Page 124 - Copper and Bronze in Art: Corrosion, Colorants, Getty Museum Conservation, By David Scott

bonate in agar gel and found that small amounts of impurities affected its development. To study
the phenomenon, the researchers added sodium carbonate solution to an agar gel containing
dissolved copper (II) acetate. An impressive array of copper carbonate bands developed as the
carbonate solution diffused through the agar. The authors explain the periodic precipitation of
copper (II) carbonate as the formation of a precipitate by flocculation of preformed peptized
sols. The copper carbonate precipitate forms because the colloid is coagulated by an excess of
the diffusing electrolyte. The carbonate is absent when the colloidal sol in the vicinity of the pre­
cipitate is coagulated and adsorbed by the precipitate.
Similar processes may be occurring to produce the complex corrosion crusts of banded
malachite and cuprite. In some cases, as many as seventy alternating layers of the two minerals
have been observed in bronze corrosion crusts. This strongly suggests that sols or gels form dur­
ing corrosion underground, allowing this unusual diffusion phenomenon to occur.

Isotope ratios to determine Ratios of the stable carbon and oxygen isotopes, 1 3 C / C and
1 2
corrosion environment 1 8 0 / 1 6 0 , in malachite corrosion crusts were investigated by
i
Smith (i978) in a pilot study to see f they could be used to deter­
mine the burial environment of bronzes. Patina samples containing malachite were examined
from bronzes found in British Bronze Age hoards; from Etruscan bronzes from Polledrara,
Italy; from Roman objects found in Sabratha, Libya; and from Shang dynasty and Han dynasty
bronzes from China.
The results were complex but showed that all the British bronzes had delta O results 5
1 8
between -1 and -4 % 0, while the other samples showed a wide spread in values from o to -11 % 0
with no correlation between patina and site.
When corroded by waters containing oxygen and carbon dioxide, the copper that is
attacked by the water and that goes into solution is mostly in the form of cupric bicarbonate
(Ives and Rawson 1962). Fractionation can occur between the carbon dioxide and the bicar­
bonate ions, resulting in C becoming more concentrated in the bicarbonate solution. Con­
1 3
sequently, malachite deposited from bicarbonate solutions might have delta C ratios more
1 3
positive than the biogenic C 0 2 values of-26 % 0. This was confirmed by Smith's pilot study. The
degree of corrosion of the bronze objects studied and the delta C of the malachite patina
1 3
appeared to be related: severely corroded objects had the most negative delta C ratios, and
1 3
moderately corroded objects the least negative. This would indicate that the consistency of the
British data is related to the C 0 2 concentration in the soil.
Smith (i978) suggests that one possible application of these isotope ratios would be to assist
with authenticity studies. Comparing the delta O and delta C ratios of malachite from the
1 8
1 3
patina of an unknown bronze object with the same ratios from an ancient bronze of known
provenance might make it possible to draw some conclusions about origins. This application,
however, has yet to be verified by any practical studies.

BASI C C O P P E R CARBONATE S
107