Journal Pre-proof

Precipitation of zinc from cyanide leach solutions using

Trimercapto-s-triazine (TMT)

E.Y. Yazici, F. Ahlatci, E. Yilmaz, O. Celep, H. Deveci

PII: S0304-386X(19)30450-5
DOI: https://doi.org/10.1016/j.hydromet.2019.105206
Reference: HYDROM 105206

To appear in: Hydrometallurgy

Received date: 20 May 2019

Revised date: 21 October 2019
Accepted date: 9 November 2019

Please cite this article as: E.Y. Yazici, F. Ahlatci, E. Yilmaz, et al., Precipitation of
zinc from cyanide leach solutions using Trimercapto-s-triazine (TMT),
Hydrometallurgy(2019), https://doi.org/10.1016/j.hydromet.2019.105206

This is a PDF file of an article that has undergone enhancements after acceptance, such
as the addition of a cover page and metadata, and formatting for readability, but it is
not yet the definitive version of record. This version will undergo additional copyediting,
typesetting and review before it is published in its final form, but we are providing this
version to give early visibility of the article. Please note that, during the production
process, errors may be discovered which could affect the content, and all legal disclaimers
that apply to the journal pertain.

© 2019 Published by Elsevier.

 Journal Pre-proof

Precipitation of Zinc from Cyanide Leach

Solutions using Trimercapto-s-triazine (TMT)

E. Y. Yazici *, F. Ahlatci, E. Yilmaz, O. Celep, H. Deveci

Hydromet B&PM Research Group, Div. of Mineral&Coal Processing, Dept. of Mining Eng.,

f
oo
Karadeniz Technical University (KTU), 61080, Trabzon, Turkey

pr
Abstract
e-
Cyanide leach solutions produced from gold/silver ores can be contaminated with zinc due to
Pr

the gold/silver cementation in Merrill-Crowe process and/or the dissolution of zinc minerals

present in the ore. Zinc levels in cyanide solutions should be controlled to avoid potential
al

problems in leaching and downstream processes associated with the accumulation of zinc. In
rn

this study, the application of a novel method was demonstrated for the precipitation of zinc
u

from synthetic/real cyanide solutions using Trimercapto-s-triazine (TMT), a non-toxic organic

Jo

reagent. Experimental studies have shown that precipitation of zinc with TMT is a fast

process occurring in a short period of time i.e., ≤15 min. Increasing the concentration of TMT

(33-90 mM) and pH (10-12.5) was found to improve the precipitation of zinc, which was

unaffected by increasing the temperature (20-60°C). Response surface methodology was

adopted for detailed investigation of the effect of [TMT]/[Zn] ratio (2.70-16.25), [CN-]/[Zn]

ratio (3.00-8.00) and initial concentration of Zn (50-950 mg/L) on the precipitation of zinc.

High recoveries (≤95.4%) were obtained under suitable conditions. Precipitation tests were

also performed using synthetic multi-metal and real leach solutions. The complete

recovery/dissolution of zinc from the precipitates was also demonstrated using different acids
 Journal Pre-proof

(HCl and H2SO4). The findings have shown that TMT can be effectively used as an

environmentally friendly reagent for removal of zinc from cyanide leach solutions.

Keywords: Cyanide Leaching, Zinc, Gold/Silver, Precipitation, Trimercapto-s-triazine

(TMT), Experimental design.

f
oo
*
Corresponding author (eyazici@ktu.edu.tr)

Postal address: Karadeniz Teknik Univ. Maden Muh. Bol. 61080, Trabzon, Turkey

Phone: +90 462 377 3683

pr
e-
Fax: +90 462 325 7405
Pr
al
u rn
Jo
 Journal Pre-proof

1. Introduction

Dissolution of gold from ores in cyanide solutions was first applied in 1887 (Habashi, 1987).

Since then, major portion of gold from its primary resources is produced using cyanide

leaching (Celep, 2015; Marsden and House, 2006). Due to the depletion of free milling gold

ores, the mining industry has turned its attention to refractory ores, which are not particularly

amenable to direct cyanide leaching due to several reasons (e.g., physical encapsulation of

gold in sulphide/silicate minerals, abundance of cyanide/oxygen consuming minerals and/or

organic/inorganic carbon). One of the main challenges in gold extraction is inherently non-

f
oo
selective nature of cyanide leaching (Marsden and House, 2006). Dissolution of impurity

elements (e.g., copper and zinc) from their minerals results in excess reagent consumption and
pr
low gold recoveries in leaching as well as technical problems in downstream solution
e-
purification and metal recovery stages (Bas et al., 2015; Henley, 1975; La Brooy et al., 1994;
Pr

Yazici et al., 2015; Yılmaz, 2016). Zinc minerals are partially soluble in cyanide solutions

(Table 1) (Adams, 2016; Sceresini, 2005; Shantz and Reich, 1978). Dissolution of some zinc
al

minerals and complexation of zinc ions in cyanide solutions are expressed by the Eqs. 1-3
rn

(Habashi, 2009; Marsden and House, 2006). Zinc-cyanide complexes have low stability and
u

hence classified as weak acid dissociable cyanide (WAD; logK<30) (Miltzarek et al., 2002).
Jo

One mole of zinc ion bounds 3-4 moles of cyanide ions (CN -) (Eqs. 4-5).

Table 1.

ZnS + 4CN- → Zn(CN)42- + S2- (1)

2ZnS + 10CN- + O2(g) + 2H2O → 2Zn(CN)42- + 2CNS- + 4OH- (∆G= –136 kcal) (2)

ZnO + 4CN- +H2O → Zn(CN)42- + 2OH- (3)

2Zn + 8CN- + O2(g) + 2H2O → 2Zn(CN)42- + 4OH- (4)

 Journal Pre-proof

Zn2++3CN-↔Zn(CN)3- (logβ3=16.05) (5)

Zn2++4CN-↔Zn(CN)42- (logβ4=19.62) (6)

Leach solutions can be also contaminated by zinc due to the application of Merrill-Crowe

process (cementation with zinc powder) for recovery of gold/silver from pregnant leach

solutions (PLSs) (Kurama and Çatalsarik, 2000; Walton, 2016). Merrill-Crowe processes

particularly suitable for gold ores containing high silver (>10 g/t Ag) (Marsden and House,

2006; Lapidus and Millard, 2016) since high presence of silver in cyanide leach solutions

f
oo
adversely affects the rate of gold adsorption onto activated carbons (Lin et al., 2010). In

Merrill-Crowe process, metallic zinc is added into the clarified and de-aerated pregnant leach
pr
solution. During the process, gold/silver is cemented/precipitated onto zinc powder/shavings
e-
with the concurrent release of zinc in the form of cyanide complexes (Zn(CN)3 -, Zn(CN)42-)
Pr

while (Eq. 7) (Marsden and House, 2006; Walton, 2016). Sufficient level of free cyanide

should be maintained in Merrill-Crowe process to prevent passivation of the surface of zinc

al

metal due to precipitation of zinc as zinc hydroxide (Eq. 8) (Walton, 2016).

u rn

Au(CN)2 - + Zn0 + H2O + 2CN- → Au0 +OH- + 0.5H2 + Zn(CN)42- (7)

Jo

Zn(CN)42- + 2OH- ↔ Zn(OH)2 + 4CN- (8)

In this process, theoretical consumption of zinc powder is 1 mole Zn per 1 mole gold/silver

(Eq. 8). However, in practice, 5 to 30-fold higher amount of zinc than the stoichiometric ratio

is required for efficient recovery of precious metals (Au+Ag) with a typical zinc consumption

of 1-10 g/g. Barren solutions after zinc cementation typically contain <0.03 mg/L Au and high

levels of zinc (up to 474 mg/L) (Fleming, 2016; Kurama and Çatalarik, 2000; Marsden and

House, 2006; Ritcey, 2005).

 Journal Pre-proof

Zinc can built-up in the leaching/adsorption (CIL/CIP) circuit through recycling of effluent

into the plant resulting in various technical problems in leaching and downstream processes.

