XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 1479
works have employed the use of Raman to predict gold rob-
bing using the Raman ratio and the D/G area ratio, while
correlating this value to the gold-rob #(Helm et. al, 2009
Ren et. al, 2017 Ng et. al, 2022). Carbon spectra feature
a characteristic D band, or “defect” band, at 1,330 cm–1
and G band at 1,600 cm–1, as well as a secondary disorder
band at 1,500 cm–1. More crystalline graphite displays tall,
narrow G bands, while spectra for activated and amorphous
carbon exhibit wider D and G bands, and taller D bands in
comparison to graphite (Ren et. al, 2017 Ng et. al, 2022).
Although amorphous carbon is acknowledged to be more
“disordered” than crystalline graphite within these studies,
the size of the D peak in Raman spectra does not accu-
rately indicate the origin or quantity of defects within the
crystal structure. Both the Raman ratio and D/G area ratio
fail to represent carbon sorption tendencies sufficiently,
highlighting the limitations of using Raman analysis alone
for organic carbon characterization (Ren et. al, 2017 Ng
et. al, 2022). While Raman spectroscopy is valuable for
analyzing the chemical properties and identifying poten-
tial origins of CM, it cannot fully characterize organic
carbon, nor predict gold robbing accurately. Therefore, a
comprehensive understanding of characterization requires
the integration of complementary methodologies beyond
Raman spectroscopy.
METALLURGICAL RESPONSE OF A
POLYMETALLIC COMPLEX ORE
The deposits considered for this study contain lead and zinc
sulfides as well as gold and silver hosted in various mineral
phases. The samples selected have variable amounts of CM
in the range of 0.01 to 0.53%, determined by leaching car-
bonate from the sample with hydrochloric acid and ana-
lyzing the residue with a carbon/sulfur analyzer. Figure 1
shows the CM content in the selected samples and the
corresponding gold robbing number, typically utilized to
quantify the gold robbing characteristics of refractory ores
(Miller, Wan and Diaz 2016). This number can be used
for an initial grouping of samples into four classifications
of gold robbing: non, mildly, moderately and highly gold
robbing.
Further metallurgical characterization was based on
bench top leach tests on a pyrite concentrate obtained fol-
lowing sequential lead, zinc, and pyrite flotation stages, as
shown in Figure 2. Prior to the flotation stages, the samples
were prepared by crushing and grinding in a laboratory rod
Table 1. Gold robbing classification
Property Non Gold Robbing Mildly Gold Robbing Moderately Gold Robbing Highly Gold Robbing
Gold-rob #0 0–1.0 1.0–2.5 2.5
Figure 1. Gold robbing number vs. carbonaceous matter content in the selected samples
works have employed the use of Raman to predict gold rob-
bing using the Raman ratio and the D/G area ratio, while
correlating this value to the gold-rob #(Helm et. al, 2009
Ren et. al, 2017 Ng et. al, 2022). Carbon spectra feature
a characteristic D band, or “defect” band, at 1,330 cm–1
and G band at 1,600 cm–1, as well as a secondary disorder
band at 1,500 cm–1. More crystalline graphite displays tall,
narrow G bands, while spectra for activated and amorphous
carbon exhibit wider D and G bands, and taller D bands in
comparison to graphite (Ren et. al, 2017 Ng et. al, 2022).
Although amorphous carbon is acknowledged to be more
“disordered” than crystalline graphite within these studies,
the size of the D peak in Raman spectra does not accu-
rately indicate the origin or quantity of defects within the
crystal structure. Both the Raman ratio and D/G area ratio
fail to represent carbon sorption tendencies sufficiently,
highlighting the limitations of using Raman analysis alone
for organic carbon characterization (Ren et. al, 2017 Ng
et. al, 2022). While Raman spectroscopy is valuable for
analyzing the chemical properties and identifying poten-
tial origins of CM, it cannot fully characterize organic
carbon, nor predict gold robbing accurately. Therefore, a
comprehensive understanding of characterization requires
the integration of complementary methodologies beyond
Raman spectroscopy.
METALLURGICAL RESPONSE OF A
POLYMETALLIC COMPLEX ORE
The deposits considered for this study contain lead and zinc
sulfides as well as gold and silver hosted in various mineral
phases. The samples selected have variable amounts of CM
in the range of 0.01 to 0.53%, determined by leaching car-
bonate from the sample with hydrochloric acid and ana-
lyzing the residue with a carbon/sulfur analyzer. Figure 1
shows the CM content in the selected samples and the
corresponding gold robbing number, typically utilized to
quantify the gold robbing characteristics of refractory ores
(Miller, Wan and Diaz 2016). This number can be used
for an initial grouping of samples into four classifications
of gold robbing: non, mildly, moderately and highly gold
robbing.
Further metallurgical characterization was based on
bench top leach tests on a pyrite concentrate obtained fol-
lowing sequential lead, zinc, and pyrite flotation stages, as
shown in Figure 2. Prior to the flotation stages, the samples
were prepared by crushing and grinding in a laboratory rod
Table 1. Gold robbing classification
Property Non Gold Robbing Mildly Gold Robbing Moderately Gold Robbing Highly Gold Robbing
Gold-rob #0 0–1.0 1.0–2.5 2.5
Figure 1. Gold robbing number vs. carbonaceous matter content in the selected samples