New Cyanotype with surprisingly short curve / high contrast

[koraks]

I haven't done much cyanotype in years, but out of curiosity, I have been doing some step wedges lately. I'm running into something odd with my New Cyanotype chemistry. It produces a very short H/D curve, i.e. it prints with very high contrast. This is weird, because I recall when I last used this formula (some 8 years ago or so), I saw the long-scaled curve typical of New Cyanotype, although I didn't do measurements back then.

Here's a collection of curves I've just plotted based on several experiments:

In terms of chemistry, I mixed up three small batches of the New Cyanotype chemistry, and all produce the same results, so it's not a mixing error - or at least not a spurious one. My approach is based on Mike Ware's notes here: https://www.mikeware.co.uk/mikeware/New_Cyanotype_Process.html
* to 2g potassium ferricyanide I add as much hot water to allow it to dissolve (2-3ml), and keep warm in a hot water jacket.
* to 6g ferric ammonium oxalate, I add 6ml hot water and dissolve
* mix together both solutions while still hot
* allow to cool down gradually
* filter the liquid and discard the crystallized out emerald green potassium iron oxalate.
The filtrate ends up being about the volume that Ware recommends, so I didn't add additional water to the stock.
I did not add dichromate, citric acid, a wetting agent or anything else to the stock solution.

I coat with a brush at about 0.5ml for a 5x7" area. I've used several papers (see chart above) without prior treatment. Coated paper is dried with a hairdryer and exposed immediately, and processed immediately after exposure.

Exposure is done with 365nm + 395nm LEDs, but I've also done a test with UV-BL tubes, producing pretty much the same H/D curve for all intents and purposes.

I've tried diluting the stock solution with water, which seems to increase printing speed (consistent with Ware's notes) and reduced dmax, but otherwise no apparent difference in curve shape/length.

I've tried adding a pinch of citric acid to the first wash bath or straight tap water, but see not difference; the prints clear in the same way.

I've tried getting the blue to develop with hydrogen peroxide vs. air-dried/naturally developed blue, and the end result is indistinguishable.

The different papers I've tested produce some differences in dmax, but curve shape is overall similar. The sagging pattern at the shoulder of some of the curves above is probably due to mottling which some papers are prone to; esp. the Schut Laurier paper tends to give some lighter blotches in the darker areas.

The somewhat long toe in some of the prints (e.g. the "Simili Japon #2" plot above) is probably due to incomplete clearing. I limited wash times to avoid bleaching as a result of the slight alkalinity of our moderately hard tap water.

I have also tried adding citric acid to the sensitizer prior to coating. In large quantities, this produced some instant fogging upon drying the paper, but this was below the dmin threshold and washed out in processing entirely. In smaller quantities, there was no apparent effect on the printing qualities or curve.

It's kind of puzzling, since Ware notes an exposure range of around 1.8logE for a full tonal scale print, and I get a full range from dmin to dmax on the paper with only around 0.6logE exposure range. In fact, the curve I get with the New chemistry is steeper than with Classic!

I'm sure I must be doing something phenomenally stupid, but I'm evidently overlooking it.

 

[nmp]

I haven't done much cyanotype in years, but out of curiosity, I have been doing some step wedges lately. I'm running into something odd with my New Cyanotype chemistry. It produces a very short H/D curve, i.e. it prints with very high contrast. This is weird, because I recall when I last used this formula (some 8 years ago or so), I saw the long-scaled curve typical of New Cyanotype, although I didn't do measurements back then.