During CIL/CIP process, the presence of zinc cyanide complexes may interfere with the

adsorption of gold onto activated carbon. Sheya and Palmer (1989) investigated the effect of

metals including Zn, Co, Ag, Ca, Cd, Co, Hg, Ni, and Sb on gold adsorption from cyanide

solutions. The authors found that despite the improving role of these metals tested, only zinc

was noted to have an adverse effect (by up to ~10%) on gold adsorption through competitive

adsorption of zinc and gold cyanide complexes. Zinc can also create some problems in the

f
oo
flotation of gold. In some gold plants, cyanide effluents containing WAD cyanide complexes

such as zinc- and copper-cyanides etc. are recycled to flotation circuits, which could adversely
pr
affect the flotation of gold. Guo et al. (2016) demonstrated using modern surface
e-
characterisation techniques that free cyanide does not depress liberated gold; however, zinc
Pr

cyanide ([CN-]/[Zn] (molar)=2) acts as a depressant for free gold particles in flotation through

hindering the adsorption of flotation collector (potassium amyl xanthate, PAX) or Aerophine
al

promoter (3418A). During the application of the Merrill-Crowe process, high zinc
rn

concentrations can severely lower gold precipitation rates due to passivation of zinc surface
u

through formation of insoluble zinc hydroxide (Marsden and House, 2006). In addition to
Jo

operational problems during the process, zinc also causes analytical problems which could

affect the process. In cyanide leaching plants, free cyanide level in leach solutions is

commonly controlled by silver nitrate titration. The presence of zinc in solutions leads to an

overestimation of free cyanide due to the low stability of zinc-cyanide complexes compared

with silver-cyanide (Breuer et al., 2011; Koc et al., 2014; Van der Merwe and Breuer, 2011).

Koc et al. (2014), demonstrated that analysed concentrations of free cyanide in the

absence/presence of zinc ([CN-]/[Zn] (molar): 0-10) were close to each other. Overestimation

of free cyanide would result in the insufficient addition of cyanide into the leach solution,
 Journal Pre-proof

which may adversely affect the performance of cyanide leaching as well as cementation.

Therefore, zinc level in barren solutions should be controlled (Estay et al., 2018; Kurama and

Çatalsarik, 2000). Removal of zinc would also lead to release (recovery) of complexed

cyanide with zinc. Recovery of cyanide is of economic importance due to the fact that the cost

of recovered cyanide is 2-3 fold less than new cyanide (SGS, 2013; Fleming, 2016).

Furthermore, a cyanide recovery plant could pay back its capital cost in less than 2 years

(Mudder and Botz, 2001; Miltzarek et al., 2002; Fleming, 2003). There are several processes

used at industrial scale i.e. ion-exchange (Paterson, 2017), AVR (Acidification, Volatilisation

f
oo
and Reneutralisation) and SART (Sulphidisation, Acidification, Recycling and Thickening)

(Deveci et al., 2006; Estay et al., 2012; Fleming, 2010; Goldstone and Mudder, 1991; Gonen
pr
et al., 2004; Yazici et al., 2009; Yılmaz et al., 2019a,b). SART process was particularly
e-
developed for copper-bearing gold ores. The process is based on the precipitation of copper
Pr

using sulphide salts (Na2S or NaHS) in acidic conditions (pH 4-5) and concurrent recovery of

cyanide that complexed with copper (Goldstone and Mudder, 1991). Other metals such as
al

zinc also precipitates in sulphide form with copper (Estay, 2018).

u rn

Zinc can be precipitated using sulphide salts in the form of ZnS from alkaline cyanide
Jo

solutions. This is commonly used for controlling zinc levels in cyanide leaching plants that

use Merrill-Crowe process. Based on sulphide precipitation, Velardena process was

developed for removal of zinc and regeneration of cyanide from high zinc containing alkaline

cyanide solutions (Eq. 9). The treated solution can be directly recycled to the leach circuit

without solid/liquid separation due to the limited dissolution of ZnS in cyanide solutions

(Table 1) (Marsden and House, 2006). Precipitation of zinc leads to recovery of up to 4 mol

of free cyanide complexed with 1 mol zinc (Eq. 9). In a recent study, Koc et al. (2016),

reported that zinc/silver can be selectively precipitated over copper (≤%11) from PLSs (29
 Journal Pre-proof

mg/L Au, 41 mg/L Ag, 859 mg/L Cu, 20 mg/L Fe, 770 mg/L Zn; pH 11.5) derived from

cyanide leaching (3 g/L NaCN, pH 10.5-11, 25°C, 24 h.) of a silver-rich pyritic gold

concentrate (62 g/t Au, 110 g/t Ag, 0.30% Cu and 3.14% Zn) by the addition of Na2S. The

researchers reported a gold loss of <21% implying that the addition of sulphide salts should

be controlled to limit the loss of gold. Alternative methods for precipitation of zinc and

concurrent recovery of cyanide are needed due to inherent disadvantages (toxicity, corrosive

properties, potential passivation effect on gold/silver, formation of thiocyanate, requirement

for careful control of dosage etc.) of sulphide precipitation.

f
oo
Zn(CN)42- + S2- → ZnS + 4CN- (9)
pr
e-
Utilisation of Trimercapto-s-triazine (C3N3S33-; TMT), an organosulphide reagent, for
Pr

precipitation of silver from cyanide solutions was reported in a recent study (Yazici et al.,

2017). TMT has been previously used for the removal of toxic heavy metals including copper,
al

cadmium, mercury and lead, from industrial effluents through formation of compounds with
rn

high stability (Dongmei et al., 2006; Matlock et al., 2002). TMT is a non-toxic reagent with
u

very low acute toxicity (LD50: 7878 mg TMT (15%) per kg). It also shows no toxic effect for
Jo

fish (LC0 13720 mg TMT (15%) per L) and water fleas (Daphnia magna, EC50: 4224 mg

TMT (15%)/ per L) (Evonik Industries, 2018). To the authors’ knowledge, there has been no

study on the precipitation of zinc from cyanide solutions using Trimercapto-s-triazine.

In this study, the precipitation of zinc from synthetic/real cyanide solutions using

Trimercapto-s-triazine (TMT) was investigated. The effect of reaction period (0-120 min.),

concentration of TMT (33-90 mM), pH (10-12.5) and temperature (20-60°C) on precipitation

of zinc was tested using classical one-factor-at-a-time approach. A more detailed research was
 Journal Pre-proof

further performed using statistical design of experiments. The influence of [TMT]/[Zn] ratio

(molar) (2.70-16.25), [CN-]/[Zn] (molar) ratio (3.00-8.00) and initial concentration of Zn (50-

950 mg/L) was investigated in detail using response surface methodology. The performance

of the proposed method was tested on synthetic solutions containing multi-metal and pregnant

leach solutions. The Zn-TMT precipitates were characterised using a laser diffraction particle

size analyser, stereo microscope, electron microscope (SEM/EDS) and FT-IR spectrometer.

The recovery of zinc from Zn-TMT precipitates was also studied.

f
oo
2. Experimental

2.1. Materials and Methods

pr
A trimercapto-s-triazine (TMT) trisodium salt (C3N3S3Na3) solution (pH 12.3) with an active
e-
C3N3S33- content of 15% w/w, kindly supplied by SNF Floerger (CAS No: 17766-26-6), was
Pr

used for precipitation of zinc from synthetic/real cyanide solutions. The density, viscosity and

freezing point of the TMT solution was 1.12 g/cm3 (20°C), 2cPs (20°C) and -3°C,
al

respectively (SNF Floerger, 2011). Fig. 1 shows the structure of TMT (C3N3S33-) and its form
rn

after bounding metals.

u
Jo

Fig. 1.

Powder/granular form of sodium cyanide (≥99.0% NaCN) and zinc sulphate (≥99.5%

ZnSO4.7H2O) was used in preparation of synthetic zinc-cyanide solutions. A stock solution of

gold (1000 mg/L Au) in 2 M HCl, silver nitrate (≥99.5% AgNO3) and copper(I) cyanide

(≥99.1% CuCN) were used for preparation of synthetic solutions containing Zn, Cu, Ag

and/or Au. Deionised-distilled water was used in preparation of all solutions.

 Journal Pre-proof

Zinc-cyanide solutions (10 mL) were added in 50-mL polypropylene (PP) Falcon tubes.

Solution pHs were adjusted to the required level using 20% NaOH or 10% H 2SO4 prior to

addition of the required amount of TMT. Agitation of the horizontally placed tubes was

performed using an orbital shaker (DAIHAN Wisecube WIS-20) with a digital temperature

control operating at a stirring speed of 180 rpm and 25°C. A non-ionic flocculant (SNF-FA

920 PWG), kindly provided by SNF Floerger, was added into the final solution after the

mixing period to allow easy collection of the precipitates. Non-ionic flocculant was used

based on the findings presented in a previous study (Yilmaz et al., 2016). The flocculant (0.3

f
oo
mL) was added from the stock solution (0.1 g/L) prepared by deionised-distilled water, prior

to centrifugation (Nüve NF 400) at 4000 rpm for 10 min. After the filtration using cellulose
pr
nitrate membrane filters (0.45 µm, Merck), clear solutions were diluted (if required) with
e-
sodium cyanide solution (1.5 g/L NaCN) for analysis of metals by Atomic Absorption
Pr

Spectrometer (Perkin Elmer AAnalyst 400). Initial and residual concentrations of zinc were

used in the calculation of precipitation of zinc (%).

al
rn

Precipitation of zinc by TMT also frees (recovers) the cyanide bound with zinc. Analysis of
u

free cyanide could not be performed due to that the interference of residual TMT in the
Jo

solution with the analysis. Therefore, recovery of cyanide in precipitation tests was calculated

using zinc-cyanide speciation data produced by MEDUSA software (2009).

