Ammonium citrate developer

A palladium print developed in potassium oxalate is brown. The same print developed in a citrate bath is grey, and the grey is close enough to platinum that people mistake one for the other. That is the whole commercial argument for this developer, and every source read for this page makes it in the same direction and none of them measures it.

What none of them did for a long time was say how much salt goes in the bottle. The firm that sells it prints the chemical formula, the molar mass and the size of the bottle, and sells the powder already weighed so that you never find out. Ware’s Platinomicon names the bath twice and gives no strength. Three published accounts by working printers name it and give no strength. One does — and the figures below are his.

Ingredient	Quantity	Form the source specifies
Ammonium citrate (dibasic)	500 g	Mougin writes only "ammonium citrate". The course reads it as the dibasic salt, diammonium hydrogen citrate (NH4)2HC6H5O7, CAS 3012-65-5, formula weight 226.18 — which is the formula and weight the supplier prints on the bottle, the salt Ware calls "mildly acidic diammonium citrate", and the salt the preparation in Part 3 of Mougin's own text actually makes. The tribasic salt sold under the same name is a different substance at 243.22 and its solution is faintly alkaline; the argument is under Mixing. 500 g of the dibasic salt is 2.21 mol
Water	1500 mL, added	at 50 °C; "Water, at 50 C or more, 1500 cc." An added volume and not a make-up volume, so the finished bath is more than 1500 mL and the nominal 33.3 g per 100 mL overstates the true strength. The temperature is for dissolving and nothing else: the salt is soluble in about one part of water, so a third of a gram per millilitre is nowhere near saturation and goes in cold if you are willing to wait. Distilled or deionised water, because calcium arrives in a bath whose whole business is carrying iron away. Above 50 C the make-up step leaves Level A and carries the Level B controls named under Safety.
Added to the working bath, not to the stock
Hydrogen peroxide	0.5 mL of a 3% solution per 100 mL of working solution	medium contrast negative, density range 1.35 — Optional, and Mougin's own method rather than received practice — he marks the section with his own symbol for a personal interpretation. He prints one "oxidizer" column after naming two oxidisers, a 3 per cent hydrogen peroxide solution and a 4 per cent potassium dichromate solution; only the peroxide is published here. The same table gives 1.0 cc to the potassium oxalate bath for the same negative, so this developer takes half the dose
Hydrogen peroxide	1 mL of a 3% solution per 100 mL of working solution	moderate contrast negative, density range 1.20 — Twice the previous dose, for a negative about half a stop flatter. The potassium oxalate bath is given 2 cc for the same negative
Hydrogen peroxide	2 mL of a 3% solution per 100 mL of working solution	low contrast negative, density range 1.05 — The top of Mougin's published range for this bath. For a contrasty negative of density range 1.8 he prints 0 cc, and that row is the reason the dose is described here as optional

Purpose

To dissolve the iron(II) that light made, so it can reach the palladium and reduce it, and to do that at a lower reducing power than oxalate. A siderotype developer contains no developing agent. The reducing agent was made during the exposure, in the paper, out of ferric oxalate; it is iron(II) oxalate, it is almost insoluble, and until a ligand picks it up it cannot react with anything. The sodium citrate developer page sets that chemistry out in full and this page does not repeat it.

The second purpose is what separates this entry from that one. To do the same job with the acid ammonium salt, which carries its own acid and therefore needs no second solid weighed into the bath. Willis’s palladiotype developer is trisodium citrate plus citric acid at a tenth of its weight, because the trisodium salt alone gives a solution near pH 8 and iron(III) hydrolyses above pH 4. Diammonium hydrogen citrate still has one carboxyl group protonated, and its solution is mildly acidic without help. That is why the formula above is one line.

The third purpose is practical rather than chemical and belongs to the papers this developer gets chosen for. Ware records that on the lightest Japanese washi, around 10 gsm, alkaline tetrasodium EDTA damages the fibres, so “the use of mildly acidic diammonium citrate as developer and clearing agent is preferred”. A bath that develops and clears, at a pH the paper survives, is doing two jobs that would otherwise need two chemistries.

Recommended uses

Palladium prints where you want grey rather than brown. This is the use every source names. If a palladium print looks like a warm sepia and you wanted it to look like platinum, the developer is the lever, and it is a cheaper one than buying platinum.

Any paper where the oxalate developer is fogging the high values. Ware’s comparative tests find that the oxalate bath “can reduce the Pd(II) to some extent, thereby causing a brownish grey fogging of the highlights by palladium metal, quite distinct from the yellow stain of iron”, worsened by humid conditions. A citrate bath does not do that, and the argument for why is on the sodium citrate page.

