Diffusion, Swelling and the Journey Into the Emulsion

Every reaction in this course happens inside a layer of jelly a few hundredths of a millimetre thick. The developer is not in a beaker with the silver halide; it is outside, and it has to get in. The thiosulfate that dissolves the unused halide has to get in and then bring the dissolved silver back out. The wash water has to remove what the fixer left behind. Agitation, edge effects, wash times, reticulation and the whole strange behaviour of stand development are one subject, and it is transport.

Diffusion, in one sentence and then in one equation

OpenStax’s definition is careful in a way that matters: diffusion is dispersal “in response to differences in concentration”, but the particles “are unaware of any concentration gradient, they simply move randomly”. There is no force pulling molecules down a gradient. There are simply more of them on one side, so more of them wander across from that side than back, and the net flow is one-way until the difference is gone.

The equation that puts a number on it is Fick’s first law:

J = −D (dc/dx)

Fick's first law of diffusion

J is the flux — how much substance crosses a unit area per unit time, in moles per square metre per second. dc/dx is the concentration gradient: how steeply the concentration changes with distance. D is the diffusion coefficient, a property of that substance in that medium at that temperature. The minus sign says the flow runs from high concentration towards low.

Everything a darkroom does to transport is a change to one of those three terms.

Term	What raises it	Photographic lever
The gradient, dc/dx	a bigger concentration difference, or a shorter distance across which it falls	fresh solution at the surface; a thin emulsion rather than a thick one; a resin-coated paper rather than a fibre one
The diffusion coefficient, D	higher temperature; lower viscosity; a smaller molecule	warmer baths; less swelling and less hardening; the size of the ion you are moving
The area	more surface in contact	agitation reaching every part of the sheet

Two useful facts about D, both from the LibreTexts treatment. Ions at room temperature usually have a diffusion coefficient between 0.6 × 10⁻⁹ and 2 × 10⁻⁹ m² s⁻¹. And the Stokes-Einstein relation, D = kT ÷ (6πηa), says D rises with temperature, falls with the viscosity η of the medium, and falls with the radius a of the moving species. A larger molecule in a stiffer gel moves more slowly, which is a sentence you will want again when fibre-base paper takes half an hour to wash.

🧮How long is the journey into the emulsion?

The random-walk relation D = λ² ÷ 2τ rearranges to a rule of thumb for how long a species takes to cover a distance x by diffusion alone: t ≈ x² ÷ 2D.

Take an illustrative layer 10 micrometres deep — 1 × 10⁻⁵ m, which is the order of an emulsion coating, though Part IV is where emulsion thicknesses are established properly — and the free-solution ion coefficient of 1 × 10⁻⁹ m² s⁻¹ above:

t ≈ (1 × 10⁻⁵)² ÷ (2 × 1 × 10⁻⁹) = 10⁻¹⁰ ÷ 2 × 10⁻⁹ = 0.05 seconds.

Note the square. Doubling the depth quadruples the time; a path twenty times longer is four hundred times slower, which on the same assumption is about 20 seconds rather than a twentieth of one.

Three honest caveats. The 10 micrometres is an illustrative distance rather than a measured coating thickness. The diffusion coefficient is for ions in free solution, and diffusion through a swollen gel is slower, so treat the answer as a lower bound and an order of magnitude rather than a measurement: the course has found no source giving a measured diffusion coefficient for a developing agent in photographic gelatin. And notice what the answer implies even so: the first arrival of developer at the bottom of an emulsion layer takes a fraction of a second, not minutes. Development takes minutes for a different reason, which the next section is about.

The emulsion is a swollen gel, and that is why anything can get in at all

Dry gelatin is a hard, glassy layer that nothing diffuses through usefully. Put it in water and, as Kodak’s 1928 primer puts it, “it swells as if, instead of the gelatin dissolving in the water, the water dissolves in the gelatin”. The layer takes up water, thickens, and becomes a mesh with solution in the gaps. The swelling is what makes the layer permeable.

The primer is equally clear about what changes the swelling.

• Acid and alkali. “A small quantity of either an acid or alkali will produce a considerable increase in the swelling, and since the developer is alkaline and the fixing bath is acid, both these solutions have a great tendency to swell the gelatin.”
• Temperature. “especially when they are warm.”
• Hardening. Alum and similar hardeners cross-link the gelatin so that it swells less. The primer’s own analogy is a good one: “Gelatin is like a sponge; the effect of hardening it is to contract all the network of the sponge.”

