EDTA, citric acid and hydrochloric acid clearing sequence

A platinum print comes out of the developer finished as a picture and unfinished as an object. The image is metal, and metal is not the problem. The problem is the iron that never became metal — most of a sensitiser’s worth of it, sitting in the paper, invisible on the day and quite capable of turning the highlights yellow twenty years later. Between the developer and the wash there are three trays of dilute acid whose whole job is to take it out. This page is what goes in those trays: the hydrochloric acid of the trade papers, the acidified citrate Willis published for palladium, and the citric acid Photographers’ Formulary still ship with their kits. Six published baths in three chemistries, and in each chemistry the strongest bath is more than three times the weakest — 1 in 60 against 1 in 200 for the mineral acid, 5 per cent against 1.5 for the citric. Not because the iron changed, but because the image did.

Bath A: dilute hydrochloric acid for black platinotypes, 1 in 60 — the historical clearing bath, and the strongest of the six

Ingredient	Quantity	Form the source specifies
Hydrochloric acid	1 mL of a concentrated hydrochloric acid of specific gravity not less than 1.16, colourless, and not the commercial or muriatic grade — Ware identifies the concentrate of this tradition as 36 per cent w/w, about 12 M solution	Wall's proportion, "one part of hydrochloric acid with 60 parts of water". Ware writes the same bath as the concentrated acid diluted 1:60 and computes 0.2 M at a pH below 1, which is a dilution to sixty volumes rather than an addition of sixty; both readings are kept and the 1.6 per cent between them is not worth an argument
Water	60 mL, added	Water added, as Wall states it, and the acid goes into the water and never the reverse. No source states a tray volume for this bath; the ratio is recorded at its own scale and the arithmetic for a working tray is set out under Mixing.

Bath B: dilute hydrochloric acid for palladiotypes, 1 in 200 — the same acid at a third of the strength, because palladium dissolves and platinum does not

Ingredient	Quantity	Form the source specifies
Hydrochloric acid	1 mL of a concentrated, ca. 36 per cent w/w, ca. 12 M solution	Anderson's figure as Ware reports it, confirmed by McCabe and by Gottlieb in 1993 — palladiotypes "must be cleared in 1:200 hydrochloric acid, not the 1:60 acid used for platinum, which otherwise dissolves palladium and bleaches the image partially"
Water	to make 200 mL	A make-up volume, because Ware's "1:200" is the same statement as his "1:60", which he converts to a molarity. As with bath A, the ratio is at its own scale.

Bath C: Willis's acidified citrate clearing bath for palladiotypes, at working strength — the manufacturer's own palladiotype bath, and the only clearing bath Willis recommended for palladium

Ingredient	Quantity	Form the source specifies
Trisodium citrate dihydrate	25 g	Trisodium citrate dihydrate, FW 294.10, as Ware names it. Ware's working strength is 2.5 per cent w/v, reached by diluting a 20 per cent stock eight times; per cent w/v is grams of solute in 100 cc of the made-up solution, which is Ware's own definition
Citric acid (anhydrous)	11 g	Ware's 1.1 per cent w/v, from the 9 per cent stock at the same eightfold dilution. Willis's developer carries citric acid too, at 2.2 per cent w/v against the same 20 per cent citrate, so the clearing stock is the developer with four times the acid, and the bath is that stock with eight times the water
Water	to make 1000 mL	A make-up volume: both figures are per cent w/v, which is defined against the volume of the finished solution. Willis's own instruction is to make a stock at 20 per cent citrate and 9 per cent citric acid and dilute it eight times for use; the working strengths are recorded here because they are what the tray holds and Ware publishes both.

Bath D: Wall's citric acid clearing bath for platinotypes, 1 oz to 20 oz — the period alternative to the hydrochloric acid, and the ancestor of every kit bath

Ingredient	Quantity	Form the source specifies
Citric acid (anhydrous)	28.35 g	1 ounce avoirdupois, converted at Ware's Appendix VIII figure of 437.5 grains = 28.3495 g. Wall's line is "citric acid, in the proportion of 1 oz. to 20 ozs. of water, may be used. This softens the paper in less degree than does the hydrochloric acid"
Water	568 mL, added	20 imperial fluid ounces, converted at Ware's Appendix VIII figure of 568.261 cm3, and stated as water added because Wall gives a proportion of acid to water and no final volume. A nominal 5.0 per cent w/v, which is two and a half times the strength of the modern platinum kit's bath.

Bath E: Photographers' Formulary dilute citric acid for platinum, 2 per cent w/v — the modern kit bath for a platinum print

Ingredient	Quantity	Form the source specifies
Citric acid (anhydrous)	30 g	The whole of the 30 g the kit supplies, dissolved in 1000 mL of warm water and then brought to 1500 mL with 500 mL of cold — a nominal 2.0 per cent w/v
Water	to make 1500 mL	at 49 °C; The sheet's instruction is 1000 mL of water at 120 degrees to dissolve the acid, then 500 mL of cold water "to bring final solution to 1500 ml", which is a make-up volume. The platinum sheet prints that temperature as 120 degrees C and the palladium sheet, whose paragraph is otherwise word for word the same, prints 120 degrees F; the Fahrenheit reading is taken here and 120 F is 48.9 C. Nothing in the formula depends on it — the water is warm only to dissolve the acid faster.

Bath F: Photographers' Formulary dilute citric acid for palladium, 1.5 per cent w/v — the same bath at three quarters of the strength, for the same reason bath B is weaker than bath A

Ingredient	Quantity	Form the source specifies
Citric acid (anhydrous)	30 g	The same 30 g, taken to 2000 mL instead of 1500 — a nominal 1.5 per cent w/v. The sheet says why in its own words: "this dilute citric acid is much weaker than that used to clear platinum prints. Palladium metal can be etched from a print by a more concentrated solution of acid"
Water	to make 2000 mL	at 49 °C; 1000 mL at 120 degrees F to dissolve, then 1000 mL of cold water to bring the bath to 2000 mL.

Used in this order — clearing a black platinotype in dilute hydrochloric acid, the trade's own procedure

1. Bath A: dilute hydrochloric acid for black platinotypes, 1 in 60 — about 10 minutes — Ware gives about ten minutes in each of three baths in one chapter and five to ten in another. Pizzighelli and Hübl give about ten minutes a change and name the end point rather than the time.
2. Bath A: dilute hydrochloric acid for black platinotypes, 1 in 60 — about 10 minutes
3. Bath A: dilute hydrochloric acid for black platinotypes, 1 in 60 — about 10 minutes — The third bath is the one that has to come out clean. Pizzighelli and Hübl's rule is to go on changing until the solution no longer turns yellow.
4. Water — about 20 to 30 minutes — Ware gives 20 to 30 minutes and, in the chapter on staining, notes that washing was sometimes carried "for as much as two hours". Wall's 1912 instruction is shorter — three or four changes for ten or fifteen minutes — and Pizzighelli and Hübl test the last wash water with litmus for neutrality.

Clear the developed print for about ten minutes in each of three successive baths of hydrochloric acid, the concentrated acid diluted 1 in 60, then wash in water for about twenty to thirty minutes.

Wall's 1924 printing gives four baths rather than three, at the same 1:60. Wall's 1912 printing gives "not less than three" and no times at all. Irving Penn, coating far more heavily than the trade, used the same three baths at 1:64 and needed six to ten hours in them.

