See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/267746281

Interference of Metals with the Determination of Free Cyanide

Conference Paper · October 2014

DOI: 10.13140/2.1.3302.3682

CITATIONS READS

2 4,622

5 authors, including:

Elif Yilmaz Firat Ahlatci

Karadeniz Technical University Eti Mine
14 PUBLICATIONS 142 CITATIONS 18 PUBLICATIONS 87 CITATIONS

SEE PROFILE SEE PROFILE

Oktay Celep Ersin Yener Yazıcı

Karadeniz Technical University Karadeniz Technical University
110 PUBLICATIONS 698 CITATIONS 130 PUBLICATIONS 2,233 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

SILVER RECOVERY BY THIOSULFATE LEACHING AND ULTRA-FINE GRINDING FROM REFRACTORY ORE AND PLANT TAILINGS View project

Project Title : SILVER RECOVERY BY THIOSULFATE LEACHING AND ULTRA-FINE GRINDING FROM REFRACTORY ORE AND PLANT TAILINGS View project

All content following this page was uploaded by Ersin Yener Yazıcı on 04 November 2014.

The user has requested enhancement of the downloaded file.

 Proceedings of 14th International Mineral Processing Symposium – Kuşadası, Turkey, 2014

INTERFERENCE OF METALS WITH THE DETERMINATION OF

FREE CYANIDE
Elif Koç1, Fırat Ahlatcı1, Oktay Celep1, Ersin Y. Yazici1 and Haci Deveci1,a

1. Hydromet B&PM Group, Mineral&Coal Process. Div., Dept. of Mining Eng., Karadeniz Technical
University, 61080, Trabzon, Turkey
a. Corresponding author ([email protected])

ABSTRACT: During cyanide leaching of gold ores, a minimum level of free cyanide
(>50-200 mg/L) should be maintained to ensure the extraction of gold available for
leaching. Silver nitrate titration (SNT) using rhodanine as the indicator is often used to
determine free cyanide concentration in leaching solutions. However, the leach solutions
also contain cyanide complexes of metals, which can interfere with analysis of cyanide by
silver nitrate titration. In this study, effect of concentration of metals (Cu, Zn, Fe, Ni and
Ag) and their speciation on the determination of free cyanide with SNT was examined. The
findings were compared with the theoretical calculations using an equilibrium modelling
software. The presence of copper, zinc and iron was found to lead to the over-estimation of
free cyanide. Effect of silver and nickel was determined to be insignificant particularly at
CN/Metal ratios of ≥4. It can be inferred that SNT method is not a reliable method and can
be misleading in the analysis of complex cyanide leach solutions for free cyanide.

1. INTRODUCTION excess silver added starts to react with

Cyanide leaching is the most preferred rhodanine to give a change in colour of
method for the extraction of gold from the solution (from yellow to salmon pink)
ores due to the high complexing ability of indicating the end point (Eq. 2) [Breuer et
cyanide with gold. In cyanide leaching, al., 2011; Mudder and Botz, 2001].
the maintenance of certain levels of free
cyanide (i.e. >50-200 ppm CN-) is of vital Ag+ + 2CN- → Ag(CN)2- (1)
importance for leaching kinetics and
ensuring the solubilisation of gold as Rh(yellow) + Ag+ ↔ Rh-Ag(Pink) (2)
Au(CN)2-. The depletion of cyanide i.e.
insufficiently low levels of free cyanide Table 1. Methods for analysis of cyanide
may result in precipitation of gold from solutions [Breuer and Rumball,
cyanide complex as AuCN. Therefore, 2007].
the concentration of free cyanide should Method Species measured
be continuously monitored to keep the SNT (rhodanine
CN-
cyanide at the required levels [Marsden or KI indicator)
and House, 2006; Bas et al., 2012a,b]. SNT
CN-,S2-,estimate of Cu
(potentiometric)
There are numerous methods for analysis Amperometric CN-, estimate of Cu
of cyanide from solutions (Table 1). Flow injection/
ligand exchange/ WAD CN, total CN
Analysis of free cyanide is often carried
UV digestion
out using silver nitrate titration (SNT) in Distillation WAD CN, total CN
which KI or preferably rhodanine is used Metal cyanides, OCN-
as the indicator. This method relies on the IC1 and HPLC2
SCN-, S2-, S2O32-
fact that silver forms stable complexes 1
Ion Chromatography
2
with cyanide (Eq. 1). As soon as all the High Performance Liquid Chromatography
free cyanide complexes with silver, the

