Randolph Herr said in his note, "The Curse Of Perflex," to the
05.05.06 MMD:

  > We have come a long way since then, scientifically speaking, and I
  > am certain that there must be a plastic film out there that will be
  > fine for pouches. If a plastic film could be found, then we would have
  > the perfect airtightness and perfect flexibility, that would make
  > these pouches out-perform any leather.  Obviously, any material
  > suggested for this use should be laboratory tested to avoid another
  > Perflex disaster.  I have heard that it was the air pollution that
  > destroyed the Perflex, but this is just anecdotal.

Yep.  We've discussed this before, and I once suggested a program of
laboratory testing for all of the materials commonly used in instrument
restoration.  I believe the material in question at the time was
rubberized cloth.

There is, unfortunately, no such thing as artificial time.  I once built
the world's simplest and cheapest museum exhibit, entitled "Very Slow
Chemical Reactions."  It consisted of a plastic tray upon which I placed
four light-brown, translucent rubber feet removed from a piece of
chemistry equipment we were restoring.  The feet were gradually
deteriorating -- liquefying, actually -- and you could see streaks of
discoloration extending through the feet that were still nominally
sound.

As the weeks wore on, the more-deteriorated foot liquefied completely
into a puddle, another became completely riddled with streaks and
started to liquefy at one corner, and the streaks advanced visibly
through the others.  Last I looked, they were all pretty well liquefied.

Someone told me that these were probably an early version of
polyurethane rubber.  I don't see how the deterioration could have been
predicted by any testing method.  There are some testing methods that
work for rubber compounds, mainly exposure to intense ozone
concentrations, heat, cold, and sunlight.  But my sense is that my
polyurethane feet survived all that, and still deteriorated.

Another example of unexpected and perhaps unpredictable material
breakdown got me through graduate school.  The foam-plastic suspensions
that surrounded the cones of most loudspeakers of the 1980's
deteriorated into gum by the 1990's.  This was unfortunate for many
people, but I was running a stereo repair shop at the time.

So the moral, I guess, is to test -- but verify with real time.  We
know that traditional materials will last until eaten by mice or
bacteria.  The history of any plastics that you'd expect to use should
be investigated back to their first industrial application.  It's 2005
now, so any material that's been holding up pretty well since, say, 1965
ought to be reasonable for instrument reconstruction.

Once chosen, a particular lot of material should -- must, I say -- be
authenticated by its manufacturer as to quality and date of manufacture.
If this sounds like the sort of precaution taken with aircraft parts,
it should: while we're not sending these magnificent devices into orbit,
they should be as reliable as a piece of aerospace equipment, if we
don't want to be rebuilding them in our lifetimes.

M. Kinsler