Used in this order — clearing a palladiotype in dilute hydrochloric acid, the American practice

1. Bath B: dilute hydrochloric acid for palladiotypes, 1 in 200 — No time is published at this strength. Ware records the strength, the reason for it and the people who confirmed it, and nothing else; the count of baths and the times were carried over from the platinotype procedure by workers who were, on Anderson's evidence, already using the wrong developer.
2. Water — The wash, as for a platinotype.

Palladiotypes must be cleared in 1:200 hydrochloric acid, not the 1:60 acid used for platinum, which otherwise dissolves palladium and bleaches the image partially.

This is the tradition Ware describes rather than the one Willis published. Its two departures from the manufacturer's instructions — an oxalate developer instead of citrate, and hydrochloric acid instead of acidified citrate — partly cancel, because the acid etches away the palladium fog the oxalate produced. That is an explanation of why nobody noticed, not a recommendation.

Used in this order — clearing a palladiotype in acidified citrate, Willis's own procedure

1. Bath C: Willis's acidified citrate clearing bath for palladiotypes, at working strength — 10 minutes
2. Bath C: Willis's acidified citrate clearing bath for palladiotypes, at working strength — 15 minutes
3. Bath C: Willis's acidified citrate clearing bath for palladiotypes, at working strength — 20 minutes — The times lengthen down the line because the last bath is doing the hardest and slowest part of the job on the least iron.
4. Water — 30 minutes — Wall's 1924 printing gives 10 to 15 minutes running, or several changes of 10 minutes each.

Three sequential baths of the eightfold-diluted citrate clearing solution, of ten, fifteen and twenty minutes, followed by a final wash in running water for thirty minutes.

Ware gives "at least 10 minutes each" in section 6.16 and the 10, 15 and 20 minute sequence in section 2.9; Wall's 1924 printing gives the same three times independently. More dilute solutions, about four times rather than eight, were recommended for the Sepia Vellum paper.

Used in this order — clearing a platinotype in citric acid, Wall's published alternative to the mineral acid

1. Bath D: Wall's citric acid clearing bath for platinotypes, 1 oz to 20 oz — Wall gives the bath and the series and no times. The series is his own — "a series of baths (not less than three)" — and applies to either acid.
2. Bath D: Wall's citric acid clearing bath for platinotypes, 1 oz to 20 oz
3. Bath D: Wall's citric acid clearing bath for platinotypes, 1 oz to 20 oz
4. Water — three or four changes for ten or fifteen minutes

Citric acid, in the proportion of 1 oz. to 20 ozs. of water, may be used in the same series of not less than three baths. This softens the paper in less degree than does the hydrochloric acid.

The only period source read for this page that offers citric acid as a straight substitution for the hydrochloric acid in the platinotype line, and the only one that gives a reason. Wall states no times for the acid baths in either chemistry.

Used in this order — clearing a platinum print in the Photographers' Formulary kit

1. Bath E: Photographers' Formulary dilute citric acid for platinum, 2 per cent w/v — 5 minutes — With intermittent agitation, and never left to float unattended, because a trapped air bubble clears nothing under it.
2. Bath E: Photographers' Formulary dilute citric acid for platinum, 2 per cent w/v — 5 minutes
3. Bath E: Photographers' Formulary dilute citric acid for platinum, 2 per cent w/v — 5 minutes — "The third tray should always contain a clear acid solution." When the first goes cloudy and yellow it is discarded, the second and third move up, and a fresh bath is mixed for the third position.
4. Water — one hour, with a complete exchange of water every five minutes — Water no colder than 68 degrees F, and the sheet recommends distilled water. That is twelve changes, or a flow that turns the bath over every five minutes.

Transfer the drained print to the first of three trays of dilute citric acid; five minutes in each with intermittent agitation, draining between trays; then wash for an hour with a complete exchange of water every five minutes.

The sheet calls the step "etching", which is a candid name for it: what the bath removes is the residual ferric oxalate, and what it will remove if pushed is the image. Its own warning is that the step "is a critically important part of the process and is usually passed over too casually", and that a print not properly etched "will darken with age and in this way can be destroyed".

Used in this order — clearing a palladium print in the Photographers' Formulary kit

1. Bath F: Photographers' Formulary dilute citric acid for palladium, 1.5 per cent w/v — 5 minutes
2. Bath F: Photographers' Formulary dilute citric acid for palladium, 1.5 per cent w/v — 5 minutes
3. Bath F: Photographers' Formulary dilute citric acid for palladium, 1.5 per cent w/v — 5 minutes
4. Water — one hour, with a complete exchange of water every five minutes

The identical three-tray procedure at five minutes a tray, with the same rotation, in the weaker bath. This dilute citric acid is much weaker than that used to clear platinum prints; palladium metal can be etched from a print by a more concentrated solution of acid.

Same procedure, same times, different bath. The two sheets are the clearest modern statement that a clearing bath is chosen for the metal and not for the iron.

Purpose

To dissolve the unreduced iron out of a developed platinum or palladium print, while the print is still wet, using the least acid that will do it.

Ware’s inventory of what is actually in the sensitiser layer when the print leaves the developer is worth having in front of you, because it explains why one tray is not enough. There is the nanoparticle noble metal that is the image; there are aquated iron(III) oxalato-complexes; there is free oxalate; there are ammonium or potassium cations; there is unreacted tetrachloroplatinate or tetrachloropalladate; and there are chloride ions. Most of that is soluble and leaves in the wash. The iron(III) does not, for two reasons that this page turns on.

The first is that some of it is chemically bound to the paper. Ware’s account is that iron(III) coordinates to the vicinal hydroxyl groups of cellulose — plausibly as a cis-bidentate complex across one glucose unit, possibly facially terdentate across two — strongly enough to be called chemisorbed rather than merely wetted. He notes the same binding has been developed as a way of permanently dyeing cellulose, and that the calico-printers of the 1850s used it deliberately as a mordant. Water does not lift it.

The second is that iron(III) does not sit still. Above about pH 4 it hydrolyses, and Ware sets out the four stages with their timescales: rapid reversible mono- and dimers; then, in minutes, a “red cationic polymer” of 2 to 4 nm particles of about a hundred iron atoms each; then, over days to weeks, 20 to 50 nm polymers that agglomerate irreversibly into rods and rafts; and finally, over years, crystallisation to goethite and haematite.

[Fe(H₂O)₆]³⁺ → Fe(OH)₃ + 3 H₂O + 3 H⁺ → FeO(OH) + 4 H₂O + 3 H⁺

Hydrolysis, and the irreversible end of it

Freshly formed iron(III) hydroxide redissolves in dilute acid. Goethite does not — Ware calls it “quite insoluble in dilute acids”. So the clearing bath has a deadline as well as a job.

!Important

Every one of these baths works on iron that is still reachable. Once the print has dried with hydrolysed iron in it, the chemistry has moved on, and no ordinary bath will touch it. Clear completely, in the wet, the first time.

Recommended uses

A developed-out platinotype or palladiotype, of any period and any recipe: the paper coated with ferric oxalate and a noble-metal salt, exposed under ultraviolet, developed in potassium oxalate or in a citrate developer, and now carrying iron it must lose. This is the clearing step of the platinotype and palladiotype processes as they were actually practised.