1027
Table 2: Stability and classification of some researchers [Breuer and Rumball,
metal-cyanide complexes [Marsden and 2007; Breuer et al., 2011; van der Merwe
House, 2006]. and Breuer, 2013] investigated the effect
Stability of copper, zinc and/or sulphide species on
Reaction equation
constant SNT using rhodanine as indicator.
Strong Acid Dissociable Cyanides (SAD) However, other metals present in cyanide
Co3++6CN-↔Co(CN)63- logβ6=64.0 leach solutions are also of concern for the
3+ - 3-
Fe +6CN ↔ Fe(CN)6 logβ6=43.6 reliability of SNT.
Fe2++6CN-↔Fe(CN)64- logβ6=35.4
Au++2CN-↔ Au(CN)2- logβ2=38.8 This study was undertaken to investigate
Weak Acid Dissociable Cyanides (WAD) the effect of concentration of metals (Cu,
Ni2++4CN-↔Ni(CN)42- logβ4=30.22
Zn, Fe, Ni and Ag) and their speciation
Cu++4CN-↔Cu(CN)43- logβ4=23.10
on the determination of free cyanide with
Cu++3CN-↔Cu(CN)32- logβ3=21.66
+ -
Ag +2CN ↔Ag(CN)2 -
logβ2=20.48
SNT. Their effect was quantified based
Cu++2CN-↔Cu(CN)2- logβ2=16.26 on comparison of the experimental data
Zn2++3CN-↔Zn(CN)3- logβ3=16.05 with theoretical calculations.
2+ - 2-
Zn +4CN ↔Zn(CN)4 logβ4=19.62
Cd++4CN-↔Cd(CN)43- logβ4=17.92 2. EXPERIMENTAL
Free Cyanide (CN-, HCN) NaCN (≥95%, Merck) was used in
CN-+H+↔HCN (pKa 9.3 at 20°C) preparing a stock cyanide solution of 0.1
M NaCN at pH 12. Stock solutions (0.1
Pregnant solutions of cyanide leaching M) of metals were prepared using
are often contaminated by the impurity FeSO4.7H2O, Fe2(SO4)3.nH2O, ZnCl2,
metals such as copper [Bas et al., 2012a; NiSO4.6H2O and AgNO3. The stock
Marsden and House, 2006; Sceresini, solution of copper (0.1 M) was prepared
2005]. Cyanide also forms complexes using CuCl2.2H2O saturated with NaCl.
with these impurity metals solubilised Metal-cyanide solutions at required
from host/gangue minerals, increasing the strengths were prepared using metal and
consumption of cyanide in leaching. To cyanide stock solutions. Distilled-
illustrate, at typical cyanide leaching deionised water was used for the
conditions, copper dominantly presents in preparation of all the solutions. pH of the
cuprous form (Cu+), which readily forms solutions was adjusted to 12 using NaOH
cyano-complexes of Cu(CN)2-, Cu(CN)32- to avoid formation of HCN(aq). Table 3
and Cu(CN)43- depending on the shows metal-cyanide solutions prepared
concentration of cyanide and copper, pH and analysed for free cyanide with SNT.
and temperature [Lu et al., 2002; Xie et
al., 2013]. Metal-cyanide complexes are Table 3: Metal-cyanide solutions
classified with respect to their stability prepared and analysed for CN- with SNT.
constants as strong (logK>30) and weak Metals
Experimental conditions
(logK≤30) complexes (Table 2) [van der tested [CN-], M
CN/Metal
Merwe and Breuer, 2013]. ratio (molar)
0.01
Cu 0.02 2-20
The presence of impurity metals can
0.03
interfere with SNT leading to false
0.02
estimation of free cyanide in leaching Zn 4-10
0.03
circuits. Although the interference of Fe(II)/Fe(III) 0.03 6-12
metals with SNT is well-known, there are Ni 0.03 4-10
limited studies quantifying the effect of Ag 0.01 2-10
these metals on the analysis. Recently,

