All things Chiba and Dichromate alternatives.

[imgprojts]

---

I had the day off today, so I spent some time experimenting with carrageenan and reading about the success of the fish glue Chiba process. That got me thinking it might be worth starting a post dedicated entirely to Chiba methods. There’s a lot to explore, especially about avoiding dichromates and finding alternative materials that respond to other sensitizers. I had AI format this to make it more readable, I am an engineer and a horrible writer.

Here’s a list of materials I’ve tried that can form images in a gum-like process. For all of these, I’ve used safe sensitizers such as ferric ammonium citrate and ferric ammonium oxalate.
PVA SBQ seems to behave most like gum. SBQ is pricey, but you don’t need much of it, and it’s not carcinogenic. The SBQ process is chemically different from Chiba, though.

Processes that need thermal control during hydrogen peroxide development​

• Fish glue – Has produced some impressive images. I haven't tried it
• Rabbit glue
• Gelatin – Your house will smell like dead animal and you'll have this hint of dead animal with you at work if you don't use gloves. I love jello and I expected so much better.
• CMC (carboxymethyl cellulose) – Must stay above 20 °C at all times per Habib Saidane, which is easy in warm climates but tricky here in Washington State for most of the year.
• Iota carrageenan + gelatin – Worked when mixed and exposed in cold as well as warm water. Several papers suggest that lambda and kappa carrageenans also respond to UV with Fe³⁺ salt

Processes that develop at room temperature (near 20–25 °C)​

• PVA – Works well, you add citric acid and keep the temperature below 40 °C. Produces consistent images, though tonal range is limited. Cyanotype also develops nicely in it.
• PVA + CMC – Same as above: add citric acid and stay below 40 °C or all of it will harden and you'll get no image.
• PVA + Sodium Benzoate and citric acid forms a white precipitate gel that is great for mixing and paintting using a brush. Cyanotype also develops nicely in it.
• PVA + dilute boric acid – This follows the classic slime formula. A 10–20% boric acid solution in water works; higher concentrations cause precipitation.
• Iota carrageenan + gelatin – Works when mixed and exposed in cold or warm water.
• PVA mixed into carrageenan gelatin – Foams heavily and the foam doesn’t dissipate quickly (or ever). I lack the patience for that, but once dry it does form an image.

Things that don’t work​

• Ferric ammonium oxalate + gelatin, causes problems such as separating from the pigment.
• Heating PVA+Citric acid crosslinks PVA so you can't try it with a heat gun or develop it in hot water.
• Developing in water only. My water is alkaline and usually just washes off images on command.
• Gelatin and PVA don’t mix well, also separate from the pigment.
• PVA + CMC doesn’t hold pigments in suspension. The pigments settle in the pits of cold press paper after a few minutes. But it is good as a sizing because it doesn't dissolve very easily.

I recently found a great paper source: the Prudiut Store on Amazon sells 300-sheet packs of A4 300 GSM 100% cotton paper for about $58. It doesn’t need acidification for cyanotype or Chiba. It just works. They ship cheaply, so expect some corner damage (mine arrived fine, though). The box even came with a bonus bag of air, which I found amusing. like the packer thought, “You look like you could use this balloon!”

For all these processes, you need to size your paper and apply a clear coat after each layer. Otherwise, the layer underneath can re-dissolve. I use the same PVA or PVA–CMC mix for both sizing and overcoating. These aren’t sensitized; just let them dry and apply the next layer quickly. Be gentle during washing, too much agitation will destroy the image. I’ve found it best to lay the paper face-down in water after peroxide treatment and leave it still. PVA, once hardened properly, can take ages to dissolve, so you can rinse it with hot water or even under the faucet without damage. The resulting image has a rubbery feel wet and a chalky mat feel when dry. You can even spray with boric acid after development and heat it to lock the image more permanently.

Other notes​

• At 365 nm and lower wavelengths, sodium benzoate works as a photoinitiator. At longer wavelengths, it doesn’t do much.
• Ammonium persulfate behaves similarly and has some effects that I have not explored yet. Its supposed to develop the image as it is exposed, needing only water afterwards but that was not the case when I tried.
• I also tried a TEOA + riboflavin dye mix with little success.
• You can mix any of the gels above with small amounts of clear water-based 3D printer resin and get an image, but I wouldn’t recommend it. The chemicals are carcinogenic, which defeats the purpose of finding safer alternatives to dichromates. Additionally they smell.