Choose the bath from the metal, not from the tradition. That is the single most useful sentence on the page, and both the historical and the modern evidence say it:

• A pure platinum print tolerates the strongest bath here. Bath A, the 1 in 60 hydrochloric acid, is what most surviving historic platinotypes went through.
• A palladium or platinum-palladium print does not. Anderson’s rule, as Ware reports it and as McCabe and Gottlieb confirmed in 1993, is 1 in 200 — bath B — and Ware’s flat statement is that palladiotypes “cannot be treated so brutally without incurring intolerable image loss”. Willis himself never recommended hydrochloric acid for palladium at all; he published bath C.
• If you are working from a modern kit, the bath is already chosen for you and the two sheets differ from each other on purpose: 2 per cent citric acid for platinum, 1.5 per cent for palladium.

Working from a kit you bought. Photographers’ Formulary supply the citric acid and the instruction; their product entry and Bostick and Sullivan’s are separate pages (Photographers’ Formulary, Bostick and Sullivan). Bostick and Sullivan ship EDTA and sodium bisulfite instead, and that bath is discussed under Variants.

Reconstructing historical practice. A conservator or a historian who needs to know what a print in front of them went through will find the whole spread here, from Pizzighelli and Hübl in 1886 to the kit sheet of last year, with the disagreements left in.

When another formula is preferable

The modern EDTA and sulphite sequence, for any print you are making today. This is not a close call, and the evidence is not the course’s. Ware publishes a three-bath sequence — acidic disodium EDTA, then a sulphite reduction, then alkaline tetrasodium EDTA — for platinum-palladium prints, and it has its own entry: the EDTA and sodium sulfite clearing sequence. Matthew Clarke and Dana Hemmenway compared clearing procedures by X-ray fluorescence at the National Gallery of Art and found that sequence left less residual iron than any other combination they tested — comparable with, or less than, the iron already present in the uncoated paper — and accelerated ageing of the cleared papers produced no perceptible yellow stain. Ware’s conclusion is unambiguous: the iron stain “can be avoided entirely by printmakers today, even in palladium printing, by using — instead of hydrochloric acid — for the clearing baths, a reducing agent in concert with a modern chelating agent such as EDTA”.

So why does this page exist? Three reasons, and none of them is nostalgia.

1. It is what the historical prints went through, and no one can read a stained platinotype without knowing what was and was not done to it.
2. It is what one of the two surviving kit suppliers still ships. Photographers’ Formulary’s platinum and palladium kits contain 30 g of citric acid and instructions for three trays of it. A reader who bought one of those kits owns this formula whether or not they own any EDTA.
3. It teaches the mechanism from the other end. Acid clearing works by suppressing hydrolysis and dissolving what has hydrolysed. Chelation works by wrapping the iron in a ligand. Comparing them is the fastest route to understanding either.

A citrate bath in preference to a mineral acid, on any palladium print. Willis’s reason was not about the iron. His distributor’s notice put it plainly: the palladiotype developer and clearing baths “have no tendency to injure the beautiful surface of the paper employed for coating”, where undue immersion of a matt platinotype in the oxalate and acid baths does destroy its bloom. Ware’s reading is that a chemist as painstaking as Willis would not have changed his clearing agent if the old one had served.

Nothing on this page, if the print was made by the Malde-Ware print-out method. In that process the first EDTA bath is the developer as well as the clearing bath, and substituting an acid tray would change the process rather than the finish. See Ware’s print-out platino-palladiotype.

Mixing

Four of these six baths are made by dissolving a weighed solid in water, and there is little to say about them beyond doing it. The other two are made by diluting a corrosive concentrate, and there is a great deal to say about that.

The acid baths, and the one rule that matters

⚠Acid into water, never water into acid

Concentrated hydrochloric acid is classified Danger, with H314 — causes severe skin burns and eye damage — in 99.9 per cent of the 5,109 company reports aggregated by ECHA that carry hazard codes, and H331, toxic if inhaled, in 49.4 per cent of them. Princeton’s guidance for photographic work states the rule for cleaning acids without qualification: add acid to water, slowly, with stirring, and never the reverse. Dilution generates heat, and water poured onto concentrated acid can boil and throw it back.

Open the concentrate outdoors or under extraction, in chemical splash goggles, gloves and an apron, with running water within reach. The dilution is the step that needs the ventilation; the tray of finished 1.7 per cent bath is a far smaller problem than the bottle it came from.

The sources state a ratio, not a tray. Wall’s 1912 line is “one part of hydrochloric acid with 60 parts of water”; Ware’s is “the concentrated acid (36 % w/w) was diluted 1:60”. Neither says how big the dish is, so the formula above records the ratio at its own scale and the scaling is done here, where it can be shown and labelled.

🧮Filling a real tray

For 1.5 litres of bath A — enough for a 10 by 8 inch dish — take Wall’s proportion of 1 part in 60 parts of water. Sixty-one parts make 1500 mL, so one part is 1500 ÷ 61 = 24.6 mL of concentrated acid into about 1475 mL of water. Read Ware’s way instead, as a dilution to sixty volumes, and it is 1500 ÷ 60 = 25.0 mL. The two readings differ by 1.6 per cent, which is less than the difference between one bottle of “concentrated” acid and another.

For 1.5 litres of bath B, the palladiotype strength, it is 1500 ÷ 200 = 7.5 mL.

This arithmetic is the course’s, not the sources’. What the sources publish is the ratio.

Wall adds two specifications that are easy to skip and should not be. The acid’s specific gravity should be not less than 1.16, and if it is lower, more of it should be used; and it should be colourless, with commercial or muriatic acid ruled out “on no account”. Both are about impurity. Yellow-brown technical acid carries iron, which is the last thing to add to a bath whose entire job is removing iron, and a weak concentrate silently gives a weak bath.

The citrate bath

Willis’s bath is published as a stock at 20 per cent w/v trisodium citrate dihydrate with 9 per cent w/v citric acid, diluted eight times for use. Solution C above is the diluted bath, at Ware’s own published working strengths, because that is what the print is put into. To make the stock instead: 200 g of trisodium citrate dihydrate and 90 g of citric acid, water to make 1000 mL, then 1 part of that stock to 7 parts of water at the tray. Dissolve the citrate first; it is the bulkier solid and the more soluble.

The kit baths

Photographers’ Formulary’s instruction is to dissolve the 30 g of citric acid in a litre of warm water and then bring the bath to its final volume with cold — 1500 mL for platinum, 2000 mL for palladium. The warm water is a convenience and nothing more; citric acid dissolves readily cold, given a minute.

✕A misprint worth knowing about

The platinum kit sheet says “place 1000 ml of water at 120 ° C”. Water at 120 °C is steam under pressure. The palladium sheet’s paragraph is word for word the same and prints 120 °F, which is 48.9 °C — comfortably hot tap water. The Fahrenheit reading is the one this page takes, and the reason is that the two sheets are otherwise identical documents.

Water

Use water without lime in it. Wall’s 1924 instruction for the Satista clearing bath is blunt — “the water used for the clearing bath must be free from lime” — and Bostick and Sullivan say in modern words that “water hardness will sometimes affect clearing time”. The chemistry is in the mechanism section below: calcium precipitates oxalate and pushes the iron complex apart.

