Photochromic or photoreactive autochrome

[jsmoove]

Is it possible to digitally print an autochrome filter/plate such as:
https://cs.nyu.edu/courses/spring16/CSCI-UA.0380-004/slides/09/image-processing.html#/16
using photochromic or photoreactive ink (example: Inkodye) in an inkjet printer,
then expose an image using a projector with a UV bulb/blacklight bulb?
See: http://www.inkodye.com/help/advanced/projector
Would this work? Would each "pixel" of RGB be activated separately or would the UV just expose the entire thing?
I probably have my order mixed up about how the autochrome process works, but im curious to know if this would work at all.

 

[jim10219]

According to their website, inkodye is no longer made. Also, I don't think it would work in an inkjet printer, as ink for inkjets need special magnetic particles mixed in with the ink in order to shoot it out of the print head. Furthermore, a projector or enlarger wouldn't produce enough UV light to expose an image. That's why for UV sensitive processes, people use contact printing methods and special UV lights or the sun. Most projectors don't have the proper hookups for a good UV bulb, and the typical CFL style blacklights don't put out enough UV light to be useful for this type of printing.

I'm not entirely sure of what you're trying to do, but as I understand it, it's probably a dead end. Do some research on cyanotypes and gum bichromates. They may offer some alternative routes to what you're trying to accomplish.

 

[nmp]

Is it possible to digitally print an autochrome filter/plate such as:
https://cs.nyu.edu/courses/spring16/CSCI-UA.0380-004/slides/09/image-processing.html#/16
using photochromic or photoreactive ink (example: Inkodye) in an inkjet printer,
then expose an image using a projector with a UV bulb/blacklight bulb?
See: http://www.inkodye.com/help/advanced/projector
Would this work? Would each "pixel" of RGB be activated separately or would the UV just expose the entire thing?
I probably have my order mixed up about how the autochrome process works, but im curious to know if this would work at all.

Looks like UV curable/developable ink. You have to do blanket UV exposure after printing the random pattern. Enlarger or projector won't work (nor is needed) as the lens system will absorb most of the UV. Just put it out in the sun. If I understand correctly, you want to fill inkjet cartridges with these inks and print the linked pattern on glass/plastic, then activate by exposing to UV. Sounds like a daunting task!

 

[jsmoove]

@jim10219 Oh I didn't notice they weren't making it anymore. I guess Solarfast then. Or any water based light sensitive vat dye. But I guess without the magnetic particles it wouldn't be possible. Photochromic inks are available in inkjet form, but they are reversible not permanent. My goal is to expose a color image on fabric. I don't know if there is such a thing as a one-layer panchromatic liquid light.
@nmp Hey. What do you mean by a blanket exposure? Yes you have it right, but instead of glass or plastic I want it directly on fabric. Are you sure a projector wouldn't work? Maybe a camera obscura then with the image on the outside? Im just curious about doing it the non contact printing way. Im mostly curious to know if the individual pixels would expose or if the whole thing would.
My idea was to project a color image onto a shirt, and the image would be transferred to the shirt. One layer.

 

[nmp]

@jim10219 Oh I didn't notice they weren't making it anymore. I guess Solarfast then. Or any water based light sensitive vat dye. But I guess without the magnetic particles it wouldn't be possible. Photochromic inks are available in inkjet form, but they are reversible not permanent. My goal is to expose a color image on fabric. I don't know if there is such a thing as a one-layer panchromatic liquid light.
@nmp Hey. What do you mean by a blanket exposure? Yes you have it right, but instead of glass or plastic I want it directly on fabric. Are you sure a projector wouldn't work? Maybe a camera obscura then with the image on the outside? Im just curious about doing it the non contact printing way. Im mostly curious to know if the individual pixels would expose or if the whole thing would.
My idea was to project a color image onto a shirt, and the image would be transferred to the shirt. One layer.