Precipitation tests were also carried out at different [CN]/[MetalT] ratios (molar) of 3-6 using

synthetic solutions containing Zn-Ag (500 mg/L Zn and 20 mg/L Ag) and Zn-Cu-Ag-Au (500

mg/L Zn, 500 mg/L Cu, 20 mg/L Ag, 15 mg/L Au) to mimic zinc-bearing leach solutions of

silver ores. Real pregnant leach solutions (PLSs) was also used in the precipitation tests. PLSs

were derived from leaching of a silver ore (145-159 g/t Ag, 0.72% Zn and 0,014% Cu; d80=35
 Journal Pre-proof

µm) and a pyritic gold/silver concentrate (62 g/t Au, 110 g/t Ag, 0.30% Cu and 3.14% Zn;

d80: 40 µm) assaying under the conditions of 3 g/L NaCN, 35% w/v solids ratio over 24 h. at

pH 10.5-11 (by 5 M NaOH). The cyanide leaching of the ore or concentrate was performed in

a 10-L reactor equipped with a mechanical stirrer and PTFE-coated four-blade propeller for

mixing. Air was sparged into the pulp (DO2=11.6±0.02 mg/L) during leaching. Following the

solid/liquid separation, the PLS was analysed for metals by AAS. The concentration of metals

in the PLS generated was 39.4 mg/L Ag, 967 mg/L Zn and 7.80 mg/L Cu for the silver ore,

and 37.9 mg/L Ag, 29.4 mg/L Au, 733 mg/L Cu and 427 mg/L Zn for the pyritic gold

f
oo
concentrate.

pr
2.2 Preparation and characterisation of precipitates
e-
Based on the fact that TMT precipitates in acidic conditions (<pH 4) in the form of
Pr

(H3C3N3S3)0, a 10-mL TMT solution (%15 w/w) was added into 5-mL concentrated H2SO4

solution (96%). A rapid occurrence of light-yellow coloured particles was observed after the
al

contact of the TMT and acid solutions. Preparation of zinc-TMT precipitate was performed by
rn

the addition of 0.5 mL TMT solution (%15 w/w) into a zinc-cyanide solution (1.5 g/L NaCN,
u

500 mg/L Zn, pH 11.5). After addition of TMT, the solution was then placed into the orbital
Jo

shaker for mixing over a period of 60 min. to allow precipitation of Zn-TMT. After that, the

solution containing precipitate was centrifuged for 5 min. at 4100 rpm. The aliquot was

discarded and the precipitate at the bottom was washed twice through centrifugation using

deionised-distilled water to remove any residual zinc-cyanide solution on the Zn-TMT

precipitate. The particle size analysis of precipitates (H3TMT and Zn-TMT) collected was

carried out after centrifugation. The precipitates were dried for 6 h. in an oven (60°C) and

then cooled in a desiccator.

 Journal Pre-proof

The produced dry Zn-TMT precipitates were examined under a stereomicroscope (Leica

MZ16) and a Zeiss EVO LS10 scanning electron microscope (KTU, Department of

Metallurgical and Materials Eng.) coupled with an Energy Dispersive Spectrometry unit

(EDS). A Fourier Transform Infrared (FT-IR) Spectrometer (Perkin Elmer Frontier) (KTU,

Department of Chemistry) was also used to obtain FT-IR spectra of a precipitate sample in the

wavelength range of 550-4000. Particle size distribution of the H3TMT and Zn-TMT

precipitates was analysed using a laser diffraction analyser (Malvern Mastersizer Hydro

2000MU).

f
oo
2.3. Recovery of zinc from Zn-TMT precipitates
pr
Recovery of zinc from Zn-TMT precipitates in acid solutions was investigated using different
e-
acids. The type (HCl or H2SO4) and concentration (1-3 M) of mineral acids on the extraction
Pr

of zinc from Zn-TMT precipitates was tested. A certain amount of Zn-TMT precipitates and a

10-mL acid solution (HCl or H2SO4) were added into 50-mL polypropylene Falcon tubes
al

prior to mixing over a period of 1 min. at 25°C. After the reaction period, the tubes were
rn

centrifuged for solid/liquid separation. The clear solutions were then analysed for zinc by
u

AAS.
Jo

2.4. Statistical Design of Experiments

Central composite design (CCD), a type of RSM, was adopted in this study to evaluate the

effect of [TMT]/[Zn] ratio (molar) (2.70-16.25), [CN-]/[Zn] ratio (molar) (3.0-8.0) and initial

concentration of zinc (50-950 mg/L Zn) on the precipitation of zinc from synthetic cyanide

solutions. In this design, there are five levels [i.e., minimum (-1.68), low (1), center (0), high

(+1) and maximum (+1.68) as coded levels]. Table 2 shows the experimental design layout

including the factors with their corresponding levels. The statistical analysis of the
 Journal Pre-proof

experimental data was evaluated using a software (Design-Expert, 2010). Probability values

(P-value) were used to determine the statistical significance of parameters. Simply, the null

hypothesis was rejected provided that the Pvalue was smaller than the selected confidence

level (e.g. 95%, α=0.05), which confirmed the statistical significance of the parameter

(Montgomery, 2001; Antony, 2003). Tests were performed in duplicates to estimate the

reproducibility. The relative standard deviation (RSD) of the experimental data was calculated

to be 0.77%. The mean values of duplicate tests were presented.

f
oo
Table 2.

pr
e-
3. Results and Discussion
Pr

3.1. Effect of reaction time, concentration of TMT, pH and temperature

al

The influence of reaction time, concentration of TMT, pH and temperature were studied using
rn

one-factor-at-a-time approach. Fig. 2 illustrates the effect of concentration of TMT (33-90

u

mM) and time (5-120 min.) on the precipitation of zinc. The results have shown that the
Jo

precipitation of zinc occurs rapidly over a period of 5-15 min. at all TMT concentrations

tested. After 5-15 min., precipitation efficiencies almost reached a plateau. Increasing the

concentration of TMT from 33 to 48 mM induced a limited improvement (by 4.20%) in zinc

precipitation. Further increasing the TMT concentration to 63 mM allowed 74.5%

precipitation of zinc over the same period of 60 min. At the highest level of TMT (i.e., 90

mM) majority of zinc (97.7%) was precipitated.

Fig. 2.
 Journal Pre-proof

Fig. 3 shows the influence of pH on the precipitation of zinc. Precipitation of zinc appeared to

be positively influenced by the increase in pH. An improvement by 17.9% in zinc

precipitation was obtained with raising the pH from 10 to 12.5 (Fig. 3).

Fig. 3.

The effect of pH on zinc precipitation is linked essentially with the speciation and binding

properties of TMT (C3N3S33-) at different pHs. At low pHs of below 5, TMT is in solid-state

f
oo
(H3C3N3S30) while by the increase in pH, it starts to dissolves in anionic form i.e., H2C3N3S3-

(pH 5-8), HC3N3S32- (pH 8-10) and C3N3S33- (>pH 12.5) (Fig. 3). Reactions for divalent metal
pr
ions such as zinc with TMT under different pH conditions can be expressed by Eqs. 9-11.
e-
Fig. 4.
Pr

 

Me 2  2 H 2 C3 N 3S3  Me(H 2 C3 N 3S3 ) 2 (pH 5-8) (9)
al

Me 2  HC3 N 3S3   MeHC3 N 3S3

2
(pH 8-10) (10)
rn

3Me 2  2C3 N 3S3   Me3 (C3 N 3S3 ) 2

3
(>pH 12.5) (11)
u
Jo

Zinc can be present as Zn(CN)3- and/or Zn(CN)4 2- depending on the pH in alkaline cyanide

solutions as illustrated in speciation diagram of zinc vs. pH (Fig. 5). At high pHs of >11.5

zinc hydroxide complexes (e.g., Zn(OH)3-, Zn(OH)42-) are starting to dominate instead of

zinc-cyanide complexes (Fig. 5). Eqs. 12-15 shows the possible precipitation reactions of

zinc-cyanide and zinc hydroxide complexes with TMT in cyanide solutions under alkaline

conditions. More negative free energy values (i.e., more favourable reactions) for

precipitation of zinc from its hydroxide complexes (Eqs. 14-15) compared with cyanide

complexes (Eqs. 12-13) could explain the increased precipitation of zinc at high pHs. By
 Journal Pre-proof

binding of TMT with zinc, cyanide is simultaneously freed (Eqs. 12-13). Theoretically, 1

mole of TMT precipitates 1.5 moles of zinc, which implies that the stoichiometric ratio for

precipitation of zinc by TMT ([TMT]/[Zn]) is 0.67 (Eqs. 12-15).

