Journal of Scientific & Industrial Research

Vol. 65, January 2006, pp. 65-67

Comparison of different extraction methods to determine gold in
geological samples
P V Sunder Raju*
National Geophysical Research Institute, Hyderabad 500 007
Received 03 May 2005; revised 25 August 2005; accepted 14 October 2005
Different petro-chemically varied lithological samples were collected from the Chitradurga schist belt, Karnataka, to
identify auriferous zones. The efficiencies of gold estimation at very low levels by different extraction methods were
compared. The extraction methods were fire-assay (FA), cyanide leach, and methyl isobutyl ketone (MIBK) solvent
extraction. The rock samples include banded iron formations (silicate, oxides, carbonate and sulphide facies), ferruginous
cherts, massive milky white quartz veins and highly sheared fractured translucent dark grey to bluish grey quartz veins.
After separation and pre-concentration of gold, MIBK solvent extraction procedures have been studied using flame atomic
absorption spectrophotometer (F-AAS).
Keywords: Chitradurga, F-AAS, MIBK
IPC Code: C22B11/00

Introduction
In nature, gold occurs as native or refractory type.
Refractory gold ores are not free milling and cannot
be leached without some form of physical, chemical
or pyro metallurgical pretreatment. Refractory gold is
found in association with sulphides (pyrite,
arsenopyrite and chalcopyrite), whereas native gold is
associated with tellurides, bismuthides, silver,
antimony and copper. In general, fire-assay (FA),
cyanide leach, open aquaregia and microwave
(closed) are some of the important extraction
methods. The present study is an attempt to find any
difference in the Au content of the splits of large
sample after crushing, to understand the efficiency of
the analytical methods based on aquaregia leach,
NaCN leach and methyl isobutyl ketone (MIBK)
extraction, and compare these methods with the
classical and widely accepted lead fire assay method.

Experimental Methods
Sampling and Powdering
Samples were collected from geologically
favorable hosts for gold mineralization from
Chitradurga and Gadag schist belts. Bulk samples
(10-15 kg) were collected and powdered
(200-250 mesh). Samples were coned, quartered and
systematically sampled to give a weight of 20 g. Prior
to chemical dissolution, the samples
1
were mixed with
ammonium nitrate and subjected to open air roasting
at 600-650

C in a porcelain crucible.

Instrumentation
SpectrAA 220 (Varian, Australia) flame atomic
absorption spectrophotometer (F-AAS) with
deuterium background correction was utilized. A
model Z 300 bench top centrifuge (Hermle Labor
Technik, Germany) was employed for centrifuging
sample solutions in MIBK extraction. Gold speck
obtained by cyanidation process was weighed using a
model UMT-2 microbalance (Metler, Switzerland).
The instrumental parameters are as follows: lamp
type, hollow cathode lamp; flame type, air acetylene;
background correction, deuterium; wavelength,
242.8 nm; slit width, 1.0 nm; lamp current, 4 mA;
measurement time, 1.5 sec; airflow, 3.5 l/min; and
acetylene flow, 1.5 l/min.

Anaytical Methods
Lead Fire Assay
Finely powdered sample (80-90%<75 m), mixed
with litharge and glass fluxes, was analyzed by FA
method, which can measure gold concentration in the
range ppb-ppm. The sample, in a crucible, was
charged at 850C in a muffle furnace and temperature
raised and maintained well above 1000C for
30-40 min until complete fusion occurred. The lead
______________
*Tel: 040-23434700; Fax: 040-23434651
E-mail: [email protected]
J SCI IND RES VOL 65 JANUARY 2006


66
circulates through the molten mass and collects the
precious metal forming Au-Ag-Pb alloy, which is
recovered as a button after cooling and separation
from the glassy slag containing base metal and other
impurities. Bone ash and marborite cupel heated to
1000C are charged with lead button. This process
results in a precious metal bead, which is parted by
dissolving in nitric acid. Alternatively, lead is
dissolved in aqua regia and analyzed for gold and
silver by F-AAS. Lead - FA analysis was carried out
on 50 g samples at the Chemistry Laboratory of
Chitradurga gold unit of Karnataka. Gold specks were
weighed using a model UMTZ microbalance,
Switzerland
2
.

