Effect of Prewash On Developer Activity

RalphLambrecht said:

or is it because the prewash makes it harder for the developer to penetrate the emulsion?

Click to expand...

It's the same thing, though. With a pre-wash, the gelatin soaks up water instead of developer. Then when the developer is introduced, it will start to diffuse through the waterlogged emulsion. So at first, you're starting with an infinitely dilute developer inside the emulsion which reaches equilibrium over time with the developer around it. How rapidly this proceeds will depend on many factors, but I suspect that a good order of magnitude will be around 15 seconds or so. On a 3 minute development time, that's a significant delay.

RalphLambrecht said:

The prewash water has already entered the emulsion, making it harder for the developer to get into the emulsion due to swelling, which impairs development.

Click to expand...

I think you mean the same as what I explained above, but I also find the formulation unfortunate because it's technically incorrect if interpreted literally.

RalphLambrecht said:

The question is: What is the cause of decreased development after prewash? Is it because the minimal amount of residual water dilutes the developer, or is it because the prewash makes it harder for the developer to penetrate the emulsion? It's likely a bit of both, but I wanted to point out that there is more than just implee dilution going on.

Click to expand...

Yes, I don't dispute that both mechanisms are involved. But... earlier you stated to me, "It's not due to the developer's dilution." So this is why I responded to your post.

Anyway... as ic-racer determined, via weight, in his particular configuration the working tank developer got diluted by about 10%. A not insignificant amount in my view.

Anyway... I think that only a real-world test, with the same configuration, would answer things conclusively. I say this because of potential drastic differences. I'm mostly from the world of continuous machine processing where "exit squeegees" strip off all surface liquids. Then as film enters developer, submerged turbulator bars are spraying against the surface of the film. So IF a prewash step was built in I would guess the effect of already-wet film would be close to negligible. At the other extreme, wet film might be simply put into developer and left stagnant. In which case the effect is likely significant, perhaps more than the developer dilution.

dokko said:

I'd be curious though what happens if you simply extend the development time of the film with prewash a little bit to get the same overall density.

Click to expand...

I did some testing on the effect of pre-soaking B&W film a couple years ago and posted the results here. After I completed the tests and analyzed the data, I came to a few conclusions:

1) Pre-soaking does lower overall density compared to not pre-soaking (not really in dispute here).

2) The reduction in average density is, within a certain range, proportional to the duration of the pre-soak (i.e., a 1-minute presoak has a little impact on density, while a 5-minute presoak has a significant impact).

3) The loss of density can be compensated for by extending development time (e.g., 20%) but the film gamma will likely be different compared to film developed without a presoak (i.e., the "control" film). In other words, if you extend development time with an aim to produce the same Zone V density seen in the control film, the shadow and midtone densities (Zones 1-6) may match the control but the highlight densities (Zones 8-10) may become considerably higher than the control. The strength of this upward shift in gamma likely varies with film type and developer (I tested using Tmax 100 and XT-3 1:1) and it's possible it might not happen for all combinations.

4) The reduction in density is a function of two mechanism: A) dilution from water carried over from the pre-soak; and B) a lower rate of diffusion of the developer into the emulsion. I personally suspected A was the principle mechanism before I did the test but I now lean more toward B. This change of heart was due to what I observed in #2 above. The amount of dilution caused by carryover of water from the pre-soak is going to be similar between a 1-minute and a 5-minute pre-soak, and yet, the 5-minute pre-soak will exhibit much lower density. It stands to reason, then, that dilution, while a contributing factor, probably isn't the principle mechanism.

Scott J. said:

2) The reduction in average density is, within a certain range, proportional to the duration of the pre-soak (i.e., a 1-minute presoak has a little impact on density, while a 5-minute presoak has a significant impact).

Click to expand...

That's interesting, because it cannot be explained by either the dilution argument, nor by the difficult-diffusion argument. So it's an observation that basically goes against either of the accepted/discussed explanations.