Fig. 5.

3 [Zn(CN) 3 ]  2C3 N 3S3   [Zn 3 C3 N 3S3 2 ](s) + 9CN-

- 3
 (12)
  
Zinc-cyanidecomplex precipitate

f
oo
(G0(25°C)= –247 kJ)

3 [Zn(CN) 4 ]  2C3 N 3S3   [Zn 3 C3 N 3S3 2 ](s) + 12CN-

2- 3

Zinc-cyanidecomplex
  
precipitate
pr (13)
e-
(G0(25°C)= +9.91kJ, G0(37°C)<0)
Pr

 2C3 N 3S3   [Zn 3 C3 N 3S3 2 ](s) + 9OH-

- 3
3 [Zn(OH) 3 ] (14)
   
Zinc-hydroxidecomplex precipitate
al

(G0(25°C)= –269 kJ)

rn

 2C3 N 3S3   [Zn 3 C3 N 3S3 2 ](s) + 12OH-

2- 3
3 [Zn(OH) 4 ] (15)
   
u

Zinc-hydroxidecomplex precipitate
Jo

(G0(25°C)= –228 kJ)

The effect of temperature (20-60°C) on precipitation performance of zinc was also tested (Fig.

6). The results have shown that varying the temperature from 20 to 60°C brought no

considerable influence on precipitation efficiency. The mean value of precipitation of zinc

was 64.7% with a relative standard deviation (RSD) of ±5.13% (Fig. 6). These implied that

Zn-TMT compounds are highly stable even at high temperatures.

 Journal Pre-proof

Fig. 6.

3.2. Statistical Experiment Design for Parameters

Central Composite Design (CCD) was adopted to investigate the effect of [TMT]/[Zn] ratio

(molar), [CN-]/[Zn] ratio (molar)and initial concentration of zinc on precipitation of zinc over

the period of 60 min. The experimental conditions in each run with corresponding results for

zinc precipitation (%) are shown in Table 3. Corresponding cyanide recoveries (%) were also

calculated and presented in Table 3.

f
oo
Table 3.
pr
e-
Precipitation results (Table 3) were varied between 11.1% and 98.6% Zn depending on the
Pr

experimental conditions. A second-order polynomial model (Eq. 16) was established using

the coded values and response data i.e., precipitation of zinc (%) (Table 3) with a coefficient
al

of multiple determinations (R2) of 84.4%. Statistical analysis of the data was carried out by
rn

the evaluation of the analysis of variance (ANOVA) (Table 4) of the mathematical model.
u
Jo

Precipitation of Zinc (%)

= 70.81 + 24.80*A – 6.33*B + 3.42*C – 1.91*A*B – 3.18*A*C – 4.45*B*C

– 3.87 *A2 – 0.06*B2 – 7.01*C2 (16)

Table 4.

P-value of the model for zinc precipitation (i.e., 0.0017) (Table 4) indicated that the model

was statistically significant at a confidence level of 99.5% (α=0.005). The linear effect of

[TMT]/[Zn] ratio was also significant at the same confidence level of 99.5% (Eq. 16). Other
 Journal Pre-proof

linear, interaction and quadratic effects of parameters were statistically insignificant i.e. the P-

values of higher than α=0.005 (99.5% confidence level). The linear effect of [CN-]/[Zn] ratio

was found to be statistically significant at a confidence level of 90% (i.e., P=0.0801 < α=0.1).

The regression coefficients in the model (Eq. 16) reflect the mode and magnitude of the

linear/quadratic/interaction effect of parameters on the precipitation of zinc. The coefficient of

the linear term of [TMT]/[Zn] ratio (i.e., +24.8) revealed its larger influence compared to

others as confirmed by the ANOVA (Table 4). The positive mode of the linear term of

[TMT]/[Zn] ratio indicated that increasing the TMT]/[Zn] ratio (i.e., concentration of TMT)

f
oo
improves the precipitation of zinc. The linear term of [CN-]/[Zn] ratio was 6.33 with a

negative mode (Eq. 16) which indicated that increased concentration of cyanide adversely

affects the precipitation of zinc.

pr
e-
Pr

The effect of parameters on the precipitation of zinc can be visualised by three-dimensional

(3D) response surface plots drawn by Design-Expert (2010). Fig. 8a,b clearly shows that
al

increasing [TMT]/[Zn] ratio (i.e., concentration of TMT) had substantial contribution to the
rn

precipitation of zinc independent of [CN-]/[Zn] ratio and initial concentration of zinc. For
u

instance, increasing the [TMT]/[Zn] ratio from 2.7 (Exp. 9) to 16.25 (Exp. 10) under the
Jo

conditions of [CN-]/[Zn]: 5.5 and 500 mg/L Zn, substantially increased the fraction of

precipitated silver by 8.6-fold (i.e., from 11.1% to 95.4%) (Table 3).

Increasing the [CN-]/[Zn] ratio exerted a negative influence on the precipitation of zinc at

certain conditions (Fig. 8a,c). To exemplify, precipitation of zinc was reduced from 80.8% to

48.0% by raising the [CN-]/[Zn] ratio from 3 to 8 at [TMT]/[Zn]: 9.48 and 500 mg/L Zn

(Table 3). The speciation diagram of cyanide species vs. concentration of cyanide (Fig. 7)

shows that at high [CN-]/[Zn] ratios Zn(CN)42- and free cyanide (CN-) become dominant
 Journal Pre-proof

species rather than Zn(CN)3-. The adverse effect of high [CN-]/[Zn] ratio on the precipitation

of zinc could be apparently linked with the depletion of Zn(CN)3 - under the increased levels

of free cyanide (Fig. 7). This is in agreement with the thermodynamic feasibility of

precipitation reactions (Eqs. 12-13). In compliance with the current results, some researchers

(Dongmei et al., 2006; Yazici et al., 2011) also reported that increasing the concentration of

ligands adversely affects the precipitation performance of metals by TMT. On the contrary, in

a recent study, Yazici et al. (2017) found that the concentration of cyanide had no significant

impact on precipitation of silver from cyanide solutions.

f
oo
Fig. 7.
pr
e-
The influence of the initial concentration of zinc was observed to be negligible as statistically
Pr

confirmed by the magnitude of regression coefficients in the model (Eq. 16) and ANOVA

(Table 4). Similar trends were also observed from response surface plots of cyanide
al

recoveries (not presented). These results demonstrated that TMT is capable of removing zinc
rn

from cyanide solutions under suitable conditions.

u
Jo

Fig. 8.

3.3. Precipitation Tests from Synthetic Multi-Metal Solutions and Real Pregnant Leach

Solutions

Synthetic solutions containing multi-metal and pregnant leach solutions are used to test the

efficiency of the process. Precipitation tests were performed using synthetic multi-metal

solutions containing Zn-Ag at different [CN-]/[Metal] ratios (molar) (3-6) at a constant TMT

concentration of 63 mM ([TMT]/[Metal] (molar): 8). The results (Fig. 9) obtained from the
 Journal Pre-proof

Zn-Ag synthetic solution containing 500 mg/L Zn and 20 mg/L Ag showed that silver also

precipitates with zinc. This is consistent with the findings presented in a previous study

(Yazici et al., 2017), which demonstrated that TMT is an effective reagent for precipitation of

silver from cyanide solutions even at a low level of silver (i.e., complete precipitation of silver

from cyanide solutions containing 1 mg/L Ag). Precipitation of silver by TMT from cyanide

solutions is thermodynamically feasible as expressed in Eqs. 17-18 (HSC Chemistry, 2011).

The highest rate of precipitation for zinc (79.5%) and silver (89.5%) was obtained at the

lowest [CN-]/[Metal] ratio of 3 (Fig. 9). Increasing the ratio of [CN -]/[Metal] from 3 to 6

f
oo
imposed a 6-12% reduction in precipitation of zinc and silver (Fig. 9).

