Trying Zerochrome-SbQ (PVA-SbQ)

[AndrewBurns]

So far I've mostly been printing with toned cyanotype, and a small amount of experimentation with carbon printing using DAS (which worked but my exposure unit is really the wrong wavelength to make it work well...). I'm currently printing using an LCD screen as a digital negative, and I'm putting together a 'UV projector' (basically a UV enlarger with an LCD screen in place of the negative). The big downside with using LCD screens, particularly when enlarged, is that they block 90-95% of the UV light you have to play with, and so you either need a LOT of UV light (which runs into thermal limitations) or you need to expose for a very long time.

Unless your printing technique is very fast...

Which brings me to Zerochrome SbQ, which is a non-toxic direct-pigment process similar to gum printing, except using polyvinyl alcohol modified with a light-sensitive compound called SbQ. Conveniently enough, a 'raw' version of the PVA-SbQ emulsion is available to directly purchase from a Chinese company in 1kg+ quantities for a reasonable price. This process seems to have all of the attractions of gum printing without the toxicity, and notably for my particular case it's fast, like 10 times faster than gum-bichromate.

I ordered a 1kg container of raw PVA-SbQ from China which should be enough to last me a long time. Based on the Zerochrome guidance I'm diluting my raw emulsion 1:5 with distilled water for a clear working solution and 1:8 with distilled water + pigment for a pigmented working solution.

Like gum, each layer of PVA-SbQ has a pretty narrow tonal range, so typically a good tonal range and dense blacks are built up by stacking multiple layers of emulsion. Calvin Grier has developed a very nice system for his Printmaker's Friend product of stacking multiple layers of emulsion with different pigment loads and different exposure times to generate a full tonal scale. A good explination of how it works is available on his website: https://printmakersfriend.com/

In my case I'm going to divide my image into 6 layers, layer 1 has maximum pigment load and each subsequent layer has half the previous layer. Similarly layer 6 gets the maximum exposure time, and every previous layer gets roughly half the exposure time. A table of the pigment:clear ratio and % exposure time for my 6 layers is as follows:

Layer	% Black Emulsion	% Clear Emulsion	% Exposure
1​	3.125	96.875	100
2​	6.25	93.75	58
3​	12.5	87.5	33.6
4​	25	75	19.5
5​	50	50	11.3
6​	100	0	6.6​

My target pigment concentration in the lightest layer (layer 6) is about 0.8% india ink, which means that the pigment concentation of my darkest layer (layer 1) needs to be a whopping 24% india ink! Here are my 6 layer formulations, plus a clear layer, painted onto some watercolour paper. Ignore the pencil annotations, they didn't end up lining up to the actual layers because my brush was wider than expected.

You can imagine that a stack of those 6 different layers should result in a nice tonal range!

The next step is to find your target exposure time. From Calvin's website he suggests how to do this with the lightest emulsion layer. I set up an automatic exposure test with my LCD printing rig that exposes the same test pattern 10 times, with 6 seconds between each exposure. So in the following print exposure 1 was 6 seconds and exposure 10 was 60 seconds.

You can see a few interesting things here, firstly as exposure goes up the dynamic range of the tonal scale reduces significantly, starting at something like 5 to 6 steps out of 21 at the 6 second exposure, down to about 3 steps for anything above 36 seconds. Secondly there's a lot of irradiation/halation, particularly at longer exposures, presumably because very low pigment density is allowing UV light to travel horizontally in the emulsion and to harden PVA outside of where it should.

Based on this test I'm choosing 18 seconds (step 3) as my target exposure for this layer, which is CRAZY FAST. For reference, my classic cyanotype exposure time is 240 seconds. And keep in mind that every other layer is exposed for less time, so layer 1 is only exposed for just over 1 second. Also this is with a 405nm light, no short wavelengths needed. In fact this stuff is probably quite sensitive to even blue light, so you really need to work with it under at least a yellow safelight.

Last night I had a go at exposing 6 test strips following the pigment densities and exposure times I gave in the table above. Each test strip was watercolour paper sized with two layers of liquitex matte varnish diluted 1:1 with water and then coated with a clear layer of PVA-SbQ (1:5 diluted) before coating the pigment layer. Results were as follows:

As you can see, there were some good and bad things about the test.