I am not sure I agree with Jim that you need magnetic particles in the ink in order to make it inkjet-able. You can pretty much put anything in the cartridge and have it printed. People are hacking printers to print edible food coloring on "frosting sheets" for cakes and stuff.

Regarding the projector/enlarger, I am pretty sure it won't work (as it clearly states in the first link you provide.) If it did, nobody would need to make a contact print for alternative process - just put the negative in the enlarger and shoot. You would need lens made of quartz to make it transparent to UV light, a little too expensive.

By blanket exposure, I meant without any negative.

I am really now quite confused about your perceived process. Can you please describe in steps how and what you are trying to do. Printing on a fabric does not sound like an autochrome.

:Niranjan.

 

[removed account4]

According to their website, inkodye is no longer made.

yup
they are now a packaging company or were doing packaging the last time i was in touch with them ...
i don't think anyone else makes a casava root based photo sensitive dye ..
=
js good luck with your project !

 

[jsmoove]

@nmp Right, but something like a photochromic ink is exposed in about 15 seconds of UV light, so with a projector + UV bulb it might be a lot longer..but not too much longer I would think?
What I was thinking is covering an entire piece of fabric with small tiny RGB colours using a dye (which would be invisible at first) then exposing the entirety of the fabric with a color image.
I probably have it completely wrong though. The autochrome goes in the direction: light-->image-->color grains-->panchromatic emulsion, I think? So is it possible to do UV light-->image-->multicolor photochromic grain?
Im probably confused lol.

 

[nmp]

@nmp Right, but something like a photochromic ink is exposed in about 15 seconds of UV light, so with a projector + UV bulb it might be a lot longer..but not too much longer I would think?
What I was thinking is covering an entire piece of fabric with small tiny RGB colours using a dye (which would be invisible at first) then exposing the entirety of the fabric with a color image.
I probably have it completely wrong though. The autochrome goes in the direction: light-->image-->color grains-->panchromatic emulsion, I think? So is it possible to do UV light-->image-->multicolor photochromic grain?
Im probably confused lol.

No sure if it is still completely clear to me:

1. What is the final product look like? Is it a color image on a tee-shirt, for example?

2. Where does the inkjet printing come in?

3. Do you want to project a color negative or a b&w negative with UV source? Either way you will be projecting a monochrome UV image not a colored one.

4. Photochromic inks are reversible, however the inkodyn like inks are permanent once exposed. Those two will behave differently, so they may not be interchangeable.

5. I am not an expert in autochromes (may be The Photochemist who is will chime in) but what you are describing does not seem to resemble its principle.

So many questions, so little time....

:Niranjan.

 

[jsmoove]

@nmp Yes, just a regular ol' tshirt, with color images...patterns..styles, etc. Or multiple shirts at once depending on how far the projector is.
Sorry by inkjet I mean just any printer that you can load ink into that can handle fabrics.
You'd have a pre-printed shirt that you "activate" with your own projector, becomes visible once exposed.
I think monochrome UV image would work? Because itd just be the intensity of light that would activate the ink "pixels".
Though now that im thinking about it, I think the light would just hit the pixels and youd just get the bunch of pixels instead of your image.
Im unclear if this is true though.
I think something that can go from non-visible to visible permanently would be best, so probs not photochromic, unless an irreversible photochromic. I guess that is a photoreactive or photosensitive then?
The other way I had looked at it is, use the regular autochrome filter, but cover the shirt in a one-layer panchromatic liquid emulsion. Which I am not sure exists.
Definitely curious !

 

[nmp]

@nmp Yes, just a regular ol' tshirt, with color images...patterns..styles, etc. Or multiple shirts at once depending on how far the projector is.
Sorry by inkjet I mean just any printer that you can load ink into that can handle fabrics.
You'd have a pre-printed shirt that you "activate" with your own projector, becomes visible once exposed.
I think monochrome UV image would work? Because itd just be the intensity of light that would activate the ink "pixels".
Though now that im thinking about it, I think the light would just hit the pixels and youd just get the bunch of pixels instead of your image.
Im unclear if this is true though.
I think something that can go from non-visible to visible permanently would be best, so probs not photochromic, unless an irreversible photochromic. I guess that is a photoreactive or photosensitive then?
The other way I had looked at it is, use the regular autochrome filter, but cover the shirt in a one-layer panchromatic liquid emulsion. Which I am not sure exists.
Definitely curious !