Lightweight and unsized papers, and Japanese washi in particular. This is Ware’s own reason for naming the salt at all, credited to the printmaker Gilles Lorin: on a 10 gsm gampi sheet the alkaline EDTA clearing bath attacks the fibres, and a mildly acidic citrate that can serve as both developer and clearing agent avoids the problem. It is a practitioner’s finding communicated privately to a Tier 1 author, not a published comparison, and the course marks the difference.

Where you want one bottle to do the developing and the clearing. The same salt at the same kind of strength appears in Ware’s account as a clearing agent as well as a developer. That is not a licence to skip the clearing sequence; it means the clearing sequence can be built from the same chemistry rather than from a harsher one.

Reading the modern kit literature. Bostick and Sullivan sell four developers for platinum and palladium and their kit instruction sheet describes exactly one of them. Knowing that the ammonium citrate bottle is the cool-toned option, and what is in it, makes the catalogue legible.

When another formula is preferable

• When the metal is platinum and you want it black rather than grainy. Every source read for this page that works platinum seriously develops it hot. Mougin puts platinum at 50 to 100 °C to avoid graininess and prints this bath under a heading of 15 to 20 °C. Use the potassium oxalate developer, which the whole platinum literature is written around.
• When you are using hexachloroplatinate(IV) as a contrast agent. Ware is unambiguous: it “cannot be employed with any chemistry involving ammonium cations, in sensitizer or developer, because ammonium hexachloroplatinate(IV), (NH₄)₂PtCl₆, has a very low solubility and will crystallise out”. This is the one incompatibility that belongs to the cation rather than to the anion, and it is the strongest single reason to prefer the sodium citrate bath over this one.
• When you want the warmest palladium the process gives. Oxalate produces a warmer colour and, in Ware’s tests on the sodium bath, higher densities. Mougin’s own ranking puts potassium oxalate at the warm end and sodium citrate at 7 °C at the cold end, with this bath between them.
• When you want the coldest tone the process gives. That is the sodium citrate bath worked at 7 °C, on Mougin’s ranking, and it is his own preferred developer.
• When the print is a kallitype and the image is silver. Use the sodium citrate, Rochelle salt or borax developers, whose alkalinity and tone controls are written for silver rather than for palladium.
• When you are printing out rather than developing. Ware’s print-out platino-palladiotype forms its image during exposure and needs clearing, not developing.
• When you bought a kit and want to follow its sheet. Both surviving kit makers ship potassium oxalate and write their times, temperatures and clearing sequences around it. The Photographers’ Formulary and Bostick and Sullivan pages cover those.
• When you are reproducing a period Palladiotype. Willis’s own paper was developed in acidified trisodium citrate, and the 1886 Pizzighelli and Hübl formulations are the other historical route. Neither used the ammonium salt.

Mixing

Weigh, dissolve in warm water, check the pH, and stop. There is no second solid, no order to get wrong and no exotherm. The salt is soluble in about one part of water, so a third of a gram per millilitre is a third of the way to saturation and dissolves without persuading. Mougin’s “at 50 °C or more” buys speed and nothing else.

The mixingOrder above is the course’s, not Mougin’s, and with one solid it is barely an order: water first, so that the salt goes into a volume rather than a volume onto a cake.

!Two salts are sold as ammonium citrate, and only one of them is this

This is the first thing to check and the easiest to get wrong, because the label is often just “ammonium citrate”.

The dibasic salt — diammonium hydrogen citrate, (NH₄)₂HC₆H₅O₇, CAS 3012-65-5, formula weight 226.18, EC 221-146-3. Two of citric acid’s three carboxyl groups are neutralised and one still carries its proton, so a solution of it is mildly acidic: the safety data sheet the supplier files for it gives pH 5.2 at 50 g/L at 20 °C.

The tribasic salt — triammonium citrate, CAS 3458-72-8, formula weight 243.22, EC 222-394-5. All three are neutralised and its solution is faintly alkaline. It has its own encyclopaedia entry precisely because of this confusion, and PubChem’s record for it carries an annotation moved across from “Ammonium citrate”.

Why it decides the bath rather than tweaking it. The pH is the reason the developer was chosen. A bath made from the tribasic salt does not sit at pH 5.5 to 6; it sits above neutral, and above pH 4 iron(III) hydrolyses in the sheet, which is the yellow highlight stain that cannot be washed out. Substituting one salt for the other because the jar says “ammonium citrate” changes the pH of the bath without changing a single figure in the formula. The CAS number on the safety data sheet settles it.

🧮Which salt Mougin meant, and how the course knows

Mougin writes “ammonium citrate” and names no grade. Four lines of argument converge on the dibasic salt, and the second is the strong one.

The supplier prints it. The Bostick and Sullivan listing for this very developer gives Chem Formula (NH₄)₂HC₆H₅O₇ and FWeight 226.19, and the safety data sheet it files under Ammonium Citrate is a Sigma-Aldrich sheet for Ammonium citrate dibasic, CAS 3012-65-5.

