Page 344 - Copper and Bronze in Art: Corrosion, Colorants, Getty Museum Conservation, By David Scott

SOME PATINA V A R I A T I O N S

Arsenic coating as a patina In the early 1970s an arsenic-rich coating found on the sur­
face of an Anatolian bronze bull dating from about 2100 B.C.E.
i
raised the question of whether such a coating was a feature of the original casting or f it was
applied separately over part of the surface (Smith 1973). The figure is from the northeastern Ana­
tolian site of Horoztepe, now in Turkey. Smith thought the bronze contained no arsenic and that
it must have been added as a separate coating or patination. It was difficult, however, to recon­
struct how the remarkable technology of making an arsenic patina might have been devel­
oped four thousand years ago in Anatolia. Smith found it unlikely that the copper-arsenic alloy
could have been applied in the molten state; the microstructure revealed no sign of a eutectic
alpha+domeykite (Cu 3As) phase, which makes application in the molten state improbable.
More recently, Beale has offered a different approach, showing that such bronze bulls were
cast in four or five pieces. An arsenic-rich surface, therefore, could be related to arsenic in the
8
alloy used in the casting. Beale found that the hindquarters of one such bull is cast in an arseni­
cal copper while the front legs are made of leaded tin bronze. X-ray diffraction analysis of the
panels to each flank and the hindquarters of the bull showed the presence of the copper-arsenic
phases whitneyite and domeykite, suggesting that surface segregation of arsenic is an explana­
tion for the presence of the domeykite phase. Inverse segregation on casting is probably respon­
sible for the coating in this case, since during the casting process some of the lower melting
point constituents, such as arsenic-rich phase, could be carried to the outer surface of the mold.
At the time that Smith was writing, it was thought that the coating must have been applied
by a cementation process. f this were the case, areas of the surface that were to have been left
I
bronze in color would have been coated with an impermeable stop-off material. Although
arsenic oxide sublimes at 193 °C, the potash and soda that would be necessary to make these
alkalies would be available from plant ash, which are easily reduced with charcoal at moderate
temperatures. Smith's laboratory replication was only partially successful in producing a
domeykite coating on a bronze replica trial piece. It used 4 parts arsenic oxide (As 2 0 3 ), 1 part
dry potassium carbonate, and 0.3 parts carbon black, all finely powdered and heated with the
copper object for one hour at 450 °C. The paucity of scientific data for Hittite bronze and bronze
technology means that we have no idea how prevalent such coatings, or segregations, may have
been in ancient Anatolia. Arsenic coatings are also found on some Egyptian bronzes (Fink and
Polushkin 1936), but no metallographic evidence has been published so far. Until that work is
available, the Egyptian technology cannot be considered properly characterized; the coatings
may simply be due, again, to surface segregation of arsenic during casting.

S O M E A S P E C T S O F B R O N Z E PATINAS
327