Hardening gelatin with eosyn and TEA

koraks · Mar 29, 2025

I'm experimenting with this and getting somewhere, but not quite where I'd like to be, yet. Specifically, I feel I need too much TEA to get the job done.

Take for instance an emulsion composed of:

Gelatin 20% w/v
TEA 20% w/v
Eosyn 1% w/v solution, 4% thereof (so effectively 0.04% w/v eosyn)
Some pigment; I use 1% Pb15:3 and then something like 20% thereof

Swell, melt and brush onto paper. Dry. Expose to intense light in visible spectrum (peak sensitivity at ca. 550nm) for a few minutes; I use a bright LED desk lamp for 10 minutes or so. Direct sunlight is also great. Then melt unexposed gelatin away in warm water. During the exposure, most of the exposed eosyn will turn colorless (or just break down, IDK).

So in principle this works. Now, what I don't like about this is how much TEA it takes. Currently I'm using a 1:1 ratio of gelatin:TEA, and if I reduce the TEA by even 50%, the whole thing breaks down. I can't imagine it should take this much TEA to get the job done.

Can anyone explain in layman's terms what the hardening process is in this case, and what factors may likely improve overall efficiency?
Thanks to @Cor for providing me with some eosyn to play with!

PS: I can find one mention on Photrio of this here: https://www.photrio.com/forum/threads/getting-dichromates-in-scandinavia-norway.142488/post-1894532
A more recent mention was on the Carbon groups.io, here: https://groups.io/g/carbon/topic/das_sensitized_gelatin/110999961
I suspect it's the same 'jim' in both cases.

koraks · Mar 30, 2025

koraks said:
Can anyone explain in layman's terms what the hardening process is in this case, and what factors may likely improve overall efficiency?

Answering part of my own question using this source: https://pmc.ncbi.nlm.nih.gov/articles/PMC5614854/

Visible light excites Eosin Y from the ground state into a triplet state. This then extracts hydrogen atoms from amine-functionalized co-initiators, such as TEA. The deprotonated TEA radical then initiates the formation of a radical center on the methacryloyl groups of GelMA.

The problem here of course is that they're not discussing gelatin as such, but methacryloylated gelatin. In this case, it's not really the gelatin that does the hardening - it's really the MA functional group added to it. The idea of GelMA is to use gelatin for its biocompatibility (i.e. it works nicely in the human body), but to add a group to it that modifies its crosslinking behavior so non-toxic agents can be used. If gelatin would crosslink particularly well without those groups, I'm sure they would use it that way.

So I'm starting to wonder if this route will work in the first place for printing. Then again, an initial experiment does show some promising results. Here's a snap of one of the tests I did yesterday - this was by far the most convincing one:

Exposure was in direct sunlight for something like half an hour. This was sufficient to also destroy the dye in the exposed areas, while the dye in the unexposed areas washed out quite well.

I'm still puzzled by what I read elsewhere about very short exposure times; this is certainly something I've not been able to replicate so far: https://groups.io/g/carbon/topic/carbrontransfer_re_non/9640857

Concerning this:

koraks said:
Currently I'm using a 1:1 ratio of gelatin:TEA, and if I reduce the TEA by even 50%, the whole thing breaks down. I can't imagine it should take this much TEA to get the job done.

I've now found the original message by 'Jim' (James?) that gives a ballpark for the amounts he uses:

It looks like a gram of dye will make 100 milliliters of sensitizer solution, and about 5 milliliters are needed for sensitising 100 milliliters of gelatin solution along with 20 milliliters of triethanolamine 1% solution, so a little goes a long way.

(the groups.io thing is an absolute nightmare to navigate, especially the archived threads...)

Working this into a ratio like I did above, it seems it would go something like this, assuming his gelatin load is around 10% w/v:

Gelatin 10% w/v
Eosin 1% w/v thereof 5% v/v, so effectively around 0.05% w/v
TEA 1% thereof 20%, so effectively around 0.2% w/v

Let's calculate this back to ratios per gelatin weight:

(w/w or v/w to dry gelatin)	My random guess	Jim's 2016 ratios
Eosin 1%	0.2%	0.5%
TEA	100%	2%

So he uses more eosin, but in particular much, much less TEA.
A similar exercise was done by a discussant ('Rudy') here: https://groups.io/g/carbon/message/7574
He concluded Jim's mix is 31mg eosin to 0.1% TEA in a 100-125ml volume, which makes 0.1g TEA and 32mg eosin. But Rudy also notes that in an academic article, the optimal ratio for hardening PEG was 0.01mmol eosin and 0.1%TEA.

