Here is an update about my findings and working methods after 2 months of working on re-creating a user-friendly version of the collotype process.
Firstly, in order to tackle the problem of not being able to achieve white highlights, I created an unsensitised plate (simply a gelatine and water solution poured onto resin-coated photographic paper), which I then treated in the usual manner, allowing it to soak in a glycerine/water bath before blotting off and inking up. It was immediately obvious that the ink was still sticking to the plate, which allowed me to rule out any sensitisation and exposure-related causes. For one reason or another, I had begun to suspect at some point that it was the glycerine bath causing the plate to take too much ink, but this doesn’t make sense as the glycerine helps keep the plate moist, and thus repel ink. I listed all of the possible parameters that I could alter in order to see if they affected the highlights, and then chose the most likely, as they were unrelated to the composition of the gelatine solution, these were: changing ink consistency and plate substrate.
Before I began experimenting with collotype, I was already well aware that traditionally collotype inks were particularly stiff, perhaps much more so than stiff lithographic inks, but having had no experience of either, I couldn’t even subjectively say whether or not the ink I was using was firm or not. I had read somewhere in relation to lithography that magnesium carbonate was added to ink to make it more tacky, and as luck would have it I already had a large bag to hand, because magnesium carbonate is commonly known as the chalk that gymnasts, weightlifters and climbers use. I measured out 7 grams of ink (a random amount), and then 0.5 grams of chalk (another random amount), and mixed them together with my palette knife spatula thing until the chalk was fully incorporated. I then transferred this small amount of modified ink to a film container for use later on.
After many tries and fails with this new ink, it got to the point one day where I got so frustrated that I just poured large lumps of chalk directly from the bag onto the ink slab without thought for measurement or finesse. It took me longer to mix the ink up, which was much stiffer this time, which also meant that it took longer to roll out, but to my surprise I had concocted a mixture that would no longer stick to the plate in the highlights! In my attempts to be methodical I was too careful, which could have been avoided by throwing a little caution to the wind, seeing as there was nothing big at stake in this case.
What was interesting to find with this newer ink mix was that the rubber brayer I was using was able to actually lift the image off of the plate. This property is taken advantage of in offset printing where the image is transferred from the plate to a rubber blanket, before being transferred to the final printing surface. The benefit of this technique is that because the plate never comes into contact with the paper, the paper is not deformed by the addition of subsequent layers, making it suitable for the registration of multiple colours. This is an interesting idea to keep in mind should I ever get to the point of trying 3 or 4 colour prints.
One of the problems with not being able to achieve white highlights is that it makes it very difficult to discern whether or not the exposure of the plates is correct. In working with altered inks it transpired that I had slowly increased my exposures over time, in order to deal with higher density negatives, but overexposing the plates meant that I had to also increase the ink stiffness again, until I reached a point where it was just too difficult to work with. In this way, I had compensated for one problem by creating another, but this error lead to me having a greater understanding of how the relationship between exposure and ink consistency functions.
Some of my early tests were conducted using watercolour paper, brush-coated with several layers of sensitising solution, and later, photographic resin-coated paper with the silver removed. This was until recently, the main method I was using, however, there are a number of problems I encountered in the process. Unlike a carbon tissue which contains sugar, the collotype solution when dry causes the paper to curl up into a tube, and must be taped down in order to lie flat, which is impractical, and results in uneven inking. Furthermore, the paper doesn’t respond well to remaining wet for long periods of time, which seemed to cause the plates to be destroyed prematurely. Interestingly though, I discovered that the pre-existing layer of gelatine that once contained silver, would eventually swell up, meaning that it may be possible to simply brush sensitiser onto fixed-out silver-gelatine RC paper, rather than go to the trouble of coating and drying a complete solution. Having already tried this same method, but on film, and with less than ideal results, this avenue will have to wait until a later date to be explored more thoroughly. Likewise, coating watercolour paper by pouring a thick layer of solution, rather than brushing on multiple thin layers, is something to be considered at a later date.
Instead of using paper or glass, I have been making my plates on thin sheets of 8×10 inch aluminium which has been prepared by grinding the surface with a waterstone, normally used for sharpening knives. Roughening the surface allows the sensitised gelatine solution to adhere without the need for subbing layers of hardened gelatine or anything else that was traditionally used, such as stale beer for example.
