Heliography on bitumen of Judea

Heliography contains no silver and forms no metal image. What is on the plate is hardened tar, and the parts the light never reached were washed away with a solvent. That single fact makes it the founding member of the other family of photochemistry — the one in which light changes the solubility of an organic layer rather than reducing a metal salt — and the family whose descendants are the photoresist, photogravure and every integrated circuit ever made. The Part I lesson owns the history and the argument; bitumen of Judea owns the material.

The chemistry

Three statements, and the course is careful about which is which.

Certain, and operational. Exposed bitumen becomes insoluble in the solvent that dissolves unexposed bitumen. That has been demonstrated every time the process has worked, from 1822 to the modern repetitions, and Jean-Louis Marignier has investigated and repeated Niépce’s work.

Certain, and worth keeping straight. The process is negative-working in the printing sense and positive-appearing in the pictorial sense. Light makes the coating stay, so the areas that received most light end up covered; on a bare polished plate those covered areas are the matt pale ones and the bared metal is dark, so the picture reads as a positive.

Not stated here: the mechanism. Mike Ware, a chemist, describes the molecular structure of bitumen as so complex and variable as to constitute a chemist’s worst nightmare. There is no formula to write. Photo-induced cross-linking of the large aromatic molecules is the general shape of the explanation and the title of Marignier’s paper on it, and this course does not paraphrase a title as though it were a mechanism.

The property that decides everything else is the absence of amplification. A silver halide crystal needs a handful of photolytic atoms to become developable and a developer then converts the whole crystal; bitumen has no such lever, so every photon that contributes does its own work on its own molecule. Ware estimates the speed of the process at less than 10⁻⁶ ISO. Niépce could improve his lenses, his supports and his coating, and did all three; he could not change the fact that what light did is all you get.

Historical workflow

Niépce’s own directions survive in Robert Hunt’s 1841 translation, and they are recorded here as history rather than as an instruction. Powdered bitumen is saturated with essential oil of lavender and gently warmed until the oil takes up the colouring matter; the varnish is rolled cold onto a highly polished plate, thin and even, giving a vermilion tint, and dried on gently warmed iron — Niépce adds that the operator should hold a small metal disc in front of his mouth while doing it, because damp ruins the coating. After exposure only a very faint outline is visible. The picture is brought out with a solvent of one part by volume of lavender oil to ten of oil of white petroleum, watched by reflected light, then drained and washed. Where the object is a printing plate, nitric acid then bites the metal the hardened bitumen does not protect.

The supports track the work: glass in 1822, on which he made copies good enough to give away; lithographic stone in 1823; copper in 1824 and 1825, where the etching finally worked; pewter from 1826, which he found gave much better results because the support was whiter; and silvered copper from 1829, the bared silver darkened with iodine vapour to improve the contrast — the step Hunt saw in 1841 had “led the way to Daguerre’s beautiful process”.

The image

The Le Gras plate at the Harry Ransom Center is a heliograph on pewter, 16.7 by 20.3 by 0.15 cm. It is not a photograph in any sense a modern eye finds comfortable: pale, matt, hardened bitumen standing against darker bared pewter, which must be lit and angled correctly to read at all. Turn it and the image collapses into a blank sheet of metal. Niépce knew this and wrote about it — viewed by reflection on the varnished side, and at a certain angle, the effect is remarkably striking, while by transmitted light it is confused and shapeless.

That is the same viewing behaviour the daguerreotype has, and for a related reason: the light and shade are a difference in how two surfaces return light, not a pattern of densities.

The tonal failure is the one that mattered to Niépce, because printing was the goal. The acid either bites or it does not, and a continuous-tone camera view has no way of instructing it. In 1827 he gave up etching his camera views for exactly that reason.

⁇The date and the exposure, neither of them settled

The date. The Harry Ransom Center, which owns the plate, dates it 1827 in its object record and its exhibition text, and Ware agrees; the Musée Nicéphore Niépce’s chronology places the Point de vue du Gras under 1827. Many older accounts give 1826, and some the range. This course writes 1827, because that is what the holding institution states, while noting that the evidence is circumstantial in every version.

The exposure. Niépce’s own directions, in Hunt’s translation, give camera exposures of six to eight hours varying with the light, and four to six hours for a contact copy of an engraving. Marignier reconstructed the process and the camera — finding a lens of 325 mm focal length and 81 mm diameter, which is f/4 — and concluded that a practical exposure was about five days. The Harry Ransom Center’s own text says “several days”. This course reports both and combines neither.

Permanence

Permanence is the one problem heliography solved, and it solved it trivially. The image substance is a hardened varnish and the material light did not act on is removed by the solvent, so there is nothing photosensitive left in the object. Compare the three answers of the 1830s: Talbot’s halide treatments stabilise, leaving the salt in the sheet with its reactivity turned down; Herschel’s thiosulfate washes out; and the hardening family removes what light did not use, which is the same convenience the cyanotype reaches by a different route.

The evidence that it lasts is analytical. In 2010 a Getty Conservation Institute team examined four surviving Niépce plates with the National Media Museum. X-ray fluorescence established the substrate of all four as fine pewter, roughly 97 per cent tin, 2 per cent lead and 1 per cent copper. Infrared spectroscopy confirmed the Le Gras image as hardened bitumen, and Le Cardinal d’Amboise as bitumen over a deeply acid-etched surface — the resist doing its other job. The surprise was the fourth: Un Clair de Lune turns out not to be bitumen at all but a photohardened resinous gum whose spectrum resembles the physautotype, a process received history dates five years after the plate. Ware notes the conflict and says the episode may yet be rewritten.

Hazards

Three classes of hazard, and each of them is why this entry gives no procedure.

Organic solvents in quantity, warmed. A volatile essential oil and a petroleum distillate, worked by hand and gently heated, with photographic materials in the same room.

Nitric acid, where the object is a plate. Its harmonised classification under the European CLP regulation, as reported by PubChem, carries the signal word Danger with H272, may intensify fire; H314, causes severe skin burns and eye damage; and H330, fatal if inhaled.

A material with no data sheet worth the name. Bitumen of Judea is a variable natural mixture rather than a compound, and the encyclopaedia entry for it carries no molecular formula, no molar mass and no CAS number — not because nobody has looked, but because there is nothing single to assign them to. A hazard assessment starts from a classification, and here there is not one to start from.

Why we study it and do not reproduce it

Because the line it started is enormous, and because the reason it failed is the most useful negative lesson in Part I.

The line. Look at the British patent abridgments for photography for 1877 to 1883, fifty years after Niépce’s death, and the class is full of specifications that are recognisably his process: a metal plate coated with bitumen or asphaltum dissolved in benzene or turpentine, exposed under a negative in a printing frame, the unaffected coating cleaned off with a solvent, and the bared metal etched or built up galvanically. Change the resin, change the solvent, keep the logic, and you have photolithography, photogravure and the manufacture of semiconductors.

The lesson. An image-forming process needs a gain stage. The two centuries between Schulze and the calotype are largely the search for one, and heliography is the case that shows what happens to a process that never finds it: no portrait, no moving subject, no changing light, and a plate that records the sum of every lighting condition that occurred while it stood open.

The course’s own reason for giving no procedure is the Level D policy: where a process depends on substances or conditions unacceptable outside a professional laboratory, the page gives the chemistry, the workflow in outline, the image characteristics and the hazards, and no quantities, no sequence and no conditions. Part XXVI owns the light-hardened colloids, with the hazard assessment written first.