2 [Ag(CN) 2 ]
- 2
pr
 HC 3 N 3S3   [Ag 2 HC 3 N 3S3 ](s) + 4CN- (G0(25°C)= ‒27.4 kJ ) (17)
 
e-
Silver-cyanidecomplex precipitate
Pr

 C3 N 3S3   [Ag 3C3 N 3S3 ](s) + 6CN-

3
(G0(25°C)= ‒68.4 kJ)
-
3 [Ag(CN) 2 ] (18)
  
Silver-cyanidecomplex precipitate
al
rn

Fig. 9.
u
Jo

The efficiency of precipitation of zinc with TMT also tested at the same TMT concentration

(63 mM) using real pregnant leach solutions (PLSs) derived from the cyanide leaching of a

zinc-bearing silver ore (Fig. 10). The addition of TMT into PLSs of the silver ore resulted in

precipitation of 66.9% Zn and 86.7% Ag (Fig. 10).

Fig. 10.

Separate tests were also performed using synthetic multi-metal (Fig. 11) and real PLSs (Fig.

12) containing Au, Ag, Co and Zn in order to demonstrate the efficiency and selectivity of the
 Journal Pre-proof

method. Fig. 11 shows the precipitation of zinc and other metals from a Zn-Cu-Ag-Au

synthetic solution containing 500 mg/L Zn, 500 mg/L Cu, 20 mg/L Ag and 15 mg/L Au.

Precipitation of zinc was 71.0-71.7% at [CN-]/[Metal] ratios of 3-4 while further increasing

the [CN-]/[Metal] ratio to 6 resulted in a significant reduction in the precipitation of zinc from

71.7% to 49.2%. Silver precipitation appeared to be not affected by the increase in [CN -

]/[Metal] ratio and followed a steady trend within the range of 88.1-93.4%. Copper was

mostly (93.6%) remained in the solution (Fig. 11). Precipitation of gold was negligible. This

can be attributed to the fact that the formation of gold-TMT compounds is not

f
oo
thermodynamically feasible (i.e., ΔG0 > 0) as presented in Eqs. 19-20 (HSC Chemistry, 2011)

Fig. 11.
pr
e-
Pr

 HC 3 N 3S3   [Au 2 HC 3 N 3S3 ](s) + 4CN- (G0(25°C)=+350 kJ) (19)

- 2
2 [Au(CN) 2 ]
 
Gold-cyanidecomplex precipitate
al

3 [Au(CN) 2 ]  C3 N 3S3   [Au 3C3 N 3S3 ](s) + 6CN-

3
(G0(25°C)=+498 kJ)
-
(20)
 
rn

Gold-cyanidecomplex precipitate
u
Jo

Fig. 12 shows the precipitation of metals from PLSs of a pyritic gold concentrate.

Precipitation of zinc and silver was 98.7% and 88.6%, respectively, while no copper and gold

precipitation was occurred.

Fig. 12.

The addition of TMT (i.e., 63 mM) was kept constant in all synthetic multi-metal and real

solutions while [TMT]/[Metal] and [CN]/[Metal] ratios were inherently varied depending on
 Journal Pre-proof

metal concentrations. The findings obtained in the tests using synthetic multi-metal (Figs. 9

and 11) and real solutions (Figs. 10 and 12) showed that the results were comparable in terms

of metals precipitation and selectivity. For instance, under the experimental conditions

([TMT]/[Metal]=4 and [CN]/[Metal]=4) (Fig. 10 and Fig. 11), consistent precipitation of

metals was observed for Zn and Ag i.e., 66.9% Zn and 86.7% Ag (Fig. 10) while 71.0% Zn

and 88.1% Ag (Fig. 11). Interestingly, in Fig. 12, high precipitation of zinc was observed (i.e.,

98.7%) compared with the data presented in Figs. 10-11. This can be ascribed to the complex

leach chemistry of real leach solutions. In all the tests (Figs. 9-12) the selectivity of Zn (and

f
oo
Ag) over gold/copper was high with consistently higher recoveries of Zn/Ag (≥71.0%) than

Cu/Au (≤6.4%).
pr
e-
The findings of precipitation tests from single-metal synthetic solutions (Table 3) showed that
Pr

[TMT]/[Zn] ratios (≥9.48) higher than the stoichiometric ratio (i.e., 0.67) were required for

effective precipitation of zinc (i.e., ≥80%) at [CN -]/[Zn]≤3. On the other hand, relatively high
al

zinc recoveries up to 98.7% were achieved from synthetic multi-metal and real leach solutions
rn

even at low [TMT]/[Metal] ratios of ≤4 ([CN -]/[Zn]=3-4) (Figs. 10-12). This could be linked
u

with the increased ionic strength and relative stability of metal ions (e.g., further
Jo

complexation of freed cyanide ions with copper particularly at low [CN -]/[Metal] ratios),

driving the precipitation reaction (Eqs. 12-13) further to the right (Eqs. 12-13). It is pertinent

to note that, the reaction in Eq. 13 become feasible at 25°C (G0= –877 kJ) in the presence of

copper (Eq. 22). Therefore, the consumption of TMT in practice would be expected to be

lower than those in single-metal synthetic solutions. The proposed method in the study could

provide an efficient alternative to existing zinc precipitation methods with its advantages

including the inherent non-toxic character of TMT and high selectivity over Cu/Au.
 Journal Pre-proof

3 [Zn(CN)3 ]  2C3 N 3S3   9CuCN2  [Zn 3 C3 N 3S3 2 ](s) + 9CuCN3

- 3  2
 (21)
  
Zinc-cyanidecomplex precipitate

(G0(25°C)= –912 kJ)

3 [Zn(CN) 4 ]  2C3 N 3S3   12CuCN2  [Zn 3 C3 N 3S3 2 ](s) + 12CuCN3

2- 3  2
 (22)
  
Zinc-cyanidecomplex precipitate

(G0(25°C)= –877 kJ)

f
oo
3.4. Characterisation of Zn-TMT Precipitates
pr
Size distributions of H3TMT and Zn-TMT precipitates are depicted in Fig. 13. The 80%
e-
passing size (d80) of the precipitates were 20 µm and 16 µm for H3TMT and Zn-TMT,
Pr

respectively (Fig. 13). The stereo-microscope images (Fig. 14) of the Zn-TMT precipitates

indicated the presence of transparent crystals.

al
rn

Fig. 13.
u
Jo

Fig. 14.

SEM-EDS analysis indicated that Zn-TMT precipitates contain around 52-53% Zn (Fig. 15).

FT-IR spectra of precipitates are also illustrated in Fig. 16. Apparently, the peaks of the Zn-

TMT precipitates were differentiated from TMT peaks with characteristic bands at 2800-

3000, 2000-2100, 1400-1500, 800-1100 and 600 cm-1. FT-IR analysis of the NaCN, NaCN-

TMT and Zn-NaCN solutions was also presented in Fig. 17. The FT-IR profile of NaCN and

Zn-NaCN solutions were appeared to be almost identical. However, only in NaCN-TMT

 Journal Pre-proof

solutions, higher intensities observed at 600 cm-1, 845 cm-1, 1218 cm-1 and 1412 cm-1 which

were indicative of the presence of TMT.

Fig. 15.

Fig. 16.

Fig. 17.

f
oo
3.5. Recovery of Zinc from Zn-TMT Precipitates
pr
The recovery of zinc from Zn-TMT precipitates and concurrent regeneration of TMT is of
e-
practical importance. Therefore, extraction of zinc using different type of mineral acids (1-3
Pr

M HCl or H2SO4) was tested. The results (Fig. 18) have shown that both acids even at the

lowest concentration tested (i.e., 1 M) allow complete recovery of zinc from Zn-TMT
al

compounds. Eqs. 23-26 shows the proposed reactions for leaching of zinc in acid solutions.
rn
u

Fig. 18.
Jo

After the acid treatment, TMT is converted to solid H 3TMT as the dominant form of TMT in

acidic conditions (<pH 5) (Eqs. 23-26 and Fig. 4). After separation of solid H3TMT by

filtration, zinc could be also recovered from the waste solution. Solid H3TMT can be

dissolved in alkaline solutions (>pH 8) and used in further precipitation of zinc.