One problem I encountered with this method was that overexposure was causing the plate to bubble, presumably because the reflected light was exposing the gelatine from the side in contact with the plate, which creates a build up of gas that cannot easily escape, as in the case of surface exposure. After reducing my exposure times I have not experienced this problem again.
Another disadvantage, although relatively small perhaps, is that the plates cannot be cut into smaller pieces like I have been accustomed to doing with paper. Instead, I just print multiple negatives on the same plate, and have also made some larger negatives specifically for the purpose of testing.
In addition to making my plates on aluminium, as part of the preparation process, I have begun using the clearing baths that are intended to remove the leftover Das after exposure, and then after washing, I leave the plates to dry completely over night again, before beginning the printing process the next day. This just ensures that I am not working with gelatine that has become excessively soft due to prolonged immersion.
An alternative collotype sensitiser
As mentioned at the start of my journey, I was looking at using an iron-based sensitiser as not just an alternative to dichromate, but as an alternative to diazidostilbene. The reason being that Das is not a particularly cheap or easy chemical to get hold of, especially in smaller quantities, and requires the clearing bath steps in order to render the plates insensitive to light. Following on from the research of Halvor Bjoerngaard and his Chiba System, which uses Ferric Ammonium Citrate as a sensitiser in carbon transfer printing, it seemed logical that the same could be substituted for use in collotype too. The particularities of FAC as a sensitiser mean that the traditional carbon transfer method has to be modified, which lead to the suspicion that its use in collotype may not be viable, however I can confirm that it does work. I am too much of a novice to be able to tell whether or not there is a noticeable difference in the results it is capable of producing, but for now it seems promising.
Unlike Das which causes the polymerization of gelatine in the presence of UV light, FAC requires a development step after exposure. It is this additional step that I think could be a positive aspect rather than a negative one, because theoretically, once we have exposed and developed a plate we don’t have to worry about it receiving additional exposure, as this shouldn’t alter the plate without further development. This would mean that we could skip the 2 clearing baths and washing steps, significantly reducing the amount of time needed to prepare a plate, but also eliminating the need for other chemicals.
Hydrogen peroxide is the developer in this new process, which is readily available and innocuous at the concentrations needed, and Ferric Ammonium Citrate is one of the chemicals used in cyanotype, which means it is relatively easy to obtain.
One of the things I have noticed that contradicts what many people say about not just collotype, but other printmaking techniques, is that from the first print, each successive one gets worse in quality and not better. At this point in time I have no idea why this is, and hope that once I have better control over the various elements that are currently causing me problems, things will eventually work themselves out.
The Problem With Shadows
As I previously demonstrated, there were instances where the darkest parts of the print came out white instead of black, because the ink had failed to transfer from the plate. I noticed that this was more prevalent when printing on thin card instead of the standard photocopier/printer paper that I am mostly using. It occurred to me that despite collotype working through an essentially lithographic process, it is in all actuality a unique combination of lithography and intaglio printing. This is because the hardened gelatine that will receive the ink for the shadows sits significantly lower than the highlights that will repel the ink. This presents an awkward problem because the roller, especially a hard rubber one, will make contact with the areas intended to be white, more easily than the black areas, which is the opposite of what we would want! So not only is there difficulty in applying ink to the shadows, there is also the problem of applying enough pressure to force the paper into those low-sitting shadow areas, without damaging the plate in the process.
Collotype is not a true planographic process, and this is something that others before me have obviously noted. One such individual was Ernest Edwards, who altered the original collotype process in a number of ways, making the process truly his own, and referring to it as Heliotype.
Here he refers to the original collotype process as discovered by Alphonse Poitevin:
“ The practical difficulties met with in attempting to work it arose first from the spongy and delicate nature of the gelatine film in those portions where light had not acted; and, next, from the difficulty of obtaining intensity in the inking, and sharpness in the image, and from other technical obstacles. Bearing in mind the character of the image on the gelatine, it will be understood that those parts of the gelatine not acted on by light are in much higher relief, from the presence of the water which they absorb, than those parts which, having been acted on by light, do not absorb water. But it is the depressions which take the ink, and the walls on each side of the depressions prevent the roller from reaching them; and, again, it is the depressions to which the pressure must be applied, in order to allow the paper to pick up the ink. It was found that when the requisite pressure was applied, the gelatine was speedily destroyed, more especially as the description of pressure employed was that used in lithography, where the force is a scraping one over the surface of the stone.”
