Erythrosin sensitising stock

Fifty millilitres of stock, made once, and thereafter counted out in drops. This is the smallest formula in the formulary by weight of active ingredient and one of the largest in consequence: three drops of it turn a blue-blind emulsion into one that records green, and change the colour of the light the darkroom may use from that moment onwards.

Ingredient	Quantity	Form the source specifies
Erythrosin	1 g	the disodium salt, which is what is sold as the dye and what both sources mean; buy on CAS 16423-68-0
Ethanol	25 mL	drinking-grade, about 95 per cent; the source specifies drinking grade because laboratory and shop grades carry added ingredients whose effect on the dye she cannot vouch for
Water	25 mL, added	Half of a 50 mL solvent, not a make-up volume: the source states 1 g of dye in 50 mL of solvent that is half distilled water and half ethanol, and gives no final volume for the finished stock. Distilled.

Purpose

To hold erythrosin at a strength at which a photographically useful dose is a countable number of drops, so that the powder is weighed once in the life of the bottle or — if the smallest jar sold is a gram — not weighed at all.

That is the whole design brief, and it is a safety decision as much as a practical one. The dose an emulsion wants is a few milligrams; nobody weighs three milligrams of a fine red powder on a domestic balance, and nobody should be opening a jar of it every time they make an emulsion. One dissolution, an amber dropper bottle, and a pipette thereafter.

What the dye is for is a separate question with a longer answer, and it belongs to the emulsion rather than to this bottle: silver bromide and iodobromide are intrinsically sensitive only to ultraviolet, violet and blue. A sensitising dye adsorbed on the crystal surface absorbs a green photon and passes the energy, or the electron, into the crystal. The result is an orthochromatic material — sensitive to blue and green, blind to red — which is where the whole of its usefulness and the whole of its inconvenience come from.

Recommended uses

Making an emulsion orthochromatic in the make. This is the use the entry is written for, and it is the one the course performs: three or four drops into the salted gelatin before precipitation, in the orthochromatic emulsion project.

As the reference strength when reading an older formula. Both published doses in this corpus are given as a volume of a two per cent solution in half alcohol and half water, so the stock is the unit the literature counts in. A formula that says “25 cc of two per cent erythrosin” is unreadable without it.

As a design exercise in dye quantity. Duffin establishes that an optimum exists, that overshooting it loses speed quickly, and that where the optimum sits depends on the particular dye and the particular emulsion together. That makes the dose a variable the maker measures rather than a constant to copy, and a stock dosed in drops is the instrument for measuring it.

When another formula is preferable

• When the material must be panchromatic, erythrosin will not do it: it buys green and stops well short of red. The dyes that reach further are pinacyanol and its relatives, and this course keeps them as study only at Level D — the corpus holds no hazard record for any of them, and they demand total darkness, which is an accessibility problem as well as a hazard-control one.
• When the plates are already coated and dried, the dye is applied by bathing rather than in the make. Wall’s 1924 route is a 1 in 5000 aqueous bath — a different concentration for a different operation, and not comparable with the doses above. This course does not offer it, because the bath as Wall gives it carries ammonia and Part V excludes ammoniacal procedures at Level B, and substituting the borax he recommends for other dye classes would be the course inventing a bath he never printed.
• When a self-screening or filtering effect is wanted rather than sensitisation, a yellow dye is the answer and Baker names four of them. That is a different chemical job — absorbing light before it reaches the crystal, rather than handing energy to it — and the requirement Baker attaches is that the filtering dye must not desensitise the grains.
• When nothing needs to see green, no dye. An undyed emulsion is easier in every way that matters: a broader safelight margin, one fewer variable, and no red powder in the room.

Mixing

One operation, and it is the only time the powder is open.

1. Buy the smallest quantity sold, and if that is about a gram, plan to use the whole jar. The objective is that the powder is weighed once or never.
2. Stage everything before opening the jar: the measured solvent, an amber dropper bottle, a beaker about twice the final volume, a plastic spoon, gloves on, and a tray under the work in still air with no fan or extractor draught running across it.
3. Make the solvent first — 25 mL of distilled water and 25 mL of drinking-grade ethanol — and then bring the dye to the solvent rather than the other way about.
4. Rinse the dye jar into the beaker in stages rather than trying to tip a fine powder cleanly. The dye that stays in the jar is dye that is not in the stock, and the rinse is how the strength stays honest.
5. Into the amber dropper bottle, labelled with the substance, the strength, the solvent, the date and the words flammable, contains ethanol, per the labelling SOP.

