Minerals Engineering 78 (2015) 83–88

Contents lists available at ScienceDirect

Minerals Engineering
journal homepage: www.elsevier.com/locate/mineng

A thermochemical study of thiol collector surface reactions on galena

and chalcopyrite
B. McFadzean a,⇑, K.P. Moller b, C.T. O’Connor a
a
Centre for Minerals Research, University of Cape Town, Private Bag X3, Rondebosch 7701, South Africa
b
Process Modelling and Optimisation, University of Cape Town, Private Bag X3, Rondebosch 7701, South Africa

a r t i c l e i n f o a b s t r a c t

Article history: The mechanism whereby thiol collectors adsorb onto a mineral surface has been widely studied. Most of
Received 20 January 2015 these studies examine the mineral surface at equilibrium state and, often, well above monolayer cover-
Revised 19 March 2015 age. However, the formation of species at the mineral surface is a complex process that is both kinetically
Accepted 10 April 2015
and thermodynamically controlled. In a typical industrial application, the dosage of collector will be
Available online 16 May 2015
around monolayer level. This study uses isothermal titration microcalorimetry to measure the heat
evolved during the reaction of sodium ethyl xanthate (SEX) and sodium isobutyl dithiocarbamate
Keywords:
(SIBDTC) and their mixtures with galena and chalcopyrite. The sensitivity of the instrument allows heats
Thiol collector
Surface chemistry
of reaction of sub-monolayer quantities of thiol collector to be measured.
Sulphide minerals The study found that the nature of the reaction between a collector and a mineral surface was depen-
Thermochemistry dent on the type of mineral and the type of collector. Proposals are made for the reactions based on pre-
viously reported results and the evidence from this investigation. It was found that when collectors
reacted with mixtures of galena and chalcopyrite, the magnitude of the reaction was lower compared
to that measured for the pure minerals alone. This may be due to galvanic interactions between the min-
erals, resulting in a passivating effect on one or both of the minerals.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction the only product that can be formed. . . when xanthate reacts with
(galena)’’. In a critical review of the literature, Finkelstein and
Microcalorimetry is an extremely sensitive surface science tool Poling (1977) reported that the oxidation of the thiol to the dithi-
that measures the heat evolved when a chemical reaction takes olate depended on the rest potential of the surface under the par-
place. It has experienced limited use in a mineral processing ticular conditions and it was concluded that it was not possible to
application despite widespread applications in pharmaceutical, form dixanthogen on a galena surface in the presence of dissolved
soil and environmental and catalysis research (e.g., Buckton oxygen. Chalcopyrite, on the other hand, was proposed to catalyse
(1995), Rong et al., (2007), Spiewak and Dumesic (1998)). the formation of dixanthogen, with the latter being considered the
Maier and Dobias (1997) and Maier et al. (1997) performed dominant surface species.
work using galena, chalcocite and sphalerite and various different Since sulphide minerals are electron conductors, when two sul-
collectors. The results reported did not indicate large differences in phide minerals come into contact with one another there will be an
the thermochemical behaviour of the collectors with galena and electrochemical interaction with a flow of electrons from the less
chalcocite at sub-monolayer coverage, but sphalerite behaved dif- cathodic mineral to a more cathodic mineral. The higher the rest
ferently. This may have been expected since it is well known that potential of the mineral, the more relatively cathodic is that min-
sphalerite requires an activator in order to promote collector eral (Nava and Gonzalez, 2006). In the case of chalcopyrite and
adsorption. Above monolayer dosages on galena, a second reaction galena, chalcopyrite has a higher rest potential than galena and
was reported to occur which they attributed to the formation of thus, galena will be oxidised i.e. be the electron donor (Rao,
dixanthogen even though Finkelstein and Poling (1977) proposed 2004). These electrons will reduce dissolved oxygen to hydroxide
that ‘‘there can no longer be any doubt that a lead xanthate is according to Eq. (1). The hydroxide may result in the formation
of metal hydroxides on the mineral surfaces.
⇑ Corresponding author. Tel.: +27 216505528. O2 þ 2H2 O þ 4e ! 4OH ð1Þ
E-mail address: [email protected] (B. McFadzean).

http://dx.doi.org/10.1016/j.mineng.2015.04.007
0892-6875/Ó 2015 Elsevier Ltd. All rights reserved.
84 B. McFadzean et al. / Minerals Engineering 78 (2015) 83–88

