The Structure of a Printing Paper: Base, Baryta, Emulsion and Supercoat

Hold a sheet of enlarging paper up to a window and you are looking at a laminate. Not a sensitised sheet of paper: a manufactured stack of four or five layers, each present for one reason, and the order they sit in decides almost every number on the packet. Fix for one minute or for ten. Wash for two minutes or for forty-five. Glaze it or never glaze it. Tone it easily or with difficulty. All of those follow from the structure, and one question in the structure settles most of them: can a processing solution get into the support, or is the support sealed?

That is the question this page answers, and it is worth answering before the emulsion chemistry because the support decides what the emulsion is allowed to do.

A fibre-base sheet, from the bottom up

Fibre-base printing paper, in section

Drawn to show the order and the rough proportion, not the thickness. No manufacturer sheet in the course's corpus publishes a layer thickness or coating weight for a printing paper, so no figure is given for any layer but the base weight, which ILFORD do publish. 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.

!What the drawing does not claim

The relative depths above are readable rather than measured. ILFORD, Foma and Kodak publish base weight in grams per square metre and publish nothing about how thick the baryta, the emulsion or the supercoat is; the conservation literature describes the layers and their order without giving coating weights either. Any cross-section you meet with micrometres on it is either citing a source this course has not found or inventing them. The order is certain, the proportions are a drawing.

The paper stock, and one term to be careful with

The base is a specially made paper, not a sheet of writing paper that happens to have been coated. The Getty Conservation Institute’s atlas traces what changed in it: the original stock was rag-based, processed wood cellulose came in after the 1920s, and the internal chemistry was modified over time by synthetic resins introduced as internal binders and sizes, put there to raise wet strength and survival of hot processing. Wet strength is the property that matters most from the printer’s side: a sheet that has to be lifted out of four trays and washed for three-quarters of an hour has to stay a sheet.

You will read everywhere that photographic base is purified alpha-cellulose, chosen so that lignin, alum-rosin sizing residues and metallic impurities are excluded. Two of those three claims are easy to believe on general paper-chemistry grounds and none of the three is stated by any manufacturer sheet in the course’s corpus.

⁇Where the alpha-cellulose claim actually comes from

The term is real and it is used precisely — but in the corpus it is used about other papers. The Library of Congress recommends that storage enclosure paper have an alpha-cellulose content of at least 87 per cent. Mike Ware’s alternative-process writing specifies nearly 100 per cent alpha-cellulose for a hand-coating sheet, where the sensitiser goes into the fibres and every impurity in them is in the image. Those are the two contexts in which the course can cite the term.

What the course states: photographic base stock is a purpose-made paper whose composition its makers do not publish; the conservation literature records rag stock giving way to processed wood cellulose and to synthetic internal sizes; and the impurity argument, which is sound for a hand-coated sheet, has not been verified as a statement about manufactured baryta base. What the course does not state: an alpha-cellulose percentage, or a list of excluded impurities, for a bought printing paper.

Sizing is what stops liquid soaking straight into the fibres, and Part XXI will make it the single most important property of a sheet, because there the sensitiser goes into the paper. Here it matters less than it does there and more than you would expect: it is part of what holds the baryta coating flat on a surface that is, close up, a mat of fibres.

Single, double and premium weight

Weight is the honest name for thickness. ILFORD publish 255 g/m² for the double-weight fibre base of MULTIGRADE FB CLASSIC, FB WARMTONE and ILFOBROM GALERIE FB, and for their resin-coated range they publish 190 g/m² coded M and 250 g/m² coded K — with a note worth reading twice, that those base weights exclude the polythene coating and the emulsion, and about 70 g/m² should be added for the total. Kodak’s paper-chemicals sheet still distinguishes SW, DW and PW — single, double and premium weight — by giving single-weight fibre paper two minutes in Hypo Clearing Agent against three for the heavier stocks.

That last detail is the whole point of weight for a chemist: a thicker base holds more solution and takes longer to clear it. Single weight is cheaper, dries flatter and washes faster; double weight handles better wet, lies flatter dry and is what almost every fibre paper now on sale actually is.