The first good thing is that you can clearly see that with each subsequent layer the pigment is darker and the 'tonal window' moves along the scale, which is what I want. Also you can see that the darker layers are sharper (most obvious in the lines and numbers) which confirms that the irradiation/halation effects in the lighter layers go away as pigment concentration goes up.

The obvious problem is that from layer 3 pigment staining suddenly gets really bad, and layers 1 and 2 stained so badly that no image was recoverable. For layer 3 I used 'spray development' where spraying water onto the print helps dislodge some of the stain, but it didn't work that well. For layer 2 I tried brushing the print under water but that just destroyed the image completely if there even was one. For layer 1 I tried washing in very hot water but that didn't help either.

I don't think this staining issue is going to be a killer though, as the 'correct' use of these layers is to stack them one on top of the other starting from lightest and going to darkest, only developing after every layer has been coated and exposed. That means that in practice the darker layers will always have several hardened lighter layers between them and the surface of the paper, and when the print is developed the layers below the dark pigment melt away which should prevent the pigment ever getting down to the paper and staining like that.

I also noticed when I was brushing on the layers, particularly the ones in the middle of the tonal scale, that the pigment appeared to be clumping and forming blotches, it was also impossible to avoid really obvious brush stroke marks. I think that both of these problems are related and to do with the emulsion not wetting nicely to the surface and trying to 'bead up', pulling pigment into clumps under surface tension rather than spreading over the surface nicely. I think in the future I'm not going to bother sizing the paper with acrylic in the first place as I don't think it helps at all, and I'm also going to try adding something like Tween 20 to the emulsion to encourage it to wet to the surface.

Overall, I'm pretty excited about this process and it seems like it has real potential. There are definitely some issues to overcome but there are other people getting good results with it, so it should just be a matter of refining my technique.

 

[koraks]

I don't think this staining issue is going to be a killer though, as the 'correct' use of these layers is to stack them one on top of the other starting from lightest and going to darkest, only developing after every layer has been coated and exposed. That means that in practice the darker layers will always have several hardened lighter layers between them and the surface of the paper, and when the print is developed the layers below the dark pigment melt away which should prevent the pigment ever getting down to the paper and staining like that.

To combat this problem in gum printing, Calvin resorted to printing a clear layer between different color layers to provide a non-staining surface for the next layer to print on.

 

[xiaruan]

Could you please tell me the channel through which you purchased PVA-SBQ? I live in China, but I haven't been able to find a way to purchase in small quantities.

 

[xiaruan]

You can try applying a layer of PVA between each layer. When I made the gum chrome salt photographic print, I would apply a layer of transparent Arabic gum between each layer to prevent color contamination between the layers.

 

[Raghu Kuvempunagar]

Could you please tell me the channel through which you purchased PVA-SBQ? I live in China, but I haven't been able to find a way to purchase in small quantities.

A bunch of friends here in India want to explore the possibility of purchasing SBQ emulsion or powder in small quantity. One seller who supposedly sells in small quantity is this:

Photoinitiators, Fluronated Chemicals

Tianjin Huiren Chemtech Co., Ltd.

Add: No.16, 5th Haitai Fazhan Road, Huayuan Industrial Park, Binhai Hi-Tech Area, Tianjin, China.

Fax: 86-22-23707858

Cellphone & Wechat & Whatsapp: +86-13032611983

Email: international@huirentech.com

We have not approached them yet. The reference was provided in another forum by someone who has purchased from the seller. If you reach out to the seller, please do share your experience as it will be useful to us also.

 

[koraks]

Could you please tell me the channel through which you purchased PVA-SBQ? I live in China, but I haven't been able to find a way to purchase in small quantities.

You could try the source mentioned in the link in #1: https://www.alternativephotography.com/zerochrome-sbq-colloid-printing/ Under 'Suppliers' you'll see the following:

We got ours from Zhejiang Rongsheng Tech Co.,Ltd in Quzhou City, Zhejiang Province,
China , whose website is https://www.rsemulsion.com/ .


This company, whose representatives are very friendly and responsive, manufactures the powdered SbQ compound, which has been outlined in the paper; as well, it is available as a premixed emulsion, which is a base for the making of silkscreen emulsion; so, it’s a semi-ready product, normally supplied to other companies that make silk screen emulsions for that specific market, normally as part of large B2B agreements.