So a film in a developer is not a fixed object. It swells in the developer, swells differently in the fixer, and contracts as it dries — and it does that on every roll you process.

The journey, in section: what has to cross what

Depths are drawn readable rather than to scale. The one thing to take away: agitation cannot reach inside the emulsion. It works entirely on the boundary layer above it. 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.

Development as a transport problem

Kodak’s 1928 primer says the speed of development “depends chiefly upon the rate at which the developer diffuses into the film”. Set that beside the calculation above, which said the trip in takes a fraction of a second, and there is an apparent contradiction. The resolution is the whole idea of this section.

The first arrival is quick. What takes minutes is keeping up. A developing grain consumes developing agent and produces bromide and spent agent, all in a space a few micrometres across. Unless those are continually exchanged with the bulk solution, the neighbourhood of the grain becomes a different solution from the one you mixed: less agent, more bromide, more oxidised product. Development then slows, not because the tank has run out but because the grain’s own square micrometre has.

Between the emulsion and the stirred bulk of the solution sits a thin boundary layer that the stirring does not reach. Everything crossing between bath and emulsion has to cross it by diffusion, and its thickness sets the gradient — a thick boundary layer means a shallow gradient and a slow exchange.

That is what agitation is for, and it is all it is for. Kodak’s process-control publication defines it in exactly those terms: “agitation is necessary to maintain uniform solution activity by removing exhausted solution from the emulsion surface and replacing it with fresh solution.” It does not mix the developer, which was mixed when you made it. It does not speed the chemistry. It thins the boundary layer.

Which is why the makers’ procedures for a small tank are so specific — Kodak’s five to seven inversion cycles in five seconds repeated at thirty-second intervals, Ilford’s four inversions in the first ten seconds of every minute — and why consistency matters more than vigour. A pattern you repeat is a boundary layer you can predict.

Ilford also puts a number on how much of development is waiting for supply. Its intermittent tank pattern leaves the film still for about five-sixths of the time; switch to the continuous agitation of a dish or a rotary tube and the instruction is to cut the development time by about 15 per cent. That figure is a direct measurement of the transport-limited fraction, and it has a second consequence: a published development time carries an agitation scheme with it, so changing your pattern changes the time you need by roughly as much as a degree and a half of temperature would.

Both makers also warn in both directions. Kodak’s troubleshooting table lists mottle against inadequate agitation and non-uniform streaks against excessive or uneven agitation, and Ilford warns that a pre-rinse before development “can lead to uneven development” — because a film already swollen with water meets the developer with a diluted boundary layer that is not the same everywhere.

Adjacency: when local exhaustion becomes a picture

Now put two areas of very different exposure next to each other, and let the local depletion diffuse sideways.

The mechanism is set out in a 1999 paper in Optica Applicata on photographic sharpness. At the boundary between a heavily exposed region and a lightly exposed one, two flows cross in opposite directions: development-inhibiting by-products “diffuse laterally from the high exposure region to the low exposure region”, while “the fresh developer diffuses laterally from the low exposure region into the high exposure region”. The consequence, in the paper’s words, is that at the edge of the dense area “the concentration of developer is higher and that of inhibitor is lower … Development runs more rapidly at the edge area, resulting in a local maximum of optical density” — the border effect. On the other side, the inhibitor flowing in “can curtail development close to the edge, resulting in a local minimum of optical density” — the fringe effect.

Two flows across one edge, and the density they leave behind

1. 1Fresh developer flows towards the dense area — the thin area is not using much of it, so there is a surplus to share
2. 2Inhibitor flows towards the thin area — bromide and spent developer released where development is heaviest
3. 3Border effect — a local density maximum on the dense side of the edge
4. 4Fringe effect — a local density minimum on the thin side; the pair is a Mackie line

Mechanism as stated by Rajkowski and Nowak, Optica Applicata 1999. The profiles are drawn to show the argument, not measured. Acutance is a measure taken from exactly this trace across a knife-edge exposure.