The baryta layer

Between the fibres and the emulsion, a fibre paper carries baryta: barium sulfate dispersed in gelatin, coated on and calendered smooth. The AIC’s conservation catalogue dates its arrival on commercial developing-out papers to 1894 and on Kodak’s to 1900; the Getty atlas traces the coating itself further back, to its introduction by José Martínez-Sánchez and Jean Laurent in 1866, before it had anything to do with silver gelatin.

It does three jobs and each of them shows in the finished print.

⚗Why barium sulfate and not something else

Barium sulfate is brilliantly white, exceptionally insoluble and chemically inert. Those three properties are the specification. The whiteness is scattering rather than absorption — a dense suspension of small particles with a refractive index well above the gelatin around them returns light in every direction — which is why a thin layer of it hides the colour of the pulp underneath. The insolubility means it survives an alkaline developer, an acid stop, a thiosulfate fixer and an hour of washing without dissolving, without reacting and without contributing a single ion to the image. It is the one component of the sheet that is present precisely because it does nothing.

The consequence for analysis is that its sulfur, not the image’s, dominates the X-ray fluorescence spectrum of a sulfide-toned fibre print — the Getty atlas warns that the sulfur signal from the baryta makes the toned silver’s own sulfur very hard to separate.

It raises the maximum black. A reflection density is a ratio between light going in and light coming back out, so anything that returns more light from the white raises the ratio at the black end of the same measurement. The baryta gives the emulsion a bright, even, non-selective ground to sit on: the light that gets through the developed silver in a shadow hits white and comes back, and the light that lands on an unexposed area comes back stronger still.

It controls how deep the image sits. The AIC catalogue draws the distinction cleanly: in a printing-out process the image can lie partly within the paper structure, while in a developing-out print it sits on top of the baryta layer. That is why a silver gelatin print looks different from a salted-paper print made from the same negative, and Part XXII will use exactly this contrast when it puts silver into fibres instead of onto them.

It carries the base tint. Getty records photographic papers made with coloured stock or with colourants added to the baryta, sold under names from snow white through pearl white and natural white to old ivory and buff. The practice survives: ILFORD publish MULTIGRADE FB CLASSIC with a white base tint, FB WARMTONE on a warm white base, FB COOLTONE on a cool white base; Foma say of FOMATONE MG Classic that “the paper base involved is coloured in compliance with the tone of the developed silver”. The base tint is a design decision made in the coating plant, and it is the reference the eye measures every tone of your print against.

Optical brighteners, and why the white can move

Some of that whiteness is not reflection at all. Optical brightening agents are fluorescent compounds that absorb ultraviolet and re-emit it as visible blue, so the sheet returns more visible light than it received and looks whiter than white. Getty records manufacturers adding them to the paper base, the baryta layer or the emulsion, and dates the practice: developed in the late 1940s, tested after 1951, in industrial production from about 1953 on Paul Messier’s research, and in general use after 1955.

Two consequences follow, and the second is the one printers meet.

The first is forensic and is Getty’s own use for the fact: a paper that fluoresces was made after 1953, so brighteners in a photograph dated earlier raise a question about the date.

The second is that a brightener is not permanent. Getty state that its concentration can be reduced by extensive washing and that later treatment with certain organic compounds, heavy metals or oxygen-containing materials can quench the fluorescence outright. A fibre print gets an hour in running water by design. The base white you measure on a freshly dried print is therefore not guaranteed to be the base white of that print in twenty years, and the change has nothing whatever to do with the silver. It is worth setting beside ILFORD’s claim for ILFOBROM GALERIE FB — a bright white base tint “which will not yellow with prolonged washing” — which is a claim about the base and not about a brightener, and is as far as the corpus goes on the subject.

Resin-coated construction

Resin-coated printing paper, in section

Drawn to show the order and the sealing, not the thickness. The layer that changed photography here is the lower one: it is what makes the sandwich a sandwich. 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.

The Getty atlas gives the origin plainly, and it is not an aesthetic one: resin-coated paper grew out of a military requirement for material that would not absorb processing solutions, so that washing and drying could be short and processing fast. Early versions really were coated with thermosetting resins, which is where the name came from; the commercial material that followed uses thin extruded layers of polyethylene, which is why the European designation PE paper describes it better than the American RC does. The first commercial RC paper was Kodak’s, in 1968.