Hope this helps.

 

[xiaruan]

A bunch of friends here in India want to explore the possibility of purchasing SBQ emulsion or powder in small quantity. One seller who supposedly sells in small quantity is this:

Photoinitiators, Fluronated Chemicals

Tianjin Huiren Chemtech Co., Ltd.

Add: No.16, 5th Haitai Fazhan Road, Huayuan Industrial Park, Binhai Hi-Tech Area, Tianjin, China.

Fax: 86-22-23707858

Cellphone & Wechat & Whatsapp: +86-13032611983

Email: international@huirentech.com

We have not approached them yet. The reference was provided in another forum by someone who has purchased from the seller. If you reach out to the seller, please do share your experience as it will be useful to us also.

You could try the source mentioned in the link in #1: https://www.alternativephotography.com/zerochrome-sbq-colloid-printing/ Under 'Suppliers' you'll see the following:

Hope this helps.

Thank you for providing the information.We are currently on the Dragon Boat Festival holiday. I will consult with these two companies after the holiday.

 

[isaac7]

Is this the same as The Print Maker’s Friend offered by Calvin at The Wet Print?

 

[koraks]

Is this the same as The Print Maker’s Friend offered by Calvin at The Wet Print?

No, it's not the same, but functionally similar.

 

[AndrewBurns]

Thank you for providing the information.We are currently on the Dragon Boat Festival holiday. I will consult with these two companies after the holiday.

Yes I went through Rongsheng to get my PVA-SbQ mixture, they also sell the raw SbQ powder if you want to synthesize the mixture yourself (although I personally wouldn't bother with that if I can just buy the finished product at a reasonable price). The Rongsheng sales person contacted me on whatsapp and has indicated that they've sold this emulsion to other people using it for artistic purposes and they seem keen to grow that market for their products.

Is this the same as The Print Maker’s Friend offered by Calvin at The Wet Print?

Calvin hasn't disclosed what his product is made from (and why would he, that's his commercial IP). I've never used PMF but from what I've seen the two systems seem to have very similar properties, however as I'm experiencing the base of the emulsion is only a small fraction of the effort required to get to consistent and high-quality prints and there's definitely a lot of value in other possible additions (e.g. surfactants, pigments etc.) plus the calibration information to get you started.

Actually I'd be interested in buying some PMF when it becomes available again. From what I've read, the early adopters of that stuff have had similar issues to me, particularly staining, and Calvin has stopped selling it until he has time to generate a comprehensive user guide and run some workshops using it. This indicates to me that while you can get some stunning results with it (see some of the prints Calvin has made with it) there's probably a lot of hidden nuance and technique that he takes for granted due to being so skilled but that other people miss.

I attempted a 6-layer print last night, always trying to run before I can walk, and I suppose predictably it was a disaster. Interestingly enough when printing a single layer with this material I never see a latent image after exposure, however when painting a new layer over a previously exposed layer a latent image is revealed.

This is the latent image of the first layer revealed after brushing over the next layer. Note the darker tones are on the left side of the print.

As I built up the layers the latent image was actually inverted compared to what the print should have been, I think this is because the exposed PVA hardens and swells slightly, meaning the exposed regions are slightly higher than the unexposed ones. Then when you paint over the next layer the pigment collects in the unexposed 'valleys' between the exposed areas, resulting in a reversed image (at least until development).

This is the latent image after brushing on the second-last layer, note that the latent image has inverted and the darker tones are on the right.

Ultimately when I went to develop the stack of 6 layers... nothing happened, it didn't seem like any pigment washed away or if it did I couldn't tell because the whole print was permanently stained black. This is exactly what I saw on my test strips so I shouldn't be so surprised but I was hoping that the melting of previous layers would prevent staining from the final ones.

For reference this was on plain watercolour paper and I put down a layer of clear PVA-SbQ before adding the pigment layers in an attempt to prevent staining. I also added a drop of Kodak photoflo to each emulsion layer to try to improve its surface wetting properties (I don't have any Tween 20 yet, but have some on order).