The preparation Mougin himself prints makes it. Part 3 of the same text gives a make-up from scratch, attributed to Richard S. Sullivan’s lab notes: 120 g of citric acid, 280 cc of water, 120 cc of 20 per cent ammonia. Work out the stoichiometry.

120 g ÷ 192.12 = 0.625 mol citric acid · by weight: 120 mL × 0.92 × 0.20 ÷ 17.03 = 1.30 mol NH₃ → 2.08 : 1 · by volume: 120 mL × 0.20 ÷ 17.03 = 1.41 mol → 2.26 : 1

Ammonia per citric acid in Sullivan's preparation, on the two readings of "20 per cent"

Whichever way the 20 per cent is read, the ratio is close to two ammonia per citric acid — the diammonium salt — and nowhere near one or three. Mougin then says to bring the pH back to 5 to 6 with citric acid or ammonia, which trims the excess. The arithmetic is the course’s, on Mougin’s published figures; the density of ammonia solution is a standard table value and the formula weights are Ware’s Appendix V.

Ware names it. His word for the developer is “mildly acidic diammonium citrate”, and mildly acidic is exactly what the dibasic salt is and the tribasic salt is not.

The pH target only works with it. A dibasic solution reaches pH 5.2 at 50 g/L unaided. Getting to 5.5 to 6 is a nudge. Getting there from a tribasic solution would mean titrating in citric acid until you had made the dibasic salt anyway.

🧮What 500 g in 1500 cc actually is, and how it compares

Nominally 33.3 g per 100 mL. Truly less, because Mougin adds 1500 cc of water to the salt rather than making up to 1500 cc, so the finished bath is more than 1500 mL. The direction is certain; the size of the correction is not, because no source read for this page gives a density for the salt or for the solution.

Against the other citrate bath. Ware converts Willis’s palladiotype developer to trisodium citrate dihydrate at 20 per cent w/v of the made-up solution.

ammonium: 500 g ÷ 226.18 ÷ 1.5 L = 1.47 mol/L · sodium (Willis, via Ware): 200 g ÷ 294.10 = 0.68 mol/L

Citrate molarity of the two baths, nominal, from the published figures

The ammonium bath is about twice the citrate molarity of the sodium one. That comparison is the course’s arithmetic on two published figures and it carries a caveat that matters: Ware’s is per 100 mL of made-up solution and Mougin’s is per volume of water added, so the real gap is smaller than the numbers say. It is not smaller by a factor of two. The reason the ammonium salt can be taken so much further is solubility — HSDB gives it as soluble in about one part of water, so a saturated solution is near 50 per cent by weight and 33 per cent is comfortably inside it.

And against Sullivan’s from-scratch preparation. 0.625 mol of citrate becomes about 141 g of the diammonium salt in a little over 400 cc, which is roughly 35 g per 100 cc of the water it is made in, against Mougin’s 33.3. Two recipes, two authors, five per cent apart.

The pH is the step people skip. Mougin says to check it and keep it between 5 and 6, and both of his printed recipes end with the same instruction. The salt puts you close on its own; hard water, an old tub of deliquesced solid, or a batch of a different grade will not. A cheap meter or a narrow-range paper is the difference between a bath that clears and one that stains.

Behaviour

It develops in seconds and then stops mattering. Every source agrees. Auerbach: development is visible instantly, leave the print 30 or 40 seconds. Van Keuren: almost instantaneous, no longer than a minute. Mougin: one minute, and “timing is not critical and doesn’t cause any increase of contrast”. The Photographic Materials Group says the same of the process generally, and adds the sentence that governs the whole tray: contrast “is achieved mostly by exposure, not the developer”.

Which means the only thing that can go wrong is how the liquid arrives. Two sources give two different tricks for the same problem, and both exist because a bath that finishes in seconds records the order in which it wetted the sheet. Van Keuren slips the print in face up and covers the surface as fast as possible. Auerbach dams the whole volume at one end of the tray and drops the tray level so it arrives as one wave. Bostick and Sullivan’s sheet, writing about the oxalate bath, says to pour rapidly enough to break the air bubbles.

It gets colder as it gets colder. Mougin states the two relationships for palladium generally: contrast increases as the temperature falls, and image colour cools with it, over a range from 7 °C to 100 °C. This bath is printed for 15 to 20 °C, which is between the hot oxalate practice and the cold sodium citrate one.

It loads up with what the prints bring in. This is the one behaviour the sources disagree about, and the disagreement is under Storage.

It is not a vigorous bath, and that is the point. Ware’s redox table puts the citrato-iron couple at +0.372 V against +0.02 V for the oxalato couple. The lower driving force is why it does not put down palladium of its own accord in the high values, and it is the same trade that every restrainer in this formulary makes by a different route.