That pair of lines — bright one side of an edge, dark the other — is a Mackie line, and the extra local contrast it produces at every edge is why a negative with strong adjacency effects looks sharper than its resolution alone would justify. Acutance is the measure taken from exactly this trace across a knife-edge exposure. The same mechanism, acting over a larger area, is what people mean by a compensating developer: highlights slow themselves down by exhausting their own neighbourhood while shadows keep going in fresh solution. Everything about how to get more or less of it — dilution, agitation, developer choice — belongs to Part VIII.

The same physics, producing faults

Local depletion is a picture when it happens at the scale of an edge and a fault when it happens at the scale of a frame.

Bromide drag. Solution loaded with bromide and spent developer is denser than fresh developer, so it sinks. In a tank left still, it flows downward off dense areas and across whatever is below them, retarding development there and leaving streaks that follow the flow rather than the image.

Surge marks at the sprocket holes. With 35 mm film in a spiral tank, agitation drives solution through the perforations, so the emulsion beside each hole is resupplied more vigorously than the emulsion between them, and develops further. The result is a regular series of marks aligned with the perforations. The course states the mechanism as the flow argument above; it has not found a manufacturer publication naming the sprocket-hole surge mark specifically, so the attribution is presented as mechanism rather than as a sourced identification. What manufacturers do document is the category: Kodak’s troubleshooting table lists “streaks of non-uniform density” against “excessive or uneven developer agitation”, and “mottle — areas of non-uniform density” against “inadequate developer agitation”.

Notice that both of Kodak’s entries are about unevenness, and that they point in opposite directions. Too little agitation and depletion patterns survive; too much or too irregular, and the flow itself prints. Neither is cured by changing the developer.

Fixing and washing: the same journey, outward

Fixing is diffusion in both directions. Thiosulfate travels in; the silver-thiosulfate complexes travel out. Washing is what removes the rest, and Ilford states its purpose exactly: washing “removes residual chemicals from the photographic process; in particular thiosulphate from the fix process can cause long-term image degradation if not effectively removed.”

Kodak’s 1928 primer analysed washing better than most modern accounts, and three of its conclusions are worth having.

Washing is diffusion, not dissolving. “The rate of washing depends largely upon the rate of diffusion of the hypo out of the film into the water providing the water in contact with the film is continuously removed. This diffusion rate has nothing to do with solubility.”

It follows a halving law. “The quantity of hypo remaining in the gelatin is continually halved in the same period of time as the washing proceeds.” The primer gives the half-time as about 15 seconds for an average negative held directly under a running tap, and about 30 seconds in most trays and tanks.

And it stops if the water does not change. “The process will then stop unless the water in the vessel is changed.”

The halving law, and what happens when the water stops changing

• Under a running tap, half-time 15 s
• Tray or tank, water renewed, half-time 30 s
• Tray, water left standing

Show the numbers behind this plot

Three curves of the fraction of thiosulfate remaining in a negative against time. The first, for a negative held under a running tap with a half-time of 15 seconds, plunges: it is at 0.25 after half a minute, 0.06 after one minute, 0.004 after two minutes and below one part in a million after five. The second, for a tray or tank with continuously renewed water and a half-time of 30 seconds, follows the same shape more slowly: 0.5 after half a minute, 0.25 after one minute, 0.06 after two minutes and about one part in a thousand after five. The third curve is the same tray with the water left standing: it follows the 30-second curve for the first half minute and then flattens completely at about 0.35 and never falls further, because the water immediately around the film has come into equilibrium with the film and there is no longer any gradient to drive diffusion. The vertical axis is drawn on a logarithmic-looking scale of fractions so that the halving is visible as a constant step. The teaching point is that the removal of the last traces depends entirely on whether the water in contact with the film is being replaced, and not on how long the film sits in water.

