Chrome alum subbing dip

One salt, one dish, and a sentence of instructions that is longer than the formula. This is the whole of what a 1941 plate manual gives for putting a substratum on glass, and the most useful thing about the entry is the shape of the gap beside it: the compound gelatin sub that everybody has heard of is not in this corpus, and this page will not invent one.

Ingredient	Quantity	Form the source specifies
Chrome alum	20 g	the dodecahydrate crystals, which is what every photographic formula in this corpus means
Water	to make 1000 mL	A two per cent solution, which is the only quantity Baker states. He gives no temperature; the salt dissolves 1 part in 10 of cold water on Wall's figure and there is no reason to warm it, because a warmed chrome alum solution turns from violet to green and Kodak attributes that to a change in the salt itself.

Purpose

To leave a film of chromium(III) salt on cleaned glass so that a gelatin emulsion coated over it holds on through the wet baths instead of lifting at the edges. Baker’s own word is substratuming; the modern word is subbing, and the older English one, which describes the failure rather than the remedy, is frilling — the emulsion rolling back from the edges of a plate in the developer or the fixer.

The dip is not a hardener for a coated layer. It goes on bare glass, after the final rinse and before the emulsion, and the plate is racked without rinsing so the salt stays where it was put. Anything that hardens an already-coated emulsion is a different operation with a different entry — a chrome alum hardening bath or a potassium alum hardening fixer such as F-5.

Recommended uses

Glass that has to keep. Baker is preparing plates for a commercial coating head and for keeping tests measured in weeks, and the neighbouring pages of the same chapter make his priorities plain: dust and bacteria are the plate maker’s enemies, mould spores travel on dust, and a speck on gelatin-coated glass finds an ideal medium. A plate that will sit in a box for a month has more to lose from a poorly anchored layer than a plate that is coated on Saturday and developed on Sunday.

Plates that frill. This is the only test worth running, and it is free. Coat, dry, and process. If the layer lifts at an edge or rolls back in the fixer, the plate wanted a sub or a hardening bath. If it does not, it did not.

Plates that will be processed warm, soaked long, or handled wet. Every one of those is a case where an unhardened, unanchored hand coat is at its most tender.

Reused glass, with a caution that is not about adhesion. Baker records Clerc’s suggestion that ultra-microscopic particles, probably of metallic silver in solid solution in the glass, cause the ghost of a previous image to appear in a new one on reused plates. Subbing does not address that. It is recorded here because a reader who is cleaning and re-subbing old plates should know the failure exists.

When another formula is preferable

• When nothing is frilling, no formula at all. The tested modern practice this course follows for plates cleans thin glass and coats it directly on a moistened plastic-wrap bed, with no sub anywhere in the procedure, and the plates process and print. That is a published position from a working practitioner and it is set out beside Baker’s on the glass plate lesson.
• When the layer needs hardening rather than anchoring, an alum bath after development, or an acid hardening fixer. Adhesion to the support and cohesion of the layer are two different failures with two different remedies, and a sub does nothing for the second.
• When the support is paper, none of this applies. Paper holds gelatin by soaking into the fibre; the problem a sub solves is peculiar to a surface into which nothing can soak.
• When the support is plastic film, this dip is not the answer either. A polyester base needs a compound subbing layer that this course has read no formula for and does not attempt, which is one of the reasons film base is out of scope for Part V.

Mixing

One weighing, in a room with no draught, and then a bottle you keep.

1. Weigh 20 g of chrome alum for each litre of dip, or 6 g for the 300 mL a dish actually needs. Weigh over a tray in still air with no fan running; the hazard here is dust and the exposure limit is a dust limit.
2. Dissolve in cold water and make up to volume. It goes in readily — Wall’s figure is 1 part in 10 of cold water — and there is no reason to warm it.
3. Do not warm it afterwards either. A cold solution is violet; Kodak records that it turns green on heating and attributes the change to the composition of the salt, so a green dip is no longer the thing you weighed.
4. Label the bottle with the substance, the strength and the date, per the labelling SOP.

🧮What two per cent means here, and what it does not

Baker writes a strength and no volume, so the arithmetic is one line and the reader should be able to see it rather than take the litre on trust.

20 g ÷ 1000 mL = 2 g per 100 mL = 2 % w/v

Two per cent by weight in volume

What it is not: two per cent of the gelatin. That figure belongs to the in-emulsion hardener, where the dose is a fraction of the dry gelatin weight and 2 per cent is the top of Duffin’s band. The two numbers are the same digits about entirely different things, and confusing them puts four times too much chromium into an emulsion. The coating lesson holds the in-emulsion rule; this page holds the dip.