🧮Two per cent, in drops, and the two published doses side by side

The stock:

1 g ÷ 50 mL = 2 g per 100 mL = 2 % w/v

Strength

The dose the course performs, at the conventional twenty drops to the millilitre:

0.15–0.20 mL × 20 mg/mL = 3.0–4.0 mg of dye

Three to four drops

3.0–4.0 mg ÷ 0.0294 mol = 102–136 mg/mol

Per mole of silver, for a 5 g silver nitrate make

Baker’s, for a make carrying five hundred grams of silver nitrate:

25 mL × 20 mg/mL = 500 mg of dye

Twenty-five millilitres of the same stock

500 mg ÷ 2.943 mol = 170 mg/mol

Per mole of silver

A factor of about 1.7 between a Tier 1 manufacturing manual and a tested Tier 2 practice, seventy-two years apart. For a dye dose that is close agreement, and it is the strongest cross-tier corroboration anywhere in Part V’s evidence. It is emphatically not an optimum: Duffin is explicit that the position of the maximum depends on the dye and the emulsion together, so both figures are places to start from and neither is an answer.

Twenty drops to the millilitre is a convention, not a measurement. Drop size depends on the dropper, the surface tension of a half-ethanol solution and how the bottle is held. If the dose matters to you, count drops into a small measuring cylinder once and calibrate your own dropper; that single measurement removes the largest uncertainty in the whole dose.

Behaviour

It is sparingly soluble in water and the solvent is half the formula. The only solubility figure in the corpus is 0.7 mg/mL, and it belongs to the free acid rather than to the disodium salt that is sold as the dye; a 2 per cent stock is 20 mg/mL, nearly thirty times that figure. The encyclopaedia entry records the tension and does not resolve it, because no measured solubility for the salt was found. What is not in doubt is the practical consequence: both published stocks are half alcohol, and a stock made in water alone is not the stock either source printed.

It stains everything it touches, and that is useful. Skin, cloth, worktop, tray, dropper. Unlike most contamination it shows exactly where the substance went, which makes a stain a legible record of a control that failed rather than merely a mess.

Its spectral shape is not flat. Wall’s description is the classic one: erythrosine gives a strong yellow, orange and green sensitiveness, but a marked want of sensitiveness in the blue-green. That dip is the signature of the eosine dyes and it is why the later cyanines replaced them. An emulsion dyed with it is not evenly sensitive across the visible; it has a hole in the middle-left of its added band.

There is an optimum and overshooting it costs speed. Duffin’s account is that sensitivity rises with the quantity of dye adsorbed up to a maximum which is reached before the crystals are fully covered, and that a rapid loss of useful speed follows if too much is added. He adds that commercial practice often uses less than the amount giving maximum speed, for the sake of the other properties. More dye is not more sensitivity.

A dyed emulsion fogs more readily than an undyed one, which is why the restrainer appears in these formulas at all and why a lengthened ripening on a dyed make is worth watching.

Image characteristics

The stock has none. The emulsion it dyes gains one property and loses one working condition, and both are visible.

Green sensitivity, measurable in a session. The same emulsion exposed unfiltered, through yellow and through green, compared by step count, is the only experiment in this course where spectral sensitivity is something the maker measures rather than reads. A red-filtered strip records nothing, which is the same fact as the safelight being usable.

No effect on image colour. Nothing in the corpus attributes tone to the sensitising dye. The evidence on image colour in these emulsions points elsewhere entirely — to the rate of silver deposition and hence to the developer.

Base fog, in the wrong direction. A dyed make and an undyed make from the same recipe are not equally clean, and the published remedy in this literature is a small bromide restrainer rather than less dye.

And the property that is not photographic at all: the material may no longer be handled under the safelight the earlier work used. That is dealt with under Interactions, because it is the largest single consequence of using this bottle.

