Coating, Drying and Hardening

An emulsion that is chemically perfect and unevenly coated is a bad photographic material. Duffin puts it as the first sentence of his coating chapter: the quality of the final photographic result depends on the evenness, cleanliness and thickness of the coated layer. Those three words are the whole of this lesson, and the third one is where beginners lose most: too thin, and a print that could reach a full black never gets there, and the maker blames the emulsion.

Substrates

Paper. Two kinds, and this part uses one of them. A commercial photographic paper is baryta paper — a paper base carrying a coat of barium sulfate in gelatin, which gives the smooth white reflecting ground under the image — or resin-coated paper, whose polyethylene layers keep processing solutions out of the fibres. Neither is straightforward to buy uncoated in small quantities, and neither is what the projects here use.

What the projects use is sized watercolour paper, and it is a real choice rather than a substitution. The tested Tier 2 practice names 90 lb hot-press — hot-press meaning smooth — in two specific brands, and the sizing is what stops the emulsion sinking into the fibres. It also means the paper’s manufacturer is not making it for you: the same practitioner records a batch of a named paper whose sizing recipe had been changed by the maker, which caused blistering during processing that took weeks and half a dozen emulsion batches to diagnose, and a later lot of a different paper that produced mottled fading late in the wash. Keep the lot and the brand in the batch record. A change you did not make is still a change.

Glass is the other substrate, and the plate lesson owns it. Its advantages are dimensional stability and a surface that takes an even coat; its problems are that nothing soaks in — so the emulsion must be washed, as the washing lesson explains — and that it must be scrupulously clean.

Film base is out of scope, and the reason is manufacturing rather than difficulty. Duffin’s account of a coating line is a support drawn around a roller through a temperature-controlled trough, chilled by cooled rollers or by carefully controlled cold air, then passed through a series of drying zones at controlled temperature and humidity. A polyester sheet also needs a subbing layer to make gelatin adhere to plastic at all. A hand coater can do glass and paper; continuous film base is a machine’s job, and this course says so rather than pretending otherwise.

Preparing glass, and one correction to make first

!There is no chrome alum gelatin subbing solution, and this page will not invent one

Part V’s manifest asks for “subbing with chrome alum gelatin”. No source this course has read gives a chrome-alum-plus-gelatin subbing formula. What Baker gives, on page 140, is a plain dip: “Substratuming is done by dipping them into a dish containing a two per cent solution of chrome alum after the final rinse and racking them without further rinsing. They should be dried in a perfectly dust-free cupboard.” There is no gelatin in it.

The course publishes what the source gives, and says that is what it is doing. Compound subbing layers carrying gelatin certainly exist in industry — that is what a film base needs — but this course has not read a formula for one, so it prints none.

Cleaning comes in two published options and the course can only publish one of them.

Baker’s first option is Level D and is not given here. It is a solution of potassium dichromate with sulfuric acid added slowly to it — chromic acid — applied with a brush. That is chromium(VI) plus a concentrated strong acid, and this course’s chromium policy excludes chromium(VI) at any level, anywhere, with no procedure. Baker’s own text calls the chromic acid very corrosive, and he gives the alternative himself in the next sentence, which is why the exclusion costs nothing.

Baker’s second option is usable at Level B: brush the glass with a hot two per cent solution of caustic soda, then wash thoroughly under the tap. Sodium hydroxide is corrosive and this solution is hot, so it is eye protection, gloves and care with a hot alkali — not a Level A operation.

The tested modern practice uses neither. The practitioner whose glass procedure this course follows cleans the glass and coats it directly, with no sub at all, on a moistened plastic-wrap bed over levelled glass. Her plates process and print.

⁇Does a hand-coated plate need subbing? The sources disagree, and that is useful

Baker (Tier 1, 1941): a 2 per cent chrome alum dip after the final rinse, racked without further rinsing, dried in a dust-free cupboard.

Ross (Tier 2, tested, from 2013): clean 1/16 inch glass, no sub, coat directly.

This is less a contradiction than a difference in what is being asked of the plate. Baker is preparing plates for a commercial coating head and for keeping tests measured in weeks; the modern practitioner is coating by hand and processing soon afterwards. The course therefore says both, and names the test that decides it for you: frilling during fixing. If the layer lifts at the edges or rolls back in the fixer, your plates need either a sub or a hardening bath. If it does not, they do not.

Coating methods, and what each one does to the layer

Four hand methods appear in the literature and in practice, and they differ in how much control you have over thickness and edges.

Pour and tilt. Flood the plate, tilt it to the four corners, and drain the excess off one corner — the wet-collodion gesture. It needs no equipment and it puts a wedge of emulsion on the plate: thicker where you drained last. Historically standard, and the least even.

