Hydrometallurgy 198 (2020) 105527

Contents lists available at ScienceDirect

Hydrometallurgy
journal homepage: www.elsevier.com/locate/hydromet

Gold leaching from oxide ores in alkaline glycine solutions in the presence
of permanganate
E.A. Oraby a, b, J.J. Eksteen a, *, G.M. O’Connor a
a
Western Australian School of Mines, Minerals, Energy and Chemical Engineering Curtin University, GPO Box U1987, Perth, Western Australia 6845, Australia
b
Mining and Metallurgical Engineering, Faculty of Engineering, Assiut University, Egypt

A R T I C L E I N F O A B S T R A C T

Keywords: In this study, the effect of using strong oxidants with alkaline glycine solutions on the leach rate and recovery of
Glycine gold was studied in the absence of cyanide. Prior research indicated that when using only alkaline glycine and air
Gold recovery at ambient temperature in the absence of catalysts, leaching times would be too long for conventional agitated
Leaching
tank leaching of ores. The ores used in this study were from oxidised geological domains, which would not be
Permanganate
Alkaline
expected to excessively consume oxidant. Ores with reactive sulfides or other reducing minerals would be ex­
pected to rapidly consume an oxidant, making the process ineffective. The influence of the oxidants such as
potassium permanganate, hydrogen peroxide, sodium chlorate, potassium iodide, and potassium ferricyanide
were initially evaluated in a screening test. The highest gold extractions were observed with potassium per­
manganate, where a recovery of 85.1% was achieved at 1.5 kg/t glycine and 3 kg/t permanganate at 30% solids,
pH 10.5 and ambient temperature using a bottle roll process. This compared to a 87.4% recovery against a
conventional cyanidation benchmark with 30% solids, pH 10.5 and ambient temperature using the bottle roll.
Other oxidants only achieved a recovery of 11% at best, so permanganate was the focus of the optimisation work.
The solution pH had very little effect on gold recovery between 9.5 and 11.5. The ratio of glycine to perman­
ganate concentration is significant, if potassium permanganate is in a molar excess, the recovery is greatly
reduced due to oxidation of the glycine. The main products of this reaction, when lime was used as pH modifier,
were determined to be calcium carbonate, calcium oxalate, trace amounts of ammonia and manganese dioxide.
Gold is recoverable from leach solutions by conventional adsorption onto activated carbon.

1. Introduction mineralogy or conditions. These lixiviants have the potential for use in
districts where cyanidation is restricted; some have faster kinetics; some
Cyanide has been used for gold extraction for over 100 years and is are more suitable for treating refractory ore and some have higher
the default lixiviant in the gold industry. It is cheap and effective for gold selectivity than cyanide (Marsden and House, 2006). However, there are
dissolution, producing a highly stable gold-cyanide complex which is also many challenges such as high cost, high reagent consumption, and
easily recovered from solution by adsorption onto activated carbon. often low leach rate, or complex downstream recovery.
However, the high toxicity of cyanide, increasing regulatory restrictions, The use of glycine and other amino acids such as glutamic acid as
and poor selectivity with low-grade ores are proving to be challenging lixiviants for base and precious metals was introduced and patented by
problems for the gold industry. Many alternative non-cyanide lixiviants researchers at the Curtin University (Oraby and Eksteen, 2014a; Oraby
have been studied and tested in the laboratory over the past few de­ and Eksteen, 2014b; Oraby and Eksteen, 2015; Eksteen and Oraby,
cades, with only the calcium thiosulfate system being commercialised at 2015; Tanda et al., 2017). The most effective application of glycine for
an industrial scale. Thiosulfate, glycine, thiourea, halide and thiocya­ tank leaching is with a cyanide-starved process, commercialised under
nate have shown some promise for some selected ores (Oraby and Eks­ the trade name GlyCat™, where cyanide plays a catalysing role (Oraby
teen, 2015; Feng and Van Deventer, 2010; Marsden and House, 2006; and Eksteen, 2015; Oraby et al., 2017). Gold and/or copper can be
Aylmore, 2005) and it is clear that a fit-for-purpose approach is required effectively recovered by carbon adsorption, resin adsorption and sulfide
as there is no panacea lixiviant that is equally suitable for every precipitation (Deng et al., 2019, 2020a, 2020b; Tauetsile et al., 2018a,

* Corresponding author.
E-mail address: jacques.eksteen@curtin.edu.au (J.J. Eksteen).

https://doi.org/10.1016/j.hydromet.2020.105527
Received 21 June 2020; Received in revised form 24 October 2020; Accepted 14 November 2020
Available online 19 November 2020
0304-386X/© 2020 Elsevier B.V. All rights reserved.
E.A. Oraby et al. Hydrometallurgy 198 (2020) 105527

