Page 246 - Copper and Bronze in Art: Corrosion, Colorants, Getty Museum Conservation, By David Scott

underwater finds from the work of MacLeod (i982, 985) is illustrated in PLATE 9 . PLATE 4 0 is
1
3
a scanning electron photomicrograph of a leaded brass nail from the Rapid showing extensive
corrosion producing covellite, chalcocite, and a variety of lead sulfides.
Nord, Lindahl, and Tronner (1993) described yet another sulfide found on a copper alloy
compass ring salvaged from the wreck of the S S Kronau, which sank in the Baltic Sea in 1676.
The object had an inner, grayish corrosion layer with occasional bluish tinges that was identified
by X-ray diffraction as spionkopite, Cu 39 S 28 . This product was only characterized as a natural
mineral in 1 9 8 0 (Goble i 9 8 o ) . Covellite was also identified from this particular Baltic site. Less
reducing sediments favor covellite formation over spionkopite, which has a stoichiometry that
approximates to Cu 14 S.
Corrosion during burial or in stagnant conditions, such as a hot spring, may deposit cop­
per sulfides on other materials or react with contiguous objects to form mixed cation products,
as the work of Daubree (1881) has shown. Another example is the Anglian helmet from the Cop-
pergate burial site at York, England. The helmet is a crested Scandinavian type analogous to
one found at Sutton Hoo, 6 Suffolk, England, and has a complex construction of iron plates
reinforced and decorated with copper alloy bindings (Tweddle 1992). The helmet was well pre­
served because it was buried in an anaerobic deposit. This oxygen-deficient environment cre­
ated unusual corrosive events resulting in siderite, FeCO s , identified on the iron plates, and
chalcopyrite, CuFeS 2, and bornite, Cu 5 FeS 4 , on the copper alloy components. All of the min­
eral identifications were determined by X-ray diffraction. Bornite was also reported by Organ

(198i) as part of the encrustation on a Classical bronze horse in the collections of the Metropol­
itan Museum of Art.
In an extensive study, Duncan and Ganiaris ( i 9 8 7 ) described the corrosion products found
on first-century C.E. Roman bronzes from waterfront sites along the Thames in London, where
land reclaimed from the river produced areas of anaerobic activity, rich in sulfate-reducing
bacteria. Medieval sites such as York, England, and Trondheim, Norway, may possess similar
environmental conditions. The bronzes from the Thames were sometimes found to be covered
with a dull, golden, sulfidic layer over a black corrosion crust, such as that exhibited by the
8
fourteenth-century bronze trumpet shown in PLATE 41. Oddy and Meeks (19 i ) described a sim­
ilar golden layer on a Roman brass bust in the British Museum as "pseudo-gilding," but Duncan
and Ganiaris, who studied the same object, determined that this layer was, in fact, a natural cor­
rosion layer rather than a deliberately gilded surface. Analysis showed that these golden layers
are composed of chalcopyrite, CuFeS 2, or pyrrhotite, Fe (1 . x) S, often in combination. PLATE 42
shows a partially cleaned Roman coin with a black sulfide layer under chalcopyrites; PLATE 43
illustrates a medieval key covered by botryoidal chalcopyrites and revealing some of the metal­
lic core underneath. The minerals in the black crusts on sheet-copper objects from waterfront
sites were identified as chalcocite, Cu 2 S; covellite, CuS; and rare occurrences of geerite, Cu 16 S.

C O P P E R S U L F I D E S
229