The glass rod, or puddle pusher. A length of glass rod, its ends taped and plugged with putty so the emulsion cannot escape past them, drawn across the sheet in front of a poured puddle. The plugged ends are the whole trick: emulsion that escapes them runs down the sides and dries as a thick dark border. This is the method the paper projects use, and a 9 inch rod for an 11 × 15 inch sheet is the tested combination.

Brush and foam brush. Familiar from the iron processes, and it leaves brush edges, which some makers want as a visual signature. It is the least even of the four for a large flat tone and the most controllable for a shaped coating.

Spoon-spread on a levelled bed, which is what the tested plate practice uses: pour the measured volume onto the plate and spread it with the back of a round-bowled spoon, chasing bubbles to the edges. The instruction that matters is do not rework a plate — the emulsion sets fast, and a second pass drags a set skin.

The coating bed and the puddle-pusher stroke

1. 1Levelled plate glass — level it by pouring a puddle of water and watching which way it runs, not by trusting a spirit level
2. 2Water film and plastic wrap — the water holds the wrap down; the wrap gives a clean, non-absorbent bed
3. 3Wetted, squeegeed paper — wet enough not to buckle within half an hour, and no wetter
4. 4Emulsion pot in a bath at coating temperature — with a thermometer in the emulsion, not in the bath
5. 5The rod, ends taped and plugged — if emulsion gets past the ends it dries as a thick dark border
6. 6One stroke, no stopping, no second pass — the layer begins to set as it is laid; a restart leaves a line

Drawn as an arrangement rather than a measured plan; no dimension here is to scale. The bed and the stroke are the two things worth rehearsing with plain gelatin before any silver is involved.

Levelling deserves its own paragraph, because it is the commonest silent fault. A spirit level gets you started and is not precise enough. The tested method is to pour a puddle of water on the glass and watch which way it wants to go, adjust, and repeat; when you can add water to the puddle and it stays round as it grows, the surface is level. The final test comes later, in the fixer: if one edge or corner clears far later than the rest, your bed was not level and the layer is wedge-shaped.

Temperature, viscosity and setting

The thermometer is a coating instrument. Melted gelatin changes viscosity with temperature, and viscosity is what decides how a poured layer flows and how quickly it stops flowing — so a maker standardises on one coating temperature and holds it from the first sheet to the last.

Duffin gives the industrial window and, more usefully, the reason for it. Trough coating runs at “a closely-controlled temperature, usually between 36° and 40 °C … The temperature range is chosen so that it is above the setting point of the gelatin but is sufficiently near to it to permit ready setting of the layer soon after coating.” That sentence is the whole of coating temperature: high enough to flow, low enough to stop flowing almost at once. The tested domestic windows sit just above it — 40 to 42 °C for the chloride paper, the low-to-mid 30s to the low 40s for the plate emulsion.

✕The two ways coating temperature goes wrong, and how each one looks

Too warm and the emulsion behaves, in the practitioner’s phrase, “just like cooking oil poured into a very hot pan”. After processing you see irregular islands of low density and matte surface sheen surrounded by deep black, and on thin paper, black spots on the back where the emulsion soaked through. The good news is that it is visible under the safelight while the sheet is still wet, so you can adjust before the next one.

Too cool is not the answer. A too-viscous emulsion coats thickly and unevenly, and thick coats bring their own troubles: a stubborn curl on drying that will not flatten even under a contact printing frame, which costs you sharpness where the negative loses contact, and — if the emulsion is hardened — thick areas that fix incompletely and darken later in room light.

Both faults are read from the same instrument. Put the thermometer in the emulsion, not in the bath.

Setting itself is not a chemical reaction. Duffin: the gelatin molecules probably exist as random coils in the sol state, and on cooling they form regular patterns linked by hydrogen bonding rather than ordinary chemical bonds. He reports Janus’s evidence that the guanidino groups of arginine play the predominant role, from two experiments: destroying those groups with hypobromite gives a gelatin that no longer forms gels at all, and converting lysine’s terminal amino groups into guanidino groups gives a gelatin of higher rigidity and a much higher setting rate. Because a hydrogen bond carries less energy than a covalent one, “it will therefore easily be broken by the application of energy, e.g. from a moderate temperature rise” — which is why a set layer melts, and why a hardener, which makes covalent or ionic links instead, does not let it.

Coating weight, which is the number that matters

A hand-coated sheet in section, and the two thicknesses that matter

Depths drawn readable rather than to scale; the real figures are in the labels. The two absences — no supercoat, no baryta — are as important as the layers that are present. Layer depths are drawn to be readable, not to scale: on real film the base is roughly a hundred times the emulsion, and drawn honestly the emulsion would vanish. Any thickness given in the labels is the real one.