There are a few things that could be the issue here, and probably more than one of them are true. In no particular order:

• I could be brushing too hard or too vigorously between layers, causing the layers to blend together and stain (particularly the unhardened regions)
• I might not be waiting long enough for each layer to fully dry before coating the next, which is likely as I'm working in my garage at night and it's nearly winter here so the layers were taking a fair while, particularly as the pigment load went up
• My pigment (Pebeo india ink) might not be reacting well with the PVA at higher pigment loads, it does seem to be clumping a bit and I have been told that india ink in higher concentrations can cause bad staining

Ultimately it seems that I need to methodically work through things rather than trying to jump straight to the finish line. For example I need to be able to make a dark test strip that doesn't stain (or at least not as badly). And I need to try overlaying my first two or three layers that currently do work well to see if my brush technique is causing problems or not.

 

[AndrewBurns]

Not much of an update, I've been trying a few things and learning a lot, but doesn't feel like I'm any closer to getting a good print.

I've been comparing one or two coats of Liquitex matte acrylic varnish diluted 1:1 with water vs. Golden 'So flat' matte titanium white acrylic paint diluted 1:4 with water as a paper size. The white acrylic paint seems to be the better of the two as it leaves the paper with a more natural feeling roughness that I think helps with wetability of the emulsion. By comparison the matte acrylic varnish seems to leave the paper slightly smoother and more plasticy, which makes the emulsion want to bead up more.

I tried printing exposure test strips with these different sizing techniques and was able to get the higher pigment-load strips to develop at least, although with a reasonable amount of staining (I think the developing water was too cold, it's winter over here).

Today I was trying to generate some spectrophotometer density readings for my various layers but ran into issues with getting consistent layer densities. I think the problem I'm having is that the layer density as spread on the paper depends a lot on how thick of a coating you put down, how much is absorbed into the paper etc. So the sizing, the type of brush being used and the brushing technique all seem to make a difference.

Take for example the following two sheets of paper, both sized with the white acrylic paint, both coated with the same emulsion, but the one on the right is visibly darker with much more visible brush strokes. The difference is that I used a hake brush on the right side and a cheap synthetic brush on the left.

The nice soft bristles of the hake brush hold more emulsion which means it goes down thicker, and even if I dry the brush out because the bristles are softer they just push the emulsion into high spots that leave streaks in the coating. The stiffer synthetic brush on the other hand doesn't hold as much emulsion and the bristles put it down more evenly. So it would seem obvious to use the synthetic brush going forward, however I'm concerned that the stiffer bristles will damage previous layers when stacking multiple layers prior to development.

I really thought this process would be a lot easier, it sounds easy in theory (making some pigmented goop insoluble in water and then washing the rest away) but the reality is a lot more difficult, specifically in getting consistent even coatings that don't damage prior layers or stain terribly. My first attempt at a carbon transfer print with DAS sensitiser went better I think, despite only having a very unsuitable 405nm exposure system!

 

[koraks]

I'm afraid the inescapable reality is, as so often, that a brushed pigment/binder layer cannot realistically be made consistent. The alt. processes in which this appears to work are either/or limited in other ways (i.e. there's a limit to absorption etc. that will allow things to even out), or they're not practiced with the degree of rigor that highlights the problems of unevenness.

 

[PGum]

In my experience, PVOH with or without SbQ is quite sensitive to any other additives. Minerals, other colloids don’t play well with it causing pigment aggregation. You may want to seek out another pigment and see if it clears better, maybe even grind some pure pigment into your emulsion. Carbon or lampblacks are often already a challenging pigment to not have staining. I have also found that some india inks contain additives -surfactants or otherwise. They may be leaching down into your lower layers and making a mess - just one possibility!

 

[AndrewBurns]

Yeah I'm fine and even happy with a certain degree of unevenness or variation, after all these are hand made and I want that to be visible to a certain extent which is why for my cyanotypes I leave brushed borders. I think it's just going to be a matter of experimenting with brushes, technique, pigments, layering, additives etc. until I get something that does what I want. Clearly it's possible as other people have been getting decent results, but being a relatively new process there's not a great body of information available and from what I've been told it's also just not as forgiving as gum bichromate in a lot of cases.

 

[AndrewBurns]

Bit more progress whenever I get some time (which is less frequently these days after starting a new job).