Image characteristics

Colour is the reason the bath exists. Three statements, from three sources, all pointing the same way and none of them measured:

• Bostick and Sullivan, on the bottle: “Produces cooler tones than the traditonal potassium oxalate developer.”
• Jill Enfield: “This will give you more neutral gray tones when used with Palladium that will make it look more similar to Platinum.”
• Stan Klimek, quoted by Ware on Hahnemühle Platinum Rag: “a very neutral black which is so hard to get from all other papers I have used”.

Speed: equal to oxalate, in the one comparison there is. Klimek’s test on that paper was a step wedge, and it found ammonium citrate “yielded equal response with the potassium oxalate”. That is worth flagging because it cuts against the neighbouring result: Ware’s own comparisons find oxalate faster than the sodium citrate bath. Either the ammonium salt at 1.5 molar behaves differently from the sodium salt at 0.7 molar, or the paper is doing the work. No source read for this page distinguishes those two explanations, and one step wedge on one paper is not a finding.

High values stay clean. No source read here reports the brownish grey palladium fog Ware’s tests find under oxalate development.

Dmax and tonal scale belong to the paper, not to this tray. Klimek measured 1.42 and a log exposure range of 2.5 on Hahnemühle Platinum Rag with a potassium oxalate bath at room temperature; Ware reports users getting 1.7 to 1.8 on gampi and kozo against 1.40 to 1.45 on a typical Western sheet, and attributes it to the absence of sizing rather than to any developer.

The mechanism

Three steps, of which the developer owns only the middle one. The full argument is on the sodium citrate developer page, including the contradiction in Ware’s own text about whether a citrato-iron complex can reduce palladium at all. What follows is the short form, plus the part that is specific to the ammonium salt.

One: light makes an insoluble salt and leaves it where it fell

Fe₂(C₂O₄)₃ + light → 2 FeC₂O₄ + 2 CO₂

Photolysis of ferric oxalate, after Ware — the gas leaves, the iron(II) stays put

Two: the citrate picks it up

FeC₂O₄ + C₆H₅O₇³⁻ → [Fe(C₆H₅O₇)]⁻ + C₂O₄²⁻

Citrate displacing oxalate from the photolytic iron(II). The course's construction on the pattern of Ware's oxalate equation; no source read prints it

Citrate is a stronger chelating agent than oxalate — three carboxylates and a hydroxyl on one small molecule — so it does by ligand exchange what the oxalate bath does by common-ion mass action.

Three: the mobilised iron(II) reduces the palladium

2 [Fe(C₆H₅O₇)₂]⁴⁻ + [PdCl₄]²⁻ → 2 [Fe(C₆H₅O₇)₂]³⁻ + Pd + 4 Cl⁻

Reduction of tetrachloropalladate(II) by the citrato-iron(II) complex, written for the bis complex that a 1.5 molar bath makes likely. The course's construction after Ware's oxalate equation

The palladium comes out as metal lodged among the cellulose fibres, and that is the photograph.

The part that is specific to this salt: the acid is already in it

HC₆H₅O₇²⁻ → C₆H₅O₇³⁻ + H⁺

The proton the dibasic salt still carries, which is what holds the bath at pH 5 to 6

Willis’s palladiotype bath is trisodium citrate plus citric acid at a tenth of the citrate’s weight, and the acid is there because a trisodium citrate solution sits near pH 8 while iron(III) hydrolyses above pH 4:

Fe³⁺ + 3 H₂O → Fe(OH)₃ + 3 H⁺

Hydrolysis of iron(III), which begins above pH 4 and ends, over years, as goethite

The diammonium salt supplies the acid and the ligand in the same crystal. One solid, one weighing, one buffer — and it is genuinely a buffer, because both the conjugate acid and the conjugate base of citric acid’s third dissociation are present from the moment it dissolves.

⚗The ammonium ion is not quite a spectator

In every equation above the cation could be anything. Three times it cannot.

It buffers on its own account. Ammonium is a weak acid in its own right, and at pH 5 to 6 the bath holds a large reservoir of NH₄⁺ several units below its pK_a. Push the bath alkaline and it pushes back:

NH₄⁺ + OH⁻ → NH₃ + H₂O

The second buffer in the bottle, which the sodium bath does not have

It is volatile, which is why the bath smells and why it drifts. The product of that reaction leaves. A sodium citrate bath losing water gets stronger; an ammonium citrate bath losing water and ammonia gets stronger and drifts upward in pH. Mougin’s instruction to maintain the pH rather than to set it is doing work here. No source read for this page measures that drift, and the course does not put a rate on it.

It has one hard incompatibility, and it belongs to the cation. Ware: hexachloroplatinate(IV) “cannot be employed with any chemistry involving ammonium cations, in sensitizer or developer, because ammonium hexachloroplatinate(IV), (NH₄)₂PtCl₆, has a very low solubility and will crystallise out.”