Coating weight is usually written as milligrams of silver halide per square decimetre, and Baker gives a target for one class of material: chloride and chlorobromide plates “should have a very thin film, not more than 25 to 40 milligrams of silver halide per square decimetre”, which is 2.5 to 4.0 g per square metre. That is the single most useful number in this part, because it turns “coat it thin” into something you can compute.

🧮Working out what you actually put down

Start from the batch. The chloride paper emulsion carries 5 g of silver nitrate in a finished volume of about 200 ml.

5 ÷ 169.87 = 0.0294 mol of silver. As silver chloride, at 143.32 g/mol, that is 4.22 g of silver chloride in 200 ml, so the emulsion is about 21 g of silver chloride per litre.

An 11 × 15 inch sheet is 165 in², and 1 in² is 6.4516 cm², so the sheet is 1,064 cm² = 10.6 square decimetres.

Now try two coating volumes.

• 20 ml on that sheet: 20 ml × 21 g/L = 0.42 g = 420 mg, over 10.6 dm² = 40 mg per square decimetre. That is the top of Baker’s band.
• 50 ml on the same sheet: 1,050 mg over 10.6 dm² = 99 mg per square decimetre, two and a half times his ceiling.

The general form, worth writing in the front of the batch book:

coating weight (mg/dm²) = volume laid down (ml) × silver halide concentration (g/L) × 100 ÷ area (dm²)

Coating weight

The volume laid down is in millilitres, the concentration in grams of silver halide per litre, the area in square decimetres, and the factor of 100 converts grams to milligrams and litres to millilitres together. Compute it once for your own emulsion and you have a number you can hold constant while you change something else.

⁇How heavily does hand coating actually coat? The sources do not agree, and one of them does not agree with itself

The course’s research pass concluded that hand coating lays down roughly two to three times the commercial coating weight. For plates that holds up: the tested practice is 15 ml per 4 × 5 inch plate, which is 0.75 ml per square inch, against Baker’s “normal thickness” of 10 or 12 cc to a 5 × 7 plate, which is 0.29 to 0.34 ml per square inch — a factor of 2.2 to 2.6.

For paper the evidence is not that clean, and this page says so. The same tutorial gives two figures that cannot both describe the same batch. The recipe says 50 to 55 ml per 11 × 15 inch sheet. The tools page says a batch coats ten such sheets. But the batch is only about 200 ml, which is four sheets at 50 ml, not ten. On the ten-sheet reading the rate is about 190 ml per square metre, which sits right on Baker’s commercial figure of one litre to 60 to 80 square feet — that is 135 to 180 ml per square metre. On the per-sheet reading it is nearly 500 ml per square metre, about three times commercial.

The likely reconciliation is that 50 to 55 ml is the volume taken up in the scoop, with the surplus running off the far edge and being available again — she does describe taking too much and wasting it at the bottom — but the source does not say that, and the course does not assert it. What it does instead is tell you to settle it for yourself in ten minutes: weigh the emulsion pot before and after coating one sheet, divide by the sheet area, and you have your own coating weight, measured. Write it down. It is the single most useful number your batch record will contain.

Why the number matters photographically. Thickness moves two things. More silver halide per unit area intercepts more light and holds more developable silver, so maximum density rises with coating weight until the layer is thick enough that nothing deeper is contributing. And speed rises with it for the same reason, up to a point. Baker’s warning is the practical form and it is exactly the mistake a first-time maker makes: it is easy to mistake the effect of too-thin coating for a poor maximum black — “low vigor in the old literature” — and to go back and change the emulsion when the coating was the problem.

His acceptance figures are worth adopting as the course’s own: fog on a trial coating on glass should not exceed 0.02 density, and after a keeping test of ten days at 105 °F below 65 per cent relative humidity, incubated plates should show no more than 0.02 extra fog compared with controls.

Drying

Drying is not waiting. Duffin found it to have an important influence on the properties of the emulsion, and gives the numbers: during drying the water content goes from about 85 per cent — only about 15 per cent being gelatin, silver halide and minor additives — down to about 5 per cent, because the gelatin retains 10 to 12 per cent of its own weight of water and will not give it up.

The rate matters too. “As the water is drawn off, the gelatin molecules initially held by the hydrogen bonds, produced in setting, will be drawn closer together and form a closer-packed array. Generally speaking the more slowly drying is carried out, the more orderly will be the array of the gelatin molecules.” Maximum order comes from drying at room temperature, which is economically impossible in a factory and entirely possible for you.