2018b). The GlyCat™ process has attracted much attention from in­ /
dustry and a pilot campaign was performed recently based on this 2MnO−4 + 2(NH2 CH 2 COO)− + 1 2 O2 →2MnO2 ↓ + 2C2 O−4 2 + 2NH3 (aq) + H2 O
technology (Seaman et al., 2019) to recover gold from a low-grade (4)
gold‑copper concentrate high in sulfide sulfur. Ammonia oxidation:
On its own, and under similar conditions, the leach rate of gold with
glycine is significantly slower than with cyanide, but still shows promise 2MnO−4 + 2NH3 (aq)→2MnO2 ↓ + N2 ↑ + 2OH − + 2H2 O (5)
for heap, dump and in situ operations. For a cyanide-free glycine oper­ Oxalate oxidation:
ation to be viable at ambient temperature in an agitated tank leach pro­
cess, the leach kinetics would need to be significantly accelerated. Some C2 O−4 2 →2CO2 ↑ + 2e− (6)
improvement has been reported with a combination of high temperature
Oxygen reduction:
and high pH, as recently shown for a Western Australian paleochannel
ore, but the rate is still lower than with cyanide (Oraby and Eksteen, O2 + 2H2 O + 4e− →4OH − (7)
2015; Oraby et al., 2017). This work used dilute glycine and conven­
The mechanism and reaction between glycine and permanganate in
tional bottle roll aeration (i.e. no oxygen sparging).
the presence of air, and the effect of this on the gold leach rate are
The reaction of gold with glycine is shown in Eq. 1.
discussed in this paper. The effects of glycine concentration, perman­
[ ]
4Au + 8NH2 CH 2 COOH + 4NaOH + O2 →4Na Au(NH 2 CH 2 COO)2 + 6H2 O ganate concentration, solution pH, and oxidant type and pH modifier
(1) were summarised.

Recently the use of alkaline glycine only, but at elevated tempera­ 2. Materials and methods
tures (60 ◦ C), high oxygen addition rate (1 L/min for a fluid volume of 1
L), 1.25 M glycine (93.8 g/L), pH of 12, and high stirring rate (900 rpm), 2.1. Samples
gold recoveries in excess of 90% were obtained. The leaching was con­
ducted for 24 h at 10 wt% solids pulp of gold ore which contains 5.4 g/t Quantitative X-ray diffraction (QXRD) and X-ray fluorescence (XRF)
(Altinkaya et al., 2020), thereby validating the earlier work done by analyses of the four samples tested are shown in Tables 1 and 2. Quartz is
Eksteen and Oraby (2015), but on ore rather than gold sheet. the dominant mineral in all samples except ore A, which also had major
Another way to improve the reaction rate could be with appropriate dolomite. The elemental analysis confirmed that silicon is the main
oxidants or catalysts. The use of an alkali-stable transition metal- element and that there was very low sulphide sulfur in all tested ores.
complex or a mixture of different complexes as oxidants were found to The fire assay analysis showed that the feed samples have a relatively
be synergistic catalysts with the alkaline amino acid system (Eksteen and high gold content ranging from 1.32 to 3.74 g/t.
Oraby, 2017). Adding one or more of the alkali-stable transition metal
complexes presents advantages which include (1) accelerated leach
rates of gold; (2) introduces a cyanide-free leaching system; and (3) 2.2. Sample preparation
reduce the leach temperatures while achieving high leach rates at room
temperature. Four gold oxide ore samples were ground to 100% passing 75 μm.
Potassium permanganate (“Permanganate”) is a strong oxidant that The mineralogical composition of the samples was determined by
may be applicable to oxidised gold ores. For ores with reactive sul­ Quantitative X-ray Diffraction (Q-XRD).To calculate the final gold
phides, it is expected that permanganate consumption would be exces­ extraction, the feed and final leach residues were analysed for gold by
sive due to reaction with abundant minerals such as reactive sulfides. fire assay and for base metals and other elements by X-ray Fluorescence
Permanganate is manufactured industrially from the reaction between (XRF).
manganese dioxide or manganese ores (pyrolusite) and potassium hy­
droxide in heated air or oxygen to produce manganate according to Eq. 2 2.3. Leaching
(Garcia, and Manuel D.O.G. (Oviedo, ES), 1991; Popoff, 1919). Through
an electrolytic oxidation, the manganate is then converted to potassium All experiments were carried out using solutions prepared from
permanganate in an alkaline medium as shown in Eq. 3. analytical grade reagents and deionised water. Unless specified, all ex­
periments were conducted using a bottle roller at room temperature.
2MnO2 + 4KOH + O2 →2K2 MnO4 + 2H2 O (2)
The ore sample, glycine and permanganate were placed in a 2.0 L plastic
2K2 MnO4 + 2H2 O→2KMnO4 + 2KOH + H2 (3) bottle with a 5 mm hole in the lid to maintain aeration. The pH was
adjusted initially with NaOH to the target pH and the bottle was rolled
Potassium permanganate is well known as one of the most effective on at 100 rpm. Further pH adjustments over the duration of the exper­
and commonly used oxidizing agents in different industries (Stewart and iments were made with lime (Ca(OH)2). Solution samples were obtained
Wiberg, 1965; Pérez-Benito et al., 2012). The oxidisation of organic at different times by filtering 10 mL of slurry with a 0.45 μm filter paper
species including amino acids by permanganate is also well known and and the solids were returned to the leach bottle. The filtrate was
many studies have addressed these reactions (Fawzy and Musleh, 2014;
Jose et al., 2005; Halligudi et al., 2000; Insausti et al., 1992). The Table 1
oxidation and loss of glycine as a complexing agent for gold will Mineralogical analysis of the tested gold ore samples.
therefore be discussed in this study.
Mineral name Ore A Ore B Ore C Ore D
For effective leaching, glycine will be added at more than the stoi­
chiometric ratio with permanganate dictated by Eq. 4. This is to ensure Mineral, %

sufficient free glycine remains in the system to complex gold. The final Quartz 35 94 81 100
by-products from glycinate ion oxidisation would be mainly oxalate and Mica group 14 <1
Talc
ammonia. In the presence of sufficient permanganate, ammonia is oxi­
<1
Kaolinite-serpentine group 5
dised to nitrogen as shown in Eq. 5. Oxalate is mainly oxidised to form Dolomite 41
carbon dioxide as shown in Eq. 6. The conventional oxygen reduction Plagioclase 10
reaction is given in Eq. 7. Alunite 15
Jarosite 4
Glycine oxidation:
Rutile Group <1