It has also been found that the galvanic interaction affected the surfaces. They found no precipitated colloidal PbX2 and theorised
floatability of the ‘‘noble’’ mineral, chalcopyrite, significantly with- that the oxidised surface prevented dissolution of the lead ions.
out having much effect on the floatability of the ‘‘active’’ mineral, With respect to chalcopyrite oxidation, the copper and iron ions
galena (Yelloji Rao and Natarajan, 1989). In a study of the galvanic in chalcopyrite have been found to oxidise at different rates
interactions between pyrite and pyrrhotite, Moslemi et al. (2012) (Buckley et al., 2014; Goh et al., 2006) with the iron oxidising faster
found that pyrite, having the highest rest potential, acted as the than the copper. Thus, it was found that iron oxide species were
cathode. They noted that the surfaces of cathodes were passivated more predominant in chalcopyrite than copper oxide species.
by oxide or hydroxide compounds, which are inherently hydrophi- It has been conceded in this and previous work (McFadzean and
lic, while the oxidation of the anode, pyrrhotite, resulted in a O’Connor, 2014) that the sulphide surfaces are likely oxidised.
hydrophobic surface. However, the extent of oxidation and the impact on the flotation
Majima (1969) found that galena, sphalerite and covellite oxi- performance remains debatable. In an extension of this work,
dised 8–20 times faster when pyrite (acting as the cathode) was which is being prepared for publication, the galena and chalcopy-
present. Building on this study, it was subsequently found that rite minerals were floated in a microfloation cell, with collectorless
the dissolution of metal ions increased for the anodic mineral recoveries for galena of 97% and chalcopyrite of 65%. This shows
and decreased for the cathodic mineral (Sui et al., 1995). These dis- that both minerals still have excellent natural hydrophobicity.
solved ions transferred between two minerals in contact, but the Previous work on samples prepared in a similar manner
larger proportion of Fe remained on the parent mineral (pyrite or (McFadzean and O’Connor, 2014) found large concentrations of
pyrrhotite) while 50% or more of the Zn (from sphalerite) and Pb lead ions in solution ([Pb]max of 9.13  105 mol/m2), which sug-
(from galena) transferred to pyrite. gests that the surface was not so oxidised that it prevented the dis-
The technique used in the present investigation has been previ- solution of the lead ions, as found in the study by O’Dea et al.
ously reported in a study of the reaction between galena and a (2001). It should be borne in mind that the microcalorimetry sam-
variety of collectors and collector mixtures (McFadzean and ples were sonicated prior to placement in the calorimeter, which
O’Connor, 2014). This paper extends those findings with a view has been shown to remove oxidation products from mineral sul-
to comparing the differences in the thermochemical behaviour phide surfaces (Clarke et al., 1995). In addition, consideration
between different sulphide minerals, viz. chalcopyrite and galena, should be given to the fact that different mineral processing plants
with different thiol collectors. The enthalpy spectra obtained for with different circuit configurations would exhibit a range of
each titration are also deconvoluted in order to attempt to distin- degrees of oxidation of the sulphide minerals. Introduction of
guish between the different reactions occurring between the min- HPGR (and other dry grinding techniques) has made it more
eral surface and the collector. important and relevant to know the effects of dry grinding on
flotation.