Four practical variables:

• Airflow, which must be enough to carry moisture away and gentle enough not to carry dust in. The tested arrangement is fabric taped over an air vent, which cuts dust without blocking flow.
• Humidity and temperature. In a cool damp room the sheets will not dry at all without help; a small radiator at about 24 °C is the tested answer, not a heat gun.
• Time. Three hours on the glass before the paper is moved is the tested figure for that emulsion in that room, and it is a starting point rather than a constant. Move a sheet before the emulsion has stopped being able to run and you get a flow mark that no amount of later care removes.
• Dust. Baker makes the drying cupboard the thing on which success with test emulsions depends: dust and bacteria are the plate maker’s enemies, mould spores travel on dust, and a speck of dust on gelatin-coated glass may deposit a spore that finds an ideal medium — especially if plates dry slowly in a humid, warm atmosphere. Note the tension with the paragraph above: slow drying gives a better layer and a worse mould risk, and the resolution is cool and slow, not warm and slow.

Hardening: the ruling, and what it costs

This is where the course has already decided, and the page states the decision, its evidence and its consequences rather than leaving you to infer them. The chromium policy lives once, in safety classification, and is not restated in different words here.

The default is no hardener in the emulsion

Because that is what the course’s own tested source does, and she gives reasons. Unhardened emulsion tones more readily and spots more readily. And the argument that decides it for a beginner: hardening makes it hard for processing chemicals to penetrate, especially the fixer, so the thick patches that hand coating inevitably produces fix incompletely and darken later in room light. Unhardened, “your processing chemistry will penetrate even the thickest areas”. Duffin says the same thing from the manufacturer’s side: excessive hardening of the layers will interfere with developer penetration.

The consequence is stated, not absorbed. An unhardened wet layer is tender. That means, every time:

• handle it by the corners while it is wet, and only by the corners;
• never squeegee it — pull it from the last wash and hang it by a corner;
• no warm processing, which swells the layer further;
• no long soaks, which rules out the multi-soak work a silvergum print or heavy retouching needs.

When it is dry it is no longer tender. The rules apply to the wet layer only, and they are the price of a layer that fixes all the way through.

Where processing needs the layer to hold: harden in the bath

If your emulsion frills, lifts or reticulates, the remedy is a potassium alum acid hardening fixer, and the practitioner reaches for exactly that: she records an emulsion that bubbled, frilled at the edges and lifted clean off the paper in one particular developer, and that “a hardening fixer prevented this completely”; for plates, “if you find you have frilling problems, switch to a hardening fixer in your second fixing tray.”

The published baths are Kodak’s, and the course names them rather than reprinting them here: F-5, an acid hardening fixer carrying 15 g of potassium alum per litre alongside hypo, sulfite, acetic and boric acids, and F-53, an acid-hardener stock made up into F-54 for paper or F-54a for films and plates. Two of Kodak’s own notes travel with them: prolonged immersion at high temperature is harmful, and the mixing order is not optional — hypo first, then sulfite, then acid, then alum, with the hardener stock added slowly to cold hypo solution with vigorous stirring, or the hypo decomposes and the alum precipitates.

Why potassium alum rather than chrome alum for this job? Three reasons and they all point the same way. Potassium alum is the only candidate in this comparison with no GHS hazard classification at all — 43 of 43 notifiers report it does not meet the criteria — where chrome alum carries three irritation statements. A hardening fixer is an item the student is buying anyway. And the chrome alum baths, SB-3 and F-16, use 30 to 60 g of chrome alum per litre, which is one to two orders of magnitude more chromium than an in-emulsion dose would be.

✦Chrome alum baths keep badly, and tell you so by colour

If you do use one, Kodak’s SB-4 gives the behaviour: freshly made it is violet-blue and keeps indefinitely while unused; a partially used bath deteriorates in a few days and turns yellow-green; and after the equivalent of twenty 10 × 8 inch films per gallon it should be replaced or scum markings result. F-16 “rapidly loses its hardening properties, with or without use”. A bath you can read by colour is a rare gift in this subject — use it.

Where a lesson genuinely needs an in-emulsion hardener: chrome alum

The coating lesson cannot teach hardening without hardening something, and any formula whose published finals already contain a hardener keeps it. In those cases the hardener is chrome alum, and the dose is Duffin’s: 0.5 to 2 per cent of the dry gelatin weight, introduced carefully into the molten emulsion immediately before coating.

Three properties from the same page govern how you use it. The reaction is fast — “as soon as drying occurs the hardening reaction takes place”. It is pH-dependent, effective at the usual coating pH of about 6 and far less effective at higher pH, so an emulsion finished alkaline will not harden with it. And its failure mode is coagulation of the emulsion from a temporary excessive local concentration, which is why the addition is slow and the emulsion is moving.

Two independent working confirmations bracket Duffin’s range: Baker’s page 108 finals work out at 0.625 per cent of the gelatin, and Wall’s rule — one part of chrome alum to 250 or 300 parts of dry gelatine — is 0.33 to 0.40 per cent. Wall also states the principle worth memorising: “the quantity used depends on the total quantity of the gelatine and not on the bulk of the emulsion.” That is what makes the dose scalable by arithmetic when a commercial formula is reduced to student size.