⚗Two opposite verdicts on oxalic acid, and both are right

Wall’s remedy for hard water is to dissolve 0.5 g of oxalic acid in a litre, let it stand for some hours and decant from the precipitate. Ware, writing about a different job, says that using oxalic acid to strip a chalk buffer out of a paper is “obviously chemically inappropriate because calcium oxalate is as insoluble as the calcium carbonate it is intended to remove”, and recommends 1 to 2 per cent v/v hydrochloric acid, or 5 to 10 per cent w/v sulphamic acid, instead.

Both statements are correct and the course states the reconciliation as its own reading rather than as either author’s: in a jug you can throw the calcium oxalate down and pour the clear liquid off it; in a sheet of paper you cannot, and the precipitate stays exactly where the carbonate was.

Behaviour

The first tray goes yellow, then cloudy. Photographers’ Formulary describe it exactly: the acid in the first tray “will soon become cloudy and yellowed in appearance … due to any remaining Ferric Oxalate being etched from the paper”. That colour is the point of the whole exercise. It is iron leaving.

The trays rotate. When the first bath is loaded, discard it, move the second and third up a place, and mix a fresh bath for the third position. The rule the sheet gives is the one to remember: the third tray should always contain a clear acid solution. The last bath a print sees is the one that decides how much iron is left in it, and a print finishing in an exhausted tray has not been cleared, it has been rinsed in dilute iron.

The end point is a colour, not a clock. Pizzighelli and Hübl, writing in 1886, give the rule that outlives every schedule on this page: change the acid “until it no longer turns yellow”. Their own practice was three changes of about ten minutes. Their fault table repeats it from the other end — whites with a yellowish tinge after drying, second cause, “insufficient immersion in hydrochloric acid”, remedy, change the bath two or three times “until the last change no longer turns yellow at the end of ten minutes”.

The bath tells you when it is too weak. Wall’s test for the hydrochloric acid bath is a good one: “a white opalescence of the bath shows necessity for more acid”. An opalescent bath is one in which hydrolysis is winning — colloidal iron(III) hydroxide scattering light — and the answer is more acid, not more time.

The print lightens. All of it lightens while wet, and a platinum print recovers on drying. A palladium print in too strong a bath does not fully recover, because some of what left was palladium.

Times. Ten minutes a bath is the historical figure for the acid; five minutes a tray is the modern kit figure for citric acid; Willis’s citrate baths run 10, 15 and 20 minutes, lengthening down the line because the last bath works on the least and most stubborn iron. Penn, whose precious-metal coating weight Ware puts at about twice Willis’s, needed six to ten hours, and the sources are candid that he paid for them.

📓What to write down

The metal and its proportion; which bath and at what strength; the time in each tray and which tray was freshest; the colour of the first bath at the end; whether the masked border was clear against the uncoated margin, and under what light you judged it; the wash time and, if you have it, the pH of the last wash water. Two of those — the border and the bath colour — are the only evidence you will ever have that the clearing worked, and neither survives the print drying.

Image characteristics

Platinum is essentially untouched. Ware states it plainly: there is no convenient reagent for dissolving platinum images. The 1 in 60 acid removes iron from a platinotype and leaves the picture.

Palladium is not. “Palladium is slightly attacked by dilute hydrochloric acid in air,” and that single qualifier — in air — is the whole mechanism, set out below. On a palladiotype the acid bath etches the high values first, which reads as cleaner highlights and, on a fogged print, as an improvement. Ware’s account of American practice is that two mistakes cancelled: workers used a potassium oxalate developer that fogged the highlights with palladium metal, then cleared in an acid strong enough to etch that fog away. The prints looked well and the discrepancy went unexamined for decades.

Prolonged clearing costs density and costs paper. Penn’s six-to-ten-hour baths caused density loss in his palladium-containing images and “the disintegration of many of the fine-art papers that he tested” — which is why he ended up on BFK Rives and Arches Aquarelle.

Surface matters too. Willis’s objection to the oxalate and acid line for palladiotype was that undue immersion “does tend to destroy their natural bloom” on the matt papers. Wall’s reason for offering citric acid as a substitute is the same one from the other direction: it “softens the paper in less degree than does the hydrochloric acid”.

And what a clearing bath cannot do is change the tone or the contrast. Nothing here is a toner and nothing here is a reducer in the photographic sense. If the print looks contrastier after clearing, what has happened is that something was removed from the highlights.

The mechanism

What acid actually does to iron

The acid is not dissolving iron the way it would dissolve a carbonate. It is doing two things at once.

It suppresses hydrolysis. Iron(III) in water is a strong enough Lewis acid to pull protons off its own coordinated water molecules, and above about pH 4 the products polymerise. Hold the pH near or below 1 and that equilibrium is pushed hard the other way: the iron stays as the aquated cation and as soluble chloro- and oxalato-complexes, and never starts down the road to goethite.

It redissolves what has already hydrolysed. Freshly formed iron(III) hydroxide is soluble in dilute acid. This is why the deadline exists, and why the bath belongs immediately after the developer rather than after the wash.

Fe(OH)₃ + 3 H⁺ + 3 H₂O → [Fe(H₂O)₆]³⁺

What the acid undoes, while it still can

There is a third party, and it comes out of the tap. Calcium — from a hard supply, or from a chalk buffer in a modern “acid-free” art paper — precipitates calcium oxalate, which strips oxalate off the iron and starts the hydrolysis it was preventing:

Ca²⁺ + [Fe(C₂O₄)₃]³⁻ + 2 H₂O → CaC₂O₄ + [Fe(C₂O₄)₂(H₂O)₂]⁻

Ware: calcium drives the iron complex apart

Ware adds that the hydroxide liberated in the paper by that reaction promotes the hydrolysis further, which is why a chalk-buffered paper is the wrong support for an iron sensitiser and why practitioners pre-treat it with acid before they ever coat.

Why three baths and not one

Because each bath removes a share of what is in front of it, and no bath removes all of it. A print carrying its residue into a fresh tray leaves most of it there and carries the rest into the next; the third tray is working on a small fraction of a small fraction. One tray held for thirty minutes cannot do the same job, because it reaches an equilibrium with the paper and stops. This is the same reasoning that puts three baths in a fixing line and three changes in a wash, and it is the reason the rotation rule exists: the freshest bath must be the last one.

The number is not a ritual. Wall’s 1912 instruction is “not less than three”; his 1924 book gives four; Pizzighelli and Hübl give “twice or three times” and then tell you the real rule, which is to keep going until the bath stops going yellow.

Why hydrochloric acid takes palladium and leaves platinum

This is the most useful piece of chemistry on the page, and it is a redox argument with a complexation twist.

Ware’s Table 11.1 gives the standard potentials of the noble metals as their tetrachloro complexes:

E(PtCl₄²⁻ / Pt, 4Cl⁻) = +0.73 V

Platinum

E(PdCl₄²⁻ / Pd, 4Cl⁻) = +0.62 V

Palladium

Palladium is 0.11 V easier to oxidise than platinum. On its own that is a modest difference; what makes it decisive is what is available to do the oxidising. Neither acid nor chloride can oxidise a noble metal. Dissolved oxygen can. Ware’s Argyronomicon gives the oxygen couple as pH-dependent:

E(O₂, 4H⁺ / 2H₂O) = +1.229 − 0.059 pH

Oxygen, in acid

A 0.2 mol/L solution of a strong acid sits at about pH 0.7, which puts the oxygen couple at roughly +1.19 V — comfortably above both metals. That arithmetic is the course’s; the expression is Ware’s.