Looking at another reference provided by Jim, I find this:

20% gelatin with 0.0289 mM Eosin Y and 90 mM TEOA

OK, that actually talks about gelatin and mentions specific amounts - yay!
eosin = ca. 692g/mol, so 0.0289mmol = 20mg/liter, or 0.2mg/g gelatin (20% w/v gelatin = 200g/liter)
TEOA = 149 g/mol, so 90mmol = 13.41g/liter, or 0.067g/g gelatin

koraks · Mar 31, 2025

After having run some more experiments, I've not made much progress. Some observations:

It's hard to figure out exactly what "Jim" did (i.e. which exact ratios he used) and what results he got. There are two sets of TEA:eosin ratios that he may have used, and he mentions a 1-minute exposure time (which he believed was overexposed). None of the ratios I've tested give any hardening whatsoever with a 1-minute exposure. The best hardening I obtained was with the dramatically high 1:1 ratio of gelatin:TEA with also a fairly large amount of eosin. Even then, a 1-minute exposure time seems exceedingly optimistic - and it's certainly not convenient or economic.

Moreover, reading the literature some more, much of it focuses on hardening not gelatin, but GelMA, i.e. modified gelatin with a methacryloyl group attached to it. In those cases, it's actually that group that is involved in the crosslinking. The gelatin is just used as a backbone for biocompatibility. Had it been possible to effectively and swiftly harden unmodified gelatin using dyes like eosin, surely, the biomedics would have opted for that route?

To be sure, I've tested also a porcine gel just to make sure that for some freak reason, it doesn't work with a bovine gel. No luck and no difference noted.

I also note that the other guy on groups.io didn't seem to have been able to replicate Jim's success. I did find one mention of @keesbran who noted that a student of his had been experimenting with riboflavin in a similar way, which did allow for hardening, but apparently also very, very slowly. That aligns with my experience so far.

Mind you, I can sort of see how the eosin-approach might work for e.g. hardening final support papers for a single transfer dichromate-based process. You could indeed sensitize the gelatin with eosin and TEA, then hang them up for a day or so during which they will effectively harden. If there's an excess of TEA, this will allow all the eosin to decolorize (the TEA is essential for this!) and you end up with no dye stain. What the TEA may or may not do in the long term to the stability of the gelatin, I really don't know. But eosin + TEA might be a feasible (non-toxic, eco-friendly) alternative to e.g. formalin, chrome alum, glutaraldehyde etc. for this kind of hardening.

For actual printing/image formation...I'm far from convinced at this point.

To be continued...maybe?

imgprojts · Jun 19, 2025

This is very interesting Koraks. I literally just learned about this and once again Koraks is somehow trying it too lol.

koraks · Jun 19, 2025

Well, more accurately: I was trying it - but frankly, I've given up for now. If you make any progress on this front, I'd be happy to hear about it! My conclusion so far is that it'll work, but relies on the use of GelMA instead of plain gelatin to make the process somewhat efficient. I'd be very happy to be proven wrong... I do expect that plenty of other molecules will work similarly to GelMA, but my interest has been mostly (exclusively) in finding something that works for plain old gelatin.

imgprojts · Jun 19, 2025

For what its worth, I read that this particular system works best with green light where as if you switch to riboflavin as the photoinitiator then it will work with UVA and blue light best.

I've been busy looking at PVA + sodium benzoate + ferric ammonium oxalate. Its really working! But I bet it would work with riboflavin+ TEOA.

What I've noticed so far though is that gum will always have this inferior resolution compared to gelatine because you loose the highlights when developing. However, I wonder if I could expose the PVA, then stick it to gelatine or a gelatin coated paper and finally develop it in cold water. I noticed that I need to spray it quickly with dilute peroxide before dunking it into dilute citric acid for developing. So why not introduce the acid in clear gelatin and bond the gelatin tissue to the exposed PVA. Finally spray the sandwich with peroxide, wait and then dissolve in cold water. Since neither the gelatin nor the hardened PVA will easily dissolve, we should see a much better image than via the normal development.

Anyway, that's how I stumbled on this thread. PVA + FAO + SB works with a 3 minute exposure on a 50W 380nm light. However it translates to 15 to 30 minutes on my DLP projector. I wanted to see if there was something more sensitive. So riboflavin+TEOA seems to might do the job? I gotta get some of those things to try.

koraks · Jun 20, 2025

imgprojts said:
For what its worth, I read that this particular system works best with green light where as if you switch to riboflavin as the photoinitiator then it will work with UVA and blue light best.