2 NH₄⁺ + [PtCl₆]²⁻ → (NH₄)₂PtCl₆

Why that contrast agent and this developer cannot share a process, after Ware

Function of every ingredient

Ammonium citrate (dibasic), 500 g per 1500 cc of water — the whole formula, and three jobs in one solid.

It dissolves the latent image. Iron(II) oxalate is soluble to 0.022 g per 100 cc, which is to say it is a solid sitting where the photon left it. Citrate chelates it into a mobile complex, and mobility is the entire developing action: nothing in the bottle reduces anything.

It carries the unexposed iron out. Most of the sheet was never exposed and every part of it holds iron(III) that has to leave before the print dries. The developing tray is where most of that happens, and the clearing sequence finishes it. The less iron the clearing baths inherit, the longer they last and the less likely a yellow stain becomes.

It acidifies the bath by itself. One of three carboxyl groups is still protonated, which is why the supplier’s safety data sheet records pH 5.2 at 50 g/L and why Mougin needs no second solid where Willis needed citric acid at a tenth of the citrate. This is the most useful single thing to know about the formula, because it explains the shape of it.

What more or less of it does. More citrate means faster solubilisation and a larger reservoir for the unexposed iron, at a cost in money and in the amount of salt a print carries into the wash. Less means a bath that mobilises the image more slowly than the paper absorbs the developer — the failure Willis named for the oxalate bath as the rapidity of solution overtaking the rapidity of reduction. Neither end of that is quantified anywhere read for this page. What is published is the floor beneath both: Mougin’s 50 cc for an 8 by 10, which is about 16 g of salt per print.

What it interacts with. Iron, which is the job; calcium, which it will sequester out of hard water and which has no business in a bath whose purpose is to carry metal away; and platinum(IV), through the cation, which is the incompatibility above.

Hydrogen peroxide, 0 to 2 cc of a 3 per cent solution per 100 cc of developer — optional, and a contrast control rather than a constituent. It reoxidises part of the iron(II) that light made, so that less of it is available to reduce palladium, which truncates the exposure scale and separates the tones. It is the same job potassium chlorate does in the sensitiser, moved into the tray, and Mougin’s stated reason for moving it is that it saves handling a chlorate that is “an unstable product, dangerous to manipulate, and difficult to weigh”. More of it means higher contrast and a shorter scale; the top of his published range for this bath is 2 cc per 100 cc for a negative of density range 1.05, and for a contrasty negative he prints none at all. His table gives the potassium oxalate bath twice the dose at every step, which is a measurable statement that this developer is the gentler of the two. Mougin prints one oxidiser column after naming two oxidisers, the second being a 4 per cent potassium dichromate solution; only the peroxide is published here, under the chromium policy.

Water is not an encyclopaedia ingredient but it is two decisions. Distilled or deionised, because calcium in a citrate bath is sequestered rather than removed and arrives in a solution whose whole business is carrying metal out of paper. And warm only for convenience: the salt dissolves cold, and 50 °C is Mougin’s floor for speed rather than a requirement of the chemistry.

Interactions

Citrate against iron, in both oxidation states. The bath has to hold iron(II) long enough for it to find palladium, and iron(III) long enough to carry it out of the sheet. Citrate does both, which is why one bottle can serve as developer and clearing agent on the thin papers Ware describes.

The ammonium ion against platinum(IV). Covered under The mechanism, and it is absolute: a hexachloroplatinate(IV) contrast agent and this developer cannot be used in the same process.

Acid against the palladium image. The developer is mildly acidic on purpose, but that direction has a limit which belongs to the clearing baths rather than to this one. Hydrochloric acid at 1:60 dissolves palladium and bleaches the image; the palladium tradition uses 1:200, and Mougin prefers oxalic acid because “it has no effect on the palladium metal”. A bath at pH 5 to 6 is nowhere near that line, and the reason to know where the line is, is that the developer decides how much work the clearing baths have left.

Ammonia against the pH you set. The bath’s own volatile component leaves in use, and the instruction in the source is to maintain rather than to set. See The mechanism.

Iron and palladium against the bath’s future. Mougin’s argument for one-shot working is an interaction rather than a shelf life: the ferrous oxalate the prints shed accumulates and “in strong concentration eventually veils the print in an indelible way”.

Variants

The premeasured powder, which is what most readers will actually have. Bostick and Sullivan sell this developer as a weighed charge of powder in a bottle: fill to the line with distilled water, shake, use. It comes in 1 L and 4 L, and the listing prints the chemical formula and the formula weight and no mass and no per cent w/v. Nothing in the kit instruction sheet mentions the bath at all — that sheet describes only the potassium oxalate developer the kits ship. Whether the powder is at Mougin’s strength is not knowable from anything published, and this page does not guess. What can be said is that the identity on the bottle and the identity Mougin’s own recipe produces are the same salt.