🧮A worked dose, and the one weighing you will do

A student make containing 10 g of gelatin needs, at 0.5 to 2 per cent of that gelatin, 50 to 200 mg of chrome alum.

Nobody weighs 50 mg. Make a 5 per cent w/v stock — 5 g in 100 ml — once, from the smallest pack you can buy, and then 50 to 200 mg is 1 to 4 ml of stock, measured with a syringe. The powder is weighed once in the life of the bottle, or not at all if you make up a whole small pack.

Label the stock with the substance name, both CAS numbers, the strength and the date. Weigh over a tray in still air with no fan running: the hazard that matters is dust, and the exposure limit is a dust limit. Nitrile gloves and eye protection while the powder is open, because H319 — causes serious eye irritation — appears in 100 per cent of the notified reports.

And add it slowly, to a moving emulsion. Duffin’s failure mode is not hypothetical.

☠What is excluded, and in which order the reasons come

Formaldehyde. Excluded, and the emulsion literature excluded it first. Wall in 1929: “Under no circumstances should formaldehyde be used as a hardener. This inevitably, according to the writer’s experience, leads eventually to fog on keeping.” Duffin in 1966 reaches the same conclusion independently and blames the same cause, adding that its volatility loses it during coating and can contaminate other materials. He then generalises the problem to the whole aldehyde class: they are reducing agents, naturally prone to fog, and the emulsion must be balanced with an active stabiliser to control it — and this course has no stabiliser dose from any source at any tier. Only after all that does the hazard record arrive: skin sensitiser, suspected carcinogen, and an EH40 long-term and short-term limit that are the same figure, 2 ppm, carrying the Carc notation. There is a third objection nobody mentions: Duffin says formaldehyde’s full hardening effect is not complete until two or three weeks after drying, which a factory absorbs between coating and dispatch and a weekend maker does not.

Glutaraldehyde. Excluded on its exposure limit alone: 0.05 ppm long-term and short-term with the Sen notation, forty times tighter than formaldehyde’s and the lowest figure anywhere in this part, for a substance carrying both H334 respiratory sensitisation and H317 skin sensitisation in more than 99.9 per cent of notified reports. A respiratory sensitiser at that limit is not controllable by a domestic room’s ventilation, and the control failure is permanent.

Glyoxal. Excluded twice over. On evidence: the practitioner who uses it publishes no dose, and Duffin’s only quantity statement for the substituted aldehydes contradicts his own kinetic finding for glyoxal, so there is nothing to print. On hazard: the harmonised classification under Regulation (EC) No 1272/2008 carries H317 skin sensitisation and H341 suspected of causing genetic defects, and glyoxal appears nowhere in EH40 under any of its names, so there is no limit to work to.

Tannin is kept as history. Wall records Wratten’s ratio of 1 of tannin to 286 of gelatine, “very effective; but it must be used with caution, as it tends to give a yellow stain during development, if used in too large a quantity”. The stain is why it lost.

Testing whether a layer is hardened

The manifest asks for a melting-point test, and the honest position is this: the principle is sound and no source this course has read gives a procedure or a criterion for one. What a hardened layer does is refuse to melt at a temperature that would melt an unhardened one — that is the definition — so warming a scrap of coated material in water and noting when the layer softens or lifts ranks two coatings against each other.

The course therefore offers it as a comparative test with its own control and no published number: coat a strip unhardened and a strip hardened from the same batch, dry both, and warm each in a beaker of water raised slowly with a thermometer in it, noting the temperature at which the layer first lifts from the edge. The hardened strip should go higher. Record both temperatures; do not report either as a constant, because no source gives one and yours will depend on your gelatin, your coating weight and how long the layer had been dry.

Keeping

Unwashed and undigested emulsion keeps a few hours to a few days refrigerated; the chloride paper emulsion can be held up to a week in a lightproof container, and Duffin’s industrial figure for undigested material is “only a few days”.

Finished emulsion, digested and stabilised, keeps far better: below 5 °C, “a stable finished emulsion [can] be kept for several months”. Washed noodles keep several days refrigerated, and are never frozen.

Remelts. Nobody in the corpus quantifies a limit, and the mechanism from the gelatin lesson says there must be one: each melt cuts chains permanently and leaves the solution thinner, and enough cycles gives a solution that will not set. Count your remelts in the batch record so that when a batch coats thinly you can tell whether that was it.

Coated stock. Dry, dark, lightproof, and away from humidity swings. The Image Permanence Institute’s account of what goes wrong with gelatin plates is a humidity story: high relative humidity promotes mould in the binder, which solubilises it and destroys the image and makes the layer stick to what it touches; low humidity shrinks it and worsens lifting and flaking; and rapid swings do the most damage of all. Their recommendation is below 18 °C at 30 to 40 per cent relative humidity.