The recto polyethylene is filled with titanium dioxide, and that pigment is doing baryta’s job: Getty describe it as simulating the visual effect of the baryta layer of a fibre paper. The emulsion is coated straight onto that pigmented plastic and overcoated with hardened gelatin. The emulsion never touches a fibre.

⁇Two claims about the back of an RC sheet that the corpus does not support

It is often said that the rear polyethylene layer is there for antistatic and anticurl reasons, and that RC prints do not curl. Neither survives contact with the sources.

The rear layer. The only function any Tier 1 or Tier 2 source in the corpus gives it is Getty’s: the verso film is matte polyethylene, made to resemble ordinary paper so that it can be written on. Nothing in the corpus attributes an antistatic or an anticurl function to it. The course states the sealing and the writable surface, and leaves the rest unclaimed.

Curl. ILFORD’s own RC sheet warns that prolonged immersion can cause edge penetration and print curl with resin coated papers, and sets a limit of fifteen minutes of total wet time to avoid it. So RC prints curl too — from water finding its way in at the cut edges, which is the one route the sandwich leaves open. What is true is the comparison: a fibre print curls as a matter of course on drying, and ILFORD’s fibre sheets tell you to squeegee both sides and clip prints back-to-back to minimise it, while an RC print dries flat unless you leave it in the water too long.

The one difference that drives everything else

Everything above reduces to a single sentence. Solution enters a fibre base and cannot enter a polyethylene one. Put ILFORD’s own published times side by side and watch that sentence turn into the working day.

Step	MULTIGRADE RC	MULTIGRADE FB CLASSIC
Develop, MULTIGRADE 1+9 at 20 °C	1 min	1 min 30 s – 3 min
Stop, ILFOSTOP 1+19	10 s	10 s
Fix, RAPID FIXER 1+4	30 s	1 min
Wash, running water above 5 °C	2 min	30–45 min
Total wet time	under 4 min	up to 50 min

The developer difference is modest and belongs to the emulsion. The fix and the wash differ by factors of two and twenty, and that is structural. Fixer has to reach the emulsion in both cases; in a fibre print, spent fixer and its dissolved silver complexes then have to come back out of the base, where they went by capillary action and where nothing but diffusion into flowing water will find them. Kodak’s paper-chemicals sheet says the same thing in its own units: Rapid Fixer for two minutes on RC paper, five to ten minutes on fibre.

That single difference propagates:

• Drying. A fibre print is squeegeed both sides, clipped back-to-back and air-dried, glazed or heat-dried. An RC print must never be glazed or dried on a drum or flatbed glazer, because the polyethylene sticks to the glazing surface. ILFORD say so directly.
• Toning. A toner has to reach the image, and on a fibre print it also reaches the base. Part XX will find both halves of that mattering.
• Permanence. The whole permanence literature that Part XII cites was written about fibre prints, because the failure mode it studies — residual thiosulfate left in a support — is a failure mode a sealed support does not have. RC paper has ageing questions of its own, in the plastic, and they are not the same questions.

Surface, and why it settles an argument before the chemistry speaks

Papers are sold in glossy, pearl, satin, semi-matt, matt and textured finishes. ILFORD code them: 1 glossy, 25 satin, 44 pearl on the RC range; 1K glossy and 5K matt on MULTIGRADE FB CLASSIC; 24K semi-matt on FB WARMTONE. Getty describe how the finishes are made — matting agents such as starches, barium sulfate or glass particles put into the emulsion, heavy calendering of the paper, or texturing of the stock or the baryta.

Why the same silver reads as two different densities

1. 1Incident beam at 45° — the same on both sheets
2. 2Specular reflection — leaves at 45° on the far side; on a glossy sheet it misses the detector entirely
3. 3Scattered light — a matt surface throws it everywhere, and some of it goes straight up
4. 4Detector on the normal — sees image light on the glossy sheet, image light plus surface scatter on the matt one

The silver is identical in both drawings. Only the surface differs, and the number differs with it — which is why a Dmax comparison between two finishes is a statement about finishes.