So why is platinum safe? Thermodynamics permits the oxidation of both, and what separates them is speed. Ware’s own account of the two metals turns on that contrast: platinum is a third-row metal whose complexes react slowly, and the whole difficulty of platinum printing, he writes, arises from “the relative slowness of the chemical reactions of platinum complexes”, where palladium “affords much speedier reactions”. That the same contrast is what protects a platinum image in an acid bath is the course’s reading of his chemistry, not a statement he makes, and it should be read as an explanation offered rather than a result established.

The chloride is not a spectator either. It is what lets the reaction proceed at all, because it carries the metal away as a stable anionic complex and so removes the product from the equilibrium:

2 Pd + 8 Cl⁻ + O₂ + 4 H⁺ → 2 [PdCl₄]²⁻ + 2 H₂O

Why the qualifier is 'in air': oxygen oxidises, chloride carries away

That equation is the course’s, written out from two statements Ware makes separately — that palladium is attacked by dilute hydrochloric acid in air, and that the couple governing palladium in chloride is the tetrachloropalladate one. He does not print it.

Three consequences follow, and all three are in the historical record. More acid etches more, which is Anderson’s 1:200 and the kit sheets’ 1.5 per cent. A citrate bath etches less, because it holds almost no free chloride and is nowhere near as acidic. And the highlights go first, because that is where the least metal is.

What acid cannot do at all

Here is the limit, and it is why the modern sequence exists. Chelating agents alone strip only about 50 per cent of the iron(III) off cellulose — Ware cites measurements on cellulose pulp using ligands considerably more powerful than EDTA — and a mineral acid, which has no chelating power to speak of, does no better on the chemisorbed fraction. Henry Chapman Jones demonstrated this in 1895: after four hours in dilute hydrochloric acid, unexposed platinotype paper still held iron detectable by discoloration under hydrogen sulphide. The Getty Conservation Institute’s atlas confirms it with modern instruments — “even after applying the best fixing and clearing procedures, enough iron is left in the platinotype photograph to allow for its detection using XRF”.

What the residue then does is slow and, on Ware’s proposal, structural. A single iron(III) centre with six oxygen donors is a weak absorber: its charge-transfer band is in the ultraviolet and its d-d transitions are spin-forbidden. But two such centres on adjacent cellulose chains can lose a water molecule and bridge:

Cell-Fe-OH + HO-Fe-Cell → Cell-Fe-O-Fe-Cell + H₂O

Ware's proposal for the chromophore of the slow yellow stain

The bridged pair absorbs strongly in the visible through metal-to-metal charge transfer, and that is the brown. It is a proposal, not a measurement, and Ware presents it as one. What is not in doubt is the observation it explains: platinotypes that were clear when made turn yellow over months or years.

Deeper: what the citrate bath adds that the mineral acid has not got

Citric acid is triprotic, with pK values close together — 2.87, 4.35 and 5.68 in the IUPAC compilation — so a citric solution buffers broadly instead of sharply. Strip the protons and what is left is a chelator: several oxygen donors on one small molecule, closing a ring around a metal ion.

That gives the citrate bath two mechanisms where the hydrochloric acid bath has one. It holds the pH low enough to suppress hydrolysis, as the mineral acid does, and it complexes the iron and carries it off, which the mineral acid does not. It also brings almost no free chloride to the tray, which is half of the reason it leaves a palladium image alone.

There is a related fact of Ware’s worth having beside it, although it is about sensitisers rather than clearing baths. He gives the citrato-iron couple as E(Fe(III)Cit/Fe(II)Cit) = +0.372 V against the oxalato couple’s +0.02 V, and draws the consequence: iron(II) citrate “does not reduce platinum(II) or palladium(II), although it will reduce gold(III) and silver(I), under the printing conditions”. That is why citrate is the iron salt of Van Dyke and argyrotype and oxalate is the iron salt of platinotype — and it is also a small reassurance about a citrate clearing bath loaded with iron(II), which has no power to plate anything back out.

Wall’s terse “this softens the paper in less degree” is the practical consequence of the acidity rather than the chelation. The course’s reasoning, not Wall’s: acid-catalysed hydrolysis of cellulose goes roughly with the hydrogen ion concentration, so a bath at pH 2 should attack the sheet something like twenty times more slowly than one at pH 0.7. Wall states the observation and offers no explanation, and neither bath’s pH has been measured for this page.

⁇How many baths, how long, and at what strength

Every figure on this page has at least two printings and some of them disagree.

Three baths or four. Wall 1912 says “not less than three”; Wall 1924 says four; Ware and Pizzighelli and Hübl say three. Nothing turns on it, because all four sources agree the real criterion is the bath ceasing to discolour.

Ten minutes or five to ten. Ware gives about ten minutes in section 2.9 and five to ten in section 9.2, both citing the trade practice.

Willis’s citrate stock. Ware converts Willis and Clements to 20 per cent w/v citrate with 9 per cent citric acid; Wall’s 1924 book prints 232.5 g and 93 g to the litre, a sixth stronger in citrate. Both are printings of the same manufacturer’s instructions. The course records Ware’s, because he states his conversion and his source, and names Wall’s here rather than averaging them.

The wash. Ware gives 20 to 30 minutes and elsewhere “as much as two hours”; Wall 1912 says Platinotype papers “do not require a long after-washing”, three or four changes over ten or fifteen minutes; Pizzighelli and Hübl give ten to fifteen minutes and a litmus test. The modern kits ask for an hour. The course takes no side and notes only that the 1912 and 1886 advice was written when the acid was the last thing in the paper and nobody was measuring residual iron.

Function of every ingredient

Three substances fill six trays, and the interesting thing about them is how differently the same job is done.

Hydrochloric acid — baths A and B

What it is. An aqueous solution of hydrogen chloride gas. The concentrate of this tradition is 36 to 37 per cent w/w, about 12 mol/L; Wall specifies a specific gravity of not less than 1.16 and a colourless liquid, ruling out the commercial and muriatic grades because they carry impurities — including iron.

Why it is there. It is the cheapest way to hold a tray at the pH below 1 that Ware records for this bath. A strong acid is completely ionised in water, so there is no reserve of undissociated molecules and no buffering; that is exactly wrong for a stop bath and exactly right here, where the object is to hold the pH below the hydrolysis threshold of iron(III) and to redissolve any hydroxide that has already formed.

What it does chemically. Two jobs, described above: suppression of hydrolysis, and dissolution of hydrolysed iron. A third, less often stated, is that its chloride ion forms soluble chloro-complexes with iron(III), which helps carry the metal out of the fibres.

Photographic consequence. A cleared white in the masked border, and an image whose highlights stay where they were put.

More of it. More acid clears faster and further — and etches. The whole spread of published strengths, from Anderson’s 1 in 300 to the trade’s 1 in 60, is a series of positions on that trade-off. The palladiotype figure of 1 in 200 is where the trade-off was struck once palladium was in the paper.

