Page 78 - Copper and Bronze in Art: Corrosion, Colorants, Getty Museum Conservation, By David Scott

I TESTING MATERIAL DAMAGE There are a number of tests
available to assess material damage. A common and useful series of evaluations are the so-called
Oddy tests for general material damage to metallic coupons of lead, copper, and silver. This is
supplemented with the sodium azide test for sulfides, the chromotropic acid test for formalde­
hyde, and the indicator-paper test for pH measurements. Further details concerning these tech­
niques can be found in Oddy (1973), Daniels and Ward (1982), Green and Thickett (1995), and
Lee and Thickett (i996). Lee and Thickett provide a particularly easy to read and comprehen­
sive account of the subject. A number of surveys of museum pollutants, primarily of organic
acids and aldehydes, have been conducted by the Getty Conservation Institute. The critical
indoor pollutant concentrations below which corrosion can be assumed to be very slow or non­
existent are difficult to establish for metallic antiquities. The work of Grzywacz (1993), how­
ever, suggests that the concentrations of organic acids and aldehydes should be kept in the range
I
of 10-20 ppb. f the concentration reaches 100 ppb, this is generally considered to be too high.
For hydrogen sulfide, the levels are in the low parts per trillion, but since detection of this gas
in the museum environment is always a worry because of its very high reactivity with works of
art, it is doubtful f there is any acceptable level of hydrogen sulfide.
i
I ASSESSING CORROSION RATES I N INDOOR ENVIRONMENTS
Rice and colleagues (198i) made indoor corrosion rate measurements on copper and other
alloys at eight locations in the United States; pollutant and relative humidity data were recorded
from six of them. The tests were carried out for only two years, however, from 1973 to 1975,
and this is not really long enough to gauge long-term effects. The pollutants measured
included sulfur dioxide, nitrogen dioxide, ammonia, reduced sulfur gases (H 2 S, S, and CH 3 SH),
chlorine gases, and airborne dust. The corrosion rates found for copper showed a general
decrease with time, particularly for the less corrosive environments. The indoor copper cor­
rosion rate distribution for the measurements was a lognormal function with an excellent
correlation coefficient of 0.97. The rate of corrosion indoors is usually orders of magnitude
19
less than the rate outdoors, unless high levels (hundreds of ppb) of organic acids are generated
in the indoor environment. The functional dependence of the corrosion rate, r, on relative
humidity, RH, is of the form r = ae 4,6RH , with little evidence for a critical relative humidity in
a complex environment.
Sulfur dioxide and ozone were shown to have a significant influence on the rate of corro­
sion, while nitrogen oxides, chlorine, and ammonia had substantially less influence. Although
the study by Rice and coworkers (198i) provides a useful insight into indoor corrosion rates, it
is too theoretical for application to museums. Careful routine observation of objects, combined
with testing of all materials used in museum construction, display, and storage, is a practical
approach to dealing with corrosion possibilities engendered by museum pollutants.
More recent studies have attempted to classify indoor atmospheres by degrees of corrosiv-
ity. The Instrument Society of America (ISA-S71.04), for example, proposes an evaluation based

C O R R O S I O N AN D E N V I R O N M E N T
61