The dip itself, as Baker gives it: the cleaned plates go into a dish of the solution after the final rinse and are racked without further rinsing, then dried in a perfectly dust-free cupboard. No time is published, no temperature is published, and no drying time is published. The instruction is a dip and not a soak.

!Baker's first cleaning option is not printed here

The same page gives two ways of cleaning the glass. The first is potassium dichromate with sulphuric acid added to it — chromic acid, which is chromium(VI) — and this course excludes chromium(VI) everywhere at every level. The chromium policy holds that ruling once and this page does not restate it in different words.

Baker himself gives the alternative in his next sentence — brush the glass with a hot two per cent solution of caustic soda and wash thoroughly under the tap — so the exclusion costs the reader nothing but the loss of one option. Caustic soda hot and concentrated is corrosive and is its own Level B operation with gloves and eye protection; see sodium hydroxide.

Behaviour

It is used up by being used. A dish of dip that plates have been carried through picks up whatever was on them. In a subbing dish that is rinse water, which dilutes it; the destructive contaminant, sulfite, is not present unless something that has held fixer has been in the dish.

It stains. The solution is violet by reflected light and reddish by transmitted light, which is Wall’s own identification test and a useful check that the bottle holds what the label says. It also marks fingers, cloth and bench, and a chromium stain on a worktop is the visible record of where the solution went.

No published time. The source’s word is dipping, and the plate comes out and goes on the rack. This course does not supply a number for an operation whose source gives none.

The dried film is invisible. There is no way to look at a subbed plate and a bare one and tell them apart, which is exactly why the batch record matters: on a hand-coated plate, “subbed or not” is one of the variables that decides whether the layer survives the fixer, and it cannot be recovered by inspection afterwards.

Image characteristics

A sub has none of its own, and saying so is more useful than inventing some. It is under the layer, it is a salt film and not a pigment, and it contributes no density, no colour and no grain.

What it changes is whether there is an image at all after processing. The failure it prevents is mechanical: an emulsion that lifts takes its silver with it, and a plate that frills at one edge loses that edge entirely. The nearest thing to a tonal effect is indirect and belongs to the hardening question rather than to this one — a layer hardened so hard that fixer cannot get through it bronzes later in room light, which is the argument the coating lesson makes for not hardening emulsions by default.

The mechanism

What is established. Chrome alum tans gelatin, and it does it well: Wall’s 1924 table of alums as hardening agents gives the weight that renders 100 parts of dry gelatine insoluble in hot water as 6 parts for potash alum and 2 parts for chrome alum, and the qualitative statement from the conservation literature is stronger still — adding potassium chrome alum to a gelatin solution gives a gel that is completely insoluble. Both statements are on the chrome alum page with their sources.

What is not. How the chromium(III) ion actually binds gelatin at the molecular level is not settled by any source meeting this course’s standard, and the encyclopaedia entry says so and stops. The period literature calls it tanning and stops as well. Duffin’s 1966 manufacturing text offers an account and hedges it — the action of chromium is probably to form ionic links between carboxyl groups on different chains — and the course’s own chemical page declined to adopt a hedged proposal as a mechanism. This page takes the same position, because a formula entry is not the place to quietly settle something the encyclopaedia left open.

What that leaves for the sub specifically, which is a narrower question than the tanning one and is worse served by the sources: nobody in this corpus explains why a film of salt dried onto glass makes a gelatin layer coated over it adhere. The plausible account — that the salt is available to cross-link the first gelatin that touches it, anchoring the bottom of the layer to a surface it cannot otherwise grip — is a plausible account and is labelled as one here. It is not in Baker, it is not in Wall, and it is not in the encyclopaedia.

Function of every ingredient

Chrome alum, 20 g to the litre. The only ingredient, and the only reason the dish is there. It is chromium(III) potassium sulfate dodecahydrate, the violet crystal, and it is the strong tanning agent of the period formulary — about three times as effective as potash alum by weight on Wall’s table. In this formula, at this quantity, it is doing something none of the other chrome alum entries in the formulary asks of it: it is not hardening a coated layer or a print in a tray, it is being left behind on bare glass at the end of a rinse, to be there when gelatin arrives. More of it does not anchor a plate better in any way this corpus can show, and it puts more chromium into the rinse water and more salt onto a surface where a crystalline bloom would be a coating defect. Less approaches doing nothing, which the tested modern alternative shows is often an acceptable outcome. The form matters as much as the number: the dodecahydrate is what the crystal jar holds and what every formula in this corpus means, and the anhydrous salt at 283.22 against 499.4 would carry the chromium of 1.76 g of crystals in every gram. It is also insoluble in alcohol, so nothing here is helped by adding spirit to speed the drying.