2
E.A. Oraby et al. Hydrometallurgy 198 (2020) 105527

Table 2
Elemental analysis of the tested gold ore samples.
Sample Au Si Al Fe Ca Mg S Mn As Cu Ni Zn

UNITS ppm % % % % % % % % % % %

Ore A 3.74 23.23 4.21 2.91 12.50 1.21 0.05 0.08 0.06 0.01 0.02 0.03
Ore B 2.45 43.1 1.18 1.08 0.12 0.28 0.246 0.01 0.002 0.003 0.006 0.003
Ore C 1.32 36.9 3.16 1.53 0.04 0.03 2.78 0.008 0.015 0.001 0.03 0.001
Ore D 2.50 44.8 0.12 1.90 0.02 0.02 0.034 0.014 0.01 0.008 0.02 0.002

analysed for gold by using atomic absorption spectrometry (AAS, Agi­

lent 55B AAS model). The final leach filtrate was analysed for a suite of Table 4
Gold recovery by glycine-permanganate mix of the tested gold ore samples at
different elements by determined by Inductively Coupled Plasma (ICP)
30% solids, pH 10.5 and room temperature (~22.1 ◦ C).
Optical Emission Spectrometry (OES) at Bureau Veritas Minerals Pty
Ltd. in Western Australia. All the solid residues were analysed in Cyanidation Time Glycine, g/L Glycine consumed, kg/t Au recovery, %

duplicate for gold analysis by fire assay. Ore A 48 2.0 1.41 85.4
Ore B 48 7.5 0.75 48.5
Ore C 48 7.5 1.00 77.3
2.4. Reagent analysis Ore D 48 7.5 1.10 83.2

Permanganate reduction by glycine has been analysed by Agilent

Carry 60 UV–Vis spectrometer. The by-products from the reaction be­ glycine can be referred to the smaller molecular weight of cyanide
tween glycine and permanganate have been analysed by Thermo Sci­ compared with glycine. The molar volume of HCN is 39.3 cm3/mol
entific™ DIONEX™ (ICS-3000) ion chromatography (IC). whereas that of glycine is 64.7 cm3/mol at room temperature, and their
diffusion coefficients in water are 1.72 × 10− 5 cm2/s (Lötter, 2006) and
1.06 × 10− 5 cm2/s (Khanal et al., 2019) respectively. The smaller mo­
3. Results and discussion
lecular size and larger diffusion coefficient of HCN in water (and its
cyanide anion in equilibrium with it) compared to glycine in water (and
3.1. Baseline cyanidation and ore comparison
its glycinate anion in equilibrium with it), leads to larger liquid state
mass transfer coefficients (via calculation of the Schmidt and Sherwood
Preliminary experiments were conducted as a control test to assess
numbers, as per the correlation for agitated tanks provided by Cussler
the cyanide soluble gold in the ore samples by conventional cyanidation
(2009)). Similarly, the diffusion of the larger gold glycinate away from
processes. Table 3 shows the gold recovery/ extraction by cyanidation
the gold surface is expected to be slower than the corresponding gold
using 300 ppm NaCN, 30% solids, and pH 10.5. The results showed
cyanide complex. Steric effects on the reacting surface due to the larger
cyanide consumption was low for these types of free milling gold ores.
and more complex glycine may further impede the rate of glycine
The glycine-permanganate leaching system was applied to each ore.
attachment to the surface compared to cyanide. The precipitation of
The total permanganate added to each was 3 kgreagent/tonneore. The gold
MnO2 and Ca(C2O2) onto the reacting surface, as per Eq. 9, may lead to
recovery is similar for ores A and D as that for baseline cyanidation as
further mass transfer resistance through possible product layers.
can be seen in Table 4. For ores B and C, which had higher amounts of
In this process, permanganate was used as an oxidant to replace the
sulfide sulfur than A and D, requires further optimisation work to
use of peroxide with glycine (Eksteen and Oraby, 2015). Despite per­
improve the recovery to a comparable level as that obtained with cya­
manganate not being as strong oxidant as peroxide, it has a higher
nidation. The slightly higher gold extraction by the cyanidation than the
reduction potential than oxygen (O2) as shown in Eq. 8. The overall
proposed process can be amended by conducting more fundamental and
reaction of gold leaching in glycine and permanganate solutions can be
optimisation studies to increase the gold recovery by the new process. In
expressed by Eq. 9. The presence of a strong oxidant such as perman­
essence, all cyanide recoverable gold should also be glycine recoverable.
ganate and MnO2 overcomes the limitation of oxygen diffusion and
solubility. The better results in the presence of permanganate can also be
3.2. Leaching parameter evaluation attributed to the increase of hydroxyl radical, which acts as a strong
oxidant for gold leaching.
Due to the sample availability, Ore A was selected to demonstrate the
effect of different factors on the gold recovery. Fig. 1 shows a compar­ MnO−4 + 2H2 O + 3e(aq)→MnO2 ↓ + 4OH − E0 = + 0.56 V (SHE) (8)
ison between the baseline cyanidation and glycine-permanganate
leaching of ore A that was reported in Tables 1 and 2. The cyanidation Au + 2MnO−4 + 4(NH2 CH 2 COO)− + O2 →Au(NH 2 CH 2 COO)2
test was conducted at 30% solids, 1000 ppm NaCN, pH 10.5 and room + 2MnO2 ↓ + 2C2 O−4 2 + 2NH3 (aq) + 2OH − (9)
temperature. The glycine leach test was conducted at 30% solids, 2 g/L
As per reaction stated in Eq. 5, ammonia itself can be oxidised to
glycine, 2 g/L permanganate, pH 10.5 and room temperature. The cy­
nitrogen by permanganate.
anide extractable gold was determined to be 87.1% under the stated
conditions. While the glycine-permanganate system was slower at
3.2.1. Effect of glycine concentration
leaching gold, the final recovery was close to that for cyanidation at
Fig. 2 shows the effect of varying glycine concentration on gold re­
85.1%. The faster kinetics leach rate of cyanide only leach compared to
covery in the presence of 1.0 g/L potassium permanganate at pH 10.5.
The gold dissolution rate after 4 h of leaching was found to be propor­
Table 3 tional to the glycine concentration. From an economic perspective
Gold recovery by cyanidation of the tested gold ore samples.
glycine may need to be recycled if high concentrations are used. These
Cyanidation Time NaCN, kg/t Au recovery, % results should be contrasted against the strong concentrations of 93.75
Ore A 48 0.62 87.1 g/L (and elevated temperature of 60 ◦ C) used by Altinkaya et al. (2020).
Ore B 48 1.05 90.8 Permanganate used with glycine thus allows a significant reduction in
Ore C 48 1.21 88.4 process intensity.
Ore D 48 0.75 84.5
Table 5 summarises the salient differences between this work and

