Mike Ware - Alternative Photography
Mike Ware - Alternative Photography

Towards an Unproblematic Cyanotype Chemistry

Herschel’s Classic Cyanotype 1842

The classic cyanotype process sometimes presents problems for the user: on coating, the sensitizer solution may be poorly absorbed by the paper, which consequently loses much Prussian blue during wet-processing, leaving an image of restricted exposure scale, poor tonal gradation, and low density. In extreme instances, the picture has been known to wash away almost entirely. Another inconvenience is that two separate bottles of solution are needed to guarantee viable sensitizer storage, because it becomes short-lived when mixed; moreover, one of these solutions usually acquires a thick layer of mould growing on its surface.

Ferric Ammonium Citrate Failings

These drawbacks are due to the inconsistent nature of the light-sensitive chemical, ferric ammonium citrate, aka ammonium iron(III) citrate. This substance had its origins in early pharmacy, before cyanotype was invented by Herschel in 1842, and its method of preparation reflected the need for a product uncontaminated with any noxious substances, because it was intended to be taken internally as an iron tonic – often referred to by apothecaries of the day as Ferri et Ammoniae Citras.

The safest procedure for making it is to precipitate solid ferric hydroxide from ferric sulphate solution with ammonia, filter it off and wash it thoroughly, then dissolve it in a solution of citric acid, and finally neutralise with ammonia; the solution can be evaporated to a syrup, drying to an amorphous solid. This preparation -which still remains the one prescribed today- does not guarantee that the product is a well-defined chemical in the molecular sense. Indeed, this procedure promotes the very opposite because ferric hydroxide is a highly polymeric (“long chain”) substance, often colloidal; the ferric ammonium citrate so obtained can be extremely variable in the proportions of ammonium, iron and citrate that it contains: in different samples its iron content is found to range from 14% to 28% w/w, and its colour varies correspondingly from light green to dark brown; no single chemical formula can be written for the actual molecular species that are present. The solid cannot be obtained as crystals –which usually indicate molecular purity- but only as an amorphous powder or glassy flakes. The solid is highly deliquescent and on storage may compact into an intractable mass. Chemists refer to it as an “ill-characterised substance” - a condemnatory term that reflects these variable properties.

The inconsistency in commercial ferric ammonium citrate means that no two people, working apart, can be sure that they are using the same chemical. Cyanotype instructions usually call for the “green form” introduced by Valenta in 1897, which performs better than the “brown form” of 1840, but this is not just a simple choice between the two because there is a whole range of varieties in between. This extreme variability will create problems especially for those who employ protocols requiring precise densitometric calibration, such as PiezoDN, to make digital negatives for their cyanotype printing.

All this unreliable chemistry is bad science! It partly explains the widely varying experiences of cyanotype printers trying to use the commercial chemical: those workers fortunate enough to have found a reliable commercial source of consistent, green ferric ammonium citrate should have few problems with printing classic cyanotype quite successfully; but there are some for whom this is evidently not so, to judge from the images posted to the relevant internet groups. Even today, after 177 years of chemical manufactury since Herschel first invented the cyanotype process, it seems we still cannot get it right every time! Subscribers to the alternative photographic process forums are accustomed to reading dismayed cries for help from would-be practitioners who are failing to make permanent cyanotype images of high photographic quality with long, well-separated tonal scales and a satisfying maximum density.

Paper Problems

Our difficulties are worsened by the variability of paper itself. Clearly a correct choice of paper and its sizing is vital to success with all the iron-based processes, which depend critically on the absorption of the sensitizer solution into the interfibrillar space within the surface cellulose fibres of the paper sheet. The image substance produced by light is often in nanoparticle form, and must be securely trapped within those fibres; otherwise, if the sensitizer just remains in the coarse pores between the fibres, the tiny particles of photoproduct will be easily washed out by the wet processing, leaving an image with empty ‘blown-out’ highlights and murky unresolved shadows. Absorption of the sensitizer can be improved by the use of a surfactant or wetting agent such as Tween 20.

New Cyanotype 1995

One answer to the chemical uncertainties of ferric ammonium citrate is simply to avoid this substance altogether, and instead to use a light-sensitive ferric compound that is pure and reproducible, and essentially monomeric with molecules small enough to penetrate the cellulose structure. This was the reasoning behind my New Cyanotype process of 1995, which employed as the light-sensitive component the widely-available and well-characterised pure salt, ammonium ferric oxalate, of known formula and molecular structure and fully reproducible composition. The sensitizer is made up as a convenient 'single-bottle' solution, adding a dichromate as preservative to ensure a very long shelf-life of several years. Requiring only a very short UV exposure, three stops less than the ‘Classic’, the ‘New’ process yields a stable Prussian blue image with a smoothly graduated tonal scale of excellent colour and a maximum density verging on black. The wet-processing is extremely simple and non-critical, and can offer a small degree of contrast control.

However, my ‘New’ sensitizer requires a preparative procedure that may be found somewhat challenging by users unfamiliar with chemical manipulations such as dissolution by heating, controlled fractional crystallization, and separation by filtration. These operations can result in a rather variable yield, which causes differences in the final sensitizer concentration from one batch to another, and this variation may require the re-calibration of the most quantitatively demanding digital negative-making protocols, such as the PiezoDN procedure. Admittedly too, ‘New’ cyanotype is more expensive in chemicals than the ‘Classic’ formula. Its much higher sensitivity makes it rather susceptible to paper impurities: only the best non-buffered papers will serve. The dichromate preservative introduces a highly toxic substance (albeit in very small amount) which is now banned in many countries. Without the dichromate, the shelf-life may be reduced to a few months.