The mechanism

Adsorption first, then energy transfer. The dye does not colour the gelatin; it sits on the surface of the silver halide crystal. Duffin’s chapter on spectral sensitisation carries the whole chain — adsorption, aggregation of the dye on the surface, and then the two proposed routes by which the absorbed energy reaches the crystal, electron transfer and energy transfer — and this page points at it rather than paraphrasing a chapter into a paragraph.

⚗Why an emulsion is blind to green in the first place

Silver bromide absorbs at the blue end and stops. Its absorption edge is where its own band gap puts it, and a green photon simply does not carry enough energy to lift an electron across that gap. Adding silver iodide at a few mole per cent shifts things a little and does not change the answer.

A sensitising dye works round the crystal’s own optics rather than changing them. The dye absorbs the green photon, because the dye’s own electronic transition is where the crystal’s is not, and what reaches the crystal is either the excited electron itself or the energy of the excitation. Either way the latent image forms by the crystal’s ordinary mechanism; the dye has only widened the range of light that can start it.

That is also why there is an optimum. A crystal surface can hold only so much dye before the molecules begin to interfere with each other and with what the surface is for, and Duffin’s point X — the maximum — arrives before the surface is fully covered.

Why the solvent is half alcohol is a matter of getting the dye into solution at all, and it is the one mechanical detail both sources agree on to the letter.

Function of every ingredient

Erythrosin, 1 g. The disodium salt of tetraiodofluorescein — four iodine atoms in a formula mass of 836, which is why a red powder this fine is dense for its bulk and stains as hard as it does. In this formula, at this quantity, it is doing one thing: providing a reservoir of dye at a strength at which the emulsion’s dose is countable. More dye in the stock does not sensitise better; it makes the drop a coarser unit and pushes the maker towards a dose past Duffin’s maximum, where speed falls away. Less makes the drop too fine a unit to be worth counting, and puts the maker back on the balance. In the emulsion the same asymmetry applies with sharper consequences, and neither published dose should be read as an optimum. Buy the right substance: the record carries two CAS numbers, 15905-32-5 for the free acid that the structure depicts and 16423-68-0 for the disodium salt that is sold as the dye, and the two differ in mass, so a dose in one is not a dose in the other. Buy on the salt’s number.

Ethanol, 25 mL. Half the solvent by volume and much more than half of it by work done: erythrosin is only sparingly soluble in water, and the alcohol is what puts a two per cent solution in the bottle rather than a suspension. Drinking grade is a specification and not a convenience — the source uses it because laboratory and shop grades carry added ingredients whose effect on the dye she cannot vouch for, and denatured alcohol in particular is denatured with something. More alcohol than half would change a formula that two independent sources print identically, and would make a 50 mL bottle a more serious flammable-liquid question on a shelf near a water bath. Less risks the dye coming out of solution, which shows as cloudiness or a deposit and means the strength on the label is no longer the strength in the bottle.

Water, 25 mL, distilled, added rather than made up to. The source states a solvent volume and not a final volume, so this page does not state one either. Distilled matters here for the same reason it matters in the make: this stock is going into an emulsion by the drop, and what is dissolved in the drop goes with it.

Interactions

With the safelight, which is the interaction that reorganises the whole session. From the moment the drops go in, the material is green-sensitive and the light it may be handled under changes. The manufacturer’s filter table gives deep red for orthochromatic materials — filter 906 — against dark brown for fast blue-sensitive materials, and a current ortho film’s data sheet specifies the same deep red filter, a 15 W bulb and a distance of not less than 1.2 m. The older instruction for the bathing route says the same thing from the other end: a deep red safelight may be used until the plates are covered with the dye solution, and after that the work goes on in darkness or by a green safelight.

This is a filter change and a fog test, not darkness, and both are things this course already owns: run the safelight test again on the dyed material rather than assuming the margin that held for the undyed emulsion still holds.

With the developer, at the far end. An orthochromatic print or plate cannot be inspected under the amber or yellow-green light a blue-sensitive material tolerates. Plan the development by time and temperature rather than by looking.

With metals, in the bathing route. Wall’s instruction is that only perfectly clean glass dishes should be used, and that old porcelain dishes and metal tanks are to be avoided, because metals tend to reduce the dyes and cause fog. It is a warning about the dye rather than about the plate.