3
E.A. Oraby et al. Hydrometallurgy 198 (2020) 105527

Fig. 1. Gold extraction by cyanidation and glycine-permanganate from gold oxide ore at leach conditions of 2 g/L glycine, 30% solid content, 100% -75 μm, pH 10.5,
2.0 g/L permanganate at room temperature. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of
this article.)

Fig. 2. Effect of glycine on gold extraction from gold oxide ore by Glyleach™ process at leach conditions of glycine, 30% solid content, 100% − 75 μm, pH 10.5, 1.0
g/L permanganate at room temperature. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

demonstrates the options available for glycine leaching, in the presence

Table 5 and absence of a catalytic oxidant. It is clear that each approach has its
Comparison of process conditions required for cyanide free gold leaching.
merits and that an evaluation of the techno-economics of the whole
Leach parameter Altinkaya et al. This work process (including downstream processing and reagent recovery and
(2020)
recycling) is required to decide on the approach most fit for a given
Temperature (◦ C) 60 Ambient (23) need.
Oxygenation 1 L/min None, other than air ingress
through lid
Optimal Glycine 93.75 2.00
3.2.2. Effect of permanganate concentration
Concentration [g/] Fig. 3 shows the effect of permanganate concentration. Interestingly
Optimal pH 12.0 10.5 the leach rate is initially faster at 2 g/L permanganate however the
Agitation, rpm Overhead agitation, Bottle roll 100 rpm overall gold recovery is higher with only 1 g/L permanganate. Gold
900 rpm
extraction reaches 76.4% after 48 h in the presence of 1 g/L perman­
%Solids 10% 30%
Gold Ore Grade (g/t) 5.4 3.74 (Ore A) ganate compared to 67.9% in 2 g/L. The decrease in the overall gold
Grind d80 = 86 μm d100 = 75 μm with higher permanganate can be attributed to the oxidisation of glycine
Oxidant/ Catalyst Addition None KMnO4 (1 g/L) by excess permanganate (Insausti et al., 1992).
Recovery 90% 85% The loss of glycine by oxidation can be made up by increasing the
Leach duration (h) 24 48
glycine concentration and thus the total amount of available glycine.
This result was shown in the comparison with cyanide in Fig. 1 with
that of Altinkaya et al., 2020. In both cases the ores are essentially oxide leaching at 2 g/L glycine where the overall gold recovery was 85.1%.
gold ores. The ratio of permanganate to glycine appears to be very important, a
The comparison of the two cyanide free glycine approaches further test with 0.5 g/L glycine and 2 g /L permanganate returned a

4
E.A. Oraby et al. Hydrometallurgy 198 (2020) 105527

Fig. 3. Effect of permanganate on gold extraction from gold oxide ore by glycine at leach conditions of 1.0 g/L glycine, 30% solid content, 100% -75 μm, pH 10.5,
permanganate and room temperature. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