The effect is not small and it is published. Foma print it twice on one sheet, once under each of FOMATONE MG Classic’s technical tables: Dmax 2.0 for the glossy surface, 1.6 for the matt. Their FOMABROM sheet makes the same point in words — the curves are valid for the glossy surface, and any other surface, “namely the matt one, causes a decrease in the maximum density value”. A difference of 0.4 in density is a factor of two and a half in the light coming back out of your deepest shadow.

📷What this does to a comparison

Two papers on a table, one glossy and one matt, cannot be compared for depth of black. Neither can a developer tested on one and a developer tested on the other, nor a toner, nor a fixing regime.

The rule the course keeps: compare on one surface, or do not compare. A great deal of published enthusiasm about a chemistry that gives deeper blacks is a change of finish reported as a change of chemistry, and it is one of the easiest errors in the darkroom to make honestly.

Surface is chosen before contrast for the same reason. Contrast is adjustable on a variable-contrast sheet, in half-grade steps, at the moment of exposure. Surface is not adjustable at all: it arrives with the box, it sets the maximum black available to you, and it changes how the print behaves in every later step. ILFORD note in their finishing section that the glossy 1K surface responds more favourably to toning than the matt 5K, so the choice reaches forward into Part XX as well.

The supercoat

On top of the emulsion sits a thin layer of hardened gelatin with no silver in it — the supercoat, which Getty also call a topcoat or overcoat and record as a later addition by manufacturers protecting the emulsion against mechanical damage. It is why a scratch you can plainly see on a print may not touch the image at all: the mark stopped in a layer that has nothing in it.

For a printer, it has one further consequence. Everything that has to reach the image has to go through it first. A toner, a retouching dye, a bleach, a local reducer: each has to diffuse through a hardened gelatin skin before it meets any silver. That sets a floor under how fast toning can act and it is part of why toning is a matter of minutes rather than seconds. Note the direction the argument runs: the supercoat slows the arrival, it does not stop it, and anything that makes the gelatin harder slows it further — which is the next section.

Gelatin hardness, wet strength, and why the makers do not want a hardening fixer

Gelatin has to be hard enough to survive four trays and an hour of water, and soft enough to let solutions in and out. That is a genuine trade-off, and the printing papers have been designed with a particular answer to it, which is why every current manufacturer sheet in the corpus advises against a hardening fixer for paper.

ILFORD state it three ways across three sheets:

• The MULTIGRADE FB CLASSIC and RC sheets: a hardening fixer is not recommended because it reduces washing efficiency; RAPID FIXER and HYPAM are non-hardening.
• The FB WARMTONE sheet adds two more consequences: it may impair toning performance and it gives a cooler image tone.
• The optimum-permanence sequences say it in the imperative: do not add a hardener to the fixer, and do not exceed the fixer’s capacity or extend the fixing time, because both make washing harder.

Kodak’s toning guide, from the other direction, says a hardening fixer makes the paper emulsion less receptive to the toner solution, and recommends a non-hardening fixer for any print to be toned.

⚗One mechanism behind four symptoms

A hardener — potassium alum in the classic acid hardening fixers of Part XI, or a chrome or aldehyde hardener elsewhere — cross-links the gelatin. Cross-linked gelatin swells less in water, and a layer that swells less admits and releases dissolved substances more slowly.

Every symptom above is that one sentence seen from a different side. Thiosulfate leaves a hardened emulsion more slowly, so washing is less efficient. A toner enters it more slowly, so toning is impaired. And the tone shift ILFORD report on a warm-tone paper is consistent with a hardened layer altering how the image silver is deposited and how the print then behaves — though ILFORD publish the effect and not the mechanism, and the course does not go further than they do.

There is one place a hardener is still wanted, and ILFORD name it: if a belt print dryer must be used on FB WARMTONE, fix with a hardening fixer, because otherwise prints stick to the belt. That is the trade being made explicit — mechanical robustness bought with washing and toning performance.

The related warning travels with it. ILFORD state that there is no benefit in extending fixation beyond the recommended time, and that long fixing times can cost print quality through image etching and, on the RC sheet, affect image colour. Fixing is not a step where more is safer; that is a film habit, and Part XI already explained why it is wrong.

Reading a paper datasheet, and what each line commits the maker to

A datasheet is a document with legal weight behind some of its lines and none behind others. Learning which is which is a skill worth having, and printing papers are the easiest place to practise it.