With acid. The dye is supplied as the disodium salt; acidifying a solution of it drives it towards the sparingly soluble free acid, which is the form the low solubility figure belongs to.

With light, which it exists to absorb. Store the stock dark; a dye stock kept on a windowsill is a dye stock being used up by the window.

Variants

No course variant is offered, and there is nothing to make safer by changing the strength. The substance is unclassified, the quantity is one gram, and the exposure that matters is a single weighing that the formula is already designed to make singular.

Wall’s bathing solution is a different formula rather than a variant of this one. A 1 in 5000 aqueous solution — 200 mg per litre — with half a per cent of ammonia added, for immersing finished plates. It is a bath concentration and cannot be set beside a per-mole dose as though the two were comparable. Wall’s own assessment of the trade-off is worth having: bathed plates as a rule have a higher colour sensitivity than plates coated from a dyed emulsion, but they do not keep as well, and home-bathed plates will not keep well at all. This course does not publish the bath, because Wall’s version contains ammonia and Part V excludes it, and because substituting the borax he recommends for other dye classes into a formula he never restated would be invention.

Baker’s point of addition is the real variant, and it is a variant of the procedure rather than of the stock. He adds the dye to the silver solution just before mixing; the practice this course follows adds it to the salted gelatin before precipitation. Both are “in the make” and they are not the same operation. Whichever you use, record which, because it is one of the few variables in this part with a published disagreement attached.

The panchromatic extension is excluded rather than varied. Pinacyanol chloride and its relatives are Level D study-only in this course: no hazard record exists in the corpus for any of them, and they require total darkness.

Safety

Level B, and the level is set by a powder with no exposure limit rather than by any known toxicity.

The honest form of words is “unclassified on a thin evidence base”, and this course does not say harmless. Thirty-seven of thirty-nine notified reports say the substance does not meet GHS hazard criteria; only about five per cent of companies supplied information at all, and two notifications did lodge hazard codes that have not been read. There is no UK workplace exposure limit for it, and the regulator’s own statement is that absence from the list does not indicate that a substance is safe. Under COSHH that means exposure is reduced as far as is reasonably practicable with no figure to work to. The full record, with its sources, is on the erythrosin page.

!You will meet a headline about this substance, so meet it here first

Erythrosin is the food colour E127, and on 15 January 2025 the US Food and Drug Administration revoked its authorisation for use in food and ingested drugs. Both halves of that belong on the page or neither does.

What happened: the revocation was made as a matter of law under the Delaney Clause, a zero-tolerance provision, after two studies showed cancer in laboratory male rats exposed to high levels of the dye.

What the same notice says: the mechanism by which it causes cancer in male rats is a rat-specific hormonal one that does not occur in humans; relevant human exposures are typically far below the levels that produced the effect; and claims that its use in food puts people at risk are not supported by the available scientific information. Other countries still permit it, under the name erythrosine.

A reader who meets the ban elsewhere and finds no mention of it here would be right to stop trusting the course. A reader who meets it without the mechanism would draw a conclusion the regulator explicitly disclaims in the same document. And note what the notice is about: eating the dye at food-relevant exposures, not dissolving a gram of it once and thereafter dispensing drops.

Controls, all of them aimed at the one minute the jar is open: buy the smallest quantity and make it all up in a single operation; work over a tray in still air with no fan running; nitrile gloves and eye protection throughout; a disposable FFP2 or FFP3 mask, which this course states plainly is a precaution against an unquantified dust and not compliance with a limit, because there is no limit.

The stock itself is a flammable liquid. Fifty millilitres of roughly fifty per cent ethanol on a shelf near a water bath is a small thing that the labelling and storage rules exist to catch.

Treat staining as a signal. A red mark on a cuff means powder or solution went somewhere unplanned, and the useful response is to find out how rather than to wash it off.

Storage

An amber dropper bottle, dark, labelled and dated. The label carries the substance, CAS 16423-68-0, the strength, the solvent — half water and half ethanol — the date it was made, and the words flammable, contains ethanol.