2. Methodology 2.2. Reagents

2.1. Mineral preparation The collectors used in this study were supplied by Senmin (Pty)
Ltd. Table 1 lists the names, abbreviations, molecular structures,
Pure galena and chalcopyrite mineral samples were obtained molecular weights and purities of the collectors. In all cases the
from Ward’s Natural Science Establishment, Rochester, NY. Each sodium form of the thiol was used. R in SIBDTC refers to isobutyl
sample was received in chunks of particles of up to 150 mm diam- and R0 refers to hydrogen. SEX was received in powder form.
eter. The mineral was hammered and then pulverised using a Sieb During manufacture, SIBDTC was stabilised in a high pH solution,
mill (Ferguson Industrial Group) for 15 s at a time. The pulverised which accounts for the low purity as shown in Table 1. Reagents
material was then passed through a 38 lm sieve and the undersize were freshly prepared for each experiment using deionised water
was used for the experiments. Samples were stored under nitrogen adjusted to pH 9 using NaOH.
at 30 °C to minimize mineral oxidation and were thawed and
dried in a desiccator under nitrogen for at least 24 h prior to use. 2.3. Microcalorimetry
BET analysis showed the surface area of the 38 lm galena sample
to be 0.2023 m2/g and the chalcopyrite sample to be 0.596 m2/g. A TAM III isothermal titration microcalorimeter manufactured
The presence of various surface species with respect to different by TA Instruments was used in heat flow mode. The instrument
sample preparation methods and the reaction of xanthate with has a liquid bath which is temperature controlled to within
these surface species has been the subject of much previous 0.0001 °C. The instrument was calibrated against a known electri-
research (Fornasiero et al., 1994; O’Dea et al., 2001; Dunn, 1997; cal power input. Three constants are generated by this calibration:
Fairthorne et al., 1997; Clarke et al., 1995). Fornasiero et al. (a) the gain factor, which is the relative deviation from a tempera-
(1994) found that, under oxidising conditions, the dissolution of ture-dependent calibration constant and (b) two time constants (s1
lead and sulphide ions, and their subsequent precipitation onto and s2) that correct for the differences in heat flow monitored by
the galena surface resulted in surface species of lead oxides such the heat detector and the rate of heat actually produced by the
as Pb(OH)2, PbS2O3, PbSO4 and PbCO3. In addition, XPS measure- sample due to the heat capacity and heat exchange coefficient of
ments found that the amount of oxygen on the galena surface the microcalorimeter. In addition to these calibrations a test
was approximately the same for galena that was ground dry in reaction was run using a ligand-binding reaction of Ba2+ with
air, compared to galena ground in an inert, wet environment and 18-crown-6 (1, 4, 7, 10, 13, 26-hexaoxacyclooctadecane). The
then conditioned with air at pH9 for two hours. Interestingly, there enthalpy of reaction ‘‘tentatively’’ ascribed to this reaction is
was a higher atomic percent of oxygen for samples that were con- DH = 31.42 kJ/mol (Wadso and Goldberg, 2001). In the
ditioned at pH9 with nitrogen for two hours than for both the sam- present study the values obtained for two titrations were
ples exposed to air. These findings point to the fact that the galena 32.10 kJ/mol and 33.38 kJ/mol, which yields an average of
surface could be expected to be relatively oxidised at the pH’s used (32.74 ± 0.91) kJ/mol.
in this study whether they were ground dry or wet. O’Dea et al. All reactions were performed isothermally at 25 °C.
(2001) found that oxidised galena surfaces had relatively low con- 0.2 ± 0.0005 g of mineral were weighed out, respectively, into the
centrations of adsorbed xanthate compared to freshly cleaved stainless steel sample and reference ampoules. 0.7 mL of pH9
 B. McFadzean et al. / Minerals Engineering 78 (2015) 83–88 85

Table 1
Collectors used in this study (sodium form).

Collector Abbreviation Molecular structure Molecular weight (g/mol) Purity (%)