Less of it. Anderson’s very dilute bath is the worked example: two minutes in a 1 in 300 bath left “much residual iron in the paper”, and Ware’s judgement is that this is precisely why so many mercury-toned sepia platinotypes have gone yellow.

What it interacts with. The image metal, as above. The paper: dilute mineral acid hydrolyses cellulose, slowly, and Penn’s disintegrating sheets are the extreme case. And anything alkaline in the paper — a chalk buffer will consume acid and gas gently while doing it.

Citric acid — baths C, D, E and F

What it is. A weak triprotic organic acid, pK values 2.87, 4.35 and 5.68. PubChem’s ChEBI description names its other identity outright: a chelator.

Why it is there. Because it does the same two jobs with one molecule and does them gently. It acidifies — enough to keep iron(III) from hydrolysing, not enough to attack palladium at any useful rate — and its conjugate base chelates the iron and takes it into solution. Wall’s reason for offering it as the alternative in 1912 was the paper; the kit suppliers’ reason for standardising on it a century later is the image.

What it does chemically. Below pH 3 the free acid predominates and the bath behaves mostly as an acid; as the bath loads with iron, citrate that has lost protons takes it up as a soluble citrato-complex. Ware’s potential for that complex, +0.372 V for the iron(III)/iron(II) couple, is high enough that the bath has no reducing power over the platinum metals at all — which is exactly what you want in a bath that is going to sit in contact with the image for fifteen minutes.

Photographic consequence. Cleared highlights with less image loss than the mineral acid gives, and a paper surface that survives it.

More of it. Faster clearing and, past a point, an etched image: Photographers’ Formulary say it in one sentence — “palladium metal can be etched from a print by a more concentrated solution of acid” — and that is why their palladium bath is 1.5 per cent where their platinum bath is 2 per cent. Wall’s 5 per cent bath is for platinum only, and it is not a strength to put a palladium print into on the strength of this page.

Less of it. Slower clearing, an earlier-exhausted first tray, and — the failure that matters — a print that looks clear wet and yellows later. Bostick and Sullivan’s remedy for yellow highlights is exactly this: “increase the concentration of the clearing agent or increase the time in the bath”.

Trisodium citrate dihydrate — bath C

What it is. The fully neutralised sodium salt of the same acid, FW 294.10, freely soluble, and not classified as hazardous in the ECHA aggregation.

Why it is there. To supply a great deal of citrate at a pH the free acid alone could not reach with the same amount of chelator. Willis’s bath is 25 g/L of the salt against 11 g/L of the free acid, which is about three parts of citrate to two of acid on a molar basis: a citrate bath with enough acid in it to stay on the right side of the hydrolysis line. That is a different design from the kit baths, which are the free acid alone.

What it does chemically. The citrate anion is the ligand; the sodium is a spectator. The part-neutralised mixture also buffers, so the bath’s pH moves less as it loads with iron and with carry-over from the developer than a bath of free acid would.

Photographic consequence. The gentlest of the six baths on both the image and the paper surface, and the one Willis chose when he had a palladium image to protect.

More citrate, less acid. More chelating power and a higher pH; push the pH above 4 and the iron begins to hydrolyse in the tray, which is the failure Ware warns about for alkaline EDTA. Less citrate, more acid and you have converged on the kit bath.

What it interacts with. Calcium, competitively — citrate is a calcium chelator as well as an iron one, which is a small mercy in hard water. And the developer it follows: Willis’s clearing stock is his own developer with four times the citric acid, and the bath is that stock diluted eightfold, so carry-over between the two trays changes nothing but strength.

Water

Not an inert filler. Its hardness decides whether calcium oxalate precipitates in the paper, and Wall’s rule for the clearing bath is that the water be free from lime. Its temperature appears in the modern sheets only as a convenience for dissolving the acid, and in the wash as a floor: no colder than 68 °F. And the wash water at the end is doing the last part of the clearing job — Pizzighelli and Hübl’s reason for washing is that acid left in the paper “would not have any bad effect on the print itself, but might injure the substance of the paper, so that in the course of time it would be destroyed”.

The EDTA that is in this page’s title and not in its table

Both surviving kit suppliers ship an EDTA clearing agent with sodium bisulfite, and both publish the quantity as “2 tablespoons of each to 1 quart of water”, adding that “this is not a critical measurement”. A tablespoon of a powder is a volume, not a weight, and converting one to the other needs a bulk density that neither supplier publishes. Under Rule 1 the bath therefore stays out of the formula object and is set out under Variants, with its chemistry, its published times and its capacity. Ware’s fully quantified version, at 5 per cent w/v disodium EDTA and 2.5 per cent w/v sulphite, is a separate entry.

Interactions

With the developer. Everything that comes out of the developer tray on the print goes into the first clearing bath, and it is not a small amount: a sheet carrying 30 per cent potassium oxalate arrives loaded with oxalate, which the acid promptly protonates to oxalic acid. That is the first bath’s real work in the first minute, and it is one more reason the first bath is the one that dies.

With the paper. Two ways. A chalk-buffered sheet consumes acid and produces calcium, which precipitates oxalate and drives hydrolysis; and the acid itself hydrolyses cellulose, slowly, which is Wall’s reason for offering citric acid and Ware’s for warning about palladium prints and long baths. Ware’s fix for the buffer is to destroy it before coating — 1 to 2 per cent v/v hydrochloric acid, or 5 to 10 per cent w/v sulphamic acid for 10 to 20 minutes, followed by a long wash — and never oxalic acid, for the reason given above.

With the wash water. Alkaline supply water is a problem in every iron process: Photographers’ Formulary’s New Kallitype sheet notes that rinsing in water of neutral or slightly acidic pH extends the life of the clearing baths, while water with a basic pH will make the print difficult to clear. The mechanism is the same one on this page — pH above 4, iron(III) hydrolyses, and hydrolysed iron in the fibres is a different and harder problem.

With the image metal. Covered under the mechanism. In one sentence: the acid and the chloride are between them capable of dissolving palladium in the presence of air, and are not capable of dissolving platinum on any useful timescale.

With the sulphite bath, where there is one. Penn put a 5 per cent sodium bisulphite bath after his hydrochloric acid clearing, noting only that “the paper seems to whiten and generally clear”. Ware’s explanation, forty years later, is that it reduced residual iron(III) to iron(II), which binds cellulose more weakly and is more easily removed. If you are going to run only one extra tray after an acid line, that is the tray, and its logic is the mechanism of the whole modern sequence in miniature.

HSO₃⁻ + 2 Fe³⁺ + H₂O → HSO₄⁻ + 2 Fe²⁺ + 2 H⁺

Penn's bath, explained: hydrogen sulphite reduces iron(III)

With gold or platinum toning of other processes. Not on this page, but worth flagging: the same acid clearing tradition passed into kallitype practice, where the image is silver and everything about the trade-off changes. See the kallitype sensitiser and the iron-silver clearing entry.

Variants

Bath A at half strength, for sepia platinotype papers. Both Ware and Wall’s 1924 book give it as “half the above strength”. Wall’s 1912 rule belongs with it: black and sepia prints should never be cleared and washed together in the same dish, because a dish that has held the sepia chemistry destroys the purity of the blacks.