Sullivan’s from-scratch preparation, printed in Part 3 of Mougin’s text and attributed to Richard S. Sullivan’s lab notes: 120 g of citric acid in 280 cc of water, heated until dissolved; 120 cc of 20 per cent ammonia solution added; boiled; the pH regulated between 5 and 6 with more of either. It makes the diammonium salt in the vessel, and it answers the salt’s price and its deliquescence at once — you never store the solid. It is also a different hazard picture from weighing a powder: concentrated ammonia solution is corrosive with a sharp respiratory hazard, and the mixture is boiled. Its ICSC card governs, and the operation is not Level A.

The sodium sibling. Willis’s sodium citrate developer is the same chemistry with the other cation and an added acid, and Mougin’s ranking makes it the colder of the two when worked at 7 °C. It is also the one to use if hexachloroplatinate(IV) is in the process.

The other cations printed alongside it. Enfield lists five developers for platinum and palladium and ranks them: sodium acetate more contrasty than potassium oxalate, potassium citrate between them, sodium citrate lower and colder, ammonium citrate the neutral grey. She gives no strength for any of the four non-oxalate baths. Willis’s 1880 patent claims the whole family in one sentence — “the tartrate or citrate of soda, of potash, or of ammonia… or acetate of soda, of potash, or of ammonia or the monoammonic, the diammonic or the monosodic ortho-phosphates” — so the modern list is his list.

The 1887 ammonic citrate, which is a different formula and worth not confusing. Willis’s patent No 1681 of 2 February 1887 dissolves “one hundred (100) grains of diammonic orthophosphate, thirty five (35) grains of ammonic citrate and ten (10) grains of ammonic chloroplatinite” in one fluid ounce of water. That is a phosphate-and-platinum developer for a sensitiser carrying no platinum at all, in which the citrate is an accelerator among three ingredients. It is the earliest published quantity for ammonium citrate in a platinum developer read for this course, and it is not this bath.

Abney’s saturated solution, 1905. The earliest ammonium citrate developing bath read for this course is a different process again — “Ferrous citrate may be purchased and dissolved in a saturated solution of ammonium citrate, adding citric acid if required to give a clear picture”, for silver chloride paper. It is worth a sentence because of how it is specified: by saturation, not by weight. The habit of publishing this salt without a number is older than the trade article.

The contrast additions. Mougin’s peroxide doses are in the table above. The same table’s other oxidiser is a 4 per cent potassium dichromate solution and the course does not publish that dose; the chromium policy governs. Ware’s view of the whole category is worth carrying here, because it comes from the person who has tested most of them: with modern negative-making, “contrast-enhancing agents, such as chlorate, hexachloroplatinate(IV) or dichromate, become unnecessary when a correctly calibrated negative is made”, and he calls them “undesirable image-degrading agents”. The course’s lead contrast additions and mercuric sepia platinotype entries cover the other historical answers to the same question, at their own levels.

Not a variant: the tribasic salt. See the callout under Mixing. It is a different substance.

Safety

Level A, and the classification comes from the substance rather than from the era. The encyclopaedia entry carries the assessment in full and it is not restated here. In outline: the ECHA notifications aggregated on PubChem give Warning with H319 serious eye irritation, and H315 and H335 from a minority of notifiers; the Sigma-Aldrich sheet the supplier files gives H319 and H335 only, with no skin statement. Neither EH40 nor the NIOSH Pocket Guide sets an exposure limit for ammonium citrate or for citrates, and EH40 states that absence from its list does not indicate that a substance is without risk. The classification rubric is what Level A means here.

Splash goggles rather than glasses, for the dry salt and for the made-up bath, because serious eye irritation is the one statement nearly every notifier agrees on. Nitrile gloves: the sheet gives 0.11 mm with a 480-minute breakthrough time, tested to EN374, and the glove policy explains what those figures do and do not cover. Weigh so that no visible dust is raised.

⚠Three steps that leave Level A, and what each of them needs

Making the salt from citric acid and ammonia. Sullivan’s preparation adds 120 cc of 20 per cent ammonia solution to a hot citric acid solution and boils it. Concentrated ammonia solution is corrosive and its vapour is a sharp respiratory hazard; the operation needs extraction, goggles and an apron, and it is not a Level A step. If you are buying the salt, none of this applies to you.

Water above 50 °C. Mougin’s make-up water is “at 50 °C or more”. The Level A criterion is explicit that nothing is heated above 50 °C, so a make-up that goes hotter is a declared Level B step: a vessel that will not fail, no carrying it across the room, and nothing else on the bench. The salt dissolves in cold water given time, which is the Level A route to the same bath.