1028
 Proceedings of 14th International Mineral Processing Symposium – Kuşadası, Turkey, 2014

In the SNT analysis, a digital burette was Ag++2Cu(CN)32-→Ag(CN)2-+2Cu(CN)2-

used to measure the added volume of (ΔG0(25°C) = +31.8 kJ) (3)
AgNO3 (0.02 M). Solution of rhodanine
indicator (0.02% w/v) was prepared in Ag++2Cu(CN)43-→Ag(CN)2-+2Cu(CN)32-
acetone using p- (ΔG0(25°C) = -98.7 kJ) (4)
dimethylaminobenzylidene rhodanine.
All the SNT analyses were performed in Rh-Ag++2Cu(CN)43-↔Rh+Ag(CN)2-
triplicate and mean values were used in +2Cu(CN)32- (5)
the presentation of results. Relative
standard deviation (RSD) of the data was
calculated to be ≤5%. (a)

An equilibrium modelling software

[Medusa, 2010] was used to calculate the
theoretical concentration of free cyanide
under the condition tested. The
theoretical (expected) values were used to
compare with the SNT (actual) results.

3. RESULTS AND DISCUSSION

3.1. Effect of Copper
The influence of copper on SNT of free
cyanide was tested at three levels of (b)
cyanide i.e. 0.01, 0.02 and 0.03 M CN-
(Figure 1). The results (Figure 1a-c)
showed that, regardless from the
concentration of CN- (i.e. 0.01, 0.02 and
0.03 M), the presence of copper led to the
over-estimation of free cyanide. Other
researchers also reported similar findings
in the presence of copper [Breuer and
Rumball, 2007; Breuer et al., 2011; van
der Merwe and Breuer, 2013].

The over-estimation of free cyanide by (c)

SNT can be ascribed to the complexation
of silver with cyanide donated from
copper-cyanide complexes (Eqs. 3-4),
which is deemed to be a faster reaction
than the formation of silver-rhodanine
complex (Eq. 5). The reactions (Eqs. 3-5)
suggest that silver is consumed during
SNT, leading to the over-estimation of
free cyanide [Breuer and Rumball, 2007;
Breuer et al., 2011]. However, Gibbs free
energies of the reactions in Eqs. 3-4
suggest that only Eq. 4 is Figure 1: Effect of copper on SNT (a)
thermodynamically favoured. [CN-]: 0.01 M, (b) [CN-]: 0.02 M and (b)
[CN-]: 0.03 M (pH 12).

1029
The magnitude of over-read values was of [CN-]: 0.03 M and pH 12, (presented
found to be strongly dependent on CN/Cu at Figure 1c), is shown in Figure 2. At
ratio. The difference between the CN/Cu ratios of ≥3, the concentration of
expected (theoretical) and measured free cyanide in solution increases.
concentrations of free cyanide tended to Increasing the CN/Cu ratio from 3.5 to
decrease with increasing CN/Cu ratio. To 20, shifts the predominant species of
exemplify, in Figure 1c ([CN-]: 0.03 M), cyanide from copper-cyanide complexes
at CN/Cu ratios of 3.5 and 20, SNT (89% as Cu(CN)32- and Cu(CN)43-) to free
determined the concentration of free cyanide (82% as CN-). This is consistent
cyanide by 325% and 7.5% higher than with the higher accuracy of SNT at high
the expected values. CN/Cu ratios (Figure 1a-c). Similar to the
findings in the current study, some
Speciation of cyanide can contribute to researchers [Breuer and Rumball, 2007;
the understanding of the effect of CN/Cu Breuer et al., 2011] also reported that the
ratio. The speciation of cyanide as a magnitude of over-estimation of free
function of CN/Cu ratio, which is cyanide using SNT with rhodanine was
analogous to the experimental conditions related to the concentration of copper.
pH = 12 0
.0 C
[ N ]TO T = 3 0 .0 0 m M
in
F ra c to

Figure 2: Speciation of cyanide vs. concentration of CN- and CN/Cu molar ratio ([CN-]:
0.03 M, pH 12, 25°C) [Medusa, 2010].

Recent studies have confirmed that and Breuer, 2013]. In this regard, van der
potentiometric titration of cyanide with Merwe and Breuer [2013] reported that at
silver nitrate is the most reliable method Cu/CN ratio of 4 where the theoretical
for the determination of free cyanide concentration of free cyanide is 417
from the solutions containing copper as mg/L, SNT analysis by potentiometric
well as thiosulfate [Breuer and Rumball, and rhodanine end-points detected 502
2007; Breuer et al., 2011; van der Merwe and 920 mg/L CN-, respectively.