Scott J. said:

the film gamma will likely be different

Click to expand...

I think I would have put this a little differently given your description; it's not so much the overall slope (gamma) that's different, but (also) the curve shape - i.e. the slope in different sections of the curve. And that's an interesting detail, as it also seems to occur in the example posted in #1. Note how the curve without prewash shoulders off, while the prewashed curve doesn't.

Scott J. said:

This change of heart was due to what I observed in #2 above.

Click to expand...

But does that really pan out? After some 20 seconds or so, I don't think there's not a whole lot of additional water going to diffuse into the emulsion. It'll be fully waterlogged. I agree that the amount of carryover will be the same.

ic-racer said:

The H&D curves presented show the film (SVEMA 42L) with and without a 3 min prewash of tempered H2O .

Click to expand...

Thank you for doing this experiment and sharing the results.

To get a better understanding of the effect of pre-washing it would be great if you could repeat each measurement at least three times and plot the average curve with a standard deviation (a family of curves works equally well).

koraks said:

After some 20 seconds or so, I don't think there's not a whole lot of additional water going to diffuse into the emulsion. It'll be fully waterlogged.

Click to expand...

Do you think 100% water saturation can occur that fast? This could probably be tested by weighing different sheets of film that have been submerged for different times.

I have an inkling that there's a physical change beyond mere "wetness" that impedes the uptake of developer into the emulsion. Could be a swelling-related phenomenon that, once initiated, slows down the rate of diffusion of the developer into the emulsion. The question for me is: Is this an imbibition-dominated process (i.e., developer solution displaces pre-soak water out of the emulsion) or a diffusion-dominated process (i.e., developer molecules diffuse out of the developer solution, travel down a concentration gradient, and enter the pre-soak water)? Both are likely happening, but the relative proportions might settle the bigger question of what's going on.

Scott J. said:

Do you think 100% water saturation can occur that fast?

Click to expand...

Yes, I do expect so. I think it's an exponential increase that only slopes off very gradually once you get past 15-20 seconds or so.

Scott J. said:

Is this an imbibition-dominated process (i.e., developer solution displaces pre-soak water out of the emulsion) or a diffusion-dominated process (i.e., developer molecules diffuse out of the developer solution, travel down a concentration gradient, and enter the pre-soak water)?

Click to expand...

And/or a third option - a chemical change. If you let the film soak in water for a couple of minutes, stuff will start to dissolve from it. This may affect development. IDK by what extent, but it does seem to offer a better explanation of the difference between 1 vs. 5 minute pre-soaks than an imbibition/diffusion explanation. And then there's the possibility of stuff in the water itself doing something, but this should not play a role in your particular test as it used distilled water. If it used demineralized water instead of distilled (which is more likely), it may be a different matter!

What makes me suspect a chemical dimension to the issue is the upswept curve that you also got with a pre-soak vs. a shouldering curve without the pre-soak. That seemed to happen quite consistently in your case and it also seems to recur in @ic-racer's test reported on here. Something that holds back development in dense areas (i.e. offers a compensation effect) seems to play less of a role when a pre-soak is introduced.

Scott J. said:

I did some testing on the effect of pre-soaking B&W film a couple years ago and posted the results here.

Click to expand...

wow, that's a super interesting test, very thorough and well documented. thanks a lot for doing and posting that.

I would have expected the highlights to compress, but figure4 shows a significant increase in highlight density instead, very interesting.
A possible explanation could be, that the pre-soaked film, having a swelled emulsion, allows for faster replenishment of exhausted developer from the start.

this does not seem to happen in the test of ic-racer, but it would be plausible that in a dilute developer the effect is stronger.

Scott J. said:

I did some testing on the effect of pre-soaking B&W film a couple years ago and posted the results here. After I completed the tests and analyzed the data, I came to a few conclusions:

Click to expand...