Pigments tested​

• Failures:
• Poster paint (chalk-containing types fail completely)
• Acrylic paint (no luck)
• Successes:
• Graphite powder – very gritty
• India ink – stains paper if unsized
• Mica powders – including fluorescent ones (they fog exposure a bit, but look beautiful under UV)
• Watercolors – best overall results

Always use proper PPE and wear orange or red safety glasses! Studies show blue light is most associated with cataract formation, and UV will damage your eyes too. I tested with 365 nm UV lamps, and with 385nm using a UV projector. For negative exposures printed on inkjet transparency, anything between 365 nm and 400 nm should work. I’ve tried 395 nm and even blue light both are slower, but effective. For projection setups, 385 nm is the sweet spot since it penetrates better than 365 nm while staying within a useful range. My current setup uses a 50 W 385 nm LED (about 10 mm square), and it handles the job nicely.

Overall, this mix of safer sensitizers and alternative binders has opened up a lot of room for experimentation. The Chiba-inspired processes might not replace dichromate-based gum completely, but they’re getting close—and they’re much safer to work with. If anything this has become a passtime with no end in sight. I could probably continue experimenting until my grand kids pry my arthritic body parts off my chemical bottles. I got no grandkids yet.

 

[imgprojts]

Okay here's the gum cunundrum that is hard for me to understand... Why doesnt it just float away?

I use cold press paper because it has a texture to it which then holds the emulsion. Under the emulsion you need something to prevent the inks or pigments from staining the paper. So I use pva or pva mixed with cmc. Both immediately start dissolving in water. But thats ok if your emulsion is hardened over it. Others use gelatine or alum gelatine or gluteraldehide hardened gelatine. So okay I get it, dont let the pigment in.

With that in mind I proceede to get polypropylene synthetic paper /Yupo. I coated it in pva, pva cmc, gelatine, gelatine + boric + ammonium persulfate. Naively thinking that on a very nice flat surface I should get good coatings that easily dissolve. But nope!

When developing, I can see the emulsion harden up and then float away as a very thin pealed skin. The things I tried didnt work to retain the pva image.

Yet somehow the same thing works great on transoarencies using cyanotype formula as the sensitizer for the pva.

The gelatine one appears to work best. The stuff stays stuck and not dissolved. I bet that exposing from the back may result in better images. I dont understand how an image could even form if the light hardening the gelarine is on the surface hardening the material there and working its way down to the paper. Technically the image is only attached where it crosslinked all the way thru. Everywhere else it just hardens and floats away.

In this image, the center graphic shows what I mean. If the material is water sokuble, then the hardened layers should just float away if theres nothing to anchor it to the emulsion. So maybe there really isnt a sizing that works and the anchors are just the paper fibers.

 

[koraks]

What you're describing above is the exact reason why carbon transfer was invented after Poitevin ran into the same problems as you do, now. The trick behind carbon transfer is that even a thin layer of hardened gelatin on top of the tissue will end up in the print because that is transferred to a new support, where it in fact becomes the bottom layer. We're talking about the period 1859-1864 now.

So this is a fundamental problem of direct pigment processes; i.e. processes where hardening of colloids or other polymers is used as the image-making concept. There are a few workarounds, but they all have their drawbacks:
* use a very thin layer so there's always adherence to the base support. The drawback is that it'll become an extremely high-contrast medium and you get virtually no higher midtones.
* use a support that's sufficiently coarse (like the cold-pressed paper) so that the thickness of the layer is variable and you get at least some adherence here and there. The result is coarse/grainy tonality.
* use a halftone printing approach where you limit the tone that's created to just black and white, and construct intermediate tones by varying dot size (AM) or dot frequency (FM).
* exposing from the back does indeed work as well, but evidently not very well with Yupo as it's virtually opaque; it does work well with transparent media, as long as they are (1) thin and (2) the light is highly collimated. Otherwise you get a very fuzzy image.

So, to answer your question:

Why doesnt it just float away?