Yes, I'm aware of that. I tested with white light and in any case an ample amount of whatever would have been needed, including sunlight. Even at very high loads of initiator the process was slooooow. I didn't see anything that resembled the super fast, super efficient process that was suggested on groups.io.

imgprojts said:
I've been busy looking at PVA + sodium benzoate + ferric ammonium oxalate.

I've been following your progress with half an eye; not every detail, but I like to skim through your posts and see where you're going. Good stuff!

imgprojts said:
What I've noticed so far though is that gum will always have this inferior resolution compared to gelatine because you loose the highlights when developing.

Depends; you're aware of Calvin Grier's work. I think that demonstrates it's possible. But that's all with a halftone screen approach. Not continuous tone. Continuous tone is inherently problematic with the pigment processes. There's really no good way around that. You can sort of optimize within a certain process window, but the problem of 'tonal threshold' (Calvin's term) will always be there. But under certain conditions you can push that threshold down far enough to make an image work.

imgprojts said:
o riboflavin+TEOA seems to might do the job?

Well, probably with GelMA, but I've not tried that. I've no idea if it'll work with PVA as I've not tried that, either. I'd be most interested to see if you managed to get some results that way!

imgprojts · Jun 20, 2025

What is missing is the monomer:

PVA + riboflavin + amine + acrylate

There's an easy to obtain set of acrylates. Those used for nail polishing/painting. MMA, EMA, HEMA, and unrelated PEGDA from bio research use. Supposedly, in agreement with what you already stated, the PVA, CMC or Gelatin is just along for the ride. Those things lack the active groups that get fused together by the photoinitiators. But when mixed with material that has the functional groups then you get the long chains trapped in the crosslinking of the active material.....that's another $20 bucks to try. Man this hobby is expensive and so far not really rewarding.... Except I have no cancer yet.

Hologram2 · Jun 21, 2025

imgprojts said:
What is missing is the monomer:

PVA + riboflavin + amine + acrylate

expensive and so far not really rewarding.... Except I have no cancer yet.

Quite right. I guess the whole thing is related to what is known as "dye-sensitized photopolymerization". You may look for the groundbreaking work done by the Osters (Gerald and Gisela Oster).

koraks · Jun 21, 2025

Hologram2 said:
Quite right. I guess the whole thing is related to what is known as "dye-sensitized photopolymerization". You may look for the groundbreaking work done by the Osters (Gerald and Gisela Oster).

Thanks for the reference. You seem to be aware of this field; if I may ask, are you familiar with any combinations that involve dyes as photon capture 'devices' and gelatin as the polymerization agent?

imgprojts · Jun 21, 2025

koraks said:
Thanks for the reference. You seem to be aware of this field; if I may ask, are you familiar with any combinations that involve dyes as photon capture 'devices' and gelatin as the polymerization agent?

My god... https://patents.google.com/patent/DE102004030019A1/en

Could these inventors discombobulate my brains after reading thru?

This one is peaking my interest: US3097097A 1959-02-12 1963-07-09 Gisela K Oster Photo degrading of gel systems and photographic production of reliefs therewith

But otherwise the reference list in every one of these patents is like a mental minefield.

Hologram2 · Sunday at 4:21 AM

koraks said:
Thanks for the reference. You seem to be aware of this field; if I may ask, are you familiar with any combinations that involve dyes as photon capture 'devices' and gelatin as the polymerization agent?

I am not aware of any substantial "hardening" (crosslinking) action on gelatin by the addition of a light-sensitive dye. The Osters had a patent (US 3097096) involving thiolated gelatin + dye to make a photographic image.

Otherwise, most imaging system based on gelatin + dye rely on the addition of monomers, dichromates, or ferric components.Diazos and perhaps, SBQ might be an exception though. There is a Chinese patent (CN112876691A) describing the preparation of a light-curable gelatin + SBQ.

koraks · Sunday at 4:24 AM

Thanks @Hologram2; this confirms what I knew, just wanted to check if I had overlooked something.

Hologram2 · Sunday at 4:24 AM

imgprojts said:
My god... https://patents.google.com/patent/DE102004030019A1/en

Could these inventors discombobulate my brains after reading thru?

With this kind of photopolymers it becomes difficult to produce a surface relief. Getting those chemicals may become a challenge. Moreover, after the light exposure you are ending up with a transparent layer.

Instead you may look for one of the easily available photopolymers used for 3d printing. They have become very cheap. Some of them allow for removal of the unreacted monomers by a simple water wash. I have a feeling some of them could be tuned into something "photographic".

imgprojts said:
This one is peaking my interest: US3097097A 1959-02-12 1963-07-09 Gisela K Oster Photo degrading of gel systems and photographic production of reliefs therewith

US 3097096 might also be worth reading-

koraks · Sunday at 4:49 AM

Hologram2 said:
Instead you may look for one of the easily available photopolymers used for 3d printing.