gold recovery of only 0.5%. In this case virtually all the glycine was amorphous to the XRD.
likely to have been oxidised. While the precipitation of MnO2 is not problematic during agitated
tank leaching, it may prove to be problematic during percolation
3.2.3. Permanganate consumption leaching of larger particles, where the precipitate is expected to cause
The permanganate concentration in leach solutions can be moni­ pore blockage and restrict the flow of solution.
tored by UV–vis spectroscopy at a wavelength of 525 nm. The UV–Vis
scan of the permanganate solution is shown in Fig. 4. A straight-line 3.2.4. Effect of pH
standard calibration curve for measuring permanganate was generated Fig. 8 shows the gold recovery at different levels of pH. It can be seen
as shown in Fig. 5. leaching at the lower pH of 9.5 commences at a slower rate, however
From the calibration curve shown in Fig. 5, permanganate con­ after 48 h recovery was slightly higher than the tests at pH 10.5 and
sumption due to the reaction with glycine can be easily monitored. Fig. 6 11.5. This result suggests that a lower usage of lime and/or caustic may
shows the fast consumption of permanganate in the presence of 1.0 g/L be possible and also allows the use of poorer quality water, which can
glycine in the solution. The permanganate concentration was monitored have a buffering effect at alkaline pH due to magnesium in saline waters
by UV–Vis spectrometer using the calibration curve shown in Fig. 5. Full (typical of Western Australian gold fields).
consumption of permanganate was achieved after only 4 h of leaching. The caustic and lime consumptions during these tests are shown in
In a solution of glycine and permanganate at pH 10.5 using NaOH as Table 7. Leaching at a lower pH of 9.5 is more efficient based on gold
a pH modifier, the precipitate from reaction 4 was analysed by Scanning recovery and reagent consumption. Leaching at a natural pH of 6.5 leads
Electron Microscopy/Energy Dispersive X-Ray Spectroscopy (SEM/EDS) to a very poor gold dissolution. Only 10% gold was extracted after 48 h
as shown in Fig. 7 and Table 6. The ratio of manganese to oxygen in the at a pH of 6.5. The lower gold recovery at pH 6.5 is due to the presence of
EDS indicates that the manganese species is MnO2. Although no crys­ glycine in its zwitterion form which is not effective to leach gold as the
talline MnO2 species were detected by X-ray diffraction, it is expected anion form of glycinate at pH 9.5.
that the precipitated particles show too poor crystallinity and appears

Fig. 4. Spectrum of potassium permanganate solution in UV–Visible region at pH 10.5 in the absence of glycine.

5
E.A. Oraby et al. Hydrometallurgy 198 (2020) 105527

Fig. 5. Calibration line of potassium permanganate using in UV–Visible region at a wavelength of 525 nm.

Fig. 6. Permanganate measurement in the presence and absence of 1.0 g/L glycine at pH 10.5 and room temperature.

3.2.5. Effect of pH modifier and the propensity of free oxalate to further oxidise, there is no detec­
Using only lime as a pH modifier was tested and it was found the gold tible free (i.e. dissolved) oxalate in the system.
recovery was 81.4% compared with 85.1% using a dual (caustic-lime)
pH modifier. The lime consumption during the lime-only test was 12.6 3.2.6. Effect of oxidant type
kg/t. The lower gold recovery with high lime use may be attributed to The effect of different oxidants on the gold dissolution was studied
the precipitation of calcium oxalate (Eq. 10) on the gold surface. Eq. 4 using hydrogen peroxide, sodium chlorate, potassium ferricyanide and
showed that oxalate is one of the by-products of glycine oxidation which potassium iodide. The results are shown in Fig. 10. It can be easily seen
was also proven by ion chromatography (IC) analysis of the leach so­ that permanganate is the most effective. Adding ferricyanide showed a
lutions as shown in Table 8. Oxalate was formed as a by-product from slight gold extraction improvement but the rate may only be suitable for
the reaction between glycine and permanganate. Adding lime as a heap or in-situ leaching. Hydrogen peroxide was covered in depth in
source of calcium leads to the precipitation of calcium oxalate. prior research (Eksteen and Oraby, 2015), and was shown to require
In the absence of calcium ions to stabilise the oxalate, it can oxidise temperatures above 40 ◦ C for any significant leaching. It is therefore
further to carbon dioxide (Eq. 6) which may also lead to calcium car­ clear that potassium (or sodium) permanganate is the preferred oxidant
bonate (calcite) precipitation on the gold surface. for agitated tank leaching at room temperature. The reason for the better
performance of permanganate over the other oxidants is that it generates
C2 O4 − 2 + Ca+2 →CaC2 O4 (10)
MnO2 in a colloidal form which acts as an additional oxidant during
Fig. 9 shows the XRD pattern of the highly insoluble calcium oxalate leaching (Pratesa et al., 2019) as well as being a stronger oxidant (based
(solubility of 6.7 × 10− 4 g/L at 20 ◦ C) which was formed after treating upon half reaction E0 ranking) in alkaline media than the other oxidants.
the glycine-permanganate by-product mix solutions by lime. It can be
seen calcium oxalate was indeed formed and some residual lime was left 3.2.7. Impurity dissolution
with the precipitated solids. Whereas oxalic acid is a mild poison using The concentrations of different impurities in the final leach solutions
the NFPA 704 classification, calcium oxalate is not (i.e. calcium oxalate after leaching for 48 h are shown in Table 9. The analysis of manganese
is not poisonous even under chronic exposure). Given the excess of lime in the final leach solution indicates and confirms that manganese has

6
E.A. Oraby et al. Hydrometallurgy 198 (2020) 105527

Fig. 7. SEM image of MnO2 precipitate after the reaction between glycine and permanganate.