It does float away, but the precise thickness of the layer that can survive (Calvin Grier calls this the 'tonal threshold') depends on a multitude of factors. Under favorable conditions, you can get some gradation; this is the twilight zone that classic gum printing exploits. Note that in classic gum printing, it's customary to stack several layers precisely because of this reason; it's virtually impossible to get a full tonal range from dmax to paper white and good tonality in-between in a single layer.

 

[imgprojts]

Its weird because most paper has visible structure that is much larger than pigment so any pigmented gel will just soak an dry on relatively flat paper layers.

See, I thought paper would have "hairs" so the gel would harden around those and via that anchor it would survive the development. But I see no hairs, its mostly just a flat vertical stack of long flat strans of cellulose fiber. So then maybe the gum or gel is hardening a little under the top fiber layers? Definetly without some sort of structure below, the exposed image just floats and you're left with perfectly fine images all wrinkled and wrapped into themselves.

A hairy paper shoul give much better results, right? Maybe?. I gotta search on this.

 

[koraks]

See, I thought paper would have "hairs" so the gel would harden around those and via that anchor it would survive the development.

It does exactly that, at least to an extent. Kind of a stochastic thing; it happens in some places, not in others. This would explain why the result is more consistent (but also, consistently worse) on a medium like Yupo.

 

[AntonKL]

I think in the direct pigment processes the pigment in the gum/gelatin acts similar to a half-tone screen too:
When experimenting with brushing charcoal dust on "chiba direct carbon" prints (after the H2O2 but before the hot wash) I tried fully leaving out the pigment from the gelatin so I could skip the clear stain prevention layer. This resulted in way higher contrast and speed. Adding Ti-white or Oxide Yellow as UV absorbing pigment brought down contrast and (sadly) speed again, unlike the UV-transparent/reflectant barium sulfate with which it behaved just like the unpigmented gelatin.

 

[imgprojts]

I think in the direct pigment processes the pigment in the gum/gelatin acts similar to a half-tone screen too:
When experimenting with brushing charcoal dust on "chiba direct carbon" prints (after the H2O2 but before the hot wash) I tried fully leaving out the pigment from the gelatin so I could skip the clear stain prevention layer. This resulted in way higher contrast and speed. Adding Ti-white or Oxide Yellow as UV absorbing pigment brought down contrast and (sadly) speed again, unlike the UV-transparent/reflectant barium sulfate with which it behaved just like the unpigmented gelatin.

So you exposed just gelatin with sensitizer and then pigmented it later?

Today I actually did something similar 1ml 10%gelatin+ .5ml NewCyanotype+ 1ml 0.1% ammonium per sulfate sprayed on flat Yupo. I used ferric ammonium oxalate and this tends to create crystals. I exposed it and then developed with H2O2 spray and washed it water. I did a few other things today with PVA where the pigment is very low because I wanted to see the exposed layer's behavior. indeed the exposed stuff is not dissolved but because it has nothing to grip on it actually floats away before the unexposed areas dissolved. But this gelatin, I noticed yesterday that it does indeed hold on to the surface well. Anyway, so the H2O2 did nothing in terms of magic color change, but the step seems to be crucial to cross-linking the gelatin. During the wash, all the color disappeared in yellow liquid. I washed it until there was no more yellow. I have it sandwiched with PVA on water color paper to see if I can peal it off tomorrow and then dissolve the unhardened gelatin. But look at the image right after exposure with some exposure adjustment in gimp. Vs the original digital image.

I guess my point is that there is definitely a photographic level image in the emulsion after you expose it. Maybe adding the pigment during/after the H2O2 development would actually stick to the gelatin while its cross-linking?

 

[NedL]

... Anyway, so the H2O2 did nothing in terms of magic color change, but the step seems to be crucial to cross-linking the gelatin. .....

It is crucial! The gelatin gets weakly crosslinked only when H2O2 is applied.

By the way, I'm not convinced that the citrate ions in FAC chiba don't play some helpful role too -- I've had better hardening with chiba when I added a little citric acid, with both FAO and FAC, and I suspect there is more to it than just lower pH . I know that citric acid can promote crosslinking with starch and glycerin.. I sometimes prepare paper sized with starch and a little glycerin, and a little citric acid makes it work much better.