If you print that random "noisy" pattern using RGB inks and shoot once thru a negative in a UV projector, all you will get is an image made of the noise, not a colored picture. If you want a colored picture, you have to coat/shoot/develop three times, each for a single color - much like what people do in gum bichromate printing. (Actually that in itself in not a bad idea: print three color image on a tee-shirt with photochromic ink. Indoors/dark you can't see the image, when you go out in the Sun, the image comes out!)

I think you have several different concepts going on at the same time. You might want to simplify and work on one issue at a time. For example, if you want to check if the UV projection works, dab an ink (whether photochrome or photoreactive) on a piece of paper and project thru a slide or negative and see how long it takes to "develop." And so on...

Good luck!

:Niranjan.

 

[jsmoove]

@nmp Yeah youre right, I got the order mixed up. Is there a way of exposing only certain colours through UV then? To do each layer separately?
Yeah that's basically the idea, so one way is to print the image ahead of time, and the other way was to activate any image you like anywhere on the shirt, that's what I was hoping for.
To make an all-purpose material.
Yeah ill have to experiment one day with a UV bulb and photochromics.
Thanks man!

 

[jsmoove]

What about: http://proposal.ducr.u-tokyo.ac.jp/cgi-bin/ccr_usr/EN/detail.cgi?num=3678
Sounds intriguing

 

[J 3]

A couple things:

• Inkjet inks do not need magnetic particles (a Laser printer by contrast does use a brush of magnetic particles to coat the electrostaticly charged paper to form the image).
• A lot of inks can work in inkjets but by no means anything, nor are all inks good. If there is a pigment for instance, it might clog a print head not designed for it.
• Inkjets work by having a tiny heating element vaporize a tiny bubble of ink in a channel which then pushes out a micro-drop of ink onto the page. It's a tiny amount but whatever you use cant be degraded by this kind of abuse.
• I think (?) there might have been some early confusion about something being panchromatic (it reacts to all visible light), and a dye being colored. Most black and white film is panchromatic (a red object will be as bright as a blue one), but produces a monochromatic image. By contrast Cyanotype media reacts only to UV light, but produces an image that is blue-green - You could expose it to blue-green light all day and not get very much of an exposure. For an Autochrome image you need a panchromatic B&W emulsion, and a non-photo sensitive color filter with small areas of each color. An inkjet in theory could print this filter if it could be done in such a way to stay sharp, and not get washed away with processing of the photo emulsion, etc. You'd in my opinion have more success trying to duplicate one of the other processes where the color filter was not bound to the emulsion (e.g. a Paget Plate - https://en.wikipedia.org/wiki/Paget_process ) because of these restrictions but that doesn't help with your fabric idea.
• There are not too many panchromatic technologies out there and they are hard to work with being sensitive to any light. Other than properly treated silver halide emulsions, there are some photo-polymers and not much else without huge exposures.
• For producing a color image from a color source by analog means, you'd need something panchromatic. Otherwise you'll be exposing each color through a negative/positive with light of a different color (usually blue/UV, like gum-bichromate or carbon transfer).
• An Autochrome (color filter on top of a black and white image) is very dark. Any color image made like this has to be either dark or very pastel by physics. Autochromes have to be back lit. Furthermore to get clean colors there has to be a tight correspondence between the B&W image, and the color filter. I don't think fabric (even say thin silk backlit) would hold hold this structure well enough to make a clean image. If not, to form a bright saturated image the image bearing substance has to be colored itself. To be as bright as possible, it has to be done in a subtractive color space (cyan, yellow, magenta) rather then and additive one (red, yellow, blue) as well. Finally to produce a final image from a camera if that is the goal it has to be a positive process (e.g. red light produces a product that reflects/transmits more red light) and much more light sensitive than a process that would work in an enlarger, or in a contact printer exposed to the sun.
• If you did try to use non-panchromatic emulsions for an Autochrome like end result you'd run up into the problem that dyes for your color filter (needed for viewing) would block the UV light you needed to form the image (Especially the yellow). You could expose through the back side, or try to use a different mask for creation and viewing but registration then becomes a huge problem.