[ZnHC 3 N3S3 ](s)  2HCl(aq)  [H3C3 N3S3 ](s)  Zn 2  2Cl (G0(25°C)= –160 kJ) (23)

Zinc-TMT precipitate
 Journal Pre-proof

2
[ZnHC 3 N3S3 ](s)  H 2SO4 (aq)  [H 3C3 N3S3 ](s)  Zn 2  SO4 (G0(25°C)= –207 kJ) (24)

Zinc-TMT precipitate

[Zn 3 C3 N3S3 2 ](s)  6HCl(aq)  2[H3C3 N3S3 ](s) 3Zn 2  6Cl (G0(25°C)= +39.3 kJ) (25)
  
Zinc-TMT precipitate

[Zn 3 C3 N3S3 2 ](s)  3H2SO4 (aq)  2[H3C3 N3S3 ](s) + 3Zn 2  3SO4
2
(26)
  
Zinc-TMT precipitate

f
(G0(25°C)= –95 kJ)

oo
pr
TMT consumption could be significantly reduced by the recovery of zinc from precipitates
using acid solutions (Fig. 18) through Eqs. 23-26, followed by conversion of solid H3TMT
e-
into dissolved form using alkaline solutions.
Pr

4. Conclusions
al

A novel method was developed for the precipitation of zinc from cyanide solutions using a
rn

non-toxic organosulphide reagent i.e., Trimercapto-s-triazine (C3N3S33-; TMT). Kinetic

u

studies (5-120 min.) showed that precipitation of zinc by TMT occurs rapidly within 5-15
Jo

min. Increasing the concentration of TMT from 33 to 90 mM was found to improve the

precipitation of zinc, which reached 97.7% at 90 mM TMT. Increasing the alkalinity from pH

10 to pH 12.5 produced a 17.9% enhancement whilst varying the temperature between 20 and

60°C had no significant influence (64.7% ± 5.13% (RSD) Zn) on zinc precipitation.

Further detailed tests designed by statistical design of experiments (i.e., response surface

methodology) carried out for investigation of the effect of [TMT]/[Zn] ratio (molar) (2.70-

16.25), [CN-]/[Zn] ratio (molar) (3.00-8.00) and initial concentration of Zn (50-950 mg/L) on

precipitation of zinc in five levels. Up to 95.4% Zn was precipitated under suitable conditions.
 Journal Pre-proof

The statistical analysis of the data has shown that the most influential factor is the

[TMT]/[Zn] ratio. The [CN-]/[Zn] ratio appeared to have also a limited adverse effect on the

process. Further tests were performed using synthetic solutions containing multi-metal and

pregnant leach solutions containing Zn/Cu/Ag/Au to mimic real conditions. These tests have

also revealed that silver also precipitates with zinc while no significant precipitation of gold

and copper occurs. This implies that TMT is highly selective for zinc over gold and copper.

The detailed characterisation of Zn-TMT precipitates was also presented. The results have

also indicated that Zn-TMT precipitates could be treated with mineral acids (1-3 M HCl and

f
oo
H2SO4) for ready regeneration/recovery of TMT. This study demonstrated that the proposed

method can be an efficient and environmentally sound option for precipitation of zinc from

cyanide leach solutions.

pr
e-
Pr

Acknowledgements

The authors would like to acknowledge TUBITAK (The Scientific and Technological
al

Research Council of Turkey) for providing financial support (Project No: 213M492), SNF
rn

Floerger for kindly providing reagents and Dr. Fatih Erdemir (KTU, Dept. of
u

Metallurgical&Materials Eng.) for his assistance during characterization tests.

Jo

References

Adams, M.D., 2016. Summary of gold plants and processes. in: Adams, M.D. (Ed.),

Advances in Gold Ore Processing. Elsevier, Amsterdam, pp. 961-984.

Antony, J., 2003. Design of Experiments for Engineers and Scientists. Butterworth-

Heinemann, 166 pp.

 Journal Pre-proof

Bas, A.D., Koc, E., Yazici, E.Y., Deveci, H., 2015. Treatment of copper-rich gold ore by

cyanide leaching, ammonia pretreatment and ammoniacal cyanide leaching. T.

Nonferr. Metal. Soc. 25(2), 597-607. https://doi.org/10.1016/S1003-6326(15)63642-1.

Breuer, P.L., Sutcliffe, C.A., Meakin, R.L., 2011. Cyanide measurement by silver nitrate

titration: Comparison of rhodanine and potentiometric end-points. Hydrometallurgy.

106(3-4), 135-140. https://doi.org/10.1016/j.hydromet.2010.12.008.

Deveci, H., Yazici, E.Y., Alp, I., Uslu, T., 2006. Removal of cyanide from aqueous solutions

f
by plain and metal-impregnated granular activated carbons. Int. J. Miner. Process.

oo
79(3), 198-208. https://doi.org/10.1016/j.minpro.2006.03.002.

pr
Dongmei, L., Yunbai, L., Ping, Y., Zhigang, C., 2006. Chemistry of copper
e-
trimercaptotriazine (TMT) compounds and removal of copper from copper-ammine
Pr

species by TMT. Appl. Organomet. Chem. 20(4), 246-253.

https://doi.org/10.1002/aoc.1049.
al

Estay, H., 2018. Designing the SART process – A review. Hydrometallurgy. 176, 147-165.
rn

https://doi.org/10.1016/j.hydromet.2018.01.011.
u

Estay, H., Becker, J., Carvajal, P., Arriagada F., 2012. Predicting HCN gas generation in the
Jo

SART process. Hydrometallurgy. 113-114, 131-142.

https://doi.org/10.1016/j.hydromet.2011.12.019.

Estay, H., Gim-Krumm, M., Quilaqueo, M., 2018. Two-stage SART process: A feasible

alternative for gold cyanidation plants with high zinc and copper contents. Minerals

8(392), 1-12. https://doi.org/10.3390/min8090392.

Evonik Industries, 2018. TMT Info 15: Toxicology / Ecology / properties of precipitates,

Application Information / Info 15 E.

 Journal Pre-proof

Fleming, C.A., 2003. The economic and environmental case for recovering cyanide from gold

plant tailings, SGS Minerals Services, Technical Paper, 2003-02, 9 pp.

Fleming, C.A., 2010. Cyanide management in the gold industry, SGS Minerals Services,

Technical Paper, 2010-04, 3 pp.

Fleming, C.A., 2016. Cyanide recovery. in: M.D. Adams (Ed.), Advances in Gold Ore

Processing. Elsevier, Amsterdam, pp. 647-661.

Goldstone, A.J. and Mudder, T.I., 1991. Cyanide regeneration process, US Patent, Patent No.

f
oo
US 4994243 A, 19 February.

Gonen, N., Kabasakal, O.S. and Ozdil, G. 2004. Recovery of cyanide in gold leach waste
pr
solution by volatilization and absorption. J. Hazard. Mater. B113, 231-236.
e-
https://doi.org/10.1016/j.jhazmat.2004.06.029.
Pr

Guo, B., Peng, Y., Mai, Y., 2016. The effect of zinc cyanide on the flotation of gold from

pyritic ore. Miner. Eng. 85, 106–111

al

Habashi, F., 1987. One hundred years of cyanidation, CIM Bulletin, 80, 108-114.
rn

Henley, K.J., 1975. Gold-ore mineralogy and its relation to metallurgical treatment. Miner.
u

Sci. Eng. 7, 289-310.

Jo

Koc, E., Ahlatci, F., Celep, O., Yazici, E.Y. and Deveci, H., 2014. Interference of metals with

the determination of free cyanide. in: İpekoğlu, Ü., Arslan, V., Şen, S. (Eds.), XIV.

International Mineral Processing Symposium and Exhibition (IMPS), 15-17 October,

Kuşadası, Turkey, 1027-1034.

Koc, E., Ahlatcı, F., Kuzu, M., Yazici, E.Y., Celep, O., Deveci, H., 2016. Recovery of silver

from cyanide leach solutions of a pyritic gold concentrate by sodium sulphide

precipitation. In: Çelik, M.S., Bulut, G., Karakaş, F., Güven, O., Baştürkçü, H.,
 Journal Pre-proof

Tarsus, Z. (Eds.), XV. International Mineral Processing Symposium and Exhibition

(IMPS), Istanbul, Turkey, 823-834.

Kurama, H. and Çatalsarik, T., 2000. Removal of zinc cyanide from a leach solution by an

anionic ion-exchange resin. Desalination, 129(1), 1-6. https://doi.org/10.1016/S0011-

9164(00)00045-X.