Water, to the litre. Not neutral. Cold water is a specification and not a default — a warmed solution turns green and Kodak attributes the change to the salt itself — and clean water matters more here than in most baths, because the last thing to touch the plate before the emulsion is whatever this dish contains. Distilled water is not required by the source and is worth using anyway on a plate you intend to keep.

There is no gelatin in this formula, and that absence is a stated fact rather than an omission. The compound sub — gelatin hardened with chrome alum and coated as a layer — is described by the Image Permanence Institute as what the industry actually did, and no source read for this course gives a formula for it: not a gelatin concentration, not a ratio to the hardener, not a coating rate, not a drying schedule. The course therefore publishes the dip it has and names the formula it does not have.

Interactions

With sulfite, which is the one that ends it. Kodak’s primers record that a chrome alum solution loses hardening power in the presence of sodium sulfite at a rate that depends on both concentrations. There is no sulfite in a subbing dish by design, and the way it arrives is a dish or a measuring cylinder that has held fixer. Keep the dip’s glassware to itself.

With alkaline developer, which precipitates chromium hydroxide as a scum. That is a hazard for a hardening bath rather than for a sub, since nothing developed goes into this dish — but it is the reason a dish used for the dip should not double as a processing tray.

With the emulsion coated over it, which is the whole point and is also the untested part. Nobody has published what a chrome alum sub does to the pH at the bottom of a hand-coated layer, and chrome alum’s hardening action is itself pH-dependent, effective near the usual coating pH of about 6 and much less so above it.

With the drying cupboard. Baker treats the drying as part of the operation, not as an afterthought: the racked plates go into a perfectly dust-free cupboard, because a dust speck on a plate that will shortly carry gelatin is a mould spore’s ideal medium.

With iron and steel, as with any tray chemistry near silver bromide — not because of a reaction with the alum, but because plate-making is an operation where a rust speck becomes a defect.

Variants

No course variant is offered, and there is nothing to make safer by changing. The formula is one salt at one strength; the strength is the formula, and a reader who wants less chromium in the rinse water uses the tested alternative, which is no sub at all.

The compound gelatin substratum is the variant that does not exist here. It was real industrial practice and the conservation literature identifies its residue in objects by X-ray fluorescence a century later. What this course lacks is a published formula for it. Should one be found in a source meeting the standard, it would be a separate entry and not a modification of this one, because a coated gelatin layer and a dipped salt film are different operations rather than two strengths of the same idea.

Wall’s basic chrome alum solution is a stronger tanning agent made from the same salt — 100 g of chrome alum in 800 cm³ of hot water, ammonia added until a slight permanent precipitate persists, filtered and made to a litre. It is on the chrome alum page. It is not offered as a sub by any source, it is made with ammonia, and Part V excludes ammonia at Level B, so it is recorded here as a neighbouring formula and not as an alternative dip.

Kodak’s hardening baths are the same salt at 30 g per litre and upwards, for a different job. See the chrome alum hardening bath, and note the difference in scale: this dip carries less chromium per litre than any of them.

Safety

Level B, and the level is set by a powder rather than by a bath. A 2 per cent solution in a dish is a mild irritant; the exposure that matters is the single weighing that makes it.

The classification, from the chrome alum page rather than restated here: signal word Warning, GHS07, with skin irritation and serious eye irritation notified in every report and respiratory irritation in most, and — the fact that decides how this page reads — no sensitisation and no carcinogenicity statement notified. The UK long-term exposure limit for chromium(III) compounds is 0.5 mg/m³ as chromium, against 0.01 mg/m³ with the carcinogen and sensitiser notations for chromium(VI).

This is chromium(III) and not the chromium of dichromate. The course states which chromium it means every time, and the ruling lives once at chromium.

Controls: weigh the powder over a tray in still air, nitrile gloves and eye protection while the jar is open and while the stock is made, and buy the smallest pack you can so the powder is weighed once in the life of the bottle. Eye protection is not optional — serious eye irritation appears in 100 per cent of the notified reports.

The other hazard on this bench is the glass, and it is the more likely injury. Cut and edged plates cut hands; the 220 mesh diamond pad that smooths an edge is a safety instrument.

Storage

A stoppered, labelled bottle, out of the light and away from anything that has held fixer. Kodak’s statement is that a plain aqueous chrome alum solution keeps its hardening power indefinitely, which makes this one of the few solutions in the formulary worth making in quantity — and the qualification is that plain is doing the work. Sulfite ends it.

