Shopping for a hardener (Re: Glyoxal?)

I can't believe people continue on this issue forever... and repeating
the same debate over and over.

In my view, an ideal hardener meets the following criteria:

(1) crosslinks gelatin to reduce swelling factor to 5x or less
(2) crosslinks gelatin reasonably fast
(3) crosslinks gelatin in wide range of conditions (robust reaction)
(4) does not compromise archival properties
(5) has relatively low toxicity and manageable risk in case of accidents
(6) has low environmental load (when inactivated prior to disposal)
(7) has reasonably long shelf life
(8) is preferrably nonvolatile
(9) is preferrably inexpensive
(10) is preferrably commercially produced and available

No known hardener meets all of them.

In my view, glut solution prepared according to my recommendation
meets all but 8. On the other hand, monofunctional aldehydes (such as
formaldehyde) and shorter bifunctional aldehydes (such as glyoxal and
succinylaldehyde) lacks 2, 3, 8, and more arguable on 5 than glut, due
to larger quantity required in practice, and therefore larger quantity
of excess agent. Glyoxal is also more susceptible to aerial oxidation
and shelf life, 7, is inferior. Mucochloric acid, mucobromic acid,
etc. are also excellent gelatin hardeners, but these are even more
toxic.

I also have a few other preferred hardening agents.

Bisepoxides are another class of excellent hardening agent. You can
buy this in a bottle and you can mix them in gelatin solution
as-is. This is inferior on 2, and the costs a bit more than others,
but meets all other criteria. People who need nonvolatile hardening
agent should consider this kind of hardeners. Indeed, those who size a
lot of paper to supply for months of use, this is probably the most
ideal agent, as you can work indoors with no special ventilation while
coating and drying the paper.

Substituted s-triazine is an excellent hardening agent, which meets
all but 10. Synthesis of this compound is simple but involves a highly
irritating chemical. I strongly suspect that Fuji Photo Film has been
making this kind of compounds for their own use since 1960s and they
still use it in part, but besides, I do not know of any small photo
company who is set up to synthesize this kind of compounds.

Partially oxidized starches and sugars (aldehyde startches and
aldehyde sugars) can work well if the distance of two aldehyde groups
are comparable to that in glutaraldehyde. Volatility should be much
less but the reaction may be slower. One problem is that I don't know
of a commercial source of such compounds.

Glyoxal is a well studied compound and we have good knowledge of what
kind of reactions it gets involved. But because it can react in many
different ways, it's nontrivial to predict what is going on in
yellowing of paper. It can get oxidized, it can polymerize, it can
generate corrosive acids, etc., and combination thereof in various
sequences. The reaction is heavily influenced by humidity of the air,
pH, presence of trace impurities, etc. and all these factors only add
up on the uncertainty in the usability of the sized paper. If it works
for someone, it is relying on the luck in many unknown factors. It's
a major shortcoming of glyoxal as a hardening agent, like Chris said;
it's like a condom that fails 17% of the time. I don't know about your
life but my life is better if I can count on sizing and condom 100%.

From: Katharine Thayer <kthayer@pacifier.com>
Subject: Re: Glyoxal?
Date: Tue, 17 Jan 2006 07:27:22 -0800

> I guess I think about these things in terms of hypothesis testing.
> To you, it seems any observation of yellowing is all it takes to say
> that glyoxal yellows. I can see the logic of the reasoning, but
> it's a different kind of reasoning than the scientific, statistical
> kind of thinking that I am used to. Katharine

I don't see how your logic is more scientific or statistical than
Chris's. You are just flipping around the null and alternative
hypotheses that Chris and I have. To me, the null hypothesis is

Glyoxal is a reliable hardener free of trouble

and we have enough evidence to reject this. To me, one failure out of
20 is enough to reject this hypothesis.

For glut, s-triazine derivatives and bisepoxides, I have tried hard in
many practical conditions, but I have no evidence in contrary to these
compounds being excellent, reliable, robust hardeners.
Received on Wed Jan 18 17:37:05 2006