Page 79 - Copper and Bronze in Art: Corrosion, Colorants, Getty Museum Conservation, By David Scott

on exposure of copper samples to various liquid, solid, and gaseous contaminants. The thick­
ness of the corrosion products that develop on the copper after one month of exposure is
determined using electrolytic cathodic reduction, establishing four levels of corrosivity. Other
organizations may divide the levels differendy, but the principal information is basically the
G
same. The lowest level of severity, labeled I , may typically have an H 2 S concentration of less
than 3 ppb; S0 2 , less than 10 ppb; Cl 2 , less than ι ppb; N O x , less than 50 ppb; and 0 3 , less than
2 ppb. Some of these levels may be considered high for the museum environment. The I rat­
G
ing of 3 ppb is too high for objects, and palliative measures are often taken at exposure levels of
i
loo ppt. Likewise, nitrogen oxides are often regarded with suspicion f levels as high as 50 ppb
are found inside the museum. The lowest possible values are desirable for any of these pollu­
tants. Levels of N 0 2 at the St. Vitus Cathedral in Prague, for example, reached 15 ppb, but
ammonia levels were substantial at 20 ppb. These ammonia levels could begin to have an effect
on exposed copper and brass, so it is impossible to consider anything in isolation. Johannson,
Rendahl, and Kucera (i998) and Johannson (i998) report on an interesting research project that
applied atmospheric corrosivity measurements to forty-two indoor sites, including nine muse­
ums and churches; these reports should be consulted for further information.
I DEPOSITION OF PARTICULATE MATTER Particulate pollution,
arising from tobacco smoke, asbestos, suspended particulate matter, and particles large enough
to be precipitated, is especially important as a corrosion initiator on polished metal surfaces. In
metropolitan areas today, most of the sulfur acquired by surfaces is not supplied in gaseous form
from reactions with sulfur dioxide but rather as dry deposition. Lobnig and colleagues (1993)
draw attention to the bimodal distribution of airborne particulates; predominating the distri­
bution are fine particles smaller than 2.5 μιη in diameter and coarse particles larger than 2.5 μιη.
For environments with extremely efficient air-filtration systems, the coarse particles are less
important, since they can be removed with about 95% efficiency, while the fine particles are
removed only at levels between 10% and 70%. The most abundant ions present in these fine par­
ticles are sulfate and ammonium ions. Lobnig and colleagues (1993) exposed polished copper to
simulated particulate matter consisting of submicron particles of ammonium sulfate. After five
(
days of exposure at 75% relative humidity, which is the critical level for NH 4 ) 2 S0 4 , corrosion
of the copper had occurred with formation of cuprite and antlerite. There is little doubt that sur­
faces left uncleaned for a long period of time, perhaps several years, create unsightly corrosion
problems that are not well documented in the literature.
Current interest in the subject is driven by the perceived threat of corrosion to copper
alloys in electronic materials caused by submicron-size atmospheric particulates. Comizzoli
and coworkers (1993), for example, provide data for the average airborne concentration of ionic
species and total mass of particles in the air of Newark, New Jersey, together with the average
accumulation rates for ions on both vertical and horizontal surfaces. For sulfates, the indoor
deposition velocity varied from 0.002 to 0.010 cm/s. Because the rate of deposition of ammo-

C H A P T E R O N E
62