You mean the kind used for "resin" printing? I like that idea. In a similar vein - how about the initiator/polymer systems used for UV curing inkjet ink? I've not looked into the chemistry of those, but availability of the materials sould be pretty good provided we can somehow tap into the industrial supply chain. There may be environmental issues though...

Hologram2 · Monday at 3:50 AM

koraks said:
You mean the kind used for "resin" printing? I like that idea. In a similar vein - how about the initiator/polymer systems used for UV curing inkjet ink? I've not looked into the chemistry of those, but availability of the materials sould be pretty good provided we can somehow tap into the industrial supply chain. There may be environmental issues though...

Thanks very much for pointing to the UV cured inkjet inks. I had completely forgotten about them. Is there a way for the non-professional user to obtain them?

I have just had a quick look at some recent patents on inkjet inks. Some of the formulas look very similar to the 3d resins. Of course viscosity is much lower for the inks. From what I gathered the overwhelming majority of these materials are sensitized to UV-A radiation. That makes perfectly sense, since UV curing will be faster generally speaking. As for the 3d resins, particularly those aiming at the home consumer market, there is a trend toward 405nm and slightly above. I guess this is largely due to safety reasons.

Hologram2 · Monday at 3:52 AM

Regarding gelatin , I forgot to mention US 3515551:

"The ability of an enzyme to hydrolyze protein can be altered upon exposure to light in the presence of a photosensitive material, thus permitting the production of photographic images. The enzyme can be converted from an active state to an inactive state, or vice versa. After exposure, development with water hdydrolyzes and removes the protein from those areas where the enzyme is present in the active state."

It relies on dyes like fluorescein, rose bengal, erythrosin, eosin, riboflavin, methylene blue, thionine, acridine, acriflavine etc.

koraks · Monday at 4:35 AM

Hologram2 said:
Thanks very much for pointing to the UV cured inkjet inks. I had completely forgotten about them. Is there a way for the non-professional user to obtain them?

I've not looked into that yet; it's on my to-do list, but admittedly it's pretty far down.

Hologram2 said:
From what I gathered the overwhelming majority of these materials are sensitized to UV-A radiation.

Yes, definitely UV; 365nm is popular right now. I think a major benefit is also that this means the ink can be handled fairly well under artificial light without too much issue. I found the eosin approach quite interesting but also quickly realized it's kind of a b*tch to work with a sensitizer that's sensitive to visible light!

Hologram2 said:
there is a trend toward 405nm and slightly above. I guess this is largely due to safety reasons.

Electrical efficiency and cost play a role, too. Quantum efficiency of LEDs isn't too great at smaller wavelengths - which is often offset by the significantly higher sensitivity of initiators to such wavelengths, so sometimes/often it balances it when it comes to 'hardening power per dollar'. Still, LEDs are about 80% cheaper or so at around 400nm than around 360nm, so there's something to be said to make a photo initiator with a peak sensitivity around 400nm.

PGum · Monday at 10:16 PM

The problem with UV curable inkjet inks as that these are mostly if not all solvent-based, where there are plenty of non-polar and resin-specific photoinitiators to choose from. Irgacure, TPO etc… Some of these can be dispersed as insoluble nano particles to harden functionalized polymers in a water dispersion but not in plain old gelatin, atleast not the mass produced ones. There may be some biomedical specialty initiators that could harden gelatin, but priced prohibitively.

Riboflavin does a pretty good job at UV hardening protein colloids like gelatin but then that gelatin needs to be in gel form, where the moisture can mobilize the free radicals, not dried or it works quite poorly.

imgprojts · Tuesday at 2:19 AM

I've been playing around with this recipe this past week. I'm using PVA and CMC for now. I was able to get an image using clear-ish 3D printing resin as the monomer. The process needs a lot of work but its definitely possible. https://www.photrio.com/forum/threads/spp-saidanes-print-process.213939/post-2906650 I think it should work just fine for gelatin. With PVA the polymer activator is pretty potent. Like only 5 seconds to fully cure the entire load of ink with very heavy pigment loading. I've created two images so far but they are no-where decent. both are faded and overexposed. I am using a visible light projector. It may be easier if used with a transparency/physical negative. I'm skeptical as to what is causing the formula to work so I may do some exploration removing parts of it to see what brakes.