Two ideas I think might possibly work for analog color on fabric:

• A sun bleached dye-destruction technology rather than an actual emulsion like the Utocolor process ( https://en.wikipedia.org/wiki/Utocolor ). The result would not be light fast though, and exposures would very long.
• Building up an image in layers using non panchromatic media somehow. Say sputter (small non-covering drops) some emulsion, expose a color separation with UV, dye the correct color and then repeat.

 

[jsmoove]

@J 3 Wow thankyou !
Youre right, I was a bit confused on the panchromatic vs the dyes, im learning lots here haha.
Autochrome really isn't the route I think.
I will check out this Utocolor process, I have never heard of it, sounds really cool.
In my searches I found: https://sci-hub.tw/10.1039/b416139d
A multicolor photochromic takes on the same light, sounds kinda neat? But itd only last a short period of time.
I know nothing about chemistry though, and it doesn't sound like the layman can easily reproduce such a thing.
There was also similarily: https://sci-hub.tw/10.1007/s11998-018-0105-0
Which just came out, and it mentions inkjet.
So for this whole fabric idea im mostly wondering about projecting an image with some sort of projector.
I think itd be the future of fashion design. To have preprinted light sensitive materials that you can project your own patterns and designs on without using a printer/dye sub/screenprint.
Will return with more questions about utocolor soon.
Thanks for the help!

 

[jsmoove]

@J 3
Im reading: https://www.photrio.com/forum/threads/utocolor.15399/print
Im wondering about what Mustafa said at the bottom there, can you just blend CMY dyes and expose?

Page 702 here goes over the Utocolor process: http://www.processreversal.org/public/text/Glafkides_photographic_chemistry_vol_2_compressed.pdf

Mentions using Borax as a fixer:
https://sci-hub.tw/10.1088/0031-9120/47/4/423

But Utocolor as opposed to multicolor photochromics, one is a bleaching method that can(?) be used to expose an image on, and the other actually absorbs the light of said image. Do I have that right?

 

[J 3]

You should know my interest is more in terms of fascination with the history and technology than in what would make a successful commercial process. As such I don't automatically filter out the obscure, imperfect, or impractical. If it takes a day for me to make an image by some old obscure process, I see it as part of the fun rather than a dead end so realize that's my bias here.

I will try to read the article when I get a chance but in brief there are only a few ways to produce a full color image on a single substrate by way of the action of camera light in a single exposure (one pass).
They can be generally grouped as:

1. Processes where colors are produced by having multiple layers each of which produces an image in a component color to form the final image - This is how slide and negative color film work. The images are in some kind of negative color space, and the layers have to allow light needed to view upper layers (IE the layers need to resemble more a stained glass window or a watercolor, than a layer of house paint). These methods tend to be the most efficient with light producing the brightest images. The complexity is creating all these layers in such a way that they are robust and react correctly. Most film dyes are not exactly the right color for a perfect 3 color image for instance (its part of why slide film has the look it does).
2. Processes where the component colors are separated into distinct spots on the image in some kind of pattern larger than the wavelength of light but small enough to allow details to be seen in the image - The most famous of these was the Autochrome but there are many others - These are usually are done in a positive color space (Autochrome's was orange, violet and a greenish blue if I remember right as a compensation to insensitivity of the base emulsion to deep red light), though they can be done in a subtractive space by giving up on the maximum saturation of the image. So in general you could have a low intensity image with vibrant colors, or a bright image with weak colors, or some midway point. Autochrome chooses the former and relies on backlighting.
3. Processes that encode the color of light by recording interference patterns (light interactions on the order of the size of the wavelength of light) - The only one of these I know about is called a Lippman plate. These images are very restrictive on lighting an viewing angle, and the very fine structure producing the image is very delicate.
4. Processes where all the interactions happen within a single layer forming a mix of dyes of the appropriate color - Utocolor is a very rare example of this. The reason you've not heard of it is this method lost the color wars and didn't even have a brief moment of historical popularity. These methods would usually have to be t color space methods because all the dyes are mixed at once.