Scott J. said:

4) The reduction in density is a function of two mechanism: A) dilution from water carried over from the pre-soak; and B) a lower rate of diffusion of the developer into the emulsion. I personally suspected A was the principle mechanism before I did the test but I now lean more toward B. This change of heart was due to what I observed in #2 above. The amount of dilution caused by carryover of water from the pre-soak is going to be similar between a 1-minute and a 5-minute pre-soak, and yet, the 5-minute pre-soak will exhibit much lower density. It stands to reason, then, that dilution, while a contributing factor, probably isn't the principle mechanism

Click to expand...

Thanks for posting! That's a very interesting test, and a result I would not have expected.

I read through your testing procedure. The only thing I would normally question is exposing the test wedge inside of a camera; the light falls off going off-axis from the lens. But... presumably you picked the long (300 mm) lens on purpose, so falloff is only on the order of 1% or so.

I like to have some sort of possible explanation, mechanism, or whatever, for "unexpected" things like this, which I don't really see in this case. Since you saw a significant difference between the 1 vs 5 minute soaks it makes me question whether diffusion of the developer into the swollen gelatin is the main culprit. I vaguely recall some older Kodak studies on the swelling of gelatin - they used what they called a "swellometer" to measure changes in emulsion thickness. Essentially a small, flat glass tube (I forget what they call them [uodate - capillary tubes, maybe]) that rides on top of the emulsion - a sort of telescopic microscope observes the upward motion of the tube against a very fine graduated scale. So they can track emulsion swelling over time in different solutions. As I recall, the swelling happened fairly quickly, etc. I'm pretty sure that some abbreviated tests were shown in an older published study by Lloyd West of Kodak on photographic water quality. (This study is referenced in some "modern" pubs, I'm thinking probably in Kodak Z-131?)

I see a handful of new posts here since I started typing. I would hypothesize that it may be a latent image effect, perhaps being bled off by the water. Fwiw when we (where I worked) were testing a new film (pro portrait, for use in our studio operations), the #1 thing we did was a latent image test - we wanted to know how long we had to hold an exposed film, before processing, in order to do critical testing that would be representative of actual production film. (It really WAS a big deal for critical work, prior to digital.) As I recall we figured, for the professional portrait/wedding films that we used, much of the shift occurred in the first 4 hours (at normal office temp, ~50% humidity). So for preliminary shooting tests we might shoot and process same day, but for critical color testing we'd hold film overnight.

Anyway, in this specific case my first suspicion would be a latent image shift. (Anyone interested in such things can probably find it in Tadaaki Tani's book, Photographic Sensitivity, around 1995. Tani did a later book that I never got, but I think includes updates on the topic.)

[Update: after posting I went back to Scott's link and read it again. The density shifts somewhere mid-scale are so large it's hard to believe it could be latent image shift. Hm...]

Mr Bill said:

Anyway, in this specific case my first suspicion would be a latent image shift.

Click to expand...

That's an interesting angle; the question would IMO thus be to what extent the latent image as such is influenced by a pre-soak. The first mechanism that comes to mind would be more facile electron mobility in the wet emulsion, resulting in more latent image loss. To quote the chart I'm looking at posted (much) earlier by @Scott J. :

Scott J. said:

Click to expand...

This might explain the overall loss of density - but so do the diffusion and the dilution hypotheses. What none of the models explain is the upswept curve that results from the pre-soak. And that one actually does occur in @ic-racer's test just as well, suggesting it is a predictable effect of a presoak (in addition to the reduced rate of development).

koraks said:

Yes, I do expect so. I think it's an exponential increase that only slopes off very gradually once you get past 15-20 seconds or so.

Click to expand...

I have no experience at all with this, but that seems a bit short, no?
don't modern films usually have hardeners built in which could take a while to wash out?