Anderson’s 1 in 300 bath, two minutes, one tray, for mercury-developed sepia platinotypes. It is in dilutions above because it is historically important and it is a bad idea: Anderson used it to keep the warmth that a normal clearing bath removes, and the price was iron left in the paper. Ware’s verdict is that he was right about the first thing and that the second thing is why those prints are yellow. The mercuric chemistry is a Level D matter and has its own entry, the mercuric sepia platinotype.

Willis’s citrate stock at Wall’s strength. 232.5 g of trisodium citrate and 93 g of citric acid to the litre, diluted 1 part with 7 parts of water, against Ware’s 20 and 9 per cent at the same dilution. Same manufacturer, two printings, about a sixth apart in citrate. Neither is more authoritative than the other; the difference is smaller than the difference between two batches of paper.

Wall’s Satista clearing bath, which is the same idea with the acid taken out: sodium citrate alone, three baths of 5, 10 and 15 minutes, a short wash of not more than 10 minutes, and the lime-free water rule attached. Its strength is not reproduced here, because it cannot be read: the metric column in the course’s copy is garbled and the imperial column beside it, one and a quarter ounces to seventy-five fluid ounces, does not convert to any plausible reading of the digits that survive. Every other Wall 1924 figure used on this page has an imperial column and a metric column that agree with each other; this one does not. Recorded anyway, because it shows what the citrate does on its own, without the acid.

Bostick and Sullivan’s EDTA and sodium bisulfite bath. Two powders in one tray — an unnamed EDTA salt and sodium bisulfite — at 2 tablespoons of each to a quart, one or two baths of five minutes, capacity up to a dozen 8 by 10 prints in fresh solution, then discarded. Half an hour’s wash after it. It is a good idea implemented in one tray rather than three: the bisulfite reduces iron(III) to iron(II) and the EDTA chelates it, which is the modern mechanism, but doing both at one pH means neither is at its optimum.

⁇Which EDTA salt, and does it matter

Ware is direct: “It is important not to use tetrasodium EDTA for the first processing bath — in spite of the fact that it has been recommended by Bostick and Sullivan — it has a high pH ~10 at which the iron(III) is not effectively complexed, but rather hydrolysed, with the eventual result of yellow or brown staining.”

The kit sheets for platinum-palladium and Ziatype say “EDTA clearing agent” and do not name the salt; the same firm’s kallitype sheet names the tetrasodium salt. The disodium and tetrasodium salts are the same ligand about six units of pH apart, which is not a detail.

This is a genuine disagreement between a Tier 1 manufacturer sheet and a Tier 2 specialist, and the course does not settle it by authority. What can be said: Ware states a mechanism, makes a prediction about a stain appearing years later, and reports accelerated-ageing evidence consistent with it. What cannot be said is that anyone has published a head-to-head comparison on platinum-palladium prints.

One reading the course offers as its own, not as anyone’s finding: the tray Ware is criticising is a tetrasodium EDTA bath on its own, and the kit’s tray also contains a bisulfite. A reducing agent changes what the chelator has to grip, which is the whole argument of the modern three-bath sequence, so the two cases may not be the same case.

The modern EDTA and sulphite sequence, which is not a variant of this formula but its replacement, and has its own entry. Three trays: acidic disodium EDTA at 5 per cent, a sulphite reduction at 2.5 per cent, alkaline tetrasodium EDTA at 5 per cent, ten minutes each with half-minute rinses between, then half an hour’s wash.

Conservation treatments, which are not printing procedures. Ware records two, and this page names them without giving procedures because neither belongs in a home darkroom. The older is 4 per cent sodium disulphite followed by dilute hydrofluoric acid to remove iron(II) as the hexafluoro anion — chemically sound, and hydrofluoric acid is a substance that requires specialist handling and specific first aid. The modern one, developed at the V&A by Jacquelyn Rees and Megan Gent from Helen Burgess’s work, is sodium dithionite with tetrasodium EDTA at pH about 9; Ware’s own convenient version is one hour at room temperature in a solution 5 per cent w/v in each. These treat a stained print that already exists. They are not an alternative to clearing one properly.

Safety

Level B, on three of the rubric’s criteria at once: the handling of a concentrated acid; a procedure that generates mists or vapours, at the dilution step; and a failure mode that is a splash or a burn rather than a spoiled print.

What is not a hazard here, and why. The finished trays are not the problem. A 1.7 per cent hydrochloric acid bath is a dilute solution of a strong acid, and a 2 per cent citric acid bath is a dilute solution of a weak one; for both, the routine controls — nitrile gloves, safety glasses, tongs, an openable window — are adequate for standing over them. Citric acid itself is classified only Warning, with H319 (serious eye irritation) in 84.7 per cent of ECHA’s 4,373 company reports and H335 in 23 per cent, and 8.2 per cent of reports record no GHS hazard at all; trisodium citrate is not classified. Nothing on this page is heated, nothing is volatile at tray temperature, and nothing generates a gas in normal use. The hazard belongs to the bottle the acid is diluted from, and to that step alone. That is why the page is Level B rather than Level A, and why a printer who buys ready-made dilute acid, or works in citric acid, is running a Level A procedure with a Level B step removed.

Hydrochloric acid, the concentrate. Danger; H314, severe skin burns and eye damage, in 99.9 per cent of ECHA’s 5,109 reports carrying hazard codes; H331, toxic if inhaled, in 49.4 per cent; H335 in 59 per cent; H290, corrosive to metals, in 22.8 per cent. HSE’s EH40 sets a British workplace exposure limit for hydrogen chloride gas and aerosol mists of 1 ppm over eight hours and 5 ppm over fifteen minutes. Chemical Safety Card 0163 adds that inhalation may trigger asthma-like reactions and that high concentrations risk lung oedema.

⚠Handling

Dilution. Outdoors or under extraction. Chemical splash goggles, nitrile gloves, an apron, sleeves down. Acid into water, slowly, stirring. Running water within reach before you open the bottle.

At the tray. Gloves and eye protection, and Pizzighelli and Hübl’s 1886 advice still stands — handle the print with tongs and keep your hands out of the liquid. Ware’s own instruction for the modern sequence is the same: do not let the processing solutions reach the skin.

Powders. Ware’s note is worth repeating: dry powdery solids present a greater risk than solutions, because the dust can be inhaled, so keep it contained and wear a mask when weighing.

Never bring the concentrated acid near hypochlorite bleach (chlorine gas), near spent fixer (decomposes thiosulfate and throws out sulfur), or near metal of any kind (hydrogen).

Splashes. Ware’s advice for the platino-palladiotype line covers this page too: mop spillages promptly, and wash skin contact off immediately with plenty of cold running water. For the concentrate, eye contact is an emergency and needs fifteen minutes of irrigation and medical attention.

Storage

The concentrate in its original labelled bottle, upright, cool, out of sunlight, in secondary containment, away from metal shelving and away from anything alkaline or oxidising. Solubility of hydrogen chloride in water falls sharply as the temperature rises, which is why a bottle stored warm fumes more than one stored cool.

The dilute baths in labelled plastic — never a food container — with the strength and the date on the label, and never in metal. No source read for this page gives a keeping time for any of these baths, and there is no obvious decomposition to fear: a dilute acid solution does not go off. What limits it is what it accumulates, so the practical rule is the one the kit sheet gives, which is to judge a bath by its colour and cloudiness rather than by its age, and to rotate the trays so the last one is always fresh.