Working the bath warm. Platinum and palladium printers routinely heat an oxalate developer, and the habit transfers. This bath is published for 15 to 20 °C and no source read for this page gives a reason to heat it. A heated tray raises a vapour and turns a splash into a scald; if a session heats it, the controls are Level B.

!The used bath is a different object from the fresh one

Sarah Van Keuren, after years of running this developer in a teaching studio, is the only source read for this page who addresses it: the bath “becomes contaminated with ferric oxalate that leaches off prints developed in it and fumes from the ferric oxalate can give you quite a headache, something I know from experience”. Her studio moved the tray into a fume hood. Gary Auerbach says the same thing more briefly — the developer “becomes slightly toxic as it builds up”.

The hazard they describe is not the citrate. It is the ferric oxalate and the palladium salt the prints bring with them, and both have their own entries and their own controls. A used developer tray belongs in the ventilation you would give the sensitiser, not the ventilation you would give a citrate.

Van Keuren also writes that the citrate bath is “not in itself poisonous (unlike traditional potassium oxalate developer which is quite toxic to begin with)”. That is her comparison of two baths, and as a comparison it is supported: potassium oxalate is classified far more severely than this salt. It is not a classification of this bath, and the classification of this bath is the one above.

Ventilation to the level HSE’s COSHH essentials sheet describes for manual processing — general room ventilation for the fresh bath, and local extraction once the tray is carrying sensitiser residues.

Storage

The solid is deliquescent, and that is the practical problem. Wall’s 1912 dictionary is blunt about it: the salt “is usually met with in the form of solution, the salt itself being so deliquescent that it is an extremely difficult matter to keep it”. A tub left open takes up water from the air, and a mass weighed out of it is not the mass the formula assumed — silently, because damp granules look like dry ones. Keep it closed, dry and labelled, in a container never used for food.

Which is exactly what the premeasured powder is for. A weighed charge in a sealed bottle that is opened once, filled with water and shaken is a real answer to a deliquescent salt, and it is worth recognising as an engineering decision and not only as a way of not telling you the strength.

The made-up bath: no source read publishes a dated life, and there is nothing in the bottle to oxidise. What limits it is what the prints put in.

⁇One bath, two working printers, opposite conclusions

Reuse it for years. Van Keuren: “It is expensive but is used over and over, and replenished with fresh developer to make up for volume that is lost by use. Only once in my experience has this developer expired from overuse.” Mougin reports the same as the received view: the literature “insists on the fact that the developer is considered inexhaustible and can be used infinitely”, and says it is even claimed to improve with age provided the pH is held at 5.5.

Throw it away after every print. Mougin: “My experience contradicts these facts. Indeed the developer gets loaded with particles of palladium, but also of ferrous oxalate which in strong concentration eventually veils the print in an indelible way. It, therefore, is preferable to use one-shot baths, which certainly is more expensive, but has the merit to give quality results.”

Both are first-hand and neither is measured. What the course can add is that the two practices differ in more than the developer. Mougin uses 50 cc for an 8 by 10, so his bath reaches a given concentration of contaminant after very few prints, while a studio tray holds litres. There is also Ware’s separate finding that a re-used siderotype developer holds sensitiser in solution and stays photosensitive, so left in the light it precipitates metal that can land on later prints. A one-shot bath cannot have that problem. characteristics.reuse is recorded as either because the sources genuinely say either, and the experiment that would settle it is at the foot of this page.

Incompatibilities

Hexachloroplatinate(IV), absolutely. The ammonium salt of it is nearly insoluble and crystallises out. If your contrast control is Na₂PtCl₆, this developer is not available to you. The incompatibilities index carries the general rules; this pair is specific to platinum printing.

Strong bases. They liberate ammonia from an ammonium salt, which costs you the buffer, the pH and the air quality at once. The safety data sheet names strong bases among the incompatible materials, and tetrasodium EDTA at pH 9 to 10 is the one a platinum printer actually has on the bench.

Strong oxidisers, the general incompatibility of an organic salt, and named on the same sheet — which is worth pausing on, because Mougin deliberately adds an oxidiser to this bath. A 3 per cent peroxide solution at 2 cc per 100 cc is a long way from a strong oxidiser in contact with a solid, and the distinction is concentration and physical state, not chemistry.

Strong acids, likewise from the sheet. In this darkroom that means the hydrochloric acid clearing baths, which belong in different trays for a different reason as well: acid at clearing strength attacks the palladium image.

Metal trays. Bostick and Sullivan’s instruction for the oxalate bath — “Do not use metal trays for developing the print as this will adversely affect the print” — transfers, and for a citrate it transfers with more force: a chelating bath in a metal tray is a bath that dissolves the tray.

Hard water. Citrate sequesters calcium rather than removing it, so it arrives in a bath whose job is to carry metal out of paper and then stays there.

