Page 136 - Copper and Bronze in Art: Corrosion, Colorants, Getty Museum Conservation, By David Scott

example, f the natron was primarily a mixture of sodium carbonate and sodium bicarbonate,
i
then a gelatinous duck-egg blue precipitate forms. f the natron also contained some sodium
I
chloride and sodium sulfate, then blue-green crystalline deposits are obtained from the precip­
itate by filtering and drying. Recent analyses of samples labeled "natron" at the British Museum
indicate that in many cases, the natron is, in fact, principally composed of sodium chloride.
Under the polarized light microscope, two different crystalline reaction products can be
distinguished when natron containing sodium chloride, sulfate, carbonate, and bicarbonate is
used in the recipe: one product consists of pale blue crystal fragments, and the other contains
spherical particles with radial markings in the spherical crystalline precipitate. Both products
are highly biréfringent. In a melt-mount of refractive index 1.66, the crystalline fragments have
lower refractive indices, indicating that these grains could not be malachite. In fact, this precipi­
tate produces an X-ray diffraction pattern that matches that of chalconatronite, as shown in
APPENDIX D, TABLE ι. This replication of an ancient Chinese recipe manifestly demonstrates
the possibility of finding synthetic chalconatronite on very old works of art as an original pig­
ment and not as an alteration product of another mineral.
Banik (i989) reports identifying green pigment on a sixteenth-century illuminated manu­
script as a mixture of pseudomalachite and chalconatronite. This indicates that chalconatronite
(probably synthetic) could have been used as a pigment more commonly than the current liter­
ature suggests. Magaloni (i996) made a recent and surprising identification of chalconatronite
(probably natural) as a pigment in wall paintings from two Maya sites in Mexico: Bonampak
and Las Monjas. This finding is intriguing; further research into the etiology of these salts is
needed to determine f they are original pigments or alteration products.
i

Chalconatronite as Even when sodium sesquicarbonate cleaning is not suspected,
a corrosion product residues from aqueous cleaning solutions may produce chemi­
cal alterations. For example, a gilt-silver stag (ca. 168O-1700)
17
by Johann Ludwig Biller the Elder (i656-i732) at the J. Paul Getty Museum was observed to be
producing small green pustular nodules on the surface. The object was made with a silver alloy
that was found to contain copper. Remnants of cleaning solutions used in the past must have
created incipient corrosion of this copper content, resulting in the small green pustules. These
were identified by scanning electron microscopy-energy dispersive X-ray analysis to contain
sodium and copper. They were identified by Debye-Scherrer powder X-ray diffraction as chal­
conatronite (Scott 1997a).
One cleaning method that can cause such alteration is immersion in a 5% solution of
i
sodium sesquicarbonate, NaHC0 3 · Na 2 C0 3 . Although it is not known f Biller's stag was treated
in this manner, the method was recommended as an early treatment to remove chloride from
both iron and bronze artifacts (Scott 1921). The sesquicarbonate solution has a pH of about 10;

BASI C C O P P E R CARBONATE S
119