Keeping fog is the thing nobody has measured for hand-coated material. Baker gives the industrial test — ten days at 105 °F below 65 per cent relative humidity, with no more than 0.02 extra fog against controls — and no source in this corpus gives a rate for a hand-coated plate or sheet. Ross warns that emulsion held too long will fog and gives no figure. So date every sheet, keep a control from each batch in the dark, and develop one unexposed sheet from an old box before you commit a picture to it. That is your own keeping data, and it is the only kind that will exist.

The defect taxonomy, in one place

These are the named faults, with the stage that produces each. The break/fix page emulsion-coating-defects owns the diagnosis and the fix; this list is the map.

Defect	What you see	Where it was made
Comets	A dense or clear head with a tail drawn in the coating direction	A particle in the emulsion at the moment of coating; filtration
Repellency spots	Small circular uncoated areas	Grease or a surfactant deficit on the surface
Bubbles	Round craters in the layer; on a negative, a clear spot with a dark ring on the print	Air stirred into the emulsion, or too little wetting agent
Uneven thickness	Density varying across the sheet; one corner clearing late in the fixer	An unlevel bed, a stopped stroke, a second pass
Thick edges	A dark border along one or both long edges	Emulsion escaping past the ends of the coating rod
Frilling	The layer lifting and rolling back at the edges when wet	Swelling, no hardening, warm or long processing
Reticulation	A wrinkled network over the whole surface	A swollen layer meeting a sharp temperature change
Mottle	Irregular patchy density with a matte sheen	Emulsion coated too warm
Pepper	A fine scatter of black specks developing without exposure	Precipitation, in a chloride emulsion; sometimes visible only with an eyepiece
Slugs, sunspots	Discrete black marks on a negative	Undissolved halide salt during precipitation
Base fog	Overall density on unexposed material	Over-digestion, an old emulsion, safelight, or keeping
Bronzing	Thick areas darkening in room light weeks after processing	Hardened thick patches that fixed incompletely

The art of the surface

The last decisions on this page are not technical, and the course would be misrepresenting hand coating if it left them out. A hand-coated sheet is not a worse industrial product; it is a different object, and three of its properties are yours to choose.

The paper is a visual decision. Hot-press is smooth and holds fine detail; a rougher surface breaks the image into the texture of the sheet. A hand-coated sheet carries no baryta layer, so its white is the paper’s white rather than barium sulfate’s, and its maximum black sits on a different ground.

The edge is a decision. A rod-coated sheet has a straight, slightly heavier edge; a brushed sheet has a brush edge, which is a signature some makers want and others regard as a cliché. Both are defensible; what is not defensible is an edge you did not choose.

The size and the shape are decisions. You are coating a sheet, not buying one, so the image area does not have to be a standard format. Coat to the negative rather than trimming to a paper size.

Baker’s matt option is the historical form of the same instinct: 200 g of rice starch ground into 1000 to 1500 cc of water, stirred into the emulsion before filtering, with an equivalent quantity of water left out at make-up. The course records it without adopting it, and notes what it is — a way of choosing a surface at the emulsion stage rather than the paper stage.

Alternative route

This page declares that it needs a darkroom, and most of it does not. The distinction worth drawing is between the part of coating that is fluid mechanics and the part that is photography, and only the second needs the light off.

No darkroom, most of the page. Levelling, viscosity, the two working temperatures, the coating weight, the choice of rod against brush against card, drying, setting, hardening and the keeping test are all learnable on plain gelatin with no silver in it, in full room light, on a kitchen table. That is not a concession: it is exactly what the coating station’s own validation coat is, and it is the fastest way to make the twenty bad coatings that teach the hand. Coat, dry, weigh, cut, look at it in raking light, repeat. Every defect in the taxonomy above except fog will show up.

No darkroom, the rest of the page. Coating a sensitised layer has to happen in the dark, and there is no substitute for a space you can stand in: a changing bag will hold a plate but not an 11 × 15 inch sheet, a puddle pusher and a levelled bed. A windowless room at night, with a red LED at the distance the safelight test in the latent image made visible establishes, is the usual answer and is enough. If you have none, the commercial-emulsion route still needs one — bought emulsion is as light-sensitive as your own — so the honest position is that the coating of a sensitive layer is the one thing in Part V that cannot be routed around.

No hardener, and no wish to handle one. That is the default this page argues for, not a concession. The whole of the hardening section can be read and none of it performed, and a bath hardener applied at processing covers every case where the layer will not hold.