Have fun!

 

[imgprojts]

I suspect there is more to it than just lower pH

I totally agree. Adding citric acid chalates Ions which then help crosslink (whatever that means to a chemist). I had one such idea which I tested and seems to do something. Since you can make "slime" by adding boric acid to PVA, I wanted to add boric ions to the PVA to bring it to the brink of crosslinking. I found that you can add up to 20% boric acid to a 5% PVA and it would barely trigger precipitation. If you keep adding 20% BA, the PVA keeps getting diluted so it doesn't gelate. Anything higher than 20% will produce immediate precipiation. Sodium Benzoate + citric acid in PVA automatically makes it a sort of white paste which actually makes really good Cyanotypes if you add NewCyanotype to it. I think there something to the additional of metal Ions for sure.

This is CMC:

Note all the OH groups
Here is Iota Carragenan:

more OH groups
PVA:

Gum arabic:

The OH hydroxilic groups are where the carbon chain join during the process. I went to ask our evil friend RAM brains about it. Here is my summary:

UV exposure in PVA/FAC (ferric ammonium citrate) or FAO (ferric ammonium oxalate) emulsions triggers photoreduction of Fe³⁺ to Fe²⁺, initiating crosslinking with the polymer's OH groups. This mimics gum bichromate but uses iron instead of chromium, hardening the emulsion selectively in exposed areas for photographic prints.

UV light excites the ferric complex ([Fe³⁺-citrate] or [Fe³⁺-oxalate]), reducing it to ferrous Fe²⁺ and releasing CO₂ or other byproducts from ligand decarboxylation. The Fe²⁺ then coordinates with adjacent OH groups on PVA chains (or residual OH in the emulsion), forming Fe²⁺ bridges that link polymer strands via oxidative coupling.

Unexposed Fe³⁺ remains soluble and washes away, while exposed areas are initially lightly crosslinked by Fe²⁺. Adding H₂O₂ reoxidizes Fe²⁺ back to Fe³⁺ (Fe²⁺ + H₂O₂ → Fe³⁺ + OH⁻ + H₂O), tightening the bridges and insolubilizing the PVA matrix permanently. H₂O₂ provides O₂ for oxidation without carbon involvement, unlike the initial UV step.

Supposedly the citric acid helps absorb UV and buffers the PH when ferric is going back to ferrous. Similarly boric bonds are a little stronger so while things are pulled by the ions it pulls polymers together closer than iron so it produces tighter matrices after the peroxide does its thing. No idea if that's true or not.

coincidentally in the random links from Mr RAM brains, it found a Patent for making FAC DIY:

CN101704738A - Production method of ammonium ferric citrate - Google Patents

The invention discloses a production method of ammonium ferric citrate. The ammonium ferric citrate is prepared by water, citric acid and iron powder according to the following steps: 1. placing water and citric acid into a stirring device to be heated to 60 DEG C; 2. opening the stirring...

patents.google.com

 

[AntonKL]

So you exposed just gelatin with sensitizer and then pigmented it later?

Yes but only as an experiment, I got better looking results when dusting onto a regular direct carbon print (with black pigment in the gelatin and the stain prevention layer in place). Also the dusting pigment must not be too fine or it will cake together and prevent proper washing off.

 

[imgprojts]

Yes but only as an experiment, I got better looking results when dusting onto a regular direct carbon print (with black pigment in the gelatin and the stain prevention layer in place). Also the dusting pigment must not be too fine or it will cake together and prevent proper washing off.

Okay, so you apply the H2O2 and then extra pigment. Do you do that as dust on a humid gelatine surface or mix the dust into the H2O2 bath? Habib experimented with pigment in the H2O2 bath if I recall. But thats with a different gel, CMC.

No, actually it was video 3 but unspecified. . Video 2 shows liquid hide glue, so I'm guessing thats what it is.

 

[koraks]

The application of pigment to a gelatin matrix reminds me of bromoil and even more so, its predecessor the oil print: https://en.wikipedia.org/wiki/Oil_print_process

 

[AntonKL]

Okay, so you apply the H2O2 and then extra pigment. Do you do that as dust on a humid gelatine surface or mix the dust into the H2O2 bath? Habib experimented with pigment in the H2O2 bath if I recall. But thats with a different gel, CMC.