I've been experimenting with gelatin sizing some paper to see how that effects the coating and staining. I've been testing with some sheets coated in 4% gelatin solution today but unhardened, and I also have some more sheets that I've attempted to harden with a 1% solution of triethanolamine (something suggested on Calvin's PMF website).

I've also had a go at mixing up some PVA-SbQ using a dry lampblack pigment kindly donated to me by another member of this forum, with the hopes that it will stain less badly than Indian ink. To help disperse the pigment I used a few drops of polysorbate 20 (Tween) and sat it on a magnetic stirrer for a while, which seemed to do the trick.

I got myself a few more brushes and I seem to have found the right kind of brush that is dense and firm enough to make nice thin and even coatings but also soft enough to not damage previous layers.

My goal today was to try to figure out the pigment concentrations I need for a 6-layer stack to get me a nice linear tonal range from paper-white all the way down to an L* of 22. I've been using a spectrometer to read the L* values for each layer after coating over the previous layers, making adjustment to pigment loading where needed. I was doing this on paper sized with unhardened gelatin and also with matte white acrylic paint side-by-side to get a feel for how each sizing handled.

The results for 5 of my 6 coating layers are shown below. Gelatin sized is on the left and acrylic sized is on the right.



Clearly the gelatin sized paper coating is much less even, I think this was because the gelatin was not hardened. I noticed that as I was brushing on the emulsion it would start to set and get sticky much more quickly on the gelatin sized paper vs. acrylic, I think the gelatin was drawing moisture out of the emulsion.

As soon as the emulsion starts to get sticky it turns into a bit of a trainwreck. You can see there's a dark boarder around the edges on both sheets, this was where the emulsion started to dry and go sticky at the edge of where I was brushing and when that happens a thicker layer builds up which you can't really do anything about. It also seems to happen faster with higher pigment-load layers. The trick is to work very fast and with a reasonably wet brush, you need to get the emulsion on the paper and spread around as quickly as possible, and then you might get one slow pass to even the coating out before you have to stop or risk making a big mess. The room I was working in today was pretty cold and humid, so I can't imagine how much worse this would be if it were hot and dry. I'm thinking maybe some kind of humectant in the emulsion might give me more time to get the coating even before it sets up.

This is a plot of measured vs. target L* values for each layer in the stack, as you can see I'm fairly close to where I want to be for the first 5 layers at least. The reason the 6th layer is missing is because I didn't have a high enough pigment load emulsion mixed up. Also the 1st and 5th layers could do with being a bit darker as well.



Here's measured L* for each layer vs. the pigment concentration of that layer (in % mass of dry pigment). Keep in mind that the L* value for each layer is the sum of all previous layers too, not just that layer in isolation.



Based on the trend line of this plot I think my 5th layer needed to have a pigment concentration of about 1.1% instead of 1%, and my 6th layer will need to be about 1.5% to 1.6% pigment.

Next step is to work out exposure times for each of these layers so that the tonal range is continuous across all the layers when stacked on top of each other. Then finally I can again try to print a test pattern to linearise.

 

[PGum]

Hard to say why the underlayers are dissolving. Hopefully it is not due to a drying retarder that could have been added to the emulsion by the manufacturer.

It is a disadvantage of SbQ, that its scale is so short, and this is discussed in journal docs. The near binary switch is perfect for screen printing emulsions, and imagesetter negatives, otherwise it is this more involved coating protocol that you are working through.

Sizing: Could a fully exposed clear layer provide sufficient sizing?

 

[AndrewBurns]

Hard to say why the underlayers are dissolving. Hopefully it is not due to a drying retarder that could have been added to the emulsion by the manufacturer.

It is a disadvantage of SbQ, that its scale is so short, and this is discussed in journal docs. The near binary switch is perfect for screen printing emulsions, and imagesetter negatives, otherwise it is this more involved coating protocol that you are working through.

Sizing: Could a fully exposed clear layer provide sufficient sizing?

I don't think the underlayers were dissolving in the above test sheets, more that the layers dry too fast around the edges (particularly the higher pigment layers). It's like around the edges of where you're brushing the emulsion isn't being re-wetted by the brush as much and starts to set within seconds, resulting in a buildup of emulsion in a ring around the area you're brushing which is darker because the pigmented layer is thicker.