Table 7
Table 6 Gold recovery and caustic and lime consumption by glycine-permanganate mix
Elemental analysis (atomic %) of selected points in Fig. 7. of the tested gold ore samples using 1.0 g/L glycine, 1 g/L permanganate and
Spectrum O (%) Na (%) Al Si (%) K (%) Mn (%)
30% solid content.
pH NaOH, kg/t Lime, kg/t Gold recovery, %
Spectrum 1 61.4 4.9 0.2 0.0 2.9 30.5
Spectrum 2 59.1 5.1 0.2 0.0 3.0 32.4 9.5 1.07 0.82 79.6
Spectrum 3 67.5 3.9 0.2 0.1 2.3 26.0 10.5 2.10 2.15 76.4
11.5 4.81 3.83 69.8

been precipitated from the solutions as MnO2. Some manganese disso­

lution was observed at low pH, where the extent of reaction between Table 8
glycine and permanganate was low. Dissolved manganese was also Ion chromatography analysis of the by-products of 2 g/L glycine oxidisation by
observed by increasing the permanganate concentration from 1.0 g/L to 1.0 g/L permanganate at pH 10.5.
2.0 g/L in the presence of 1.0 g/L glycine. The presence of soluble Sample ID Concentration [mg/L]
manganese at 2.0 g/L permanganate is due to its excess over glycine.
Oxalate Formate Acetate Lactate

Before lime addition 281 <10 <10 <10

After lime addition 63 <10 <10 <10

Fig. 8. Effect of pH on gold extraction from gold oxide ore by glycine at leach conditions of 1.0 g/L glycine, 30% solid content, 100% -75 μm, pH, 1.0 g/L per­
manganate and room temperature. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

7
E.A. Oraby et al. Hydrometallurgy 198 (2020) 105527

Fig. 9. XRD patterns of the precipitated solids after treating glycine-permanganate solution mix with Ca(OH)2.

Fig. 10. Effect of oxidant on gold extraction from gold oxide ore by glycine at leach conditions of 1.0 g/L glycine, 30% solid content, 100% -75 μm, pH 10.5, 1.0 g/L
oxidant and room temperature. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Table 9
Impurity dissolution during cyanide and glycine leaching of gold oxide ore at different leach conditions.
Leach conditions Cu Fe Si S Ni Zn Ca Mn

Glycine, g/L pH permanganate, g/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L

0.5 10.5 1.0 BDL BDL 32.0 18.0 1.2 BDL 7.5 BDL
1.0 10.5 1.0 0.5 BDL 26.0 18.0 1.8 BDL 10.5 BDL
2.0 10.5 1.0 1.5 BDL 30.0 16.0 2.6 BDL 6.5 BDL
1.0 6.5 1.0 5.0 BDL 0.5 28.0 2.5 BDL 1.6 158
1.0 9.5 1.0 1.0 BDL 12.0 18.0 3.0 BDL 4.5 BDL
1.0 11.5 1.0 BDL BDL 114 16.0 1.6 BDL 11.6 BDL
1.0 10.5 0.5 0.5 BDL 15.0 19.0 BDL BDL 0.6 BDL
1.0 10.5 2.0 4.5 0.5 24.0 86.0 1.6 BDL 7.5 35.1
2.0 10.5 2.0 5.0 0.5 20.0 26.0 1.6 BDL 10.5 BDL
1.0 g/L cyanide 10.5 – 9.5 2.0 24.0 40.0 19.8 1.6 91.0 BDL