Series	Time washing, minutes	Fraction of the original thiosulfate remaining
Under a running tap, half-time 15 s	0.00	1.00
Under a running tap, half-time 15 s	0.25	0.50
Under a running tap, half-time 15 s	0.50	0.25
Under a running tap, half-time 15 s	0.75	0.13
Under a running tap, half-time 15 s	1.00	0.06
Under a running tap, half-time 15 s	1.50	0.02
Under a running tap, half-time 15 s	2.00	0.00
Under a running tap, half-time 15 s	3.00	0.00
Under a running tap, half-time 15 s	5.00	0.00
Tray or tank, water renewed, half-time 30 s	0.00	1.00
Tray or tank, water renewed, half-time 30 s	0.50	0.50
Tray or tank, water renewed, half-time 30 s	1.00	0.25
Tray or tank, water renewed, half-time 30 s	1.50	0.13
Tray or tank, water renewed, half-time 30 s	2.00	0.06
Tray or tank, water renewed, half-time 30 s	3.00	0.02
Tray or tank, water renewed, half-time 30 s	4.00	0.00
Tray or tank, water renewed, half-time 30 s	5.00	0.00
Tray, water left standing	0.00	1.00
Tray, water left standing	0.25	0.68
Tray, water left standing	0.50	0.50
Tray, water left standing	0.75	0.41
Tray, water left standing	1.00	0.37
Tray, water left standing	1.50	0.35
Tray, water left standing	2.00	0.35
Tray, water left standing	3.00	0.35
Tray, water left standing	5.00	0.35

The first two curves are arithmetic from the half-times Kodak's 1928 primer states; the third is drawn to show the shape of what the same primer describes when it says the process stops unless the water is changed, and its plateau is illustrative rather than a measured value. None of the three is a measurement made by this course. The curve is drawn to show the shape, not measured from a real material. Your own materials will differ, and measuring them is what the sensitometry part of the course is for.

Two practical conclusions follow, and both modern manufacturers agree with them.

Changes of water beat a running tap for economy. The primer says outright that where water is to be economised, “by far the most effective way of washing is to use successive changes of small volumes of water”, repeating the procedure about six times. Ilford’s modern film method is the same idea in three steps: fill the tank and invert five times, drain and refill and invert ten times, drain and refill and invert twenty times. Each fill restores a full gradient; each set of inversions clears the boundary layer. Ilford calls it faster than running water, says it uses less, and states it still gives negatives suitable for long-term storage.

Warm water does not help. This is the primer’s most counter-intuitive result, and its reasoning is Fick’s law applied twice: warmth raises the diffusion coefficient, but it also swells the gelatin, and “its swelling hinders diffusion in about the same proportion as the rise in temperature accelerates it, so that, as a matter of fact, washing goes on at about the same rate at all ordinary temperatures.”

🔬Why fibre-base paper washes so much more slowly than resin-coated

Not because it is thicker in the sense of being a bigger object, but because of where the thiosulfate goes and how far it has to come back. Ilford’s account is explicit: resin-coated papers are “waterproofed” by their polythene coating, so the fixer only ever occupies the emulsion; fibre-based papers “absorb processing solutions more readily into the paper base and therefore need longer washing times”, and Ilford’s paper developer sheet notes the same absorbency in the tray, where fibre paper “absorbs far more liquid” and goes limp.

Now apply t ≈ x² ÷ 2D. A resin-coated print has only to clear its emulsion layer. A fibre print has to clear the whole depth of a matted cellulose base as well — a much longer path, and a more tortuous one, through a material that holds solution rather than repelling it. Because the time goes as the square of the distance, a path twenty times longer is four hundred times slower. The course has no measured base thickness from a source it has read, so the factor is left as an order-of-magnitude argument; what is sourced is the direction, the cause, and the consequence Ilford publishes. Its sequences differ accordingly: 30 seconds in vigorous running water for a resin-coated print in a dish, against a fibre-base optimum-permanence sequence of five minutes’ wash, ten minutes in Washaid and five minutes’ final wash.

Ilford also warns against the opposite error: do not over-wash resin-coated papers, because water entering the cut edge of the paper can eventually swell and damage the print. Longer is not always better.

A wash aid shortens the fibre sequence, and the mechanism is worth knowing because it is not the obvious one. Ilford describes Washaid as “a hypo-eliminator formulated to aid the efficient removal of the thiosulphate by-products of fixation by ion exchange”. It is not dissolving anything and not destroying the thiosulfate; it is offering an ion the paper will hold in preference, so that the thiosulfate is released and can diffuse out. Part XII turns all of this into wash times and tests.

Osmosis, and where the word actually belongs

Osmosis is the movement of the solvent — water — across a barrier, in response to a difference in dissolved concentration. Most of this page is not osmosis: developer entering an emulsion and thiosulfate leaving it are diffusion of the solute, and calling that osmosis is a common and unnecessary error.