As I can see the industry is definitely going more towards complicated printers, rather than optics and chemical processing but if you were looking to do the latter, trying to mine the past with modern chemistry I think would be the way to go. The basic observation of Utocolor was a subtractive dye (magenta, yellow, cyan) absorbs light of the corresponding color (green, blue, red) and reflects the rest. That action pushes an electron into an excited state, and that excited state tends to bleach out the dye (destroy the molecule changing it to something that will not reflect light anymore). The process is normally undesirable and is what causes blue jeans to fade, and low quality inkjet printouts to fade. This creates the core problem with the Utocolor concept and what a modern interpretation might seek to improve - The process requires two contradicting things. You need to have dyes that fade in light quite quickly (extreemly fugative) in order to form an image with minimal exposure, but one the image has formed you want the dyes to be extremely stable (lightfast) so that you have a durable image. Utocolor never solved this problem sufficently. The dye was made extra sensitive by a chemical bath, but it wasn't sensitive enough for short exposures. At the same time the finished images had to be stored away from bright lights to keep.

I'm not a chemist, but there is a lot more out there in terms of general tactics one might try for making a better variant.

• Solook into modern methods for stabilizing dyes.
• Find better ways of making dyes temporarily more sensitive to light so that more stable dyes can be used or shorter exposure times may be achieved. This might involve temperature, UV exposure, etc in addition to the sensitizer itself.
• create a micro-engineered product that works by the action of light in a more complicated way (say grains of pigment coated in a photo-polymer of some sort).
• Engineer it some way so that some other property is manipulated by the action of the light (say solubility) so that you can remove the non image dye in some other way.
• Use some other photo sensitive component to increase sensitivity so that you're not relying on the destruction of the dye itself to make the exposure. This is how Cibachome worked but it was a layered product that required a lot of engineering to get the layers right.

Modern approaches would tend towards the micro engineered I think. Microscopic grains of pigment with coating engineered to react with only the correct color of light. Exposure to light permanently fixes the particle to the fabric in some way, and the remainder are washed away in some way. It's not a camera emulsion nor single layer, but the one example I can think of is the Zink color technology that goes into the new Polaroid Zip printers - http://www.supercook.me/download/19103/3_

 

[J 3]

I terms of what one can achieve as an amateur, duplicating turn of the century technologies generally are not out of the realm of possibility IMHO. A lot depends on dedication of course. The equipment available back then was much worse than today, and you've got the advantage of a century of text books and a community of people you can ask via the internet. Chemistry is purer today, and a greater variety is available but at the same time dangerous chemistry has restrictions that were not in place back then. On the other hand these efforts to create a color image at the turn of the century were often well funded commercial interests. An amateur is not going to build a multi-ton, multi-layer coating machine or be able to reproduce the millions of dollars in proprietary research it took to produce the final version of Kodachrome say for example. There is however a thread on Photorio where Autochrome is being reproduced (A highly engineered product for the time) and it's getting very close to being called a duplication. Getting consistent results (something big operations are better at) seems to be the main remaining snag. The trick is to use from in intervening history what you can, and avoid what you cant. You can buy off the shelf dyes that folks from the turn of the century would have needed a PhD to produce, if they even knew of them.