1030
 Proceedings of 14th International Mineral Processing Symposium – Kuşadası, Turkey, 2014

3.2. Effect of Zinc

The effect of zinc on the determination of (a)
free cyanide by SNT was evaluated at
total cyanide concentrations of 0.02 M
(Figure 3a) and 0.03 M (Figure 3b). The
presence of zinc also led to over-
estimation of free cyanide (Figure 3a-b).
Similar to the case for copper (Eqs. 3-5),
ligand exchange can be assumed to occur
in that silver form complexes with
cyanide, which is supplied by zinc
cyanide complexes [Breuer et al., 2011].
Consistent with the findings in the
current study, the adverse effect of zinc (b)
on the analysis of free cyanide with SNT
was also reported by other researchers
[Breuer et al., 2011; van der Merwe and
Breuer, 2013].

Increasing the concentration of cyanide at

a constant CN/Zn ratio increased the
magnitude of over-estimation (Figure 3).
At a CN/Zn ratio of 10, rising the
concentration of cyanide from 0.02 M
(Figure 3a) to 0.03 M (Figure 3b) led to
41% and 111% higher measurements of Figure 3: Effect of zinc on SNT (a)[CN-]:
free cyanide, respectively, compared with 0.02 M and (b) [CN-]: 0.03 M (pH 12)
the predicted values.

(Figure 3) also reveals that SNT readings 3.3. Effect of Fe(II) and Fe(III)
were close to total cyanide levels (i.e. The influence of Fe(II) and Fe(III) on the
0.02 or 0.03 M) despite the presence of analysis of free cyanide by SNT was
zinc. It is relevant to note that pH tested at CN/Fe ratios of 6 and 12 (0.03
strongly affects the speciation of zinc at M CN-) (Figure 4). The presence of Fe(II)
cyanide solutions in that zinc is present and Fe(III) was found to induce higher
predominantly in the form of hydroxide free cyanide readings than the expected
complexes (i.e. Zn(OH)3- and Zn(OH)42-) values (Figure 4). Fe(III) interfered more
at ≥pH 12 [Medusa, 2010]. Similar to the severely than Fe(II). Iron interference can
findings in this study (Figure 3), Breuer be attributed to the precipitation of silver
et al. [2011] and van der Merwe and by ferrocyanide (Fe(CN)64-) and
Breuer [2013] reported accurate detection ferricyanide (Fe(CN)63-) as Ag4Fe(CN)6
of total cyanide in the presence of zinc (Ksp=1.6x10-41) (Eq. 6) and Ag3Fe(CN)6
using SNT with rhodanine or (Eq. 7), respectively [Marsden and
potentiometric end-point by adjusting pH House, 2006; Patnaik, 2004].
to ≥12. Breuer et al. [2011] stated that the
reliability of potentiometric SNT method 4Ag++Fe(CN)64-→Ag4Fe(CN)6 (6)
decreases in the presence of zinc at <pH
12 due to increased difficulty of detection 3Ag++Fe(CN)63-→Ag3Fe(CN)6 (7)
of the end-point.

1031
 (a)

(b) Figure 5: Effect of silver on SNT ([CN-]:

0.01 M, pH 12).

Figure 4: Effect of on SNT (a) Fe(II) and

(b) Fe(III) on SNT ([CN-]: 0.03 M, pH
12).
Figure 6: Effect of nickel on SNT ([CN-]:
0.03 M, pH 12).
3.4. Effect of Silver
The effect of silver was examined since it
is ubiquitously occurs in gold ores and
4. CONCLUSIONS
hence, present in cyanide leaching
The pregnant cyanide leach solutions of
solutions. (Figure 5) shows the influence
complex ores contain impurity metals
of CN/Ag ratio on SNT. Although an
that form cyanide complexes. A certain
over-detection of free cyanide was
level of free cyanide is required to be
observed at a CN/Ag ratio of 2, the
present in leach solution for efficient
magnitude of interference largely
leaching of gold. In most plants, silver
decreased at high CN/Ag ratios of ≥4.
nitrate titration (SNT) with rhodanine as
indicator is used for the monitoring of
3.5. Effect of Nickel
free cyanide. This study confirmed that
(Figure 6) illustrates that the influence of
the presence of metals may strongly
nickel (Ni2+) on analysis of free cyanide
interfere with SNT. The findings showed
with SNT is not significant at a CN/Ni
that the presence of copper, zinc and iron
ratio of ≥6. This can be ascribed to the
was found to lead to the over-estimation
relatively high stability of nickel-cyanide
of free cyanide particularly at low
complexes (Ni(CN)42-, logK=30.22,
CN/Metal ratios (i.e. high concentration
(Table 2) in that silver cannot extract
of metal). On the other hand, the
cyanide from such stronger complexes.