Recap

• Evenness, cleanliness, thickness. Duffin’s three words, and thickness is where beginners lose.
• Sized hot-press watercolour paper and glass; film base is out of scope because it needs a coating machine and a subbing layer.
• There is no chrome alum gelatin sub in any source read. Baker’s dip is 2 per cent chrome alum and no gelatin; the tested modern practice uses no sub at all, and frilling in the fixer is the test.
• Chromic acid cleaning is chromium(VI) and is not published here; Baker’s own caustic soda alternative is Level B.
• Coat above the setting point and near it: 36 to 40 °C industrially, 40 to 42 °C in the tested domestic practice, with the thermometer in the emulsion.
• Setting is hydrogen bonding, which is why moderate warmth undoes it and a hardener does not.
• Coating weight is mg of silver halide per dm². Baker’s target for chloride and chlorobromide plates is 25 to 40; a 5 g silver nitrate batch at 20 ml on an 11 × 15 sheet gives 40, and at 50 ml gives 99.
• The published hand-coating rates do not fully reconcile, so weigh the pot before and after and measure your own.
• Drying takes the layer from about 85 per cent water to about 5, and slower drying gives a more orderly gelatin array — cool and slow, not warm and slow, because of mould.
• Default: no hardener in the emulsion, with the consequences stated — corners only, no squeegee, no warm processing, no long soaks.
• Harden in the bath with a potassium alum fixer when the layer will not hold; chrome alum in the emulsion at 0.5 to 2 per cent of the dry gelatin only where a lesson needs it.
• Formaldehyde was excluded by Wall and Duffin before the toxicologists got to it; glutaraldehyde and glyoxal go out on hazard, and glyoxal on the absence of a dose as well.
• Keeping fog for hand-coated stock is unmeasured anywhere. Date the boxes and develop a control.

Check your understanding

Question 1. Your emulsion carries 5 g of silver nitrate in 200 ml of finished emulsion, and you lay 30 ml of it on an 11 × 15 inch sheet. Roughly what coating weight is that in milligrams of silver chloride per square decimetre, and how does it compare with Baker's figure for chloride plates?

• About 6 mg/dm², well below his 25 to 40
• About 60 mg/dm², around half as much again as the top of his 25 to 40 band
• About 600 mg/dm², more than ten times his band
• It cannot be computed without knowing the gelatin content

Show the answer and why

Answer: About 60 mg/dm², around half as much again as the top of his 25 to 40 band

5 g of silver nitrate is 5 ÷ 169.87 = 0.0294 mol, which as silver chloride at 143.32 g/mol is 4.22 g, so the emulsion is about 21 g of silver chloride per litre. 30 ml of that is 0.63 g, or 630 mg. An 11 × 15 inch sheet is 165 in² × 6.4516 cm²/in² = 1,064 cm² = 10.6 dm². 630 ÷ 10.6 = about 59 mg/dm². Baker's stated ceiling for chloride and chlorobromide plates is 25 to 40 mg/dm², described as a very thin film, so this is a heavier coat — which is the normal condition for hand coating and is why maximum black and fixing time both differ from a commercial paper's.

Question 2. Why does this course default to no hardener in the emulsion, and what four handling rules does that decision oblige?

• Because hardeners are all hazardous; the rules are to wear gloves, goggles, an apron and a mask
• Because a hardened thick patch fixes incompletely and bronzes later, and hand coating reliably produces thick patches; the rules are handle by the corners while wet, never squeegee, no warm processing and no long soaks
• Because hardening is expensive; the rules concern storage only
• Because unhardened emulsion is faster; the rules concern exposure

Show the answer and why

Answer: Because a hardened thick patch fixes incompletely and bronzes later, and hand coating reliably produces thick patches; the rules are handle by the corners while wet, never squeegee, no warm processing and no long soaks

Both a tested Tier 2 practitioner and Duffin describe the same trade-off from opposite sides — hardening makes processing chemicals, especially the fixer, penetrate with difficulty, and excessive hardening interferes with developer penetration. Since hand coating lays down a heavier and less even layer than a machine, and the outside edges of any coating are thicker than the middle, the beginner's normal condition is exactly the case where a hardened layer fixes incompletely and darkens later in room light. The price is a tender wet layer, and the four rules are that price stated as instructions rather than absorbed.

Question 3. A student reads that the course uses chrome alum and asks whether that is not the same chromium the course bans elsewhere. What is the accurate answer?