The place the word genuinely earns is the swelling and shrinking of the gelatin layer itself, which is water moving into or out of the gel as the surrounding solution changes. Move a film from a concentrated bath to plain water and the layer takes up water and expands; move it the other way and it gives water up and contracts.

Reticulation is the visible failure of that. Kodak’s 1928 primer names the cause precisely: “all solutions must be kept at the same temperature in order to avoid sudden contractions or expansions of the gelatin which may result in detaching the film from its support or in the production of reticulation, i.e., a coarse wrinkling all over the film.” The primer’s context is important and often dropped: it is discussing “when photographic materials are exposed to extreme temperatures”, and it recommends a hardener in the fixing bath as the control.

⁇Can a modern film actually be made to reticulate?

The mechanism above is well sourced and undisputed. What is disputed is whether it is reachable with a modern hardened film at ordinary darkroom temperature differences.

Kodak’s 1928 primer describes reticulation as a real risk with the materials of its day, under extreme temperatures, and prescribes alum hardening against it. Modern practitioners frequently report being unable to produce it deliberately on current films, and it is often described in current writing as largely historical. The course has not found a modern manufacturer statement establishing either position, and so does not pick one: it teaches the mechanism, which is certain, and does not claim that your film will or will not reticulate at a given temperature difference.

What both positions agree on is the control, and it is the same control that a manufacturer states for other reasons: keep the baths close in temperature.

How close is close?

Manufacturers do publish numbers for bath-to-bath temperature matching, and they are looser than the tolerance on the developer itself.

Statement	Tolerance	Source
Wash water against process temperature, film	within 5 °C	Ilford, on HP5 Plus, Rapid Fixer and the fill-and-invert method
All process solutions against the developer, film	at least within 5 °C	Ilford, on ILFOTEC DD-X
All process solutions against the temperature in use, film	within 1 °C	Ilford, on ID-11, Microphen and Perceptol
Wash and processing conditions, paper	18–24 °C throughout	Ilford, Washaid sequences
Developer temperature itself, film	20 °C, with the time compensated if it differs	the kinetics page

The second and third rows are from the same manufacturer and differ by a factor of five. The course does not reconcile them, because neither sheet says what it is protecting against: the ±1 °C on the powder-developer sheet appears in a passage about preparing solutions and is plausibly about rate as much as about swelling, while the 5 °C on the DD-X sheet is stated for the stop, fix and wash steps where rate no longer matters. Where two sheets from one maker differ, follow the one for the product in front of you, and take the tighter figure when you cannot tell.

Read the difference. The developer’s own temperature matters to a fraction of a degree, because it sets a rate. The temperature difference between baths matters at the level of a few degrees, because it is about the mechanical state of the gelatin. Two different quantities, two different tolerances, and mixing them up is why people either fuss over the stop bath or ignore the developer.

Stand development, described rather than recommended

Long, dilute, almost unagitated development is a deliberate decision to make development transport-limited. With no agitation, the boundary layer thickens and the local depletion the adjacency section described is allowed to build up over the whole frame: highlights exhaust their own neighbourhoods and slow down, shadows sit in solution that is barely used and keep going. The result is strong compensation, strong edge effects, and — because those same flows are what produce bromide drag — a real risk of streaking, which is the same physics arriving as a fault.

That is the mechanism, and it is all this page will say. Part VIII is where a method is tested and recommended or not; Part XXVIII catalogues what it goes wrong as. The point here is that stand development is not a recipe with a mysterious virtue: it is a choice about a boundary layer.

The five levers, and which one to reach for

Everything in this page is controlled by five things, and a fault is usually a signal about which one.

Lever	What it changes	Reach for it when
Agitation pattern	the boundary-layer thickness, and its uniformity	development is uneven, mottled or streaked; or you want more or less adjacency effect
Dilution	how quickly a local volume is exhausted	you want compensation, or you want a longer, more controllable time
Time	how far development proceeds, mostly in the highlights	contrast is wrong and exposure was right
Temperature	the rate, at about 10 per cent per degree	you are trying to hit a published time, or a whole session has drifted
Hardening and bath matching	how much the gelatin swells and how abruptly	the layer is physically damaged, frilling or wrinkling

The commonest mistake is reaching for time when the answer is agitation. If a negative is uneven, more time makes it more unevenly dense. Unevenness is nearly always a transport fault, and transport faults are fixed by changing how solution is delivered.

