Re: Gum hardening -- top down?

From: etienne garbaux ^lt;photographeur@nerdshack.com>
Date: 04/05/06-11:03:26 PM Z
Message-id: <p05210600c05a44c65ef9@[192.168.1.100]>

Katharine wrote:

> AFAIK, it has hardly been "established" that gum hardening occurs at
> the paper surface, as Judy suggested, but it's also interesting that
> Sandy is so certain that hardening occurs from top down that he
> claims he will remain certain, no matter what proof is offered. :--)
> I myself am waiting for further evidence to illuminate the issue.
>
> In the past, the "proof" that has been offered here for hardening at
> the paper surface is no proof at all but Mike Ware's speculation that
> the dichromate may be strongly absorbed to the paper; he thinks it
> may move down through the layer and congregate at the paper
> surface, and if this is so, that is where hardening would take
> place, because that is where the bulk of the dichromate would be
> found. But it's just a hypothesis, and he has offered no evidence
> that I know of to support this hypothesis.

As always, I will preface my comments with the disclaimer that I am not a
gum printer. I do have quite a lot of experience with carbon, carbro, and
dye-transfer, all of which involve differentially hardened gelatin.

I think it is established beyond serious question that dichromated gelatin
hardens from the exposure surface through the thickness of the coating. If
the exposure is made through the substrate, the image is adhered to the
substrate and shows clear relief after "development."

This seems quite reasonable, given (1) that hardening occurs in proportion
to exposure and (2) that actinic light is attenuated by the dichromate
itself, as well as any pigment in the emulsion. These two principles imply
that maximum hardening will always occur on and near the surface of
exposure.

Absent any countervailing evidence, one would expect the same two
principles to lead to the same conclusion with respect to gum.

For our thought experiment, one more principle needs to be stated: we
expect hardening to be proportional to the dichromate concentration in the
coating, as well as to exposure.

We are quite certain that in dichromated gelatin, the dichromate
concentration is approximately erqual throughout the thickness of the
coating. If the paper substrate really does leach the dichromate out of
gum, this could at least partly explain greater hardening at depth than at
the exposure surface. This suggests that pretreating paper with dichromate
before coating with gum could enhance the effect. It also suggests
experiments to explore the theory. (1) Coat a gum layer, allow it to dry,
scrape off some of the surface layer, and analyze the scrapings for
dichromate content. (2) Using a non-reactive surface or container, measure
the hardening effect on gum of various dichromate levels and exposures;
also, measure attenuation of actinic light per unit thickness for
unpigmented and "typically" pigmented gum. These tests should allow one to
predict how much dichromate differential it would take to offset or
overshadow the decrease in exposure with depth.

My guess is that whatever dichromate gradient might be found will not be
sufficient to explain the apparent lack of "top-down" hardening in gum
printing.

I am more inclined to think that the effect is mostly macro and mechanical.
To wit: relatively less-hardened gum can avoid being washed away when it is
"reinforced" with the matrix of paper fibers lower in the layer than toward
the top of the layer where such fibers are fewer or absent entirely. Think
of washing paint out of a brush. At the ends of the bristles, where they
are free to move, it is relatively easy to wash out the paint. But down at
the root, where they form a dense matrix, it is much more difficult -- not
because the paint is more hardened there (in fact, because air cannot
penetrate, it is probably significantly *less* hardened there), but because
it is harder to wash out even unhardened paint from the dense matrix.

Best regards,

etienne
Received on Wed Apr 5 23:04:01 2006

This archive was generated by hypermail 2.1.8 : 05/01/06-11:10:23 AM Z CST