Sodium ethyl xanthate SEX 144 99

Sodium isobutyl dithiocarbamate SIBDTC 199 31

deionised water was added to each ampoule which was sonicated 3. Results and discussion
for 30 s to disperse the mineral slurry and remove surface oxida-
tion products and the ampoules were then loaded into the Fig. 1 shows the heat evolved over time when SEX is contacted
microcalorimeter. The required concentration of collector was with finely ground galena (A) or chalcopyrite (B), or a mixture of
made up in pH9 deionised water and loaded into the titration syr- equal surface areas of galena and chalcopyrite (C). In the case of
inge while ensuring that no bubbles were present. Once the base- the SEX-galena reaction, each peak represents the addition of
line had stabilised, the experiment was started with 2 lL of 5.85  108 mol SEX, which represents the equivalent of approxi-
collector being titrated every 60 min. The amount of collector mately 0.25 of a monolayer as explained above. Since the different
was carefully calculated such that each injection represented experiments contain different surface areas of mineral, all
0.25 of a theoretical monolayer. This was based on the BET sur- microcalorimetric traces shown in this paper have had the heat
face area of the mineral and assuming that the cross-sectional area flow values normalised with respect to the SEX-galena system so
of the collector molecule is 28.8 Å2 (Grano et al., 1997). 20–25 that they can be compared to one another.
injections were performed, which resulted in a total theoretical The reaction of SEX with galena (A) shows two distinctive
monolayer coverage of approximately 4–6 theoretical monolayers sequential peaks. This is highly significant since such discrimina-
– the approximate surface coverage range of interest to commer- tion has not been previously observed in similar studies. A previous
cial processes. Blank experiments were performed to determine paper (McFadzean and O’Connor, 2014) showed that the addition of
the heat of dilution by titrating the same volumes of SEX or galena to a pH9 solution resulted in a decrease in pH and this was
SIBDTC used in the collector–mineral experiments into a pH9 solu- attributed to the uptake of hydroxide ions and the formation of oxi-
tion containing no mineral. The heat produced was negligible – dation products on the galena surface. These oxidation products are
two orders of magnitude less than what was measured in the case known to be particularly soluble compared to lead sulphide. In
of the collector experiments with mineral. addition, dissolution of lead ions in solution was measured and
found to reach a maximum of 9.13  105 mol/m2 of galena. This
was approximately 200 times in excess of the amount of xanthate
2.4. Deconvolution of microcalorimetry peaks available for complexation. Thus, the sequential peaks observed
in the microcalorimetry experiments were attributed to the reac-
Where it was suspected that a heat flow peak represented more tion of dissolved lead ions near the galena surface with SEX (Eq.
than one event or reaction occurring simultaneously or sequen- (2)) and subsequent precipitation of the lead xanthate complex
tially, a deconvolution of the microcalorimetric data was per- onto the galena (Eq. (3)).
formed. Each event was described by a gamma function and 2þ
Pb ðaqÞ þ 2EX ðaqÞ  PbðEXÞ2 ðaqÞ ð2Þ
since there were two events in each peak, each peak was repre-
sented by the sum of two 3-parameter gamma functions; PbðEXÞ2 ðaqÞ ! PbðEXÞ2 ðsurfÞ ð3Þ

a

ci bi i ðai 1Þ ðbi tÞ
f i ðt; ai ; bi ; ci Þ ¼ t e
Cðai Þ
response ðtÞ ¼ f 1 ðt; a1 ; b1 ; c1 Þ þ f 2 ðt; a2 ; b2 ; c2 Þ

where a is the shape parameter, b is the rate parameter and c is a

scaling factor for non-normalised data. The gamma function is par-
ticularly suitable since it allows non-symmetric peak fitting typical
of reactive events. The start and end of a peak was manually chosen
from the data file comprising a series of peaks generated by the
microcalorimetric experiment. Each peak was then baseline cor-
A
rected between the first point and the average of the last 10 points.
The double gamma function parameters for each peak were then B
estimated using a non-linear least squares regression solver (lsqr- C
solve, www.scilab.org). The parameters a, b and c were obtained
for each of the two events in a peak and since the area under the
non-scaled (c = 1) gamma function is unity, the value of the param-
eter c is the area under the curve and thus also the total heat Fig. 1. Microcalorimetry traces showing the heat flow evolved over time for the
released by the events in a peak. All results were systematically reaction of SEX with galena (A), chalcopyrite (B) and a mixture of equal surface
analysed in this way. areas of galena and chalcopyrite (C).
86 B. McFadzean et al. / Minerals Engineering 78 (2015) 83–88