Sodium Cyanide Leach
The sample (200g) was weighed into a bottle and
200ml of distilled water was added. The pH was
adjusted to 11 by the addition of CaCO
3
(3g). The
solution was made 2% with respect to NaCN .The
open bottle is laid on its side on a bottle roller and
allowed to roll on a bottle roller for 24 h. The bottle
with contents was weighed before and after the
cyanide leach to estimate the amount of evaporation.
The solution was filtered and heated up to 30C. The
lead acetate (2g) was added and the solution was
stirred. After 15-20 min, the contents were heated up
to 60-65C. Simultaneously, zinc dust (2g) was added
and contents were stirred. Then concentrated HCl
(10-12 ml) was added drop wise while keeping the
beaker on a hot plate for 5 min. Sponge is formed,
allowed to cool to room temperature and then the
sponge containing solution is filtered. The sponge was
washed 2 to 3 times with distilled water. Excess water
was squeezed out and spongy mass was rolled in to
lead foil after adding metallic silver granules
(0.1 - 0.5g). The rolled lead foils were cupelled. The
contents are wrapped in a lead foil and heated in
bone-ash cupell. This process of heating is designated
as cupelled or cupellation at 1000
o
C for about 1 h.
The complex bead containing gold and silver was
dissolved in 1:4 HNO
3
for~10 min and then 1:1 HNO
3

to separate gold from silver. The gold speck was
weighed on a microbalance.

MIBK Procedure
Each sample (10g) was transferred into a porcelain
crucible with ammonium nitrate (1g), mixed
thoroughly, and roasted for 1 h in a muffle furnace at
650C. Roasted samples were transferred in glass
beakers (250 ml each) and freshly prepared aqua regia
(30 ml) was added to stabilize AuCl
3
complex, during
evaporation on a hot plate. Each beaker was covered
with a watch glass and heated on a hot plate. Heating
was continued for at least 4 h and enough aqua regia
was added at regular intervals to maintain the free
acid level at about 1 cm above the sample surface
3,4
.
The watch glasses were removed and the contents
evaporated slowly until the residue became nearly
dry. 3 M HCl (40 ml) was added to each beaker and
warmed until clear solution is obtained. These
solutions were cooled and filtered using Whatman 40
filter paper. The filtrate of each sample was
transferred to a 250 ml-separating funnel. The beaker
was washed with a minimum amount of 3 M HCl and
the washings were transferred to a separating funnel
and shaken for 5 min. After the phases were clearly
separated, aqueous phase was drained off. A washing
solution (10 ml of conc. HCl+10 ml conc. HBr to
500 ml with DD water) was added and the separating
funnel was shaken for 2 min. After the phases were
completely separated, aqueous phase was discarded
and the organic phase was collected for F-AAS gold
estimation
5
.

Results and Discussion
Samples were analyzed for Au (Table 1); Lead Fire
Assay method was taken as standard and each
determination was carried out 5 times using standard
reference materials (Table 2).

Cyanidation Processes
Cyanidation process is quite time consuming and
hazardous in handling. Quite often, the silicate and
iron rich samples are not dissolved completely and
give quite erratic results. Large sample volumes are
required for this technique
6
to counteract the
heterogeneity problems and this technique is
ineffective for the leaching of encapsulated gold.

MIBK Method
MIBK method is suitable for treating multi-varied
samples especially banded iron formations (oxides,
Table 1 Comparison of gold extraction methods
Rock type MIBK Cyanidation Fire Assay
g/g g/g g/g
BIF 0.42 0.40 0.32
BIF 0.52 0.40 0.60
BIF 0.78 0.40 0.60
BIF 0.49 0.20 0.40
Metavolcanics 0.44 0.60 0.10
Metavolcanics 0.40 0.42 0.32
Metavolcanics 0.52 0.40 0.60
SUNDER RAJU: EXTRACTION METHODS TO DETERMINE GOLD IN GEOLOGICAL SAMPLES


67
sulphides, silicates and carbonates). Iron interfering
with gold wavelength during analysis was overcome
by back extraction of iron from the organic extract.
Centrifugation of colloidal materials in the organic
phase also affects aspiration of solutions in F-AAS.