I think that for this process brush and brushing technique are critical and speed is really important. It's quite different to how I've done cyanotype in the past where I have practically all the time in the world to spread the chemicals around and density is mostly set by how much the paper absorbs.

In the case of PVA-SbQ I think I'm probably going to have to start taping the paper down to a rigid substrate and also masking the edges when coating, because you need to rapidly brush over the entire sheet and you need to significantly over-brush the image area to ensure the coating is even where you want it to be. I've been using pretty thin cheap watercolour paper for testing and it really wants to buckle once I start brushing the coating on which also makes it very difficult to get an even coat.

Edit: I believe you can use a hardened clear layer as sizing, but I don't know how it goes for staining. Calvin seems to recommend hardened gelatin as the best bet for PMF and gum printing.

 

[PGum]

Is your PVOH/ SbQ translucent, or almost clear? This would help you to know that no PVA has been added, which I understand may be used in some off the shelf emulsions.

The one other issue with PVOH - those forms that are used with SbQ, is that the viscosity goes up exponentially with strength, so it doesn’t take much water loss before it sets up. Around the edge of the print, the air is dryer, not having the entire surface around it being wet, which sets up a micro-environment of high humidity as you move away from the edge.

Adding a humectant might cause problems with underlayers remaining soft, unless you were to develop each layer separately. I have used glycerin with casein and it helps since casein also has this rapid viscosity increase and setup on drying. Not much of an issue with gum however.

 

[AndrewBurns]

Is your PVOH/ SbQ translucent, or almost clear? This would help you to know that no PVA has been added, which I understand may be used in some off the shelf emulsions.

The one other issue with PVOH - those forms that are used with SbQ, is that the viscosity goes up exponentially with strength, so it doesn’t take much water loss before it sets up. Around the edge of the print, the air is dryer, not having the entire surface around it being wet, which sets up a micro-environment of high humidity as you move away from the edge.

Adding a humectant might cause problems with underlayers remaining soft, unless you were to develop each layer separately. I have used glycerin with casein and it helps since casein also has this rapid viscosity increase and setup on drying. Not much of an issue with gum however.

I've only ever looked at the raw PVOH-SbQ under yellow light given how fast it reacts, but from what I can see it's quite clear. The raw emulsion is extremely high viscosity though, similar to honey or condensed milk. I dilute it 1+8 with water before coating. To an extent you can 're-melt' an area of emulsion by brushing over it with fresh solution, but it never works that well and you risk damaging prior layers so ultimately once it starts to set you're out of luck.

I'm considering trying a coating rod as a way of getting repeatable layers of an exact thickness very rapidly but I'm not sure what kind of wet coating thickness I should be targeting. I know for carbon tissue it's generally between 0.5 and 1mm thick, however for gum/PVA-SbQ I'd imagine each coating is significantly thinner, like 100 to 200 microns or maybe even less. Has anybody had experience coating gum with a formed rod? And if so what 'size' of rod was used, in terms of wet coating thickness.

 

[PGum]

I can only weigh in on some qualitative results from a coating rod. As a proof of concept, I fashioned one from a 1/2 in dia steel rod that I wrapped with 28 guage vinyl coated hook up wire. It worked very well on hard substrates like glass and metal. Issues with paper were unevenness due to paper buckling, and pigment stain - the abrasion from the sweep pushed the pigment deeper into the paper, where it would not wash out. Perhaps if you did a brushing for a first coat, it might protect the paper from staining with subsequent rod coats if done quickly.

I don’t know of anyone who has had success with a coating rod for use with pigment - based printing on paper, but let‘s see, as it would be a great alternative if it worked.

 

[AndrewBurns]

Here's the same test sheets as before but with the 6th layer applied. I went with 2% pigment by weight in the final layer and it's still not as dark as my target (L* around 28-30 vs. 22) but good enough for now. The pigment I'm using gives a very matte black once it dries which I think is limiting my ability to get a really dark Dmax, and when I plot L* vs. pigment concentration I'm really starting to reach an asymptote so there's probably no point trying to go any darker.