Edwards’ Heliotype process uses a pre-hardened and sensitised gelatine solution, which is exposed before being transferred to a metal plate, where it will remain for printing. The advantages of this method are that the pre-hardener in the gelatine (alum chrome) toughens it up and reduces swelling, while still retaining its ability to be acted on and selectively hardened by exposure. This not only improves the life expectancy of the plate, but the reduced swelling makes inking and printing easier, and supposedly allows heliotypes to produce greater blacks than other collotype processes. The other significant distinction in Edwards’ method is that heliotypes are neither printed in a lithographic nor rotary press, but in a simple press applying vertical pressure. These differences not only make the heliotype process more desirable in terms of the quality it is capable of producing, but also in terms of practicality, especially for the individual, for whom a lithographic press imposes significant financial and spatial constraints.
As I continue to work with aluminium plates, I have it in the back of my mind that my final process will resemble heliotype more so than traditional collotype. I have yet to try the pre-hardening step, but will do so once I have other parts of the process nailed down, and I already have the necessary chemicals.
One problem I haven’t been able to overcome, but have come to understand is the issue of the rollers leaving unintended marks on the plate. Originally I thought that the traces left by the edges of the rollers might be able to be compensated for with a particular rolling technique, but it seems that this problem is simply a function of using a roller that is smaller than the plate width. In all of the processes similar to collotype, such as bromoil, and in all of the examples I have seen where ink is applied either to a printing plate, or directly to a final matrix, by brush or with a small roller, there is always a kind of patchiness apparent. This appears to just be a consequence of using a small implement to cover a large area, in the same way that painting the wall of a house with a brush is more likely to leave visible marks compared to if it was done with a large roller. In processes other than collotype, this also seems to result in a softer kind of image, which can be confusing because we may mistake this look as being a limitation of the process rather than as a result of the method by which ink was applied. Any random collotype example can show the level of detail and clarity that it is capable of producing, but only provided that the plate is inked up correctly. This demonstrates the difficulty in mastering the process, because of its many interdependent factors that all must come together simultaneously, and which rely solely on operator skill.
One thing that is apparent in using a foam roller compared to the rubber brayer is that the foam roller not only produces a noticeable grain effect that is separate from the grain structure of the collotype gelatin itself, but it also imparts quite a significantly lower contrast than the rubber roller given any particular ink stiffness. I have used this to my advantage recently, by first applying a layer of ink with the foam roller in order to fill in the blacks, but then going over the plate with the rubber roller in order to clear the highlights and also to bring back sharpness by removing the grain.
Perhaps the biggest defect in all of my test prints is the random appearance of debris, which comes at least in part, from the main roller mechanism being broken and releasing particles into the ink and onto the plate. This isn’t a huge problem at the moment, as there are more than enough additional factors that make the prints of low quality, but it’s obvious that I will need to invest in a better brayer without yet forking out the 500 euros it would cost to get a high quality litho roller. I haven’t been able to successfully use my wider rollers for applying ink because it seems that they are not perfectly flat, meaning that they neither receive nor distribute ink evenly, making them more of a hindrance than anything.
In order to come up with my own specific recipe for collotype ink I have devised a test, similar to what I used before with the unsensitised plate. In theory I should be able to take unaltered ink straight from the tin and then gradually add chalk in measured amounts, making a trial print each time before increasing chalk content until the point that the plate will no longer accept the ink. With small amounts of calibration afterwards I should be able to arrive at the ideal ink consistency that will not print on unsensitised gelatine. Once I am certain that the highlights should print white I can then finally go about making a step wedge test plate which will be able to more accurately show me what my exposures are like. Knowing that the whites are in theory where they should be will allow me to focus on analysing the shadow density and detail, which will eventually provide me with an ideal exposure time, as well as an idea about what kind of tonal scale is produced at that level. All of this information should not only allow me to standardise my workflow and to select and produce negatives of an appropriate contrast, but it should also make tackling other problems simpler.