The citrate bath is the one to watch, and this is the course’s reasoning rather than a published keeping time: it is the only one of the six whose solute is an organic salt at a pH high enough to be hospitable, so a stored bottle can grow a haze in a warm room where a tray of mineral acid cannot. Make it in quantities you will use, and look at it before you pour it.

Label the acid baths for their metal. A tray marked “1:60” and a tray marked “1:200” look identical and are not interchangeable, and the consequence of confusing them is an etched palladium print.

Incompatibilities

Hydrochloric acid and hypochlorite bleach — chlorine gas. This is the one that hurts people in domestic settings.

Hydrochloric acid and spent fixer — a few drops of a strong acid decompose thiosulfate and throw out sulfur. Acid waste and fixer waste do not share a bottle, and they do not share a drain in the same minute.

Hydrochloric acid and metal — of any kind: copper, brass, zinc, aluminium, iron, tin. The product is hydrogen. That includes the metal parts of a print washer and the underside of a sink.

Any of these baths and an alkaline anything. Neutralising a clearing bath in the tray is not clearing, it is precipitating iron in the paper.

The clearing baths and calcium. Hard water, a chalk-buffered paper and a calcium-loaded wash are all the same problem, and the product is calcium oxalate, which is insoluble and stays where it forms.

Cross-contamination between prints. Wall’s 1912 rule about black and sepia platinotypes sharing a dish is a special case of a general one: a tray that has held mercuric chemistry is not a tray for anything else.

Waste

What leaves the darkroom from this page is dilute acid carrying iron, and, from the first bath, traces of the noble metal. Two things to do with it.

Neutralise it. A clearing bath at pH below 1 is a long way outside the window sewer codes usually set: Kodak’s J-52 publication gives 5.6 to 9.4 as the pH range most frequently specified, and Kodak’s J-300 guidance for amateurs describes neutralising an acidic bath with sodium bicarbonate solution before it goes to the sewer, added slowly because the mixture foams. Do it in a ventilated place, wearing what you wore to mix the bath, in a vessel with room for the foam.

Keep the first bath separate if the print was platinum or palladium. Ware’s instruction for the modern sequence is that the spent first bath “should be saved for recovery of precious metals”, and the same reasoning applies here: the first tray holds whatever noble-metal salt did not become image, and that is not a trivial quantity at platinum prices.

ILFORD’s guidance for domestic users is to bottle wastes separately, label them and take them to a household waste and recycling centre’s chemical cupboard; failing that, small amounts flushed with plenty of water, unmixed, and never into a septic tank. Local regulation governs, and nothing on this page overrides it.

Troubleshooting

Yellow highlights on the dry print, or a yellow cast in the masked border. Insufficient clearing — Pizzighelli and Hübl’s own diagnosis in 1886, and Bostick and Sullivan’s in this decade. The remedies are the same in both: more baths, longer, or a stronger bath. Judge it against the uncoated margin outside the mask, under a bluish light, which is Ware’s instruction and the reason to mask in the first place.

The whites go yellow only after the print has been dry for months. Too late for a clearing bath. That is hydrolysed iron ageing in the fibres, and it is a conservation problem, not a darkroom one. Prevention is the only cure that belongs in this course.

Yellow whites and the clearing was thorough. Wall’s 1912 note gives a cause nobody expects: paper “blued with ultramarine, which when treated with hydrochloric acid turns yellow”. Test an unexposed off-cut in the acid before committing a print to a new paper.

The bath goes milky rather than yellow. Wall’s test: “a white opalescence of the bath shows necessity for more acid”. The bath is too weak to keep the iron in solution and it is hydrolysing in the tray.

The image weakens and does not come back on drying. Too much acid for the metal. If the print is palladium or a Pt/Pd mixture and you used a platinum-strength bath, that is the whole explanation; go to 1 in 200, or to the palladium kit’s 1.5 per cent citric acid. If the highlights cleaned up dramatically, some of what left may have been palladium fog from an oxalate developer — Ware’s self-cancelling pair of errors, which is not a technique.

Uneven clearing, in patches or rings. Air bubbles. The Formulary sheet warns specifically against leaving a print to float unattended, because a bubble held against the surface clears nothing beneath it. Agitate intermittently and start the print face down if it floats.

Clearing takes far longer than the sheet says. Three usual causes. A heavy coating, which is Penn’s case: about twice the trade’s coating weight multiplied his clearing time by a factor of twenty. Hard water — Bostick and Sullivan note that water hardness sometimes affects clearing time. Or an alkaline rinse between the developer and the clearing bath, which starts the hydrolysis before the acid can prevent it.

The paper feels soft, or the surface has lost its bloom. The acid has been at the cellulose or the size. Shorten the time, weaken the bath, or move to citric acid, which was Wall’s reason for offering it in the first place and Willis’s for using citrate on his matt palladium papers.

A test that resolves most of these. Coat a strip and do not expose it at all, then run it through the clearing line alone. If it does not come out as white as the uncoated paper beside it, the clearing line is the problem and nothing about the exposure or the developer needs adjusting yet.

Experiments

1. Find your own end point. Coat and expose four identical prints, masked. Clear the first in one bath, the second in two, the third in three, the fourth in three baths of double the time. Wash and dry all four together. Read the masked border of each against the uncoated margin with a densitometer, in blue light where the iron stain absorbs most. Plot residual border density against number of baths. The knee of that curve is your end point for that paper, that coating weight and that water — and it is the only number on this page that is genuinely yours.

2. The metal decides the bath. Coat one sheet in platinum only, one in palladium only, one in a 50:50 mixture, all at the same drop count, and expose them to the same maximum black. Cut each in half. Clear one half of each at 1 in 60 and the other at 1 in 200, for the same time. Measure maximum density before and after. The prediction from Ware’s Table 11.1 is that the platinum halves will read the same and the palladium halves will not.

3. Acid against citrate on the paper, not the image. Take a paper you like and cut six strips. Immerse two in bath A, two in bath D and two in bath E, for ten minutes and for two hours. Wash and dry. Compare surface, dimensional stability and — if you have the means — wet tensile strength against an untreated control. Wall’s claim that citric acid “softens the paper in less degree” is a testable one and he offers no evidence for it.

4. Does Penn’s tray earn its place? Two identical prints, both cleared conventionally in three acid baths. Give one an additional five minutes in 5 per cent sodium bisulfite, wash both identically, dry, and then age both in the same warm, humid place for as long as your patience allows, alongside a coated but unexposed off-cut of each. Read the borders monthly. This is the experiment Ware’s account of Penn implies and nobody appears to have published.

5. Water hardness, deliberately. Clear matched prints in bath E made with distilled water, with your tap water, and with your tap water dosed to a known calcium hardness. Time to a clear border, and residual border density after drying. If your supply is hard, this experiment will change how you work more than any other on the page.

🧪Controls that make these worth doing

Every one of these needs an uncoated margin on the same sheet as the reference white, a coated but unexposed area inside the mask as the thing being measured, and a control print that goes through the standard line on the same day in the same water. Record the coating weight, the paper batch, the water, the bath strengths, the times and the room temperature. Without the first two you are measuring paper, and without the third you are measuring the week.