Line on the sheet	What it is	What it commits the maker to
Base weight, e.g. 255 g/m²	A measurement of the product	Fully — it is a specification
Base tint, image tone, surface code	Product identification	Fully — a different tint is a different product
ISO(R) and ISO speed by filter	Measured to a named standard, stated as representative of results obtained when processed to the maker’s own recommendations	To the measurement under their conditions, not to yours
Characteristic curves	Measured, with the developer, dilution, time and temperature printed beside them	To that curve in that developer at that time
Processing summary	Recommendation	To the result being achievable, not to it being the only route
“Can also be processed in other high quality dish developers”	Permission	To nothing
Optimum permanence sequence	Recommendation with a purpose attached	To that sequence being their best advice for archival storage
Keeping time — three years for FB CLASSIC, two for the RC range	Claim, conditioned on storage as recommended	To the condition, which is doing most of the work

The habit to build is to read the conditions attached to a number before the number. ILFORD’s ISO(R) tables carry the sentence “representative of those obtained when dish/tray processing the paper to ILFORD recommendations” every time they appear. That is not a disclaimer; it is a statement that the figure is a property of a system, and your system is not theirs until you have matched the developer, the dilution, the time and the temperature.

Choosing a paper, which is a craft decision made with chemistry

What fibre buys: a base whose behaviour the permanence literature actually describes; the ability to glaze; the deep tonal separation that comes with a heavy baryta ground; a surface that takes toning well; and a sheet that feels like an object.

What resin-coated buys: a print washed and dry in ten minutes instead of two hours; flatness without a press; a wash the environment and your water bill can live with; and — the reason it belongs in this course — fast feedback. A learner who waits an hour to see whether a decision was right makes a tenth as many decisions.

✦The honest recommendation

Learn on resin-coated, finish on fibre. Every mechanism this part teaches is present in both, and the cheap fast sheet is the one that will let you meet those mechanisms twenty times in an evening. Then make the print that matters on fibre, having already learned what you are doing, and give it the hour it needs.

The one caution: the two are not interchangeable at the moment of measurement. A dry-down figure, a maximum black, an exposure and a grade are all properties of a paper-and-developer pair, and moving from RC to FB is a change of pair. Part XVIII’s experiment measures one pair; Part XIX will assume you re-measure when you change.

Recap

• A printing paper is a stack, and the order of its layers decides its processing. Fibre: stock, sizing, baryta, emulsion, supercoat. Resin-coated: matte polyethylene, paper core, pigmented polyethylene, emulsion, supercoat.
• Baryta is barium sulfate in gelatin, chosen for whiteness, insolubility and inertness. It raises maximum black, keeps the image on top of the paper rather than in it, and carries the base tint.
• Resin-coated paper is polyethylene on both faces with titanium dioxide in the front layer doing baryta’s job. It exists because someone needed a paper that would not drink.
• Solution enters fibre and not polyethylene, and that fact alone explains a fix of one minute against thirty seconds, a wash of forty-five minutes against two, the glazing rule, the toning difference and most of the permanence literature.
• Surface changes measured density — Foma publish 2.0 glossy against 1.6 matt on one paper — so papers, developers and toners are compared on one surface or not at all.
• Hardened gelatin swells less, so a hardening fixer costs washing efficiency and toning performance. Every current maker’s sheet in the corpus advises against one for paper.
• Three things the course does not claim: a layer thickness for any coating, an alpha-cellulose specification for manufactured base, and an antistatic or anticurl function for the back of an RC sheet.

Check your understanding

Question 1. A fibre print and an RC print are fixed side by side for ten minutes in the same non-hardening fixer. Which one is harmed, and by what mechanism?

• Neither — extra fixing is always safe
• The RC print, because polyethylene dissolves in thiosulfate
• The fibre print, because thiosulfate and dissolved silver complexes enter the base and are then very hard to wash out
• Both equally, because fixing time affects the emulsion and not the support

Show the answer and why

Answer: The fibre print, because thiosulfate and dissolved silver complexes enter the base and are then very hard to wash out

Both suffer from over-fixing of the emulsion — ILFORD warn of image etching and a colour shift — but the structural harm is the fibre print’s. Fixer soaks into the base by capillary action, and everything that goes in has to diffuse back out into flowing water. That is why ILFORD’s optimum-permanence sequence tells you not to exceed the fixer capacity and not to extend the fixing time: both make washing harder. The RC print’s core is sealed, so ten minutes costs it far less — though ILFORD do cap total wet time at fifteen minutes because water finds the cut edges.