Recap

Diffusion is net movement down a concentration gradient produced by nothing more than random motion, and Fick’s first law puts it as a flux proportional to the gradient and to a diffusion coefficient that rises with temperature and falls with viscosity and molecular size. Because gelatin swells in water, the emulsion is permeable at all; acid, alkali and warmth swell it more, and hardening swells it less. Time by diffusion goes as the square of the distance, so a fraction of a second gets a developer to the bottom of an emulsion layer and twenty seconds or more gets thiosulfate out of a paper base. What takes minutes is not the journey but keeping up with consumption, and the unstirred boundary layer at the surface is where that battle is fought — which is the whole and only function of agitation, in Kodak’s own words removing exhausted solution from the emulsion surface and replacing it with fresh. Let the same depletion diffuse sideways across an edge and you get the border and fringe effects that make a Mackie line and raise acutance; let it happen across a frame and you get bromide drag and surge marks. Washing is the same physics outward, and it obeys a halving law: about fifteen seconds under a tap, thirty in a tank, and it stops entirely when the water stops changing, which is why changes of water beat soaking and why warm water does not help. Osmosis proper is the water moving into and out of the gel, and reticulation is its visible failure. And the five levers you actually hold are agitation, dilution, time, temperature and hardening.

This is the last page of Part III. What follows in Part IV is silver itself: the halides, the latent image, and development as the amplification this part has only pointed at.

Check your understanding

Question 1. A film is developed with no agitation at all for thirty minutes. Describe the negative feature by feature and give the transport explanation for each.

• It will be uniformly under-developed, because the developer never reaches the film
• It will be evenly over-developed, because the film sits in developer for a long time
• Reduced overall density with strong compensation, exaggerated edge effects at every boundary, and streaks running downwards from dense areas — because local depletion is never relieved, sideways diffusion across edges is allowed to build up, and bromide-laden solution sinks and retards whatever lies beneath it
• It will be fogged, because the developer oxidises

Show the answer and why

Answer: Reduced overall density with strong compensation, exaggerated edge effects at every boundary, and streaks running downwards from dense areas — because local depletion is never relieved, sideways diffusion across edges is allowed to build up, and bromide-laden solution sinks and retards whatever lies beneath it

Take the three separately. Overall density is lower than an agitated development of the same length because each grain's neighbourhood is only resupplied by diffusion through a thick boundary layer. Edge effects are stronger because the lateral flows that produce the border and fringe effects are never washed away. And the streaks are gravity: solution loaded with bromide and spent developer is denser than fresh developer, so it sinks off the dense areas and drags across whatever is below. All three are the same mechanism at three different scales — grain, edge and frame.

Question 2. Dark streaks run downward from the sprocket holes of a 35 mm negative. Name the mechanism, and give two changes to agitation that would prevent it.

• Uneven resupply: agitation drives solution through the perforations, so the emulsion beside each hole is refreshed more than the emulsion between them and develops further; use gentler inversions, and vary the axis of inversion so no part of the film is favoured on every cycle
• Light leak; use a changing bag and a darker room
• Fixer exhaustion; mix fresh fixer and fix for longer
• Static discharge during loading; earth the spiral

Show the answer and why

Answer: Uneven resupply: agitation drives solution through the perforations, so the emulsion beside each hole is refreshed more than the emulsion between them and develops further; use gentler inversions, and vary the axis of inversion so no part of the film is favoured on every cycle

The perforations are holes in an otherwise smooth sheet, so solution driven along the film accelerates through them and thins the boundary layer locally. Kodak's troubleshooting table gives the category — streaks of non-uniform density from excessive or uneven developer agitation — and prescribes checking and adjusting agitation. Note that the fix is never "more agitation": vigorous, single-axis agitation is what produces this fault. Gentler and more varied is the direction. Note also the honest limit stated on the page: the sprocket-hole case specifically is presented as mechanism, not as a sourced manufacturer identification.

Question 3. Explain why fibre-based paper needs longer washing than resin-coated, in terms of distance and swelling, without using the word thicker.