Fire Assay
Traditional Fire Assay method was carried at
Chitradurga gold unit to confirm the true values
obtained in-other methods. This procedure is time
consuming but perfect gravimetrical results of gold
are obtained.

Conclusions
Different methods of analytical results vary on the
type of sample containing gold content especially on
refractory ores. Sodium cyanide leach results were
lower in comparison to Fire Assay and aqua regia
leach because of non-effective leaching of sulphide
and silica bearing samples containing gold. The
advantage of MIBK method is primarily associated
with its cheaper cost, than the Fire Assay fluxes. As
exploration samples require a better detection limit
than that offered by gravimetric procedures, Fire
Assay provides total fusion of gold as it dissolves and
extracts the entire platinum metal group in addition to
gold and silver. The combination of a Fire Assay
fusion with AAS technique offers the advantage of
using large quantity sub sample together with a very
sensitive technique to yield better detection limits.
Occasionally, interferences from elements like Fe etc.
especially in case of samples of banded iron
formation of different facies (sulphides, oxides,
silicate and carbonate) and quartz vein containing
refractory gold will reduce instrument sensitivity.
This study shows that both aqua regia and Sodium
Cyanide Leach methods and MIBK solvent extraction
techniques, act as potential alternatives for the Fire
Assay method. However, explorationists looking for
the best resolution of low-level gold anomalies, the
alternative chemical procedures play a vital role for
the investors to decide about the economics of the
deposit, as proved in the case of gold exploration
studies in Gadag and Chitradurga Schist Belts in
Karnataka, India.

References
1 Balaram V, Hussain S M, Uday Raj, Charan S N, Subba Rao
D V, Anjaiaha K V & Illangovan, A rapid method for the
estimation of gold, platinum, palladium and silver in rocks
and ores by ICP-MS for geochemical exploration, Atomic
Spectro, 18 (1997) 17-24.
2 Fletcher K & Horsky S, Determination of gold by
cyanidation and graphite furnace atomic absorption
spectroscopy, Geochem Explor, 30 (1988) 29.
3 Rubeska I, Thomas V A, Baby T V & Thomas O, Interim
Report CH-5, UN assisted project, Trivandrum, 1980, India
1-12.
4 Ramesh S L, Sunder Raju P V, Anjaiah K V, Rao T G,
Ramavathi Mathur, Dasaram B, Nirmal Charan S, Subba Rao
D V & Balaram V, Determination of gold in rocks, ores and
other geological materials by flame and graphite furnance
atomic absorption spectrometry techniques, Atomic Spectro,
22 (2001) 263.
5 Balaram V, SunderRaju P V, Ramesh S L, Anjaiah K V,
Dasaram B, Manikyamba C, Ram Mohan M & Sarma D S,
Rapid partial dissolution method in combination with AAS
techniques for use in geochemical exploration, Atomic
Spectro, 20 (1999) 155-160.
6 Balaram V, Rao T G, Anjaiah K V, Ramesh S L,
Manikyamba C & Naqvi S M, Analytical methods for the
rapid analysis of gold and precious metals to make a first
order appraisal of a given area for exploration, in Proc
Volume of Nat Semin Exploration and Exploitation of Gold
Resources in India (NGRI, Hyderabad) 1996, 231-239.


Table 2 Standard reference materials used
Sample MIBK-
FAAS
Certified value Type
g/g g/g
GAu 16
@
1.08 0.09 1.09 0.03 Ore from
sandstone gold
GAu 17
@
3.010.20 3.140.06 Ore from Silt
stone
GAu-18
@
9.960.57 10.000.20 Ore from
faulted zone
OX-11
#
2.890.04 2.940.03 Gold bearing
quartz vein
OX-12
#
6.490.37 6.600.08 Gold bearing
quartz+pyrites
S-2
#
1.520.03 1.530.03 Gold bearing
quartz+pyrites
WG-2
#
1.39 0.06 1.380.03 Dissemenated
gold
KH-1
#
0.84 0.01 0.850.02 Gold with
pyrites
@
Institute of Geo Physical and Geology Exploration, Lang Fai,
102849, China
#
Rock labs, New Zealand