1032
 Proceedings of 14th International Mineral Processing Symposium – Kuşadası, Turkey, 2014

influence of silver and nickel were Mudder, T.I., Botz, M.M., 2001. The Chemistry
determined to be insignificant particularly and Treatment of Cyanidation Wastes, 2nd
ed. Mining Journal Books Ltd, London.
at CN/Metal ratios of ≥4. It can be Patnaik, P., 2004. Dean’s Analytical Chemistry
concluded from these findings that, in the Handbook, 2 edition. ed. McGraw-Hill
presence of impurities (particularly Cu, Professional, New York.
Zn and Fe), SNT with rhodanine Sceresini, B., 2005. Gold-copper ores, in: Mike D.
indicator does not give accurate results Adams and B.A. Wills (Ed.), Developments
in Mineral Processing. Elsevier, 789–824.
and SNT with rhodanine is not a reliable Van der Merwe, W., Breuer, P., 2013. Cyanide
method for following free cyanide level Analysis for Complex Cyanide Solutions.
in complex leach solutions. CIM J. 4, 120–128.
Xie, F., Dreisinger, D. and Doyle, F., 2013. A
Acknowledgements: The authors would Review on Recovery of Copper and Cyanide
From Waste Cyanide Solutions. Miner.
like to express their sincere thanks and Process. Extr. Metall. Rev. 34, 387–411.
appreciations to The Scientific and
Technological Research Council of
Turkey (TUBITAK) (Project no:
213M492) for the financial support.

REFERENCES
Bas, A.D., Yazici, E.Y. and Deveci, H., 2012a.
Treatment of a Copper- rich Gold Ore by
Ammonia Assisted Cyanide Leaching.
Presented at the XXVI International Mineral
Processing Congress (IMPC2012), New
Delhi, India, 24-28 September, 356–365.
Bas, A.D., Kucuk, A., Yazici, E.Y., Deveci, H.
2012b. Assessment of Ammoniacal
Ammonium Sulphate Leaching as a
Pretreatment Process for Copper Bearing
Gold Ores. In Özdağ H., Bozkurt V. and İpek
H. (Eds.), Proceedings of XIIIth International
Mineral Processing Symposium-IMPS 2012,
October 10-12, Bodrum, Turkey, 563-569.
Breuer, P.L., Rumball, J.A., 2007. Cyanide
Measurement and Control for Complex Ores
and Concentrates. Presented at the Ninth Mill
Operators’ Conference, Fremantle, WA, 19-
21 March, 249–254.
Breuer, P.L., Sutcliffe, C.A. and Meakin, R.L.,
2011. Cyanide Measurement by Silver
Nitrate Titration: Comparison of Rhodanine
and Potentiometric End-Points,
Hydrometallurgy, 106, 135–140.
Lu, J., Dreisinger, D. And Cooper, W., 2002.
Thermodynamics of the Aqueous Copper–
Cyanide System, Hydrometallurgy, 66, 23–
36.
Marsden, J., House, I., 2006. The Chemistry of
Gold Extraction. Society for Mining,
Metallurgy, and Exploration, USA.
Medusa, 2010. Software for Chemical
Equilibrium Diagrams, 32 bit version. Royal
Institute of Technology, Sweden.

1033
 PROCEEDINGS OF 14th INTERNATIONAL
MINERAL PROCESSING SYMPOSIUM

October 15 - 17, 2014

Kuşadası – Turkey

Mineral Processing in Everything !

Editors

Üner İPEKOĞLU, Vedat ARSLAN and Sezai ŞEN

Dokuz Eylül University, Department of Mining Engineering

Turkish Mining Development Foundation

View publication stats