It can be clearly seen from this figure that the heat evolution reaction occurs. Both the galena and chalcopyrite first reactions
curve for the reaction of SEX with chalcopyrite (B) is quite different have a relatively low enthalpy of reaction compared to the second
from that with galena (A). The initial fast reaction evolves far less reactions. This means that the second reactions are more exother-
heat for chalcopyrite compared to galena and the reaction contin- mic and evolve more heat than the first reactions.
ues evolving heat for almost the entire hour before the next injec- Fig. 4 shows the heat flow data for the reaction of SIBDTC with
tion. In line with the findings of Finkelstein and Poling (1977), it is galena, chalcopyrite and a mixture of galena and chalcopyrite. The
postulated that this heat evolution curve represents an initial rapid experiments were carried out in exactly the same way as the case
reaction of SEX with surface metal ions (MS(surf)) to form the metal where SEX was used as the collector. The reaction of SIBDTC with
thiolate (MX2(surf)), shown in Eq. (4), and a subsequent slower, sur- galena (A) is a rapid reaction with a relatively low molar enthalpy
face-catalysed oxidation of the metal thiolate to dixanthogen (X2), of adsorption (Table 2). The shape of the trace suggests that only a
shown in Eq. (5). single reaction is occurring, which is confirmed by the deconvolu-
tion procedure which shows no second reaction (Figs. 3 and 5).
MSðsurfÞ þ 2XðsurfÞ  MX2ðsurfÞ ð4Þ This suggests that different mechanisms are occurring in the
SEX-galena compared to the SIBDTC-galena system. It is proposed
MX2ðsurfÞ  M2þ
ðsurfÞ þ X2ðadsÞ þ 2e
ð5Þ
that SIBDTC reacts directly with the galena mineral surface (rather
An alternative reaction mechanism considered was the surface- than with dissolved ions in solution as was proposed for the SEX
catalysed oxidation of xanthate to dixanthogen in a first reaction system) up until the formation of the first pseudo-monolayer.
and the subsequent adsorption of the dixanthogen onto the min- This would be consistent with the difference in the chemical struc-
eral surface in a second reaction. However, the amount of heat tures of DTC and xanthate. The N atom in the DTC is relatively
evolved during the second reaction is greater than the amount of more positively inductive than the negatively charged and elec-
heat evolved in the first reaction (Fig. 3) and it is therefore unlikely tron-withdrawing O atom in the xanthates. This would result in
that the second reaction could be a physisorption, which would a stronger bond between the DTC and the sulphide surface since
release far less heat than a chemical reaction. electron density is pushed towards the reactive thiol head group,
Curve C in Fig. 1 shows the heat evolution curve of the SEX reac- allowing for greater electron orbital overlap and a stronger bond
tion with a mixture of galena and chalcopyrite. Different masses of (McMurray, 1996). The oxygen atom in the xanthate ion is more
galena and chalcopyrite were added to the reaction ampoule to electronegative than the nitrogen atom in the DTC ion, thus pulling
give a total of 0.06 m2 of mineral – 0.03 m2 each of chalcopyrite electron density away from the reactive head group and resulting
and galena. The aim of these experiments was to simulate the in a weaker bond for the xanthate ion. In addition, dissolution
behaviour of a real ore, which would contain a mixture of minerals. experiments showed only trace amounts of copper and iron in
In the case of curve C it is interesting to observe that the exotherm solution, but large quantities of lead in solution (200 times more
is not simply an additive of results for each individual mineral. The than the amount of xanthate in each injection) (McFadzean and
heat evolved per mole of SEX addition is substantially lower for the O’Connor, 2014).
mixture of chalcopyrite and galena (20 kJ/mol) than for either the The reaction of SIBDTC with chalcopyrite (B), on the other hand,
pure chalcopyrite (26 kJ/mol) or galena (53 kJ/mol) alone shows two distinct reactions occurring with a large combined
(Table 2). It is postulated that galvanic interactions between the enthalpy of reaction. A comparison of the SIBDTC-chalcopyrite
chalcopyrite and the galena may cause the passivation of the min- reaction with the SIBDTC-galena reaction shows that SIBDTC has
erals, perhaps through the deposition of an iron hydroxide layer a far higher enthalpy of adsorption on chalcopyrite (150 kJ/mol)
produced via the reduction of O2, which results in a diminished than on galena (35 kJ/mol) (Table 2). This is in line with the dif-
interaction between collector and mineral surfaces. ferent reaction mechanism suggested for chalcopyrite with the for-
In order to investigate these observations in greater detail a mation of the dithiolate, rather than the metal thiol complex.
deconvolution of each peak was carried out using the procedure Curve C (the mineral mixture) has a shape that is similar to the
described in the experimental methodology. Fig. 2 shows a typical SIBDTC-chalcopyrite system than the SIBDTC-galena system. The
example of such a procedure in the case of the SEX-chalcopyrite molar enthalpy of adsorption for one monolayer of SIBDTC is
system (A) and SIBDTC-chalcopyrite system (B). It is interesting shown in Table 2. This shows that the combined minerals have a
to observe that in all cases the deconvolution fitted remarkably heat of reaction between that of either of the pure minerals. This
well with only two peaks required. Fig. 3 shows the integrated val- indicates that the SIBDTC has a different mode of interaction with
ues for each pair of peaks for approximately the first monolayer for the combined minerals than does the SEX and this will be dis-
the systems shown in Figs. 1 and 4. These values essentially decon- cussed in more detail below.
volute the molar enthalpy values, although it is not possible to The deconvoluted data is presented in Fig. 3. This shows, firstly,
determine the heat evolved per mole since the number of moles that the amount of heat evolved for SIBDTC reactions is greater
associated with each deconvoluted peak is unknown. It can be seen than for SEX reactions in every case. This is in line with the argu-
that the second SEX-chalcopyrite reaction has the highest enthalpy ment above that SIBDTC is a stronger collector than SEX. Fig. 4
of reaction. This is due to the extended time over which the shows that there is only one reaction occurring in the SIBDTC-
galena system as discussed above. The energy of the second reac-
tion is zero, showing that no secondary reaction occurs. The reac-
Table 2
tion of SIBDTC with a mixture of chalcopyrite and galena was
Cumulative molar enthalpy of reaction for one monolayer collector coverage.
deconvoluted into two peaks and therefore suggests that the
Curve Collector–mineral Cumulative molar enthalpy at one mechanism is the same as for SIBDTC on chalcopyrite (metal thiol
system monolayer collector coverage (kJ/mol)
formation, followed by oxidation to the dixanthogen). However,
A SEX galena 53 the amount of heat evolved was far less for the mineral mixture
B SEX chalcopyrite 26 (66 kJ/mol) than for chalcopyrite (150 kJ/mol), which reinforces
C SEX 20
chalcopyrite + galena
the proposal that the galvanic interaction between galena and
A SIBDTC galena 35 chalcopyrite may have a passivating effect on the surface reaction.
B SIBDTC chalcopyrite 150 It is well known that collector mixtures are widely used in
C sibdtc 66 industry and Fig. 5 shows the interaction of collector mixtures with
chalcopyrite + galena
galena and with a galena–chalcopyrite mixture. In addition, for
 B. McFadzean et al. / Minerals Engineering 78 (2015) 83–88 87