Ammonium Dicitratoferrate(III) Characterised 1998

In contrast with the ‘Classic’ chemical ammonium ferric citrate which, as we have seen, is polymeric and highly variable in its composition, a monomeric dicitrato-complex of iron(III) has only recently been isolated and fully characterised, as reported in the contemporary chemical literature.* The essential starting material for making the monomeric dicitrato complex is the widely-obtainable simple salt, ferric nitrate nonahydrate, which is a pure crystalline solid containing monomeric hydrated ferric ions. However, if this pale violet-coloured salt is dissolved in water only, hydrolysis and polymerization of the ferric species begins immediately. The authors of this preparation dissolved the ferric nitrate directly in citric acid and neutralised it by adjusting the pH to ~8; the product is highly soluble in water, so had to be precipitated as yellow crystals by the addition of a large volume of ethanol, in which it is insoluble. The crystalline solid could then be analysed and fully characterised by 3-D X-ray structure determination, and other physical methods.

*M. Matzapetakis, C.P. Raptopoulou, A. Tsohos, V. Papaefthymiou, N. Moon, and A. Salifoglou, ‘Synthesis, Spectroscopic and Structural Characterization of the First Mononuclear, Water Soluble Iron-Citrate Complex, (NH)5Fe(C6H4O7)2.2H2O. Journal of the American Chemical Society, 1998, vol. 120, pp. 13266-7.
DOI 10.1021/ja9807035

Simple Cyanotype 2019

Inspired by the publication of this chemical research just described above, I have recently explored a third option for making up cyanotype sensitizers: to prepare a photosensitive citratoferrate complex in situ by minimising the hydrolysis of iron(III) and formation of polymeric molecules. I have devised a sensitizer solution that is very easily made up within less than an hour, from widely available, inexpensive pure chemicals: citric acid, ferric nitrate, and ammonia, and potassium ferricyanide.

The reasoning behind this formulation is therefore to dissolve the ferric nitrate crystals directly into a solution already containing a two-fold molar ratio of the citrate ligand, as the free acid at low pH to suppress hydrolysis, then to neutralize the hydrogen ions carefully with ammonia, in the expectation that the main reaction product in solution will be the monomeric dicitratoferrate(III) anion. There is no need to try to isolate a solid from this – the solution can be used directly as a sensitizer. The preparation calls for no more skill than an ability to measure volumes of liquids and weigh out solids, with only moderate precision, and stir to dissolve them. It is highly reproducible and simple to scale up for commercial manufacture of large quantities of sensitizer.

It is obvious that there will also be a by-product of this reaction left in the sensitizer solution besides the free ammonium ions: viz. a high concentration of nitrate ions. These do not interfere with the photochemistry, and their presence actually seems to confer the following significant benefits on the sensitizer. Coincidentally, the addition of ammonium nitrate to improve the stability and quality of blueprint paper sensitizer was actually the subject of a U.S. patent, no. 2,113,423, as long ago as 1938.

The resulting sensitizer, which I have named ‘Simple’ Cyanotype to distinguish it from the two others, performs surprisingly well: admittedly, it is less than half the speed of my New Cyanotype, but it is still about 4 times faster than a typical Classic Cyanotype formula using commercial ferric ammonium citrate. It yields a long, well-separated tonal exposure scale of at least 2.7 (nine stops), with delicate high value gradation and a good maximum density of about 1.4. There is little runoff of image substance in the wet processing, which only uses very dilute citric acid. The ‘reversed’ shadow tones return to their full density fairly quickly, without treatment by hydrogen peroxide.

This formulation can offer a further benefit: by the addition of controlled volumes of ammonia solution, the extent of the neutralization and resulting pH have the effect of varying the contrast or Exposure Scale (ES) of the sensitizer. From the maximum of 2.7 (9 stops) it can be lowered to any value down to 1.8 (6 stops), and even lower to 1.2 (4 stops) by development in water alone. Thus the sensitizer can be prepared variously to meet the needs of the user: from the demanding requirements of PiezoDN on the one hand, to the printer of traditional silver-gelatin negatives on the other. By controlled mixing of the two solutions of the two extremes of contrast, any intermediate exposure scale could also be accurately matched. It is believed that this is the first time that a cyanotype sensitizer with fully controllable contrast has been devised.

The sensitizer solution can claim to be fairly non-toxic (it contains neither dichromate nor oxalate) so it may be safely used by children, and it is not potentially harmful to the environment. Moreover, this sensitizer shows no susceptibility to mould growth on its surface, that irksome feature of commercial ferric ammonium citrate solution. The shelf-life of this economical ‘single-bottle’ version appears to be long enough -about a month- to make it useful, but refrigeration greatly extends this. If long term stability is important it may be possible to reformulate it as an enduring ‘two-bottle’ A+B mixture like the Classic cyanotype. I hope also that it may soon become commercially available as a ready made-up solution, from a few dedicated suppliers on our respective continents. For a detailed summary of the instructions for making and using the sensitizer see the Simple Cyanotype Instructions download here.

back to top