Tuesday, August 18, 2015

Eucalyptol

One or both of my readers (hi mom) may be familiar with the b/w film developer Caffenol. This is made with instant coffee, washing soda and vitamin C powder. It is environmentally friendly, cheap, high quality, easy and cheap. Alright, I am cheap and Rodinal might be a little cheaper than Caffenol, but there is just something very cool about mixing household ingredients and getting film developer out of it. So I occasionally mix up a batch when I have time. There are examples around here.

Recently, there was a discussion of the active ingredient. It turns out NOT to be caffeine, but caffeic acid. So that begs the question (for me): What else has caffeic acid in it that might be readily available? The answer: The bark of Eucalyptus globulus (according to Wikipedia). This is an Australian native tree that happens to be a common invasive species in So. Cal. where I live. In fact, there are big ones just outside my front door. So it is very simple for me to step outside and collect some bark off the ground, and that's just what I did. I didn't weigh it or anything, this was more of a 'proof of concept' experiment than an optimization. So the best I could say about the quantity used would be 'a couple of handfuls'. I took it inside and broke it into little pieces, then I put those into my coffee grinder. Legal Disclaimer: Do not use anything you plan to later use for food prep when preparing photographic chemistry. I ground it up pretty fine and put it into a pint mason jar. Then I poured boiling water in and let it steep for about 30 minutes. I then poured the liquid through a coffee filter and discarded the used bark. I have absolutely no idea how much caffeic acid is in this tea (if any). I subsequently did some Google Scholar research and found a paper where the researchers could not detect caffeic acid in the bark of Eucalyptus globulus. That was discouraging, but I had come this far, so I decided to at least do a drop test on a piece of film.

To do a drop test, just get an old piece of film, expose it to the light and put a drop of your developer on it. If it turns black, your developer is active, if not, then it isn't. I had a little over 500mL of eucalyptus tea, so I followed the recipe for Caffenol C-L, adding 16g of Washing Soda and 10g of Vit. C. I didn't add any KBr. Then I topped off to 1L with distilled H2O. The film that needed to be developed was Kadak Recordak Dacomatic. This is copy film with a very thin emulsion layer and no anti-halation layer, so probably would be my best bet if the developer ended up being weak. So I cut the leader from one of my rolls and put a drop of the developer (heretofore known as Eucalyptol) on the emulsion. One minute later... nothing. Give it a little shake. Then it happened. The emulsion started to turn. First just a faint brown, then darker and darker. It was working! I swirled the drop of developer onto a different area of the film and there it sat. Maybe there was a little development, but not much. So there it was. A one-shot developer.

Now to decide on a developing time. I figured I would do a stand development, but I didn't want the bromide drag that comes with it on 35mm film. So I decided 1 hour semi-stand with inversions at 20, and 40 minutes. That way, if the dev was weak, at least I might get thin but scanable negs.

Here are some of the results:

N2020-Dacomatic-004 N2020-Dacomatic-003 N2020-Dacomatic-021 N2020-Dacomatic-018 N2020-Dacomatic-014

Okay, HCB I am not, but you get the idea. The tonal scale is compressed (more contrast), though part of that is attributable to the fact that I was shooting mostly in full sun, mid-day at the freakin' beach. You won't find many more contrasty scenes than that. Also the film itself is intended for copying documents, so contrast is sort of its thing. Normally, I develop this film in Adonal (Rodinal) 1:100 for 70min with inversion initially and at 30 min. That gives a pronounced grain that I find appealing. It's not for everyone, especially those who enjoy the digital aesthetic, but I like the grain. With this developer, I am going to call the grain "chunky style", but still not so obtrusive as to be distracting. Additionally, while I was doing some light dust spotting in Photoshop, I did notice some very very small pinholes in the emulsion. These appear as tiny black dots on the scan. My intuition tells me that there could be something in the eucalyptus extract that is being hard on the thin emulsion layer of this film. I am doubtful that I would see it with a regular pictorial photographic film (stay tuned).

Over all, I think this is a good alternative to caffenol for those of us who want to see just how far down we can drive the cost of developing a roll of film in a home-made developer. Here's the math. I used about $0.10 worth of washing soda, about $0.76 worth of vit. C powder, and maybe a nickel's worth of distilled water. So that's around $0.90 in ingredients, not counting the film which was practically free (I think I paid about $0.03/ft). To compare, the same amount of Adonal 1:100 would cost about $0.14. So this isn't really a money saving proposition, especially if you consider the time spent collecting, grinding, steeping and weighing. It is more about the adventure of making something that works. It is closely related to the "maker" movement you see in places like the Maker Faire and Make magazine. It lets us explore our creativity while still engaging our left-brained nature.