Solvent gels

Agnes Gall-Ortlik <gallortlik [at] yahoo__fr> writes

>We are actually working on the cleaning of an outside wall made of
>enameled stoneware plates. It was covered with polyurethane varnish
>during the 1970's and today the resin is very deteriorated ...

I think you are definitely on the right lines in focusing your
solvent gel around acetone which, in my experience, is a moderately
good solvent for swelling polyurethane coatings; and to look to
increasing the activity of this solvent by admixture with others,
possibly with greater intrinsic swelling power, seems also a very
sensible thing to pursue.

Having looked at the mixtures you have tried already, I might
suggest a few other lines of enquiry, as follows:

If acetone is too rapidly evaporating, you could try going to a less
volatile ketone, such as butanone (also known as methyl ethyl
ketone). While this is not quite as safe to use as acetone, it still
cannot be considered to be an especially hazardous solvent, having a
long-term (8 hour t.w.a.) Occupational Exposure Limit of 200ppm.  My
suggestion, though, would be to stay with acetone as the principal
solvent, but to look to increase its swelling action further by
addition of other solvents.

The polyurethane coating here is just 30 years or so old, in which
case it may not have oxidised to quite such a severe extent as you
might think: it may actually retain quite a degree of non-polar,
hydrocarbon character. Accordingly, you might consider introducing
into your solvent mixture a component which reduces the overall
polarity, ie. increases the Teas fd value of the mixture.  The
logical choice would be an aromatic solvent, such as xylene or
Shellsol A100, but these solvents are perhaps not the most pleasant
to use.  In place of a wholly aromatic solvent, though, you might
consider including in your mixture a solvent like white spirits (ca.
17-19% aromatics), at a proportion of, say, 30% or less by volume.
You might find that the reduction in polarity actually serves to
increase the swelling power of the solvent mixture.  However,
shifting the overall polarity of the solvent mixture to lower
polarity might make it more difficult to create a gel with Klucel or
MHPC, since these are relatively polar cellulose ethers. If this is
the case, then using ethylcellulose might give better results.

An alternative approach to using a hydrocarbon might be to try a
chlorinated solvent, but there are either environmental or health
risks associated with that group of solvents: dichloromethane, for
example, is now given the Health Risk Phrase R40 - Limited evidence
of a carcinogenic effect'.

As far as solvents that are considerably more active in terms of
swelling power are concerned, I am somewhat surprised that you have
considered 2-ethoxyethanol (also called cellosolve).  This certainly
cannot, as your opening comments imply, be considered to be a member
the group of less harmful solvents.  In fact, from the health and
safety point of view, 2-ethoxyethanol (together with its related
compounds, 2-ethoxyethyl acetate [ie. cellosolve acetate] and
2-methoxyethanol [ie. methyl cellosolve]) is one of the most
problematic solvents in use in conservation.  This group of glycol
ether solvents presents some serious health hazards.  Not only are
they currently allocated very low Occupational Exposure Limits
(OELs), in the EU they are included in the group of substances which
present such particular concerns that they are assigned Maximum
Exposure Limits (MELs) which are levels that must never be exceeded
in the workplace.  For information, the current UK and EU
occupational exposure levels for some of the solvents mentioned here
are:

                        MEL?   Long term (8 hr)  Short term(15 min.)
                      (Y / N)  Occ. Exp. Limit    Occ. Exp. Limit
                                (OEL or MEL)      (OEL or MEL)

acetone                  N             500                 1500
butanone                 N             200                  300
xylene                   N              50                  100
2-ethoxyethanol          Y              10                    -
2-ethoxyethyl acetate    Y              10                    -
2-methoxyethanol         Y               5                    -
1-methoxypropan-2-ol     N             100                  150

The group of cellosolve solvents I mention are now actually all
formally classed as 'toxic', which means they all carry the hazard
warning label, 'T' for 'toxic' rather than 'Xn' for 'harmful'.
Furthermore, they are all identified as Reproductive Toxins,
Category 2, by virtue of the main health risks they present, namely
'R60 - May impair fertility' and 'R61 - May cause harm to the unborn
child'.  They are flammable, and can be harmful by skin absorption,
inhalation or swallowing.  Considering conservation is populated by
an abundance of young women of child-bearing age, I find it
remarkable that solvents such as the cellosolves continue to find
any use at all, especially since there are much safer alternatives
with almost identical solubility properties.

If you do wish to use a solvent that is an ether-alcohol, then I
would suggest that, instead of 2-ethoxyethanol, you use
1-methoxypropan-2-ol, which can be obtained from VWR/Merck or other
laboratory suppliers, or from Shell as their product Methyl
Proxitol.  This solvent has exactly the same functional groups and
molecular formula as 2-ethoxyethanol, but it is an order of
magnitude safer according to its occupational exposure limits (see
table above) and it is not presently a concern with regard to
reproductive toxicity.

However, rather than an ether-alcohol as the solvent you mix with
acetone to increase the swelling power on the polyurethane coating,
you might consider one of the very strongly dipolar solvents, added
to the acetone (or acetone + white spirits) in relatively small
proportion, which could be from as little as 1% up to, say, 10-20%.
Personally I would look to something like N-methyl-2-pyrrolidone
(NMP) for such purposes.  NMP is a very strongly swelling solvent on
coatings such as cured oil films and probably therefore also
polyurethanes; it should, however, be treated with caution, both in
terms of its activity and its health hazards (Long term [8 hour] OEL
25ppm; Risk phrase 'R36/38 - irritating to eyes and skin').

I'm sure that other people would be thinking along similar lines,
but with another strongly dipolar solvent, dimethylsuphoxide (DMSO).
Personally, I have no experience of using this for polyurethane
coatings, but it may well work, though I would be wary about its
tendency to be strongly retained on the object's surface.

Regarding the differential degradation of the polyurethane over
different colour fields, I agree that light and, perhaps, thermal
effects are probably responsible, but one might also consider the
possibility of some photochemical inhibition of degradation by the
pigment in the substrate, but that would depend very much on the
specific chemical nature of the colorant.

Alan Phenix
Senior Lecturer, Conservation of Fine Art
Northumbria University
Newcastle upon Tyne
UK


                                  ***
                  Conservation DistList Instance 18:57
                  Distributed: Wednesday, June 8, 2005
                       Message Id: cdl-18-57-002
                                  ***