I did a quick search and found two resources which have some info on gelatine swelling, I don't have time to look into them in detail, but they seem interesting anyway:

Just a quick note related to the amount of emulsion swelling - if someone wanted to compare, to see if there is a change between 1 vs 5 min presoak: the poor man's method might be to weigh some wet film. Preferably strip off the surface water with a squeegee immediately before weighing. The assumption is, loosely, that the swollen gelatin is very nearly filled with water, so that extra weight would correlate to an increase in thickness.

dokko said:

I have no experience at all with this, but that seems a bit short, no?
don't modern films usually have hardeners built in which could take a while to wash out?

Click to expand...

This is not how hardening works. Hardening gelatin is a bit like drying a plate of spaghetti. On the fresh plate, the strands of spaghetti are supply and you can fork out just a few; they will disentangle. If you let the whole mess dry out, you'll find that it all becomes a solid bunch that cannot be disentangled, with the strands becoming brittle and shrinking a bit. This is how hardening gelatin also works, more or less - it makes the colloid molecules contract and fold into each other, compacting the matrix and making it less supple (and permeable to water). This effect is irreversible, so there's no such thing as "washing out the hardener".

What does happen, is if you wet the emulsion, it'll swell up to a certain point. Back to the spaghetti: if you pour water over the dried mass, it'll reconstitute a little, but not as much as when it was cooked originally. If you try to do that (the equivalent on gelatin would be heating it and/or raising pH dramatically), what happens is that the spaghettis disintegrate instead of reconstitute. So a hardened gelatin layer can be softened again, but the nature of this softness (so to speak) is fundamentally different from an unhardened gelatin matrix: in the unhardened gelatin, you have supple, flexible long colloid strands, while in a hardened gelatin that you soften through heat and/or addition of lye, you effectively break up the strands.

Note that your question is not so much about the effect of hardening; the film is hardened to the same degree in the pre-soak vs. no-soak experiments - so it really doesn't matter. The question ultimately is about how long it takes for the emulsion to swell (i.e. soak up as much water as it'll be able to). As I indicated before, this would most likely be an exponential curve (that tangentially approximates a certain limit) with most of the soaking happening in the first half a minute or so. This is also by design; you don't want to build a photographic film that swells with so much difficulty that it takes eons to soak up developer. You do notice big differences in development time across different films (e.g. Fomapan 200 vs. HP5+) which is likely in part due to this difference in hardening.

ic-racer said:

The H&D curves presented show the film (SVEMA 42L) with and without a 3 min prewash of tempered H2O .

Click to expand...

Were the pre-washed negatives with agitation or not, or would it even matter. Just curious.

The tank was spinning on a Jobo CPP-2 in the water bath in both cases. For the non-Prewash, the tank spun dry for 5 minutes. For the Prewash, the tank spun with tempered water for 3 minutes and the water was removed with the lift prior to starting development.

Mr Bill said:

Just a quick note related to the amount of emulsion swelling - if someone wanted to compare, to see if there is a change between 1 vs 5 min presoak: the poor man's method might be to weigh some wet film. Preferably strip off the surface water with a squeegee immediately before weighing. The assumption is, loosely, that the swollen gelatin is very nearly filled with water, so that extra weight would correlate to an increase in thickness.

Click to expand...

Through stupid errors with film holders over some time I have accumulated 5 sheets of ruined and undeveloped TMX. Since I had nothing else better to do.........for kicks and for what it may be worth, maybe none, I did this. I used fresh distilled water for each in a 5x7 tray, squeegeed both sides and immediately weighed them. Assuming it has any meaning at all, any differences do not seem very statistically significant.

Nice experiment @Chuck_P - I think the numbers do illustrate one thing though: that after a minute, the film has pretty much soaked up as much liquid as it ever will.

Chuck_P said:

Through stupid errors with film holders over some time I have accumulated 5 sheets of ruined and undeveloped TMX. Since I had nothing else better to do.........for kicks and for what it may be worth, maybe none, I did this. I used fresh distilled water for each in a 5x7 tray, squeegeed both sides and immediately weighed them. Assuming it has any meaning at all, any differences do not seem very statistically significant.