Question 2. The same negative is printed on a glossy and on a matt sheet of the same emulsion, at the same exposure, developed identically. What does a reflection densitometer read in the deepest shadow?

• The same density on both, because the silver is identical
• A higher density on the matt sheet, because matt surfaces absorb more light
• A lower density on the matt sheet, because its rough surface scatters light back to the detector without it passing through the image
• An unpredictable difference, because reflection density is not a defined quantity

Show the answer and why

Answer: A lower density on the matt sheet, because its rough surface scatters light back to the detector without it passing through the image

Foma publish the pair for FOMATONE MG Classic: Dmax 2.0 glossy, 1.6 matt, on one sheet with one emulsion. A reflection density is a ratio of light returned to light incident, so surface scatter that reaches the detector without ever meeting the silver lightens the reading. Nothing chemical has changed. This is why the course insists that papers, developers and toners are compared on one surface.

Question 3. A supplier’s page states that a printing paper’s base is "purified alpha-cellulose with lignin and metallic impurities removed". How should that be treated?

• As established fact, since it is a technical statement about a manufactured product
• As unverified: no manufacturer sheet in the course’s corpus states it, and the term is used in the conservation literature about enclosure and hand-coating papers
• As false, because photographic base is made of rag
• As irrelevant, because the base is sealed under baryta anyway

Show the answer and why

Answer: As unverified: no manufacturer sheet in the course’s corpus states it, and the term is used in the conservation literature about enclosure and hand-coating papers

It is plausible and it is unsourced, and those are different things. What the corpus does support is the Getty atlas’s account — rag stock giving way to processed wood cellulose after the 1920s, with synthetic resins added as internal binders and sizes for wet strength — and the Library of Congress’s 87 per cent alpha-cellulose recommendation, which is about storage enclosures. A supplier page is not a tier 1 or tier 2 source, and the honest page says what is established and stops.

Question 4. Why does ILFORD advise against a hardening fixer for their papers, and what single mechanism connects that advice to Kodak’s advice against one before toning?

• Alum reacts with silver, and toners cannot then reach the silver
• Hardener raises the fixer’s pH, which slows both washing and toning
• Cross-linked gelatin swells less in water, so dissolved substances enter and leave the emulsion more slowly
• Hardened prints are thicker, so diffusion paths are longer

Show the answer and why

Answer: Cross-linked gelatin swells less in water, so dissolved substances enter and leave the emulsion more slowly

One mechanism, four published symptoms. A hardener cross-links the gelatin; cross-linked gelatin swells less; a layer that swells less exchanges dissolved material more slowly. So thiosulfate leaves more slowly — ILFORD’s "reduces washing efficiency" — and toner enters more slowly — Kodak’s "less receptive to the toner solution". ILFORD add that it may impair toning and gives a cooler image tone on FB WARMTONE. The one case where the trade is worth making is also theirs: a belt dryer, where prints would otherwise stick.

Question 5. ILFORD publish an ISO(R) of 95 for MULTIGRADE FB CLASSIC at filter 2, with the note that the values are "representative of those obtained when dish/tray processing the paper to ILFORD recommendations". What does that note do?

• It is a legal disclaimer with no technical content
• It says the figure is a property of a whole system — paper, developer, dilution, time, temperature — rather than of the paper alone
• It means the figure is approximate to within about ten per cent
• It means the figure applies only to the glossy surface

Show the answer and why

Answer: It says the figure is a property of a whole system — paper, developer, dilution, time, temperature — rather than of the paper alone

It is the most informative sentence on the page. A range figure is measured on a processed sheet, and processing is part of the measurement: change the developer, the dilution, the time or the temperature and you have changed the system the number describes. That is exactly why Part XVIII asks you to measure your own maximum black and dry-down for your own pair, rather than adopting a published figure, and why Part XIII treats a grade number as a maker’s designation rather than a measurement.

Check my answersClear