|
Pre-Columbian Peruvian Metals
Sample 10
|
|
Tombaga Filigree
Vicus
North Coast, Peru
ca. 100 BC - 300 AD
|
|
Report by :
Laura Limata
June 5, 2003
|
| Introduction |
Figure 1: A macro photo of the object studied.
The area where the sample was removed from is labeled.
|
|
| This sample is a nose filigree, which is any delicate and
intricate ornamental work made from gold, silver, or other fine twisted
wire. It was dated to the Vicus culture of northern Peru from the period
between 100 BC and 300 AD. Grannulation is a technique that attaches tiny
gold balls to larger pieces fro ornamentation. This technique evolved
independently from the old world in places like Peru. The objective of
this investigation was to determine the joining method used for the gold
granulation of this filigree (nose piece). |
| Proceedure |
|
| A small part of the granulation from this sample was removed
from the upper right point of this sample with a jeweler’s saw as
shown in Figure 1. The piece was then mounted in epoxy and ground on 600
grit Silicon Carbide paper until all three joints between the balls could
be seen. The sample was then polished on 6 um diamond paste, 1 alumina
um, and 0.3 um alumina polishes. The sample was cleaned ultrasonically
in between each step to remove remnants of polishing compounds from the
void in the center of the sample. The sample was finished by polishing
with 0.05 um non-agglomerating alumina master prep. The sample was then
photographed with a Nikon Camera mounted to an Olympus Light Optical Microscope
(LOM). The sample was etched with an Aqua Regia solution consisting of
20 ml nitric acid, and 30 ml hydrochloric acid for four minutes. After
etching it was ultrasonically cleaned in EDTA, and then ethanol to avoid
staining. The etched sample was photographed again. The sample was also
studied under Scanning Electron Microscopy (SEM), and an Energy Dispersive
Spectrometry (EDS) scan was run to determine the composition. |
| |
|
| Figure
2: Polished, but unetched sample showing the joined balls. Picture
width is 1.5 mm. |
Figure
3: Etched sample showing potential composition differences at the
joints Picture width is 0.75 mm. |
| Results |
|
| Figure 2 shows the polished, but unetched sample.
Figure 3 shows the etched sample. Here the discoloration around the joints
indicates that the composition may be different. Figure 4 is a close up
on one of the joints that shows a moving grain boundary indicating diffusion.
The EDS scans for the matrix and the joint are in Figures 5 and 6 respectively. |
|
|
|
Figure 4: Etched
sample showing a moving grain boundary.
|
| Discussion |
|
Literature indicates that the technique of granulation developed
in South America among cultures such as the Vicus in Peru. It requires
the manufacture of small gold spheres, and a technique to attach them
to the surface. The granules were formed by taking lengths of thin gold
wire, and cutting it into small pieces. These pieces where then heated
and turned into spheres by surface tension. Another technique used in
Columbia was casting. The spheres studied here did have significant porosity
as displayed in Figure 2. This could be an indication that these spheres
were casted and not cut from wire as originally thought [1].
The granules then needed to be attached to the surface. Several techniques
were used. One is a solder, although this is not likely here since there
was not a significant compositional difference shown from the EDS scans
between the matrix and the joint. Another technique used is in situ copper
salt reduction. An organic glue was mixed with copper salts and used to
attach the spheres to the base metal. The piece was then heated allowing
the spheres to fuse. The use of a copper salt lowered the melting temperature
of the metal, making it easier to fuse. This is a possible technique for
this case because of the small amounts of copper picked up in the EDS
analysis. This copper could also have come from the base metal, which
is believed to be a tumbaga. Further analysis will be needed to see how
the copper is spread throughout the joint, and if there is a high enough
concentration. The third technique used is to simply glue the spheres,
without salts, and allow the spheres to melt on their own. This would
obviously require tight temperature control, and would be quiet difficult.
This is actually a likely formation technique for this sample, because
of the small compositional differences between the matrix and the joint.
In addition, the moving grain boundary points to the fact that higher
temperatures may have been used, which again gives this technique credibility
[1].
|
|
|
Figure 5: EDS scan taken in the
center of one of the balls. |
Figure 6: EDS scan taken at the
joint, showing a small increase in the amount of copper. |