Lapidus, G. and Millard, M., 2016. Processing of high-silver gold ores. in: M.D. Adams (Ed.),

Advances in Gold Ore Processing. Elsevier, Amsterdam, pp. 821-830.

f
oo
Marsden, J., House, I., 2006. The Chemistry of Gold Extraction, Society for Mining,

Metallurgy, and Exploration, USA, 688 pp.

pr
MEDUSA, 2009. Software for Chemical Equilibrium Diagrams, Royal Institute of
e-
Technology, Sweden.
Pr

Miltzarek, G.L., Sampaio, C.H., Cortina, J.L., 2002. Cyanide recovery in hydrometallurgical

plants: use of synthetic solutions constituted by metallic cyanide complexes. Miner.

al

Eng. 15, 75-82. https://doi.org/10.1016/S0892-6875(01)00204-7.

rn

Mudder, T. I. ve Botz, M. M., 2001. The Chemistry and Treatment of Cyanidation Wastes,
u

Mining Journal Books Ltd, London, 393 pp.

Jo

Paterson, M. R., 2017. The RECYN Process – Changing The Negative Perception of Cyanide.

Proceedings of Gold Sessions at ALTA 2017, 20-27 May, Perth, Australia.

Ritcey, G.M., 2005. Tailings management in gold plants. Hydrometallurgy. 78, 3-20.

https://doi.org/10.1016/j.hydromet.2005.01.001.

SGS, 2013. Cyanide recovery, SGS Minerals Services, T3 SGS 019, 11-2013, 6 pp.
 Journal Pre-proof

Sheya, S., Palmer, G., 1989. Effect of metal impurities on adsorption of gold by activated

carbon in cyanide solutions, Report of Investigations, US Bureau of Mines, RI:9268, 6

pp.

SNF Floerger, 2011. Metalsorb ZM3 - Product Bulletin.

Van der Merwe, W. and Breuer, P., 2011. Cyanide analysis for complex cyanide solutions,

World Gold 2011, 50th Annual Conference of Metallurgists of CIM, 2-5 October 2011,

Montreal, Canada, 581-592.

f
oo
Walton, R., 2016. Chapter 31 – Zinc Cementation. in: Adams, M.D. (Ed.), Advances in Gold

Ore Processing. Elsevier, Amsterdam, pp. 553-560.

pr
Yazici, E.Y., Deveci, H. and Alp, I., 2009. Treatment of cyanide effluents by oxidation and
e-
adsorption in batch and column studies. J. Hazard. Mater. 166(2-3), 1362-1366.
Pr

https://doi.org/10.1016/j.jhazmat.2008.12.050.

Yazici, E.Y., Deveci, H. and Yazici, R., 2011. Recovery of silver from X-ray film processing
al

effluents using trimercapto-s-triazine (TMT). Sep. Sci. Technol. 46(14), 2231-2238.

rn

https://doi.org/10.1080/01496395.2011.595032.
u

Yazici, E.Y., Yilmaz, E., Ahlatci, F., Celep, O. and Deveci, H., 2017. Recovery of silver from
Jo

cyanide leach solutions by precipitation using Trimercapto-s-triazine (TMT).

Hydrometallurgy. 174, 175-183. https://doi.org/10.1016/j.hydromet.2017.10.013.

Yılmaz, E., 2016. Metals precipitation and recovery of cyanide from cyanide leaching

solutions, MSc Thesis, Karadeniz Technical University, Dept. of Mining Eng.,

Turkey, 103 pp. (in Turkish).

Yilmaz, E., Ahlatci, F., Kuzu, M., Yazici, E.Y., Celep, O., Deveci, H., 2016. Removal of zinc

from cyanide leach solutions. in: Aydıner, K., Yaşar, S., Yaşar, Ö. (Eds.), International
 Journal Pre-proof

Black Sea Mining and Tunnelling Symposium (IBSMTS), 2-4 November, KTU,

Trabzon, Turkey, 447-454.

Yılmaz, E., Yazıcı, E. Y., Celep, O., Deveci, H., 2019a. Processes for cyanide recovery from

leach effluents, Mining, 58(1), 53-71 (in Turkish).

https://doi.org/10.30797/madencilik.537646.

Yılmaz, E., Yazıcı, E. Y., Celep, O., Deveci, H., 2019b. Recovery of cyanide from effluents:

SART process, Gumushane University Journal of Science and Technology Institute,

f
9(3), 600-615 (in Turkish). https://doi.org/10.17714/gumusfenbil.452203.

oo
pr
e-
Pr
al
u rn
Jo
 Journal Pre-proof

List of Tables

Table 1. Solubility of zinc minerals in cyanide solutions (Habashi, 2009).

Table 2. Experimental design layout with investigated parameters and their levels applied for

precipitation tests.

Table 3. Experimental design runs with corresponding results for precipitation of zinc (pH

11.5, 60 min., 25°C).

f
oo
Table 4. Analysis of variance (ANOVA) result for the second-order regression model of zinc

precipitation.
pr
e-
Pr
al
u rn
Jo
 Journal Pre-proof

Table 1.

Mineral Formula Solubility (%, 24 h.)

Smithsonite ZnCO3 40.2

Zincite ZnO 35.2

Hydrozincite 3ZnCO3.2H2O 35.1
Franklinite (Fe,Mn,Zn)O.(Fe,Mn)2O3 20.2
Sphalerite ZnS 18.4 *

f
oo
Gelamine H2Zn2SiO4 13.4
Willemite Zn2SiO4 13.1
* 2 g/L NaCN, 45°C (Mudder and Botz, 2001) pr
e-
Pr
al
u rn
Jo
 Journal Pre-proof

Table 2.

Parameters with coded levels (Actual Levels)

Exp. No
[TMT]/[Zn] [CN-]/[Zn] Zn (mg/L)

1 -1 (5.45) -1 (4.01) -1 (232)

2 1 (13.50) -1 (4.01) -1 (232)
3 -1 (5.45) 1 (6.99) -1 (232)
4 1 (13.50) 1 (6.99) -1 (232)
5 -1 (5.45) -1 (4.01) 1 (768)
6 1 (13.50) -1 (4.01) 1 (768)

f
oo
7 -1 (5.45) 1 (6.99) 1 (768)
8 1 (13.50) 1 (6.99) 1 (768)
9
10
-1.68 (2.70)
1.68 (16.25)
pr
0 (5.50)
0 (5.50)
0 (500)
0 (500)
e-
11 0 (9.48) -1.68 (3.00) 0 (500)
Pr

12 0 (9.48) 1.68 (6.99) 0 (500)

13 0 (9.48) 0 (5.50) -1.68 (50)
14 0 (9.48) 0 (5.50) 1.68 (950)
al

15-20 0 (9.48) 0 (5.50) 0 (500)

u rn
Jo
 Journal Pre-proof

Table 3.

Calculated
Precipitation
No. [TMT]/[Zn] [CN-]/[Zn] Zn (mg/L) Recovery of
of Zn (%)
Cyanide (%)
1 5.45 4.01 232 36.2 7.1
2 13.5 4.01 232 98.6 95.6
3 5.45 6.99 232 44.1 44.0
4 13.5 6.99 232 92.9 91.1

f
5 5.45 4.01 768 47.8 40.6

oo
6 13.5 4.01 768 91.6 89.1
7 5.45 6.99 768 32.0 30.2
8
9
13.5
2.7
6.99
5.5
pr
768
500
74.1
11.1
72.5
30.2
e-
10 16.25 5.5 500 95.4 93.8
Pr

11 9.48 3 500 80.8 69.8

12 9.48 8 500 48.0 47.4
13 9.48 5.5 50 22.6 0.0
al

14 9.48 5.5 950 66.1 64.4

rn

15-20 9.48 5.5 500 71.2±0.55 69.8±0.55

u
Jo
 Journal Pre-proof

Table 4.

Degree of Mean
Source Sum of Squares F-value P-value
Freedom squares
Model 10248.82 9 1138.76 7.90 0.0017
A-[TMT]/[Zn] 8402.69 1 8402.69 58.27 <0.0001
B-[CN-]/[Zn] 546.89 1 546.89 3.79 0.0801
C-Zn (mg/L) 160.12 1 160.12 1.11 0.3168
AB 29.23 1 29.2 0.20 0.6622

f
AC 80.90 1 80.90 0.56 0.4711

oo
BC 158.26 1 158.26 1.10 0.3195
A2 215.77 1 215.77 1.50 0.2493
B2
C2
0.06
708.08
1
1
pr 0.06
708.08
0.00
4.91
0.9842
0.0510
e-
Error 1442.06 10 144.21
Pr

Total 11690.88 19
al
u rn
Jo
 Journal Pre-proof

List of Figures

Fig. 1. The structure of (a) trimercapto-s-triazine trisodium (Na3C3N3S3) and (b) metalated

form (Me3C3N3S3) at >pH 12.5 (structural waters were not included) (Henke and Atwood,

1998; Henke et al., 2001).