8
E.A. Oraby et al. Hydrometallurgy 198 (2020) 105527

3.2.8. Gold recovery by carbon-in-pulp (CIP) and carbon-in-leach (CIL) Table 10

modes Gold loaded on carbon from CIP and CIP modes for gold recovery using 2 g/L
Two different leach tests were conducted to establish if gold can be fresh carbon.
recovered by carbon in the presence of glycine and its oxidisation by- Sample Ag Au Ca Cu Mn
products. Gold glycinate adsorption on carbon has been proved and UNITS ppm ppm ppm ppm ppm
studied in detail for different glycine systems and reported elsewhere
CIP Carbon 38.9 596 1300 300
(Oraby and Eksteen, 2015; Tauetsile et al., 2018a, 2018b; Eksteen et al., <0.5
CIL Carbon 66.9 610 2300 160 15.2
2018; Altinkaya et al., 2020). Leaching was conducted at 1.0 g/L glycine
and 1.0 g/L potassium permanganate at pH 10.5. Fresh activated car­
bon, PICAGOLD® G210AS (Jacobi Carbons) produced from coconut only (Adams, 2016)). In areas where cyanide is not allowed, this process
shell was used. The carbon was mechanically sized to − 2.36 + 2.00 mm. provides a feasible alternative as the process is quite robust and use the
Carbon was added at 2.0 g/L at the start of leaching to simulate CIL; or same underlying CIP/CIL technology used in conventional cyanidation,
after 24 h of leaching to simulate CIP. without the concomitant safety, health, and environmental risks.
Recovery for the CIL test was 85.8% and for CIP 85.1% based on fire *Approximate and indicative bulk values available to the authors at
assay analysis of the gold residue after leaching. This result indicates the time of writing (can vary significantly by location and purchase
that a conventional downstream recovery process using activated car­ contract). The authors do not take any responsibility for the accuracy of
bon can be used. Both CIL and CIP systems are applicable. Fig. 11 shows these prices which are only given for illustrative purposes.
the gold in solution during the CIL and CIP leach tests. Once the carbon
was added to the leach pulp after 24 h, essentially all the gold was 4. Conclusions
adsorbed and the concentration in the final solutions was only 0.008
mg/L. In the CIL mode, with the leaching and the adsorption happening Previous studies have shown that cyanide-free glycine leaching of
in parallel, gold in solution gradually increases up to 4 h after which the gold is limited by slow kinetics and so is only likely suitable for heap, vat
leaching slows down and the adsorption continues until the gold in so­ and in-situ leaching. The addition of a very low dose of cyanide can
lution reaches 0.012 mg/L after 48 h. Table 10 shows the gold and other increase the rate enough for tank leaching, but this does not completely
elements loaded onto the carbon during the CIL and CIP leach tests. The solve the environmental issues associated with cyanide. Alternatively,
presence of Mn on carbon in the CIL mode can be explained because the one needs to raise the temperature, use direct oxygenation with high
carbon was added at the start of leaching, i.e. before the Mn precipitated glycine concentration and high pH to improve the gold leach rate.
to MnO2. In this study, it was found that adding a strong oxidant such as
permanganate to a cyanide-free glycine leach significantly enhances the
3.2.9. Glycine-permanganate versus cyanidation rate and extent of gold extraction to approximate the same level of gold
The most obvious benefit of the proposed glycine-permanganate leaching obtained with cyanide after the same leach duration of 48 h.
system is the elimination of cyanide in the process while operating at The addition of permanganate should be controlled based on the stoi­
similar overall operating conditions as cyanidation, which can be used chiometric reaction with glycine to avoid the oxidation of the total
where cyanide is not allowed. The proposed process reagent cost is available glycine in solution. Different leach conditions and parameters
anticipated to be higher than cyanide (e.g. initial cyanide concentration were evaluated to investigate their impacts on the gold leach rate and
was 0.3 g/L versus glycine at 1.0 g/L). However, controlling the addition overall gold recovery from the glycine-permanganate system. It was
of permanganate (around USD2/kg*) and recycling some of the glycine shown that an excess of glycine relative to permanganate was required
(around USD1.60 per kg*) will definitely improve the process economics in all cases to ensure that sufficient residual glycine is present to com­
compared to conventional cyanidation (with cyanide costs around USD plex the gold.
2.5/kg*). Not included though, is the cost of cyanide detoxification, The highest gold recovery from the glycine permanganate system
which is not required for the glycine process. Detoxification costs range was 85.1%, compared to 87.4% for a conventional cyanidation. This was
from USD 1.5/kgWAD CN to USD 4.5/kgWAD CN (based on reagents achieved in solutions containing 1.0 g/L glycine and 1.0 g/L

Fig. 11. Gold in solutions during CIL and CIP modes for gold recovery from glycine-permanganate solutions at leach conditions of 1.0 g/L glycine, 1 g/L per­
manganate, 30% solid content, pH 10.5 in the presence of 2.0 g/L carbon. (For interpretation of the references to colour in this figure legend, the reader is referred to
the web version of this article.)