I dust it on the humid gelatin yes: After the cold H2O2 bath comes a brief cold wash, then I put it face down on a glass plate and carefully squeeze out the water. An older approach that also works with some gelatin types was just dabbing it dry with kitchen paper.
I also played around with similar wet approaches (with Na.Persulfate) but found them too wasteful for my liking.

The application of pigment to a gelatin matrix reminds me of bromoil and even more so, its predecessor the oil print: https://en.wikipedia.org/wiki/Oil_print_process

Yes, and oil prints also work with chiba system!

And there's also a number of historical dust-on processes that rely on modifying stickiness (rather than solubility). Mostly dichromate based but some used iron. They can give results so good one of them was even commonly used for duplicating negatives for a brief while, but they are also annoyingly complicated and never really caught on outside of that and printing on ceramics or enamel. See ch. 44 in this old german book.
A more recent process, I didn't try it yet, is "New Resinotype" (not the same as the resinotype currently on the alt process subforum) which seems to remove the complications by using acrylic resin rather than dichromate+stuff, it's waterless even, looks pretty easy (and also fits the overall thread topic of avoiding chromate I guess).
(Action starts at 51 minutes, Link taken from here)



Here's some ~10x15cm prints I made with a UV enlarger from 35mm negs with my approach, and the print from the experiment with oxide yellow (~8cm long edge). Note the slight staining - it is a very fine line between getting too much contrast and getting lots of staining.
The white dots are actually the sparkle effect of the charcoal powder I used.

 

[koraks]

Yes, and oil prints also work with chiba system!

Thanks so much, I never realized that, but it makes good sense - albeit that you do indeed need some kind of sealing layer on top of the gelatin as in Chiba, right?

 

[AntonKL]

The manual makes no mention of that, though I haven't done it personally

The Rawlins oil print process – a safer version – AlternativePhotography.com

www.alternativephotography.com

A japanese twitter user beat the author to it, he also uses a UV enlarger and has a lot more patience than I. This account put me on track with the dust-on approach too (he is trying to make it work as a positive process however)

https://nitter.net/Anthracosaurus_/status/1646157046100680704

 

[koraks]

Well, what do you know, it does do something;

There's an odd reversal going on which is probably because the subject didn't receive enough exposure and I suppose the matrix remained so soft in that area that it basically just inked through. This was as quick & dirty a test one could imagine, but I'm quite surprised I actually get something of a recognizable image.
Heck, I didn't even have a proper oil paint so I made one from some pigment and a random bottle of oil...

 

[AntonKL]

Wow that's really cool!
How much gelatin did you use if you got results so quickly? Or did you have gelatin paper lying around? The thickness they use in the article I linked does sound absurdly high.

 

[koraks]

Yeah, I just used some transfer paper I use for final transfer for carbon printing; this is unhardened gelatin since I use it for double transfer. I think this is around 30g/m2 dry gelatin load; I'd have to calculate that back to g/liter but it's probably 60-70% of what's used on the website. Honestly I think the matrix should be thicker, so I might give it another go at some point. It also seems like this negative is just too contrasty; I was surprised at his short exposure times with that face tanner; my exposure with 100W 365nm LED was something like 8 minutes (he mentions 1.5 or so).
Oh, also, I didn't control the FAC load in any way; just squirted some from a pipette onto it, gave it a good brushing and then blasted it they with the hairdryer. The whole process tool something like 15 minutes from reading your link to that photo above.
Btw, I wonder why he doesn't simply incorporate the FAC in the gelatin; it should be quite stable, no?

 

[AntonKL]

Ah that's still a lot more than the 6g/m2 we use with the direct carbon approach, so no quick drying times (but of course it can be prepared in advance just like you did).
With his 60ml 10% / A4 he uses a whooping 96g/m2.

Btw, I wonder why he doesn't simply incorporate the FAC in the gelatin; it should be quite stable, no?

I suppose that's what he and his target audience are used to from the dichromate version.
But also I think I did get negative interactions between the (buffered) paper and the direct carbon coating if I didn't dry it quickly enough so I'd be cautious.