Here's a collection of curves I've just plotted based on several experiments:
View attachment 368503

In terms of chemistry, I mixed up three small batches of the New Cyanotype chemistry, and all produce the same results, so it's not a mixing error - or at least not a spurious one. My approach is based on Mike Ware's notes here: https://www.mikeware.co.uk/mikeware/New_Cyanotype_Process.html
* to 2g potassium ferricyanide I add as much hot water to allow it to dissolve (2-3ml), and keep warm in a hot water jacket.
* to 6g ferric ammonium oxalate, I add 6ml hot water and dissolve
* mix together both solutions while still hot
* allow to cool down gradually
* filter the liquid and discard the crystallized out emerald green potassium iron oxalate.
The filtrate ends up being about the volume that Ware recommends, so I didn't add additional water to the stock.
I did not add dichromate, citric acid, a wetting agent or anything else to the stock solution.

I coat with a brush at about 0.5ml for a 5x7" area. I've used several papers (see chart above) without prior treatment. Coated paper is dried with a hairdryer and exposed immediately, and processed immediately after exposure.

Exposure is done with 365nm + 395nm LEDs, but I've also done a test with UV-BL tubes, producing pretty much the same H/D curve for all intents and purposes.

I've tried diluting the stock solution with water, which seems to increase printing speed (consistent with Ware's notes) and reduced dmax, but otherwise no apparent difference in curve shape/length.

I've tried adding a pinch of citric acid to the first wash bath or straight tap water, but see not difference; the prints clear in the same way.

I've tried getting the blue to develop with hydrogen peroxide vs. air-dried/naturally developed blue, and the end result is indistinguishable.

The different papers I've tested produce some differences in dmax, but curve shape is overall similar. The sagging pattern at the shoulder of some of the curves above is probably due to mottling which some papers are prone to; esp. the Schut Laurier paper tends to give some lighter blotches in the darker areas.

The somewhat long toe in some of the prints (e.g. the "Simili Japon #2" plot above) is probably due to incomplete clearing. I limited wash times to avoid bleaching as a result of the slight alkalinity of our moderately hard tap water.

I have also tried adding citric acid to the sensitizer prior to coating. In large quantities, this produced some instant fogging upon drying the paper, but this was below the dmin threshold and washed out in processing entirely. In smaller quantities, there was no apparent effect on the printing qualities or curve.

It's kind of puzzling, since Ware notes an exposure range of around 1.8logE for a full tonal scale print, and I get a full range from dmin to dmax on the paper with only around 0.6logE exposure range. In fact, the curve I get with the New chemistry is steeper than with Classic!
View attachment 368504

I'm sure I must be doing something phenomenally stupid, but I'm evidently overlooking it.

If I understand/recollect correctly, two requisites for getting the most out of New cyqnotype are - unbuffered (or neutralized) paper and "development"
with stronger mineral acids like HNO3 or HCl. Otherwise the Prussian blue formation is not as efficient. There is a thread on this subject somewhere.

:Niranjan.

 

[koraks]

Thanks; I'm using the same papers as I did 'back then' and I also never used strong mineral acids formerly, so I'm kind of puzzled as to this outcome. I may try some more papers, neutralize some paper and give HCl and/or HNO3 a try.

 

[koraks]

"development"
with stronger mineral acids like HNO3 or HCl

This made a difference!
I tried acidifying the paper with some citric acid, but that didn't do the trick.
I also tried 'developing' the prints in tap water with a tiny bit of HCl added to it - and this made a world of difference. The effect is the same on acidified and non-acidified papers (at least the ones I'm testing with), suggesting it's really the HCl in the tap water that makes all the difference.

It looks like I now get the 11-12 0.15logD steps or so you'd expect. Overall print quality is a whole lot better, too, with smoother tonality, less blotchiness etc.

Weird - I'm quite sure I never used HCl or HNO3 'back then'.

 

[nmp]

Weird - I'm quite sure I never used HCl or HNO3 'back then'.

Could be the lack of measurements then vs now....

 

[PGum]

Another thing to check against is the use of the hair dryer. Maybe allotting more time for the sensitizer to sink into the paper and dry naturally will provide better tonality.

 

[sojournermike]

Just a note to say thanks for this thread. I’ve been planning to have another go at cyanotype soon. This is useful.