As a next step I'm going to coat some sheets with 6 stripes of emulsion, one of each layer that I've just determined, and expose a step wedge onto each stripe at various exposure times. If I get my exposure times right then the tonal scale of each stripe should slightly overlap the ones before and after in the order, but if not then it should be easy to tell visually and tweak from there.

 

[AndrewBurns]

Ran an exposure test of my 6 layers tonight, each stripe was one of the 6 layer formulations with a different pigment load, from lightest to darkest. Coated on gelatin sized paper that I had attempted to harden with 1% triethanolamine.



I exposed a step wedge into each layer for a different length of time. In theory the lightest layer was supposed to receive the longest exposure and the darkest layer the shortest. Unfortunately I managed to get the paper around the wrong way and so the lightest layer got the least exposure and the darkest got the most!



The step wedge for the lightest layer was supposed to be the most 'full' and the darkest the least 'full' such that when the layers are stacked on top of each other they create a nice full tonal range, but obviously I got it back-to-front. Still, you can see what I was going for and it kinda worked, although I'll need to increase the exposure time for all of the layers I think as even the longest exposure length is a bit far from the end of the tonal scale.

Also the gelatin didn't feel hardened at all, very slimey during development even in just warm-ish water. I also had a lot of difficulty clearing the print, the PVA-SbQ clearly softened but didn't want to dissolve off the page and just sat there. I had to brush it with a soft hake brush underwater very gently to shift the unexposed PVA, and even then there was some staining. I don't know if this would be better if the gelatin was properly hardened, apparently I needed to add some potassium alum with the TEOA so I'll have to try to harden the sheets again once I get some of that.

Finally I tried to repeat the experiment with a sheet of paper sized with just matte white acrylic paint diluted 1+4 with water and it was a disaster, all of the layers stained permanently and would not budge from the paper after exposure, even with brushing. I've tried the acrylic paint size before and it worked, but that was with a layer of unexposed clear PVA-SbQ put down before the pigmented layer which I didn't do this time. Clearly that clear layer was critical to preventing staining (but I suspect it also increases halation).

 

[AndrewBurns]

Well I did a 'battle of the sizings' test yesterday, repeating the 6-layer exposure test I did previously with the following paper treatments (all on cheap watercolour paper):

• 'Hardened' (not really hardened) gelatin + an unexposed PVA-SbQ clear coat
• 1 layer of Liquitex matte acrylic varnish diluted 1+1
• As above but with an unexposed PVA-SbQ clear coat
• 1 layer of Liquitex matte acrylic varnish diluted 1+1 and then two layers of matte white acrylic paint diluted 1+4
• As above but with an unexposed PVA-SbQ clear coat

During coating it was very clear which ones had the clear coat and which didn't as the clear coat ones were harder/more annoying to coat. You could definitely feel that the clear coat would start to soften and get sticky as you were brushing on the layer above if your brushing took more than about 10 seconds or a few brush strokes.

When developing only two of the sheets with the clear coat actually cleared with only agitation in the water and even for those two the darker layers caused some permanent staining despite the clear coat. The rest of the sheets didn't clear at all, even with some pretty aggressive brushing under the water they were all permanently stained regardless of how long they sat in the water for.

The best result was the 'hardened' gelatin with the PVA-SbQ clear coat, which developed pretty nicely and only stained for the darkest two layers (the darkest layer barely developed). The staining might have been ok if the layers were stacked, because then the darkest layers wouldn't be in direct contact with the paper.

Here's two tests on gelatin sized paper, the one on the left was with no clear coat and the one on the right is with clear coat. Also I messed up the order of exposure on the left one so the darkest layers got the most exposure instead of the least, but you get the idea.



The sheet on the left (no clear coat) also did not develop with only agitation, I had to brush the unexposed PVA-SbQ off underwater, while the clear coat version only required time and agitation. The no clear coat test has about the same stain level for every layer, while the test with clear coat has virtually no staining for the lighter layers but then significant for the darker layers.

I also notice that the dynamic range of each layer in the test without clear coat is significantly greater, there's a very clear and gradual gradient from light to dark tone in each of the scales, while on the test with clear coat each layer is almost binary, going from white to full density in one or two steps. I'm pretty certain that this is due to the clear coat, because I've seen this behavior in other clear-coat tests as well.