9
E.A. Oraby et al. Hydrometallurgy 198 (2020) 105527

permanganate. Although a portion of the glycine is oxidised by per­ Eksteen, J.J., Oraby, E.A., 2015. The leaching and adsorption of gold using low
concentration amino acids and hydrogen peroxide: effect of catalytic ions, sulphide
manganate, the overall reagent consumption is low enough to be
minerals and amino acid type. Miner. Eng. 70, 36–42.
considered as a practical and economic alternative to cyanide leaching, Eksteen, J.J., Oraby, E.A., 2017. Recovery of Precious and Chalcophile Metals
while operating under comparable overall conditions (temperature, (WO2019033154).
grind, %solids, pH, carbon adsorption approach and residence time). Eksteen, J.J., Oraby, E.A., Tanda, B.C., Tauetsile, P.J., Bezuidenhout, G.A., Newton, T.,
Trask, F., Bryan, I., 2018. Towards industrial implementation of Glycine-based leach
The leaching can be conducted at low pH (pH 9.5) which will make the and adsorption technologies for Gold-Copper Ores. Can. Metall. Q. 57 (4), 390–398.
process suitable for any sites that have low-quality (such as saline) Fawzy, A., Ashour, S.S., Musleh, M.A., 2014. Base-catalyzed oxidation of L-asparagine by
water. After glycine leaching, gold can be recovered by carbon in leach alkaline permanganate and the effect of alkali-metal ion catalysts: kinetics and
mechanistic approach. React. Kinet. Mech. Catal. 111, 443–460.
(CIL) or carbon in pulp (CIP) modes. Further studies are currently being Feng, D., Van Deventer, J.S.J., 2010. Effect of thiosulphate salts on ammoniacal
conducted to have a better understanding of the gold leaching mecha­ thiosulphate leaching of gold. Hydrometallurgy 105, 120–126.
nism and the effects of different parameters on the reagent Garcia, Manuel D.O.G. (Oviedo, ES), 1991. Process for Producing Potassium Manganate
United States Industrial Quimica del Nalon, S.A. (Oviedo, ES), p. 5011672.
consumptions. Halligudi, N., Desai, S., Mavalangi, S., Nandibewoor, S., 2000. Kinetics of the oxidative
degradation of rac-serine by aqueous alkaline permanganate. Monatsh. Chem. 131,
321–332.
Declaration of Competing Interest Insausti, M.J., Mata-perez, F., Alvarez-Macho, P., 1992. Permanganate oxidation of
glycine: influence of amino acid on colloidal manganese dioxide. Int. J. Chem. Kinet.
24, 411–419.
The authors hereby declare: Jose, T.P., Nandibewoor, S.T., Tuwar, S.M., 2005. Mechanism of oxidation of L-histidine
by heptavalent manganese in alkaline medium. E J. Chem. 2, 75–85.
1. Messrs. Oraby and Eksteen are the listed inventors of a number of Khanal, S.P., Kandel, Y.P., Adhikari, N.P., 2019. Transport properties of zwitterion
glycine, diglycine, and triglycine in water. AIP Adv. 9, 065303.
patents that utilises glycine as a lixiviant for numerous precious and
Lötter, N., 2006. Cyanide volatilization from gold leaching operations and tailing storage
base metals from various materials such as ores, wastes, concen­ facilities. In: Thesis to Obtain a Master’s Degree in Metallurgical Engineering,
trates, etc., and which is background Intellectual Property to the Faculty of Engineering, Built Environment and Information Technology. University
research presented in this paper. of Pretoria, Republic of South Africa.
Marsden, J.O., House, C.I., 2006. The chemistry of gold extraction. In: Society of Mining,
2. Should processes be commercialised that utilises the patents Metallurgy and Exploration (SME), 2nd ed., p. 651 Littleton (CO), USA.
mentioned in Point no. 3 above, Messrs. Oraby and Eksteen may be Oraby, E.A., Eksteen, J.J., 2014a. A Process for Precious Metals Recovery (US Patent
eligible to receive royalty income. US20160194734).
Oraby, E.A., Eksteen, J.J., 2014b. The selective leaching of copper from a gold–copper
3. Dr. Greg O’Connor has no conflict of interest to declare. concentrate in glycine solutions. Hydrometallurgy 150, 14–19.
Oraby, E.A., Eksteen, J.J., 2015. The leaching of gold, silver and their alloys in alkaline
glycine–peroxide solutions and their adsorption on carbon. Hydrometallurgy 152,
193–203.
Acknowledgements Oraby, E.A., Eksteen, J.J., Tanda, B.C., 2017. Gold and copper leaching from gold-copper
ores and concentrates using a synergistic lixiviant mixture of glycine and cyanide.
The financial support from Curtin University is gratefully Hydrometallurgy 169, 339–345.
Pérez-Benito, J.F., Rodriguez, R.M., De Andrés, G., Brillas, E., Garrido, J.A., 2012.
acknowledged. Kinetics and mechanisms of the permanganate oxidation of L-valine in neutral
aqueous solutions. Int. J. Chem. Kinet. 21, 71–81.
Popoff, S., 1919. The course of the reaction between manganese dioxide, potassium
References
hydroxide and oxygen and the manufacture of potassium manganate. [Easton, Pa].
J. Ind. Eng. Chem. 11 (4), 1919.
Adams, M.D., 2016. Comparison of the MMS CN-DTM cyanide destruction process with Pratesa, Y., Wibisono, A., Narottama Putra, W., Sadeli, T., Zakiyuddin, A., Sri
Caro’s acid, SMBS/air and hydrogen peroxide options. In: ALTA 2016 Conference, Harjanto, S., 2019. Effect of H2O2 and MnO2 as oxidators of gold and copper leach
p. 7. processes from printed circuit boards. J. Sains Mat. Indonesia 20 (3), 132–136. April
Altinkaya, P., Wang, Z., Koroleva, I., Hamuyuni, J., Haapalainen, M., Kolehmainen, E., 2019.
Yliniemic, K., Lundström, M., 2020. Leaching and recovery of gold from ore in Seaman, B., Newton, T., Oraby, E.A., 2019. Development of a glycine-cyanide leach
cyanide-free glycine media. Miner. Eng. 158, 106610. process for gold-copper concentrate. In: Paper Presented at the ALTA 2019, Perth,
Aylmore, M.G., 2005. Alternative lixiviants to cyanide for leaching gold ores. In: Australia.
Adams, M.D. (Ed.), Advances in Gold Ore Processing. Elsevier, Oxford, pp. 501–540. Stewart, R., Wiberg, K.B. (Eds.), 1965. Oxidation in Organic Chemistry, Part A, 1965.
Cussler, E.L., 2009. Diffusion: Mass Transfer in Fluid Systems. Cambridge University Academic Press, New York. Ch.1.
Press, Cambridge, UK, p. 613. Tanda, B.C., Oraby, E.A., Eksteen, J.J., 2017. Recovery of copper from alkaline glycine
Deng, Z., Oraby, E.A., Eksteen, J.J., 2019. The sulfide precipitation behaviour of cu and leach solution using solvent extraction. Sep. Purif. Technol. 187, 389–396.
au from their aqueous alkaline glycinate and cyanide complexes. Sep. Purif. Technol. Tauetsile, P.J., Oraby, E.A., Eksteen, J.J., 2018a. Adsorption behaviour of copper and
218, 181–190. gold glycinates in alkaline media onto activated carbon. Part 1: isotherms.
Deng, Z., Oraby, E.A., Eksteen, J.J., 2020a. Cu adsorption behaviours onto chelating Hydrometallurgy 178, 202–208.
resins from glycine-cyanide solutions: isotherms, kinetics and regeneration studies. Tauetsile, P.J., Oraby, E.A., Eksteen, J.J., 2018b. Adsorption behaviour of copper and
Sep. Purif. Technol. 236, 116280. gold Glycinates in alkaline media onto activated carbon. Part 2: kinetics.
Deng, Z., Oraby, E.A., Eksteen, J.J., 2020b. Sulfide precipitation of copper from alkaline Hydrometallurgy 178, 195–201.
glycine-cyanide solutions: precipitate characterisation. Miner. Eng. 145, 106102.

10