Mike

 

[bernard_L]

Another thing to check against is the use of the hair dryer. Maybe allotting more time for the sensitizer to sink into the paper and dry naturally will provide better tonality.

The wetting agent (Tween 20 or other) is supposed to also help in that respect.

 

[koraks]

Another thing to check against is the use of the hair dryer. Maybe allotting more time for the sensitizer to sink into the paper and dry naturally will provide better tonality.

I see what you mean, and also @bernard_L. I do give the sensitizer sufficient time to soak into the paper before drying it and I work with papers that differ quite a lot in terms of absorbency to begin with; some really soak it up almost instantly (Schut Salland). The issue seems to resolve itself with the use of a strong acid rinse.

Could be the lack of measurements then vs now....

There's that - although I distinctly recall that I got a far longer tonal scale than with classic cyanotype, whereas the affected prints I reported on actually have a shorter tonal scale than the classic formula. So I'm really quite sure that it worked differently back then. I must have used different papers; indeed one of the papers I used to use a lot, I'm no longer using, so perhaps I misremembered using some of the papers back then for this process that I still use today.

 

[koraks]

To follow up on this/bring closure, some more curves to illustrate.

First, this is Classic cyanotype vs. New ("New (no HCl)"), with the New print washed in tap water with a little citric acid. This gives the very short tonal range that dumbfounded me:

Note how the 'New' curve is actually quite a nice curve, but it's even (slightly) steeper than Classic!

Here's the Classic curve compared to New cyanotype, this time with the New print washed in tap water with a little HCl added to it:

Note how the New curve is now much more stretched out and gradual, and covers an exposure range of roughly the 1.8logE that Mike Ware also notes in his process description.

I'm not sure if I'll proceed with the experiments; all I wanted to do was have a look at the difference between the Classic and the New curves and I've got that in the image above. It's a nice blue, but still..you know, blue. I know toning is an option and I've done quite a bit of it in the past, but I've never gotten anything that really pleased me.

This is the test print I made the curve above from; it's on Schut Salland paper, a 50% cotton rag paper. It's a very absorbent paper that has a cottony feel to it and is dead-matte due to its soft, fibrous texture.

This is an identically processed print on Simili Japon, from the same Dutch paper mill; this is a 100% rag paper that has a decidedly warm/yellowish paper base. It's hot-pressed, has a very slight gloss as a result and doesn't soak up sensitizer as readily as the paper above; it does give slightly higher dmax due to the sensitizer remaining mostly on top of the paper instead of being soaked into it:

The prints above are both 'developed' in a (very, very) weak HCl solution; compare with a print (on Simili Japon) that was soaked in tap water:

(Ignore the paper base looking more yellow here; it's mostly due to automatic adjustments done by the scanning software)
Note the much harsher tonality that's really more like Classic cyanotype.

The only difference between both Simili Japon prints above is the 'development'/first wash.

 

[nmp]

Very dramatic demonstration!

Interesting that the New cyanotype demonstrates significant solarization fairly quickly after reaching the Dmax while the Classic seems to be still going up. Could you have added exposure to it to get better Dmax and/or move the middle/lower part of the curve somewhat upwards increasing the DR in the process?

:Niranjan.

 

[koraks]

I'm a bit hesitant drawing conclusions about what happens at the top of the curve, especially w.r.t. the solarization part. Visually, it doesn't really look that way, but the density differences are marginal to begin with. In any case, measurement issues can play a role there; it's kind of hard to accurately hit the steps with the highest density, and in dmax, you actually see some density variation as a result of coating variations anyway.

No doubt the density of classic could be increased a tiny bit with more exposure, but it won't be much. Coincidentally, while rummaging through an old box, I found my very first classic cyanotype tests I did a decade or so ago. It included an exposure vs. dmax test and it very clearly showed only marginal gains beyond a certain point. Measurable, perhaps, but not practically meaningful.