My theory here is based on Calvin's theory about how tone builds up in gum printing with continuous-tone negatives. To paraphrase, if the layer only hardened from the top down towards the paper then you'd go from a zero thickness layer to full thickness as soon as the exposure depth reached the paper, giving you no tonal range at all. He theorises that light penetrates down through the layer and reflects back off the paper, starting some hardening from the paper upwards as well as from the top downwards.



In the case of a pigmented layer over a clear layer, the light penetrating down to the paper starts hardening the clear layer upwards while the pigmented layer is hardening from the top down. If the layers are the same thickness then by the time all of the clear layer has hardened the hardening action from the top down reaches the clear layer, and you're back to a binary on/off switch of tone.



I think it might work if the clear coat was significantly thinner than the pigmented layer, so that the clear coat fully exposes and then the bottom side of the pigmented emulsion starts to harden before the hardening from the top surface penetrates down. But I'd imagine a thinner clear coat would be more likely to melt/get damaged when brushing a layer on top and if it isn't thick enough relative to the paper texture it might not prevent staining either.



In any case it seems like right now I'm not able to get a print to develop out without huge staining without a clear coat (or brush developing, which isn't ideal) and with the clear coat I get significantly less dynamic range at least on the first layer. I think properly hardened gelatin sizing is going to be essential to prevent stain, and I've got some Hahnemuhle platinum rag on the way to see if that works better than the cheap watercolour paper I've been using.

I'm also wondering if I should reduce the dilution of my layers (currently 1+8 PVA-SbQ to water), less dilute emulsion will be more viscous and give me thicker layers that need less pigment to reach target density, which should mean less staining. I wonder if thicker layers would also be more likely to dissolve away properly when developing and resist melting from having new layers brushed over them.

 

[AndrewBurns]

Sick of paper for a while I decided to try printing on some aluminium sheet. One of the main reasons I was interested in this process was for printing on strange materials (glass, aluminium etc.).

My first attempt at a calibration print actually worked surprisingly well, despite the fact that I used completely the wrong exposure times by accident, blasted the plate with water from the tap when developing and then dropped it!



The aluminium plate is anodised and cleaned but otherwise not coated, so the PVA doesn't adhere to it very well. You can see where I hit it with the tap water and then dropped it the print was damaged, it actually looked exactly like frilling/flagging on a carbon print. After the print dried the surface was still quite delicate, however painting some acrylic varnish over the top made it significantly more robust.

It's nice to see that when I don't have to deal with paper staining my process can work just fine, however because the aluminium was uncoated it was a total nightmare to brush the coatings on. I could get literally one stroke with the brush when putting a new layer on and it was ok, but if I tried to brush a second time I'd start to rip up the previous layers. To help this I'm going to try coating a fresh sheet with a thin layer of the acrylic varnish first, to improve adhesion of the PVA. Hopefully this doesn't result in staining, however I've got a feeling that it might.

 

[koraks]

My theory here is based on Calvin's theory about how tone builds up in gum printing with continuous-tone negatives.

But what kind of negatives did you use here? Inkjet? Those aren't strictly continuous tone, but behave as something like a hybrid between continuous tone and halftone. The result is that you can get pretty odd interactions between effects characteristic of halftone and cont. tone processes.

One thing that comes to mind in particular is the issue of dot gain, and the halation effect you describe in relation to the clear undercoat will affect dot gain. Note that this also happens in the top of an uncoated paper base, depending on the surface texture of the paper. So in the real world, it's not really a case of a binary variable (e.g. clear coat or no clear coat), but a continuum that runs from a theoretically perfectly smooth and reflecting surface (something like Yupo comes reasonably close) running all the way to a paper base with a relatively heavy/thick clear coat underneath the image-forming layer.

Anyway, if I look at the steps you posted, my first guess would be that you're running into a dot gain issue. It'll be more severe for the print with the clear coat underneath the image-forming layer. What kind of light source did you use here; to what extent is it collimated? You could try and repeat the print using the sun on a clear day, with the printing frame perfectly perpendicular to the sun rays. This should limit dot gain due to imperfect collimation dramatically. You'll still have some dot gain due to scatter and this might allow you to observe the effect that Calvin refers to, assuming that it's how it works.