 

[jsmoove]

@J 3 Its the same for me, im just as fascinated and curious, even more so for the obscure stuff. Even if I don't get to experiment im happy to learn about how things work.
#4 "Processes where all the interactions happen within a single layer forming a mix of dyes of the appropriate color" seems to be the only way to go as far as im concerned, and its the most interesting to me as well.
Extremely fugitive dyes, got it. That Turmeric and Borax article I linked above more or less says the same thing about needing natural fugitive dyes.
So what im confused about is, do you literally just mix the 3 dyes together? To form a neutral or black tone?
Im assuming you cant just mix any "light sensitive" dyes together.
So heres maybe a dumb question, why cant I just mix something like Solarfast dyes together? Whats the difference?
If I can figure out what types of dyes to use inorganic or organic (organic is cooler), then I can probably narrow down the stabilizers that exist in research.
I understand how it works when its layered and subtractive,-- regular color film.
Yeah I had read about the Zink process here on Photrio not that long ago, if there were a Zink process but for fabric then that would be perfect. But yeah the layers are an issue.
I mean obviously one could just dyesub or direct to garment a shirt, but I feel like projecting an image is just that much more of a cooler concept haha.
What you say is true, theres a wealth of information out here!
What is your opinion on the multicolor photochromics?

*I found a little blurb on single layer color papers, its page 628 from: http://www.processreversal.org/public/text/Glafkides_photographic_chemistry_vol_2_compressed.pdf

 

[J 3]

I do not know how the original formula was done (layers or mixed). Either could work, especially if the dye was actually a pigment but the single mixture would be harder to make work. True dyes when in close proximity sometimes interfere with each other. I'll see if I have a recipe. The old patents in general are a mixed bag, written poorly enough to trip up a competitor but just well enough to sue if such competitor does make it to market.

 

[jsmoove]

@J 3 Here is the blurb up close:
https://imgur.com/a/IvXpygl
Sure, recipes sound like a good place to start. I imagine so, I haven't looked at any myself.

I am confused though, is Utocolor basically the same as whats written here or was it literally just mixing 3 dyes, what kind?

 

[J 3]

It is not what was mentioned in the blurb article. That sniper was about making a more conventional color chromogenic emulsion but all mixed together. It would be a heck of a feat if it could be done but would have involved a lot of advanced chemistry. It would likely be what I had called a micro-engineered emulsion - in this case a grain with a color filter shell over a silver halide grain and the chemicals that form the dye on color development, plus the chemistry that keeps the colors from cross talking. Figuring Kodak probably tried and failed to produce something like this I doubt its possible without a massive effort and a comercial lab if at all. Utocolor is a simpler process.

 

[Photo Engineer]

Making a single layer color material is daunting to say the least.

Mark Osterman, of GEM, has told me that after examination of materials, and trips to France to look at details, no one has yet duplicated Autochrome as done by the brothers Lumiere. What is being done is not identical, but is a variant in each case, and does not approach the original in quality.

PE

 

[J 3]

Mark Osterman, of GEM, has told me that after examination of materials, and trips to France to look at details, no one has yet duplicated Autochrome as done by the brothers Lumiere. What is being done is not identical, but is a variant in each case, and does not approach the original in quality.
PE

That is disappointing but at the same time leaves me even more in awe of the original. I'd hoped the modern attempts were getting simular results, but with more lightfast modern dyes. I want to see Autochrome live again. I've seen a lot but never an Autochrome in person.

 

[jsmoove]

Ah I see. Ok thanks for the clearup. So what types of dyes are needed for the Utocolor process then I wonder?

 

[jsmoove]

Again...why not use some type of immiscible photochromic dyes? Like mix C+M+Y photochromic grains or something together? To make a multicolor photochromic. It wouldn't be stable just like utocolor however.
Or like Mustafa says at the bottom of: https://www.photrio.com/forum/threads/utocolor.15399/print
Mixing CMY dyes.