• It is the same element, so the ban is inconsistent and the course should not use it
• They are different oxidation states with different hazards: chrome alum is chromium(III), notified as GHS07 Warning with skin, eye and possible respiratory irritation and no sensitisation or carcinogenicity statement, with a British exposure limit of 0.5 mg/m³, against 0.01 mg/m³ with both the carcinogen and sensitiser notations for chromium(VI), which the course never uses at any level
• Chrome alum contains no chromium
• The ban applies only to solutions above 5 per cent

Show the answer and why

Answer: They are different oxidation states with different hazards: chrome alum is chromium(III), notified as GHS07 Warning with skin, eye and possible respiratory irritation and no sensitisation or carcinogenicity statement, with a British exposure limit of 0.5 mg/m³, against 0.01 mg/m³ with both the carcinogen and sensitiser notations for chromium(VI), which the course never uses at any level

The distinction has to be explicit every time, because "chromium" as an undifferentiated word would either ban a hardener that is demonstrably manageable or licence a carcinogen that is not. The fiftyfold gap between the two rows of the same EH40 table, plus the presence and absence of the Carc and Sen notations, is the evidence. Note the second half of the ruling: chromium(VI) — dichromates and chromic acid — is used nowhere in this course, which is why Baker's dichromate glass-cleaning option is described and not published, and why his own caustic soda alternative in the next sentence is what the course gives instead.

Question 4. Duffin says industrial trough coating runs at 36 to 40 °C. What is the reason for that specific window, and what goes wrong on each side of it?

• It is the temperature at which silver halide is most stable; below it the crystals dissolve and above it they grow
• It is above the gelatin's setting point but close enough to it that the layer sets soon after coating; too warm and the emulsion spreads unevenly and soaks through, giving mottle, and too cool and it coats thickly and unevenly, curling on drying and fixing badly
• It is the temperature at which the hardener works; outside it no hardening occurs
• It is a compromise with the drying zone temperature and has no effect on the layer

Show the answer and why

Answer: It is above the gelatin's setting point but close enough to it that the layer sets soon after coating; too warm and the emulsion spreads unevenly and soaks through, giving mottle, and too cool and it coats thickly and unevenly, curling on drying and fixing badly

Duffin gives the reason explicitly: the range "is chosen so that it is above the setting point of the gelatin but is sufficiently near to it to permit ready setting of the layer soon after coating". Viscosity is the variable, and it changes with temperature, which is why a maker standardises on one coating temperature and holds it from the first sheet to the last. The two failure modes are both visible: a too-warm emulsion behaves like oil in a hot pan and leaves islands of low density with a matte sheen, and a too-cool one coats thick, curls stubbornly on drying and — if hardened — fixes incompletely in the thick areas.

Question 5. You want to know whether your hardened coating is actually harder than an unhardened one. What can this course offer, and what can it not?

• A published melting-point test with a temperature criterion for adequately hardened gelatin
• A comparative test only: coat hardened and unhardened strips from one batch, warm each in water with a thermometer and note when the layer lifts, because no source read gives a procedure or a criterion and both temperatures depend on your gelatin, coating weight and drying history
• Nothing at all; hardening cannot be tested outside a laboratory
• A measurement of the swelling ratio, which the corpus provides

Show the answer and why

Answer: A comparative test only: coat hardened and unhardened strips from one batch, warm each in water with a thermometer and note when the layer lifts, because no source read gives a procedure or a criterion and both temperatures depend on your gelatin, coating weight and drying history

The principle is sound — a hardened layer refuses to melt at a temperature that melts an unhardened one, which is more or less the definition — but no source in this course's corpus gives a method or a number. Rule 1 therefore keeps the test comparative and insists on the control strip. That is not a weakness peculiar to this test: the same gap runs through the whole hardening question, because no hardener dose has ever been tested at domestic scale. Every figure in the literature is either a fraction of the gelatin weight, which scales by arithmetic, or a commercial batch quantity, which does not.

Question 6. Two sheets from one batch dry differently: one is left in a cool room for six hours, the other is dried quickly in warm air. What does Duffin's account predict about the layers, and what is the countervailing risk?

• The quickly dried sheet has the more orderly gelatin array; the risk is that it is slower to reach room conditions
• The slowly dried sheet has the more orderly gelatin array, since maximum order comes from drying at room temperature; the countervailing risk is that slow drying in a humid warm atmosphere is exactly what lets a dust-borne mould spore establish itself in the binder
• Neither differs; drying only removes water
• The slowly dried sheet loses more water, ending below 5 per cent

Show the answer and why

Answer: The slowly dried sheet has the more orderly gelatin array, since maximum order comes from drying at room temperature; the countervailing risk is that slow drying in a humid warm atmosphere is exactly what lets a dust-borne mould spore establish itself in the binder

Duffin: as the water is drawn off the gelatin molecules are pulled into a closer-packed array, and "generally speaking the more slowly drying is carried out, the more orderly will be the array"; optical activity measurements put maximum order at room-temperature drying, which he calls economically impossible in a factory. It is entirely possible for you. Baker supplies the other half: dust and bacteria are the plate maker's enemies, and a speck of dust on gelatin-coated glass may deposit a mould spore that finds an ideal medium, "especially if plates dry slowly in a humid warm atmosphere". The resolution is cool and slow rather than warm and slow, with the dust kept out.

Check my answersClear