Label the substance, the strength, both CAS numbers and the date. The dodecahydrate is 7788-99-0 and the anhydrous salt 10141-00-1, and a bottle labelled only “chrome alum” is a bottle whose strength is guesswork by the time it is next wanted.

Watch the colour. Violet by reflected light and reddish through the bottle is what it should be. Green means it has been warmed; a yellowish-green working bath is Kodak’s own signal that a hardening bath has stopped hardening, and although that observation is about a processing bath rather than a dip, a green subbing dish is not what was weighed.

The dish is not storage. Return the dip to the bottle or discard it; a dish standing open collects dust, which is the one thing the drying cupboard exists to keep off the plates.

Incompatibilities

Sulfite in any form — fixer, a developer’s preservative, a washing aid — which destroys the hardening power. See incompatibilities.

Alkaline developer, which throws a chromium hydroxide scum that cannot be removed once dry.

Heat, in the mild sense that matters here: warming turns the violet solution green.

Alcohol, in which chrome alum is insoluble, so an alcoholic rinse is not a way to speed the drying of a subbed plate.

Dichromate and anything else carrying chromium(VI), which is not an incompatibility of the substance but a rule of this course: the two oxidation states are not interchangeable and must not share a bench, a bottle or a waste stream where they could be confused.

Waste

A chromium(III)-bearing solution, and the rinse water that follows the plates is part of it. This is the stream the course’s own coating-station route was written for: collect the spent dip and the first rinse in a labelled container rather than sending them to a drain, and follow the chromium(III) rinse route and the disposal ruling.

It is not silver-bearing, which distinguishes it from almost every other waste stream in Part V. Keep it separate from the silver stream rather than merging them: two labelled bottles are easier to hand over than one bottle of unidentified mixture.

The quantity is small and that is the point. Three hundred millilitres of dip carries 6 g of chrome alum. The option that produces no chromium waste at all is the one the tested modern practice takes, which is to coat clean glass and not sub it.

Local regulation decides how a chromium-bearing waste classifies and this course cannot tell you what it says where you are. The classification method is named on the chrome alum page; the course states the chemistry and does not compute a threshold.

Troubleshooting

The emulsion frilled anyway. Subbing is one of three variables and the other two are more likely. The glass may not have been clean — a hot caustic soda brush and a thorough rinse is the published method — or the layer may be too thick, or the fixer too warm. Change one thing, and use the coating pair described under Experiments to find out which.

The emulsion frilled and the plate was not subbed. That is the test result the sources say to act on rather than a fault. Sub the next four plates from the same melt, or fix in an acid hardening fixer such as F-53 made up as F-54, and compare.

A crystalline bloom or a haze under the coating. The dip was too strong, or the plate was rinsed and then dipped in a dish that had dried out and concentrated. Make it fresh at 2 per cent and rack wet.

Green dip. It has been warmed. Discard it and make it again cold; the colour is telling you the salt has changed.

A violet stain on the bench, the rack or your fingers. Expected, and it is also the visible record of where the solution went. Treat it the way the course treats every stain from a substance with a dust hazard: as a control that has failed somewhere upstream of the stain.

Spots or comets in the coated layer. Dust, in nine cases out of ten, and Baker’s dust-free cupboard is the remedy his page insists on. The subbing dish is one more open vessel in a room where dust is the enemy; keep it covered.

Experiments

The subbing pair, which is the only experiment that answers the question for your hands. Coat four plates from one melt: two on clean glass, two on glass dipped at 2 per cent and racked without rinsing. Process all four in the same trays on the same day and watch the edges in the fixer. Two published positions, one melt, and an answer that is yours rather than either source’s.

Vary the fixer instead of the sub. Take four unsubbed plates and fix two in a plain hypo bath and two in an acid hardening fixer. If the hardening fixer stops the frilling, you have the remedy the tested practitioner reaches for and you have not put chromium anywhere.

Push the layer until it fails. Coat one subbed and one unsubbed plate heavily, then process both warm — 24 °C rather than 20 °C — and soak them longer than any sensible procedure would. The first edge to lift tells you which variable your process is actually near the limit of.

Test the strength downwards. Dips at 2, 1 and 0.5 per cent, four plates each, judged by the same fixing test. Nothing in the corpus says 2 per cent is an optimum rather than a convention, and this is the cheapest way to find out whether it is.

Keep a subbed plate for a month before you coat it. Baker’s dip is aimed at a plate that must keep, and his own keeping test is an oven at 105 °F for ten days with a criterion of not more than 0.02 extra fog density. You will not reproduce his oven; you can compare a fresh plate and a stored one from one melt and find out whether the sub survives storage in your cupboard.