Fig. 2. Deconvolution of microcalorimetric spectra for SEX-chalcopyrite (a) and SIBDTC-chalcopyrite (b).

3.5 (a) SEX Galena Peak 1 (b)

SEX Galena Peak 2
Normalised energy of reacon (mJ)

3.0 SEX Chcpte Peak 1

SEX Chpte Peak 2
SEX Galena Chcpte Peak 1
2.5 SEX Galena Chcpte Peak 2

2.0

1.5

1.0

0.5

0.0
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Collector psuedo-monolayer coverage

Fig. 3. Energy of reaction for deconvoluted peaks 1 and 2 of the same reactions shown in Fig. 1(a) and 4(b). Lines are a guide to the eye only. Note the increase in y-axis scale
for (b).

A
A
B
C

B
D
C
E

Fig. 4. Microcalorimetry traces showing the heat flow evolved over time for the Fig. 5. Microcalorimetry traces showing the heat flow evolved over time for the
reaction of SIBDTC with galena (A), chalcopyrite (B) and a mixture of equal surface reaction of SEX-galena (A), SEX-chalcopyrite (B) SEX-galena + chalcopyrite (C),
areas of galena and chalcopyrite (C). SEX + SIBDTC-galena (D), SEX + SIBDTC-galena + chalcopyrite (E).

comparison purposes, Fig. 5 also includes the reactions of SEX with spectrum observed is very similar to that observed when pure
galena (A), chalcopyrite (B) and galena–chalcopyrite mixture (C). SIBDTC was used (curve A, Fig. 3). Once chalcopyrite is added to
Curve D shows a mixture of SEX and SIBDTC reacting with galena. the galena, the enthalpy of reaction is reduced and once again sug-
This shows that the kinetics of the reaction is rapid and the gests that there is a galvanic interaction occurring between the two
88 B. McFadzean et al. / Minerals Engineering 78 (2015) 83–88

