Ferric oxalate

Almost every platinum print ever made depended on a substance that chemistry could not properly describe. Ware’s verdict in the Platinomicon is worth quoting whole: ferric oxalate “is ill-characterised, evidently polymorphic, apparently uncrystallisable, and the molecular structure was previously unknown”, the product varying with the method of preparation and therefore with the supplier. Its price tells the same story: Ware notes that iron(III) oxalate often costs over a hundred times as much as iron(II) oxalate.

In photography

It is the light-sensitive component of the traditional siderotypes. Ware’s outline for non-chemists puts it plainly: “The key compound is ferric oxalate”, and all the iron imaging systems have the same basis. Coat paper with it, expose to ultraviolet, and the iron is reduced from Fe³⁺ to Fe²⁺ while the oxalate is oxidised away as carbon dioxide. The iron(II) left behind is a reducing agent strong enough to precipitate a noble metal from its salt, and that metal is the image.

⚗The chemistry

Ware dissects the light reaction as two iron(III) plus one oxalate giving two iron(II) and two molecules of carbon dioxide, then gives the equation as it is usually written for the solid. The crucial detail is what the photoproduct is: in dry ferric oxalate it is ferrous oxalate, which is insoluble. It stays exactly where it was made, and it cannot go and find the platinum salt. That is why the traditional process needs a developer at all, and why a dry ferric oxalate sensitiser prints out hardly at all. Ware records the further consequence that the colour change on exposure is only slight, from pale yellow-green to pale yellow-brown, and that it is not permanent, because atmospheric oxygen re-oxidises the iron(II) back to iron(III).

UV + Fe₂(C₂O₄)₃ → 2 FeC₂O₄ + 2 CO₂

Ware's equation for the photochemical reaction in the solid

Development is dissolution. Potassium oxalate supplies oxalate ions that dissolve the insoluble ferrous oxalate into a mobile complex, and only then can the iron(II) reach and reduce the platinum(II). Ware writes the balanced reaction with the potassium ions omitted because they take no part.

2 FeC₂O₄ + 4 [C₂O₄]²⁻ + [PtCl₄]²⁻ → 2 [Fe(C₂O₄)₃]³⁻ + Pt + 4 Cl⁻

Ware's development equation: oxalate dissolves the iron(II), which reduces the platinum

Note the stoichiometry: two iron(II) for one atom of platinum, because platinum(II) needs two electrons and each iron supplies one. Ware remarks that these ratios matter in formulating the sensitiser correctly.

The standard solution. Because the solid’s water content is uncertain, the working unit is a solution of stated strength. Ware’s standard is 25 per cent w/v, made by weighing 12.5 g of ferric oxalate hexahydrate with 1.0 g of oxalic acid dihydrate, adding exactly 44 cc of distilled water, and stirring constantly in the dark at room temperature for about twenty hours until the solid dissolves to a clear dark olive-green — never heated, then filtered into a labelled brown bottle. That gives 50 cc at about 1.03 molar in iron. The oxalic acid goes back to Willis’s own patents, in which he added enough to render his ferric oxalate “freely soluble”; later workers used anything from 0 to 5 g per 100 cc, and Ware adopted 2 g per 100 cc as standard after testing the range. Its second effect is on print-out: Ware found the excess acid increased the extent of print-out very slightly, converting about a tenth of the sensitiser to the trisoxalatoferrate(III) anion whose photoproduct is soluble.

What “ferric oxalate” means on a label. Photographers’ Formulary is blunt that “the photographic term ‘ferric oxalate’ is a misnomer, which has given rise to a considerable amount of confusion in the photographic literature”. Two commercial forms exist, which it calls tripotassium ferric oxalate and tri-hydrogen ferric oxalate; both are photosensitive, but “only the acidic form is sufficiently photosensitive to be useful in photography”, and the supplier does not recommend the green tripotassium solid, whose photo-activity is low. Its kits therefore ship a 20 per cent solution of the acidic form. The same sheet gives a test any printer can run: dissolve two crystals of potassium ferricyanide in about 2 ml of water and, under a red safelight, add one drop of the ferric oxalate. Good material darkens only slightly and looks yellow-brown to orange in room light; a blue cast means iron(II) is already present, and the deeper the blue the poorer the material. Then hold the tube beside a lamp: a deep Prussian blue forming within a minute on the side nearest the light is the proof that photosensitive iron(III) is still there.

Properties

The structure, at last. Ware records that infrared and Raman spectra of the hexahydrate show no free oxalate anion and no trisoxalatoferrate(III) ion, and that Mössbauer spectroscopy shows all the iron(III) centres in a similar high-spin, nearly regular octahedral environment. Because the substance seems to be uncrystallisable, no single-crystal structure exists; but in 2015 one relatively well-characterised form analysing as the tetrahydrate was studied by powder X-ray and neutron diffraction. The structure is polymeric: zig-zag chains in which octahedral Fe(III) centres are bridged by cis-bidentate oxalate, cross-linked into a two-dimensional layer by a third oxalate binding trans through one oxygen at each end, with one water molecule completing each iron’s coordination sphere and two more hydrogen-bonded in the lattice.

That single fact explains the behaviour every platinum printer complains about. A polymer does not dissolve like a salt: it dissolves slowly, to a solution whose molecularity is unknown and probably a mixture of oligomers, temperature-dependent and slow to reach equilibrium. Hence the twenty hours of stirring, and hence the drift of a stored solution.