Fig. 2. The effect of time and concentration of TMT (33-90 mM) on precipitation of zinc (500

mg/L Zn, [CN-]/[Zn]: 4, 1.5 g/L NaCN, pH 11.5, 25°C).

f
Fig. 3. Effect of pH on precipitation of zinc (63 mM TMT, 500 mg/L Zn, [CN -]/[Zn]: 4, 1.5

oo
g/L NaCN, pH 11.5, 60 min., 25°C).

pr
Fig. 4. pH dependent speciation of trimercapto-s-triazine (C3N3S3) (constructed based on data
e-
from Dongmei et al. (2006)).
Pr

Fig. 5. Speciation of zinc in cyanide solutions (500 mg/L Zn, [CN -]/[Zn]:4, 1.5 g/L NaCN,

25°C) (Medusa, 2009).

al

Fig. 6. Effect of temperature on zinc precipitation (63 mM TMT, 500 mg/L Zn, [CN -]/[Zn]:4.
rn

1.5 g/L NaCN, pH 11.5, 60 min., 25°C) (Error bars show a standard deviation of 5%).
u
Jo

Fig. 7. Speciation of cyanide species vs. concentration of cyanide (Zn: 500 mg/L, pH 11.5,

25°C) (Medusa, 2009).

Fig. 8. Three-dimensional response surface plots showing the influence of dual parameters on

precipitation of zinc (a) [TMT]/[Zn] and [CN -]/[Zn], (b) [TMT]/[Zn] and Concentration of

Zn, (c) [CN-]/[Zn] and Concentration of Zn (pH 11.5, 60 min., 25°C).

Fig. 9. Precipitation of metals by TMT from synthetic Zn-Ag solutions at different [CN -

]/[Metal] ratios (63 mM TMT, 500 mg/L Zn, 20 mg/L Ag, pH 11.5, 25°C)
 Journal Pre-proof

Fig. 10. Efficiency of precipitation of zinc from real pregnant leach solutions (39.4 mg/L Ag,

967 mg/L Zn and 7.80 mg/L Cu) of a silver ore (63 mM TMT, 3 g/L NaCN, pH 11.5, 60 min.,

25°C).

Fig. 11. Precipitation of metals by TMT from synthetic Zn-Cu-Ag-Au solutions at different

[CN‒]/[Metal] ratios (63 mM TMT, 500 mg/L Zn, 500 mg/L Cu, 20 mg/L Ag, 15 mg/L Au,

pH 11.5, 25°C).

Fig. 12. Efficiency of precipitation of zinc from real pregnant leach solutions (37.9 mg/L Ag,

f
oo
29.4 mg/L Au, 733 mg/L Cu and 427 mg/L Zn) of a pyritic gold concentrate (63 mM TMT, 3

g/L NaCN, pH 11.5, 60 min., 25°C).

pr
Fig. 13. Particle size distributions of H3TMT and Zn-TMT precipitates.
e-
Pr

Fig. 14. Stereo-microscope images of Zn-TMT precipitates.

Fig. 15. SEM image with EDS analysis of the Zn-TMT precipitates.
al
rn

Fig. 16. FT-IR spectra of TMT (H3TMT) and Zn-TMT precipitates

u

Fig. 17. FT-IR spectra of NaCN (1.5 g/L), NaCN (1.5 g/L)-TMT and NaCN (1.5 g/L)-Zn
Jo

(500 mg/L) solutions.

Fig. 18. Recovery of zinc from Zn-TMT precipitates in different mineral acids (HCl and

H2SO4) at varying concentrations (1-3 M) (1 min., 25°C).

 Journal Pre-proof

Na S N S Na Me
C C N S Me
S

N N N N
C
S
S Na Me
Fig. 1.

f
oo
pr
e-
Pr
al
urn
Jo
 Journal Pre-proof

Precipitation of Zn (%) 100

80

60

40
33 mM TMT
48 mM TMT
20
63 mM TMT

f
90 mM TMT

oo
0
0 30 60 90 120
Time (min.)
pr
Fig. 2.
e-
Pr
al
rn
u
Jo
 Journal Pre-proof

Precipitation of Zn (%) 100

80

60

40

20

f
oo
0
pH 10 pH 11 pH 12 pH 12.5
Fig. 3.
pr
e-
Pr
al
urn
Jo
 Journal Pre-proof

1,0
1.0

0,8
0.8
α0
(H3C3N3S3)0
α1
(H2C3N3S3)-
Fraction

0,6
0.6
α2
(HC3N3S3)2-

0,4
0.4 α3
(C3N3S3)3-

0,2

f
0.2

oo
0,0
0.0
0 2 44 6 pr 8 10 12
12 14
14
pH
e-
Fig. 4.
Pr
al
u rn
Jo
 Journal Pre-proof

f
oo
pr
e-
Fig. 5.
Pr
al
rn
u
Jo
 Journal Pre-proof

Precipitation of Zn (%) 100

80

60

40

20

f
oo
0
20°C 30°C 40°C 60°C
Fig. 6.
pr
e-
Pr
al
rn
u
Jo
 Journal Pre-proof

f
oo
pr
e-
Pr

Fig. 7.
al
rn
u
Jo
 Journal Pre-proof

Precipitation of Zn (%) a

B: [CN-]/[Zn] A: [TMT]/[Zn]

f
b

oo
Precipitation of Zn (%)

pr
e-
Pr
al

C: Zn (mg/L) A: [TMT]/[Zn]
rn

c
u
Precipitation of Zn (%)

Jo

C: Zn (mg/L)
B: [CN-]/[Zn]

Fig. 8.
 Journal Pre-proof

[CN-]/[Metal] ratio
3 4 6
100
Precipitation of Metals (%)

Zn Ag

80

60

40

f
oo
20

0
1.15
pr
1.54 2.30
e-
NaCN concentration (g/L)
Pr

Fig. 9.
al
rn
u
Jo
 Journal Pre-proof

Precipitation of Metals (%) 100

80

60

40

f
20

oo
0
Cu
pr
Zn Ag
e-
Fig. 10.
Pr
al
rn
u
Jo
 Journal Pre-proof

[CN-]/[Metal] ratio
3 4 6
16
Precipitation of Metals (%)

Cu Ag Zn Au

Au Concentration (mg/L)
100
12
80

60 8

40
4

f
20

oo
0 0
1.15 1.54 2.30
pr
NaCN Concentration (g/L)
e-
Fig. 11.
Pr
al
u rn
Jo
 Journal Pre-proof

Precipitation of Metals (%) 100

80

60

40

20

0
Au Cu Ag Zn

f
oo
Fig. 12.
pr
e-
Pr
al
urn
Jo
 Journal Pre-proof

Cumulative Undersize (%) 100

80

60

Zn-TMT
40
H3TMT
H3TMT

20

0
0,1
0.1 1 10 100 1000

f
oo
Particle Size (µm)

Fig. 13.
pr
e-
Pr
al
u rn
Jo
 Journal Pre-proof

200 µm 1 mm

f
oo
pr
e-
500 µm 500 µm
Pr

Fig. 14.
al
rn
u
Jo
 Journal Pre-proof

f
oo
pr
e-
cps/eV

Pr
al

Zn N
S O
rn

keV

Fig. 15.
u
Jo
 Journal Pre-proof

100
Transmitance (%)

80

60

40

20 Zn-TMT
Zn-TMT
H3TMT
H3TMT
0
4000 3500 3000 2500 2000 1500 1000 500
cm-1

f
oo
Fig. 16.

pr
e-
Pr
al
u rn
Jo
 Journal Pre-proof

100
Transmitance %

90
80
70
1.5g/L NaCN
60
1.5g/L NaCN-TMT
50 1.5g/L NaCN-500 mg/L Zn
40
4000 3500 3000 2500 2000 1500 1000 500
cm-1
Fig. 17.

f
oo
pr
e-
Pr
al
u rn
Jo
 Journal Pre-proof

HCl H2SO4
H2SO4
100
Recovery of Zn (%)

80

60

40

20

0
11M 33
M
Acid Concentration (M)

f
oo
Fig. 18.

pr
e-
Pr
al
rn
u
Jo
 Journal Pre-proof

Highlights

 Trimercapto-s-triazine (TMT) efficiently precipitates zinc from cyanide solutions

 Precipitation of zinc by TMT occurs rapidly within ≤15 min.

 [TMT]/[Zn] ratio is the most influential parameter on the precipitation process

 TMT selectively precipitates zinc from gold and copper

 TMT can be readily recovered from Zn-TMT precipitates by mineral acids

f
oo
pr
e-
Pr
al
u rn
Jo
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12
Figure 13
Figure 14
Figure 15
Figure 16
Figure 17
Figure 18