No source publishes a keeping figure, and none is invented here. What the sources do give is a practice: date the bottle and store it dark. The failure mode to watch for is the dye coming out of solution, which appears as cloudiness or a deposit at the bottom.

Do not filter a cloudy stock and carry on using the label’s strength. Filtering removes dye and leaves the number on the bottle describing a solution that no longer exists. Warm it gently and shake it; if it will not clear, make it again.

Keep it away from heat and from the water bath, both because it is half ethanol and because a stock kept warm is a stock evaporating towards a higher strength than its label.

The unused stock is a waste decision rather than a drain decision. See Waste.

Incompatibilities

Acids, which drive the soluble disodium salt towards the sparingly soluble free acid. See incompatibilities.

Metals in a dye bath — old porcelain, metal tanks, anything that has corroded — because metals tend to reduce the dyes and cause fog.

Light, which is not a chemical incompatibility but is the one that empties the bottle.

Ignition sources, for the ethanol half.

An excess of itself, which is the incompatibility nobody expects: past Duffin’s maximum, more dye costs speed rather than adding it.

Waste

Milligrams, in a stream that already carries silver. The dye that reaches the drain reaches it in the wash water of an emulsion make that is already carrying silver, nitrate and halide, and it does not get a disposal route of its own: it joins that stream, which is bottled and taken to a licensed route like every silver-bearing stream in this part. See the disposal ruling and the general chemical waste SOP.

The unused stock is the item that needs a decision. Bottle it and label it rather than pouring it away — both because it is half ethanol and because the absence of a hazard classification is not evidence of safety.

Stained cloths, gloves and paper towel go with the solid waste from the session rather than into the general bin unlabelled, on the same principle.

Local regulation decides. This course states the chemistry and applies the standing jurisdictional caveat; it does not compute a threshold for a substance that has no exposure limit anywhere.

Troubleshooting

The stock is cloudy, or has thrown a deposit. The dye has come out of solution — most likely the solvent was not half alcohol, or the bottle has been cold. Warm it gently and shake. If it will not clear, make it again, and do not filter it and assume the strength survived.

The green-filtered strip records nothing. Either the dye never reached the crystals or the stock has degraded. Check that the drops went into the salted gelatin before the silver rather than afterwards, and check the bottle’s date and that it has been stored dark.

Base fog is high on the dyed emulsion and low on the undyed one from the same recipe. Expected in kind, not in degree. The published remedy in this literature is a small bromide restrainer with the finals; a shorter ripening is the other lever. And re-run the safelight fog test, because the material that is now fogging is not the material the test was done on.

Speed went down when the dye went up. That is Duffin’s maximum, met from the far side. Go back to three or four drops and work downwards rather than upwards.

Red fingerprints on the next day’s work. A stained dropper left on the bench, or gloves removed in the wrong order. The dye is a tracer for exactly this kind of failure; clean up as soon as the drops are counted.

A red stain in the emulsion after processing. Residual dye that the processing did not remove. Record it against the dose; it is one of the visible costs of overshooting.

Experiments

The dye series, which is the whole design exercise in one afternoon. Five batches at 1, 2, 4, 8 and 16 drops, everything else held, each coated at the same temperature and volume and each read on a step wedge. Somewhere in that series is the maximum Duffin describes and the fall-off beyond it, and where it sits is a property of your dye and your emulsion rather than a number anybody can give you.

Three filters and one emulsion. Expose identical strips unfiltered, through yellow, through green and through red. The red strip should be blank, and the blank strip is the same fact as the safelight you are now allowed to use.

Calibrate your own dropper. Count drops into a small measuring cylinder until you have a millilitre. Twenty is the convention; your bottle may give sixteen or twenty-five, and the difference is the largest single uncertainty in the dose.

The two points of addition, put against each other. One make with the dye in the salted gelatin and one with it in the silver solution, at the same milligrams per mole. Two Tier 1 and Tier 2 sources disagree about where it goes and neither says the other is wrong; a matched pair on your bench is the only evidence this corpus will ever have.

Keep a dyed plate and an undyed one for a month. Wall’s observation about bathed plates is that they do not keep well; nothing in the corpus says what a dyed make does over time. Coat, dry, store both in the dark, and read them against fresh strips from the same melt.