Formula weights, and why they cannot be assumed. Ware sets the four out side by side: anhydrous Fe₂(C₂O₄)₃ at 375.76, the tetrahydrate at 447.81, the CRC Handbook’s pentahydrate at 465.83, and the hexahydrate that Alfa Aesar and Aldrich supply at 483.84. PubChem’s computed molecular weight for the anhydrous formula unit is 375.75. A recipe in grams is therefore ambiguous unless it names the hydrate: a gram of the anhydrous salt carries 29 per cent more iron than a gram of the hexahydrate, on the course’s own arithmetic from those two formula weights.

Solubility. Ware records that the solid is slow and difficult to dissolve in water although the final solubility is high. Wall’s 1912 dictionary says the opposite — “nearly insoluble in water”, soluble in oxalic acid and alkaline oxalates — and both statements are defensible, because what a photographer calls dissolving it is really complexing it with the excess oxalate. The course reports the disagreement and notes that the practical instruction, add oxalic acid and stir for a day, follows from either reading.

Handling

The hazard record here is unusually thin and the page says so. PubChem’s aggregated GHS entry for diiron trioxalate, EC 220-951-7, comes from one company and one notification: signal word Warning, the exclamation-mark pictogram, and the combined statement H302+H312, harmful if swallowed or in contact with skin. Ware’s own hazard summary for the hexahydrate is “Harmful by ingestion and skin contact. Irritating to skin and eyes.”

Why Level B. The course rubric caps Level A at a classification “at most irritant, harmful if swallowed, or corrosive at the concentrations actually handled”. Harmful in contact with skin is an acute dermal toxicity entry outside that list, exactly as it is for oxalic acid, and the same Level B criterion covering fine powders that must not be inhaled applies to weighing the solid. Two further considerations push the same way. The classification rests on a single notifier, so the course errs upward rather than downward on principle. And the substance is an oxalate: the oxalate anion’s systemic toxicity is well documented for oxalic acid and potassium oxalate, and the course applies those precautions here rather than assuming that a chelated iron salt behaves differently, while stating plainly that it has not verified an oxalate-specific toxicology for this compound. It does not reach Level C, whose criterion applies where a fume cupboard or specialist disposal is the recognised control; gloves, goggles, dust suppression, eyewash and hygiene are the controls.

⚠Safety

Gloves and goggles; weigh over a tray without raising dust; wash hands before eating and eat nothing at the bench. The rule that matters most is a storage rule rather than a personal one: keep every oxalate away from every silver salt. The NIOSH pocket guide lists silver compounds among oxalic acid’s incompatibilities and the ILO-WHO chemical safety card names explosive silver oxalate as the reason, so the ferric oxalate bottle, the oxalate developer and the silver nitrate live in different places and their wastes go into different containers. Do not warm the solution or the coated paper: Photographers’ Formulary sets 50 °C as the limit, and Ware’s preparation instructions forbid heating at all.

Waste

Three streams carry this substance out of a darkroom: the unexposed sensitiser washed off the print, the spent oxalate developer, and the clearing baths that strip the residual iron. Together they hold iron, oxalate and traces of platinum, palladium or silver. The noble metal is worth recovering. The oxalate governs the containers: oxalate waste and silver-bearing waste do not share a bottle, for the reason above. Kodak’s J-52 publication gives 5.6 to 9.4 as the pH window sewer codes most frequently set, and an acidified sensitiser sits below it. The Getty Conservation Institute’s atlas records that even after the best fixing and clearing procedures enough iron remains in a platinotype to be detected by X-ray fluorescence, which is a useful reminder that the clearing baths are where most of the iron actually ends up. ILFORD’s guidance for domestic users is to bottle wastes separately, label them and take them to a household waste and recycling centre’s chemical cupboard. Check your local regulations; they govern, and they differ.

History

Ware traces the light sensitivity of ferric oxalate to the German chemist Johann Wolfgang Döbereiner in 1831, and records that Robert Hunt advocated its photographic use in Researches on Light. Hunt’s paragraph 262 is a fair example of how close and how far the 1840s got: from “the powerful influence exerted by the oxalate of the protoxide of iron on the salts of platinum” he hoped for a photographic process, found that a paper prepared with the oxalate and neutralised platinum chloride did darken where the light fell, and then found that the darkening bleached away again unless more oxalate was applied. He had the two reagents of platinum printing and no way to fix the reaction.

Herschel did not use it. His siderotypes of 1842 ran on the much more available ammonium iron(III) citrate, and Ware notes why that choice foreclosed platinum: ferrous citrate is not a strong enough reducing agent to reduce platinum(II), so the more energetic oxalate was required before a platinotype was possible. Willis supplied it in the 1870s, and had to make his own potassium oxalate developer as well, because he could not buy it in London.

The modern movement is away from the substance again. Ware’s own work over thirty years has replaced it with the well-characterised, crystalline, cheap and stable ammonium iron(III) oxalate, whose photoproduct is soluble and which therefore prints out rather than needing development. Ferric oxalate survives where the historical practice is the point — the traditional two-stage platinotype, the kallitype — and it survives with all its awkwardness intact, which is why the first thing an experienced printer does with a new bottle is test it against ferricyanide.