• Fibre paper holds more silver, so more must be removed
• Fibre paper is hardened, and hardening slows diffusion
• Fibre paper is developed for longer, so it has more chemistry in it
• Resin coating waterproofs the base so fixer only ever occupies the emulsion, while fibre paper absorbs it into the cellulose base; the thiosulfate must then diffuse back out along a much longer and more tortuous path, and because diffusion time goes as the square of the distance a path twenty times longer is four hundred times slower

Show the answer and why

Answer: Resin coating waterproofs the base so fixer only ever occupies the emulsion, while fibre paper absorbs it into the cellulose base; the thiosulfate must then diffuse back out along a much longer and more tortuous path, and because diffusion time goes as the square of the distance a path twenty times longer is four hundred times slower

The square is what makes the difference so extreme. Ilford states the cause plainly — resin-coated papers are waterproofed by their polythene coating and fibre papers absorb processing solutions into the base — and the physics converts that into a time. The hardening answer is a real trap, and Kodak's 1928 primer disposes of it: hardening does not slow diffusion at all, unless the gelatin has been dried after being hardened, because the network contracts without the layer as a whole contracting.

Question 4. Kodak gives the washing half-time for a negative under a running tap as about 15 seconds. Roughly what fraction of the original thiosulfate remains after two minutes, and what single condition does that figure depend on absolutely?

• About a half, and it depends on the water being warm
• About one part in 250 — two minutes is eight half-times, so 1/2 to the power 8 — and it depends absolutely on the water in contact with the film being continuously replaced
• None at all, and it depends on the fixer being fresh
• About a tenth, and it depends on the film being hardened

Show the answer and why

Answer: About one part in 250 — two minutes is eight half-times, so 1/2 to the power 8 — and it depends absolutely on the water in contact with the film being continuously replaced

Two minutes is 120 seconds, which is eight periods of 15 seconds, so the fraction remaining is one over two to the eighth, about 0.004 or one part in 250. The condition is the one the primer states in the same breath: the process "will then stop unless the water in the vessel is changed", because a halving law requires a gradient and a gradient requires the outside concentration to stay low. This is exactly why several changes of a small volume outperform a long soak in a large one, and why Ilford's three-fill inversion method works at all.

Question 5. Two darkroom workers each match their bath temperatures. One holds every bath within 0.3 °C of the developer; the other holds them within 5 °C but controls the developer itself to 0.3 °C. Who is doing the right thing, and why?

• The first: everything should be as close as possible
• Neither; temperature does not matter once the film is fixed
• The second: the developer's own temperature sets a reaction rate and needs tight control, while the difference between baths is about the mechanical state of the swollen gelatin and manufacturers state it as a few degrees — Ilford asks for solutions within 5 °C of the developer
• Both are wasting effort, since the compensation chart handles temperature

Show the answer and why

Answer: The second: the developer's own temperature sets a reaction rate and needs tight control, while the difference between baths is about the mechanical state of the swollen gelatin and manufacturers state it as a few degrees — Ilford asks for solutions within 5 °C of the developer

Two different physical quantities are being controlled, and they have different tolerances for good reasons. The developer temperature is a rate variable: at about ten per cent per degree it changes how much development happens. The bath-to-bath difference is a swelling variable: an abrupt change in the layer's water content is a mechanical stress on the gelatin, and Ilford's published figure for that is within 5 °C for films, and 18 to 24 °C throughout for papers. Chasing a tenth of a degree on the stop bath buys nothing; letting the developer drift 2 °C costs about 20 per cent of the development.

Question 6. A negative comes out with irregular blotches of higher and lower density that do not correspond to anything in the subject. Which of the five levers should you change, and what would happen if you changed the wrong one?

• Agitation: the fault is unevenness, which is a transport fault; extending the time would only make the same uneven pattern more strongly developed, and changing dilution or developer would leave the pattern in place
• Time: develop for longer so the blotches even out
• Temperature: warm the developer to speed diffusion
• Hardening: use a hardening fixer

Show the answer and why

Answer: Agitation: the fault is unevenness, which is a transport fault; extending the time would only make the same uneven pattern more strongly developed, and changing dilution or developer would leave the pattern in place

Kodak lists mottle — areas of non-uniform density — against inadequate developer agitation, and the remedy it gives is to check and adjust agitation. The general rule this page is trying to install is that unevenness is nearly always about how solution is delivered, and level is about how long or how warm. Reaching for time when the answer is agitation makes the fault more visible, because both the blotch and its surroundings develop further and the difference between them grows with the highlights.

Check my answersClear