acetate violet toner produces lead-bearing waste, which is a hazardous-waste stream in its own right and the reason this course gives no lead procedure.

Paper and packaging. Sensitised offcuts and blotters carry iron and ferricyanide. Dry them, keep them out of the compost, and put them in general waste rather than recycling. Packaging that held the dichromate is contaminated packaging and goes with the dichromate.

Troubleshooting

The dried coating is blue instead of greenish-yellow. The supplier’s own diagnostic, and it means iron(II) is already present — from light striking the coated sheet, or from contamination. Check the safelight conditions, the brush, the tray and the age of stock A. See blue fog in cyanotype highlights.

The print is weak and the highlights have gone. Two causes and the sheet names both. Under-exposure — remember that the correct exposure looks over-exposed and the shadows should have begun to reverse — or over-washing, “particularly if the wash water is slightly alkaline”. See weak cyanotype maximum density.

The whole print turned blue while it dried. The supplier’s own troubleshooting row: it was not washed enough. Wash properly and dry in a darkened room.

The image faded after a few weeks and the paper looks fine. Alkali. Check the mount board, the storage sleeve and whether the print has been handled — the sheet’s warning about perspiration is not theoretical. See cyanotype bleached by an alkaline wash.

Metallic, coppery shadows instead of deeper blue. Bronzing, from the heavy exposure the sheet asks for. Bronzing and reversal is the entry; it is often a sign you are at the right end of the exposure rather than the wrong one.

Blue has crept from a dense area into the highlight beside it. Excess iron(II), which the kit’s sensitiser has slightly more of than the open formula does — see the arithmetic under The mechanism. It is worse on a heavily sized paper and worse with a long exposure.

Blue washed out of the print and coloured the water. Peptisation of the pigment out of the paper. The entry has the argument; the supplier’s own comparison sheet says this is a known weakness of the classic formula against the New Cyanotype.

Yellow stain in the highlights that will not wash out. Residual iron. Yellow iron stain in cyanotype highlights, and the supplier’s answer is the 5 per cent oxalic acid spot treatment followed by a wash.

Streaks and brush marks. Brush coating top-to-bottom then side-to-side is the supplier’s method; the failure modes are in blotchy and streaked hand coating. If the sheet was very wet, the sheet’s own advice is to rotate it end for end while it dries rather than to hang it and let it run.

The coating fogged before printing, in patches. The supplier attributes it to a drying place that is too humid, or to chemicals that are too old. See mottled coating from humidity and repellency spots if the sensitiser beaded rather than wetted.

The negative’s emulsion was damaged during exposure. The supplier’s row: the support was still damp when it was exposed. Dry the sheet properly, and see uneven contact print for the related registration problems.

Two prints from the same box do not match. Ask when each was coated and how long the sensitiser had been mixed — two to four hours is the published life. If that is not it, the unstated iron content is the variable you cannot check, which is the argument of this whole page.

Blue fingers. Gloves, and strong soap for what is already there. The supplier says so, and it is the only entry in its troubleshooting table that is purely cosmetic.

Experiments

Find out what the iron content of your jar actually is, by proxy. You cannot assay the salt at home, but you can measure the thing it controls. Coat a strip, expose a step tablet, wash, dry, and read the result on your densitometer; record the box, the catalogue number and the date. Do it again with the next box. The supplier does not publish the number that moves between them, so your own record is the only version of it you will ever have — and this is exactly the argument for keeping a formula version record for a bought product.

Put the kit against the open formula, one variable apart. Mix the classic sensitiser from bulk chemicals at 10 per cent ferricyanide, mix the kit’s at 8, coat two strips of the same paper from the same batch, expose them together under the same lamp and read both. The predicted difference is small and in a known direction. A null result is publishable and would tell you the ferricyanide is not the limiting reagent at either strength, which is worth knowing.

Measure the two-to-four hours. Mix 20 mL of sensitiser and coat a strip from it immediately, then at one, two, four, eight and twenty-four hours, keeping the mixture in the dark. Expose all six identically. The supplier’s figure is a range with no evidence attached; you can have a curve.

Test the greenish-yellow diagnostic. Deliberately fog a coated sheet — a few seconds of room light — and photograph it beside a correctly dried one. Knowing what the failure looks like before it matters is worth more than reading about it, and this is the safest deliberate failure in the whole kit.

Measure the ferricyanide first wash. The 0.2 per cent bath is on the dry-kit sheet and on neither liquid sheet, it involves no chromium, and the supplier quantifies nothing about it. Two identical prints, one washed normally and one given the 2 g/L bath first, both read on a densitometer. If it works, it is the contrast control this course would rather people used.

Measure the dilution control the supplier declines to specify. “The greater the dilution the softer your print” is an instruction with no numbers in it. Coat at full strength, at 1+1 and at 1+3, expose identically, and plot the three curves. You will have specified the supplier’s own instruction better than the supplier did.

Time the peroxide bath against the air. One print into the 50 mL-in-500 mL bath for five seconds, its twin left to dry in the dark, both read at ten minutes, one hour, six hours and twenty-four. This is the reoxidation of Prussian white measured rather than described, and it will also tell you how much of the “print looks weak when wet” effect is real.

Test the alkali sensitivity that the supplier warns about. A spare print, a fingerprint pressed firmly onto one corner, and a week. Then a second corner touched with a drop of dilute borax solution. The result is the reason cyanotypes are matted rather than handled, and it is the same reaction the brown-to-black toner uses on purpose.

Run the toner the kit prints, at both published strengths. The tannic acid toner exists as an open formula because this supplier published it, and its two printings differ tenfold in the ammonia bath. Two matched prints, one bleached at 10 mL in 100 mL and one at 10 mL in 1000 mL, both toned identically afterwards. Whether the tenfold difference is a typographical error or a real change in the method is currently an open question in this course, and a careful printer could close it.

Record all of it. A bought product earns a line in the formula version record as much as a mixed one does — the catalogue number, the purchase date, the date each stock was made up, the water used, and what the first sheet off each box did. On a product whose formula is published and whose materials are not, your own record of the boxes is the only specification you will ever have.