Collodion print-out paper

☠Level D: studied, never performed

The register classifies this row Level D, and the reason is the binder rather than the silver. Collodion is nitrocellulose dissolved in diethyl ether and alcohol, and the Level D policy covers the dangerous solvents on exactly that ground. No procedure appears here: no quantity, no sequence, no conditions.

Collodion print-out paper is the third answer to the same question the salted paper print and the albumen print answer differently: where should the silver chloride sit? In the fibres, in a protein layer, or — here — in a film of nitrated cellulose.

The chemistry

The halide is made inside the binder. The Getty Conservation Institute sets out the preparation: one portion of collodion is mixed with silver nitrate and a little citric or tartaric acid, another with alcohol-ether soluble halides — usually calcium, strontium or lithium chlorides — and the two are combined under non-actinic light, so that silver chloride forms as microcrystals well dispersed through the collodion.

Sodium chloride is not on that list, and its absence is the whole argument for the odd salts. Collodion is nitrocellulose in alcohol and ether, so a chloride that will not dissolve in alcohol and ether cannot be got into the layer. Wall’s 1924 formulary states the trade’s preference outright: the best chlorides are those of strontium and lithium, and those two are the most satisfactory in practice. His collection of collodio-chloride formulas — Hanneke, Niederstadt, Valenta, Eder, Liesegang and a dozen more — uses lithium chloride in most of them, typically 2 to 3 g against 15 to 25 g of silver nitrate in a litre of alcohol and ether, with citric acid and a little glycerine.

The citric acid is doing the job sodium citrate does in a salting solution: it is a halogen acceptor, taking the chlorine out of circulation as light frees it so that the print keeps building in the frame instead of going backwards.

!Which lithium chloride does a period formula mean?

Wall’s 1912 dictionary gives the compound as LiCl·2H₂O = 78.5 and describes octahedral crystals, very deliquescent; his 1924 formulas call for “lithium chloride, cryst.”. The modern record is the anhydrous salt, LiCl, relative molecular mass 42.4 — and the course could not establish from any source it holds whether the dihydrate Wall describes is still a commercial article.

The gap between those masses is a factor of about 1.85, the largest hydrate discrepancy anywhere in the course’s encyclopaedia. Anyone reworking a collodio-chloride formula has to establish which salt was meant before weighing anything, and the course cannot do it for them. Wall’s own handling note makes the question sharper rather than softer: the salt absorbs varying quantities of moisture, so even a jar of the correct article is of uncertain composition.

Historical workflow

Coat the paper with the collodio-chloride emulsion; dry; print out by contact under a negative; rinse; tone; fix; wash. It is a printing-out paper throughout, judged by eye and deliberately overprinted, because fixing shifts the rich print-out black to a weaker dull brown.

The toning is where it differs from albumen. Reilly groups matte collodion with the arrowroot, matte albumen and matte gelatin printing-out papers popular between 1895 and 1925, and records that platinum toning harmonised especially well with those surfaces. Matte collodion was toned in gold and then platinum, for a neutral or olive black; matte gelatin printing-out paper was toned in platinum alone, which is why so much of it is brown.

The image

Glossy and matte varieties, and the matte ones are the reason the paper survives in the record as a distinct thing rather than as a footnote to gelatin printing-out paper.

Colour follows the toning sequence, and the neutral-to-olive black of a gold-then-platinum collodion print is a look the other printing-out papers do not easily reach.

Sharpness and highlight cleanliness belong to the binder, as they do in albumen: the silver is in a film on the surface rather than among the fibres.

What this atlas cannot state is a set of identification criteria distinguishing a collodion printing-out print from a gelatin one by eye. Both are binder-on-baryta printing-out papers of the same period, and the course has not read a key it can reproduce.

Permanence

The albumen print’s failure happens here too, and Reilly says so explicitly. His chapter on the causes of highlight yellowing in albumen prints states that this kind of yellow staining can and does occur in gelatin and collodion as well as albumen prints, that the presumed mechanism is the formation of silver sulfide by reaction of binder-bound silver with labile sulfur from residual fixer or atmospheric pollution, and that the onset time depends primarily on the moisture level and temperature of storage and on the amount of residual thiosulfate and silver-thiosulfate complexes present.

That is a useful correction to a common assumption. The retained-silver problem is not a peculiarity of egg white; it is what happens when a binder holds silver that fixing cannot reach, and three different binders do it.

The countermeasures are the family’s: thorough washing, and toning with a noble metal — here gold and platinum together rather than gold alone.

The binder has its own clock. Nitrocellulose is the material that, as a film base, was the cause of many fires and, in IPI’s phrase, innumerable losses of archival records; Kodak last manufactured nitrate film in 1951. Whether a thin collodion layer on paper behaves like a nitrate film base over a century is not a question this course has a source for, and it does not assert one in either direction.

Hazards

Diethyl ether is the reason for the classification. It is a volatile, extremely flammable solvent worked in an open dish, and the Level D policy names the dangerous solvents as one of the families it covers.

Nitrocellulose is not one substance but a family, and the safety card draws the line as a threshold: the formulation with more than 12.6 per cent nitrogen is used exclusively for the manufacture of explosives. The photographic grade sits below that figure — Hardwich’s 1864 account lists four varieties in order of substitution, the most substituted being true gun-cotton and insoluble in ether and alcohol, with the less substituted ones being the ones photography used. The course states the threshold and gives no route to either side of it.

Silver nitrate at 15 to 25 g per litre in an alcohol-ether solution is a different proposition from silver nitrate in water, and the course has read no hazard assessment of that combination.

Platinum toning, where the period look is wanted, is Level C on its own formulary page and is published there as chemistry rather than offered for use.

Why we study it and do not reproduce it

Because it completes the comparison the atlas is organised around. Four binders — none, protein, nitrated cellulose and gelatin — carrying the same silver chloride chemistry, with the sharpness, the surface, the toning behaviour and the failure mode moving together each time. Leave this one out and the series has a hole in the middle of the period it mattered most.

Because the yellowing evidence is general and this is where it becomes general. Reilly’s sentence that the staining occurs in gelatin and collodion as well as albumen prints is what turns an albumen anecdote into a rule about binders and retained silver.

And because the classification is about the solvent rather than the picture, which is a distinction worth making explicitly. Nothing about the finished print is hazardous. What the course will not put in a domestic room is a dish of ether.

Part XXVI owns the collodion family — the wet plate, the ambrotype, the tintype and this — in a lesson written for the purpose. The register counts this row once, here, and notes that the design names it as a cross-reference to the alternative-processes cluster.