View attachment 391561

Click to expand...

that's interesting in it's own way to estimate the swelling of the emulsion.
from what I could find out the emulsion thickness of TMX when dry is around 0.01mm (possibly twice that at max).
so assuming those were 4x5 sheets, if they didn't swell they could soak up a max of 1.2mg of water (emulsion volume being 1.2 cubic mm).

the increase in weight in your test seems to be around 600mg, so assuming there are no water droplets on the outside of the film, that would mean the gelatin layer must have increased about 500 times in volume and therefore thickness (or 250times if the emulsion thicknes was 0.02 mm to start with).

seems like a lot, maybe I missed something in my logic or math?

edit: I just checked with three different AI models and they all said rougly 5.7times thicker, which makes a lot more sense. so there must be two factors of 10 wrong in my thinking above.

koraks said:

Nice experiment @Chuck_P - I think the numbers do illustrate one thing though: that after a minute, the film has pretty much soaked up as much liquid as it ever will.

Click to expand...

Tha's exactly what I was thinking when I got to the third sheet.

dokko said:

so assuming those were 4x5 sheets

Click to expand...

Yes, sorry, they're 4x5

This thread is wonderfully informative and interesting.
I have to chuckle though when I think of what my wife might say if she saw me weighing a bunch of wet sheets of film.
Something like: "What are you futzing with now?! Why don't you go out and take some pictures?"
I just can't figure out why she doesn't understand some things!

I'm puzzled as to what might cause the upswept curve for presoaked negatives in Scott's original test. Maybe the water-saturated emulsion dilutes developer by-products enough to affect the inhibition properties caused?

In any case, the effect seems substantial enough to make me think twice about presoaking negatives that need a lot of contraction. Next time I have such a situation I'll make two negatives, develop one with and one without presoak, adjusting development time to compensate and see.

Thanks for the tests one and all!

Doremus

dokko said:

from what I could find out the emulsion thickness of TMX when dry is around 0.01mm (possibly twice that at max).
so assuming those were 4x5 sheets, if they didn't swell they could soak up a max of 1.2mg of water (emulsion volume being 1.2 cubic mm).

Click to expand...

4x5* = ca. 10x12.5cm = 125cm2. At a height of 10um this makes 1.25cc. The actual weight gain is around 0.6g, so that seems quite nicely in the order of magnitude where we'd expect it.

Doremus Scudder said:

Maybe the water-saturated emulsion dilutes developer by-products enough to affect the inhibition properties caused?

Click to expand...

I'm thinking along those lines, yes.

Doremus Scudder said:

In any case, the effect seems substantial enough to make me think twice about presoaking negatives that need a lot of contraction.

Click to expand...

This also has implications for color film...

Could a pre-soak do something to the AHL (which then does something to development)?

Mr Bill said:

Hi, I believe it is, more than likely, mostly due to the dilution.

FWIW I have pretty extensive experience in large lab processing systems, although mainly in color. I spent years as the QC manager, with 5 or 6 full time employees, including a chemist and a decently equipped chem lab. I know from experience that if a C-41 (color film) developer has a replenishment error of about 10% this is enough to move the "control plots" from near-center to near the "action limit" specs. Meaning still within spec, but recommended to take corrective actions. I would judge such density shifts to be roughly in the same range as ic-racer has observed.

Now, I have no actual experience with the process configuration ic-racer is working with, nor have I ever investigated the effect of prewashing (out of about ten commercial processes I've worked with, NONE have recommended a prewash). So I am strictly guessing on this, but I do think that it is very plausible that the 10% (more or less) dilution is the main culprit here.

One last note... the sensitometric effects of a diluted developer seem to significantly depend on how sensitive the developing agent is to development byproducts. So the results may vary, depending...

Click to expand...

Oh, but it is not the dilution. the water carr over from the prewash is rather minimal but the impact on an already wet emulsion on the development process is huge in comparison.