Anything that has cleared a print must not go back on paper. A citrate bath that has done clearing work is carrying dissolved iron, and the encyclopaedia entry makes the point explicitly.

Waste

The bath that leaves the tray is carrying iron and palladium, and those govern, not the citrate. The citrate itself is a food-grade organic salt and the ammonium ion is fertiliser; what makes the stream a waste stream is the dissolved ferric oxalate and the trace of noble metal from the sensitiser.

The supplier’s safety data sheet says of the pure salt: “Do not let product enter drains.” Kodak’s J-52 publication gives 5.6 to 9.4 as the pH window sewer codes most frequently set, and a citrate bath at pH 5.5 to 6 sits inside it — which is exactly why pH is not the deciding question here. The dissolved metal is. Collect the spent developer with the three clearing baths, label the container, and follow the disposal caveat and your local authority; ILFORD’s advice to domestic users in the United Kingdom is a household waste and recycling centre’s chemical cupboard.

The one-shot practice makes more waste, in smaller batches. That is not an argument either way, but it is a consequence of the choice under Storage that neither source mentions.

Troubleshooting

Yellow stain in the highlights. Iron(III) left in the sheet and hydrolysed. Mougin’s own list of causes: “the paper or developer are too alkaline, or… the ferric oxalate is too old or polluted”. Check the pH of the bath first, because it is the cheapest of the three to check and the one this page is about. If the bath has been standing, ammonia has left it and the pH has drifted up.

Lap marks, or a visible line where the developer arrived. The bath finishes in seconds and records the order of wetting. Van Keuren’s answer is to cover the whole sheet as fast as possible; Auerbach’s is to dam the volume at one end and release it in one wave; Dana Sullivan’s, reported by Van Keuren, is to dilute the developer — and no dilution ratio is published by anyone, which is why none appears above.

The print comes up veiled and the veil will not clear. Mougin’s diagnosis for a bath that has done too much work: ferrous oxalate in strong concentration. Fresh developer is the test.

Brownish grey fog in the high values. That is the oxalate bath’s failure, not this one’s. If you see it in a citrate tray, the likeliest explanation is that the tray held oxalate last.

Crystals in the bath or on the print. If a hexachloroplatinate(IV) contrast agent is anywhere in the process, this is the ammonium salt of it coming out of solution. Change the developer or change the contrast agent.

The salt has caked in the tub and will not weigh true. Deliquescence. Wall warned about it in 1912. Buy the premeasured charge, or make up a stock solution and date the bottle.

A platinum print is grainy. The bath is too cold for platinum. That is not a fault in the developer; it is the wrong developer, and the oxalate one worked hot is the answer.

Nothing much happens. Check that there is palladium in the coating and that the paper was exposed. Development in this process is instantaneous by every account read here; a slow start is an exposure or a coating problem, not a developer one.

Experiments

🧪The step wedge that settles the tone claim for your paper

Coat six identical sheets from one batch of sensitiser and expose all six identically under a step tablet. Develop three in this bath at 20 °C and three in potassium oxalate at 20 °C, clear all six through the same sequence, and read them.

You are testing three claims at once and all three are published without measurement: that this bath is cooler in tone, that it holds the same speed, and that it leaves the high values clean. Klimek got equal response and a neutral black on one paper; Ware got higher density and warmer colour from oxalate on another. Your paper is the variable neither of them controlled.

Measure the pH of your own made-up bath. The safety data sheet gives pH 5.2 at 50 g/L. Mougin’s bath is more than six times that concentration and he asks for 5.5 to 6. Mix a litre, measure it, and find out how far your salt lands from his window before you develop anything in it. This is a five-minute test that prevents the commonest defect on the list above.

Watch it drift. Measure the pH of a working bath at the start of a session and at the end, and again after a week with the lid off. The ammonium ion is volatile and the source’s instruction is to maintain the pH rather than to set it, but no source read for this page publishes a rate. One printer with a meter and a notebook would have a number nobody has.

Run the one-shot argument to a conclusion. Develop twenty 8 by 10 prints through a single 500 mL tray, keeping the first, tenth and twentieth. Then repeat with 50 mL of fresh bath per print. Mougin says the first set will end veiled; Van Keuren ran a bath for years. Both cannot be right for the same paper and the same coating weight, and the disagreement is worth more than either answer.

Make the salt, and compare it with the bought powder. Sullivan’s preparation is 120 g of citric acid, 280 cc of water and 120 cc of 20 per cent ammonia, brought to pH 5 to 6 — a Level B operation, for the reasons under Safety. Make it up to Mougin’s strength, print a step wedge in it beside one developed in the premeasured product, and you will have compared a known composition against an unknown one. That is the only experiment on this page that can tell you anything about what is actually in the bottle.