Hologram2 · Tuesday at 5:51 AM

imgprojts said:
I've been playing around with this recipe this past week. I'm using PVA and CMC for now. I was able to get an image using clear-ish 3D printing resin as the monomer. The process needs a lot of work but its definitely possible. https://www.photrio.com/forum/threads/spp-saidanes-print-process.213939/post-2906650

This is very interesting!

imgprojts said:
I think it should work just fine for gelatin.

It surely does, at least for the recording of holograms. When I started the thing (www.holographyforum.org/forum/viewtopic.php?t=7636) I was amazed to make holograms that looked exactly like dichromated gelatin holograms (DCG).

I meanwhile developed it a bit further. Adding a monomer (methylene bisacrylamide) turned it into a nearly high-speed material, not far from AgX. Moreover, it was possible to spectrally sensitize it from 400 - 660nm.

imgprojts · Tuesday at 9:14 AM

Hologram2 said:
This is very interesting!

It surely does, at least for the recording of holograms. When I started the thing (www.holographyforum.org/forum/viewtopic.php?t=7636) I was amazed to make holograms that looked exactly like dichromated gelatin holograms (DCG).

I meanwhile developed it a bit further. Adding a monomer (methylene bisacrylamide) turned it into a nearly high-speed material, not far from AgX. Moreover, it was possible to spectrally sensitize it from 400 - 660nm.

Dude, the Thursday before I started testing I read thru your posts in that forum. I got excited about it but did want the option of something not activated by normal light. Plus it had to be cheap as heck. I guess I'll try the gelatin version. I'm always switching between gelatin thick tissues and gum thin. Haven't found something that really let's me get into the photo making part. Not yet.

imgprojts · Thursday at 10:06 AM

I tried the same activator I used for PVA on a Knox Gelatin sample and it did work. I tried an image but screwed up on the transfer. I exposed for 15 minutes on my visible light projector and obtained something that was over exposed. I made two simple samples and exposed one that was still wet to UV light 365nm for 1 minute. I washed both in cold water and nothing moved. Then I moved to warm water and the unexposed dissolved while the exposed did not. So I think it would work. I found a paper that describes the rate of polymerization of Riboflavin (RF) on HEMA. Accordingly, higher PH makes it faster, lower viscosity (liquid) makes it faster. https://pmc.ncbi.nlm.nih.gov/articles/PMC3794564/pdf/ISRN.PHARMACEUTICS2013-958712.pdf That same group has a second paper with RF polymerizing other substances where they put together rate to proportion correlations. Accordingly the amount of monomer has a quadratic proportion to the polymerization rate where as the sensitizer can either increase it up to a critical point and then decrease it. So the amount of polymer is the best way to control the reaction.

imgprojts · Thursday at 10:10 AM

This is what I tried:

My tissue formula (I don't think it matters much what concentrations you use)
60ml water level
4.8g gelatin
10ml pigment poster paint + mica + carbon mix
2g sugar

2 spoons of monomer
3spoons of activator
1spoon FAO

Activator: (almost bottom of the barrel dilute per some tests I did on PVA)
50ml water (less water will make it stronger, towards 1ml you will get paper yellowing under prolonged UV light)
• Riboflavin – .01 g turns liquid orange yellow but clear. hard to mix
• FAO – 0.14 g
Ph measures ~5 keep Ph at around 4
• TEOA – 0.49g - liquid becomes dark orange and has Ph ~9
Citric acid – 0.65 g -liquid is light yellow and has a Ph ~4

Hologram2 · 2025-06-27T09:34:01-0400

imgprojts said:
I tried the same activator I used for PVA on a Knox Gelatin sample and it did work. I tried an image but screwed up on the transfer. I exposed for 15 minutes on my visible light projector and obtained something that was over exposed. I made two simple samples and exposed one that was still wet to UV light 365nm for 1 minute. I washed both in cold water and nothing moved. Then I moved to warm water and the unexposed dissolved while the exposed did not. So I think it would work.

That is pretty amazing.
HEMA would be very convenient - not that toxic and (relatively) easily available, I believe.
Thanks for pointing to that paper.

Hardening gelatin with eosyn and TEA

koraks

Moderator

koraks

Moderator

koraks

Moderator

imgprojts

koraks

Moderator

imgprojts

koraks

Moderator

imgprojts

Hologram2

koraks

Moderator

imgprojts

Hologram2

koraks

Moderator

Hologram2

koraks

Moderator

Hologram2

Hologram2

koraks

Moderator

PGum

imgprojts

Hologram2

imgprojts

imgprojts

imgprojts

Hologram2

Hardening gelatin with eosyn and TEA

Moderator

Moderator

Moderator

Moderator

Moderator

Moderator

Moderator

Moderator

Moderator

Privacy & Transparency

Privacy & Transparency