Page 390 - Copper and Bronze in Art: Corrosion, Colorants, Getty Museum Conservation, By David Scott
P. 390
cases, the effects were not great. Alkaline glycerol and formic acid treatments adversely affected
coins containing lead.
Another use of Rochelle salt for cleaning an important, exposed bronze surface is worth
noting here. Around 1990 Ghiberti's famous bronze doors on the Baptistry in Florence, Italy,
were taken indoors and replaced by replicas, while the conservation studios of the Opificio delle
Pietre Dure in Florence undertook conservation work on their gypsum-encrusted surfaces. 21
Rochelle salt was shown to be the most effective treatment for this large bronze relief. The salt
was used in neutral aqueous solution, however, without any alkali additions, at a concentration
of about 30% (w/v). This solution effectively treated some of the individual gilded door panels
without disrupting the cuprite surface immediately under the gold. In other areas of the surface,
however, brochantite and antlerite existed within small pits under the amalgam gilding, result
ing in the gold surface being blistered with numerous small mounds. Some of these blistered
gilded areas burst open to reveal copper sulfates; beneath these sulfates some copper chlo
rides — such as clinoatacamite and paratacamite, or even nantokite — may exist. These open blis
ters could not be fully treated.
After treatment of the door panels, distilled water was used to wash away residues of the
Rochelle salt. The panels were then dewatered with acetone and finally heated slightly to dry.
The doors are now on exhibit in nitrogen-filled display cases in Florence.
Immersing objects in large tanks of aqueous solutions is difficult and can create practical
problems. Aqueous gels based on cleaning reagents that incorporate Polyacrylamides such as
buffered citric acid or tartrates with B TA additions may eventually prove useful for cleaning
bronze objects. 22
Washing marine finds in distilled water has also been attempted, but, although this may
wash out chlorides by hydrolysis of cuprous chloride, the process is very slow. MacLeod (i987a)
i
estimates that effective stabilization can be attained only f the washing period is from two to
four years, which is generally impractical. During such long periods of immersion, reactions
with other corrosion products might occur, and these would not be controllable. For these rea
sons, this method cannot be recommended.
Another approach to the washing process for bronze objects is immersion in sodium ses
quicarbonate solutions, subject to the patina alteration problems previously discussed. Sodium
sesquicarbonate does remove more chloride ions than distilled water, although at 5% (w/v) in
water this solution also slowly removes copper from the sound metallic regions of an object as
well as from the corrosion crust.
MacLeod (1987a) also noted that a solution of 1% (w/v) of aqueous benzotriazole with
5% (v/v) of ethanol is capable of displacing a large proportion of chloride-containing corrosion
products. Some of these treatments have not been successful, however, possibly due to insuffi
cient impregnation time. Immersion in this solution for a few months, rather than a few days,
may lead to better corrosion resistance of the treated artifacts.
C O N S E R V A T I O N T R E A T M E N T S F O R B R O N Z E O B J E C T S
373