minerals which passivates the surface. The deconvolution of the References

peaks for both D and E show no evidence of secondary reactions
occurring. It is therefore postulated that the SIBDTC reaction is Buckley, A.N., Hope, G.A., Parker, G.K., 2014. A spectroscopic investigation of the
interaction of n-octanohydroxamate collector with Cu–Fe sulfide minerals.
the overriding reaction and there is little evidence of the SEX reac- Miner. Eng. 64, 23–37.
tion – particularly the secondary reaction. Clarke, P., Fornasiero, D., Ralston, J., St, C., Smart, R., 1995. A study of the removal of
oxidation products from sulfide mineral surfaces. Miner. Eng. 8 (11), 1347–
1357.
Dunn, J.G., 1997. The oxidation of sulphide minerals. Thermochim. Acta 300, 127–
4. Conclusion 139.
Fairthorne, G., Fornasiero, D., Ralston, J., 1997. Effect of oxidation on the
The nature of the reactions was different for SEX on galena and collectorless flotation of chalcopyrite. Int. J. Miner. Process. 49, 31–48.
Finkelstein, N.P., Poling, G.W., 1977. The role of dithiolates in the flotation of
chalcopyrite. This suggests that different reaction mechanisms
sulphide minerals. Miner. Sci. Eng. 9 (4), 177–196.
were occurring. SEX on galena shows two sequential reactions. Fornasiero, D., Li, F., Ralston, J., St, C., Smart, R., 1994. Oxidation of galena surfaces. J.
Due to changes in pH, dissolution of lead from galena and based Colloid Interface Sci. 164, 333–344.
on previous findings (McFadzean and O’Connor, 2014), the reaction Goh, S.W., Buckley, A.N., Lamb, R.N., Rosenberg, R.A., Moran, D., 2006. The oxidation
states of copper and iron in mineral sulfides and the oxides formed on initial
mechanism proposed for the SEX-galena system is: exposure of chalcopyrite and bornite to air. Geochim. Cosmochim. Acta 70,
2210–2228.
2þ
Pb ðaqÞ þ 2EX ðaqÞ  PbðEXÞ2 ðaqÞ ð6Þ Grano, S., Prestidge, C.A., Ralston, J., 1997. Solution interaction of ethyl xanthate and
sulphite and its effect on galena flotation and xanthate adsorption. Int. J. Miner.
PbðEXÞ2 ðaqÞ ! PbðEXÞ2 ðsurfÞ ð7Þ Process. 52, 161–186.
Maier, G., Dobias, B., 1997. 2-Mercaptobenzothiazole and derivatives in the flotation
The exotherm of SEX-chalcopyrite could be deconvoluted to of galena, chalcocite and sphalerite: a study of flotation adsorption and
two peaks, indicating two reactions occurring. In line with propos- microcalorimetry. Miner. Eng. 10 (12), 1375–1393.
Maier, G., Qiu, X., Dobias, B., 1997. New collectors in the flotation of sulphide
als by Finkelstein and Poling (1977) and due to the fact that very minerals: a study of the electrokinetic, calorimetric and flotation properties of
little copper or iron dissolved into solution, the following reaction sphalerite, galena and chalcocite. Colloids Surf. A 122, 207–225.
mechanism was proposed: Majima, H., 1969. How oxidation affects selective flotation of complex sulphide
ores. Can. Metall. Q. 8, 269–273.
McFadzean, B., O’Connor, C.T., 2014. A thermochemical study of thiol collector
MSðsurfÞ þ 2XðsurfÞ  MX2ðsurfÞ ð8Þ surface reactions on galena. Miner. Eng. 65, 54–60.
McMurray, J., 1996. Organic Chemistry, fourth ed. Brooks/Cole Publishing
MX2ðsurfÞ  M2þ
ðsurfÞ

þ X2ðadsÞ þ 2e ð9Þ Company, New York.
Moslemi, H., Shamsi, P., Alimohammady, M., 2012. Electrochemical properties of
SIBDTC reacted differently to SEX on both galena and chalcopy- pyrite, pyrrhotite and steel: effects on grinding and flotation processes. J.
rite. A direct reaction mechanism of SIBDTC onto the galena surface Southern African Inst. Mining Metall. 112, 883–890.
Nava, D., Gonzalez, I., 2006. The influence of galena galvanic interaction on the
was proposed since only one reaction peak was observed and due reactivity of chalcopyrite flotation concentrate. Electrochem. Soc. Trans. 2 (3),
to the fact that SIBDTC is a stronger collector than SEX. SIBDTC- 255–264.
chalcopyrite deconvolution indicated that two reactions were O’Dea, A.R., Prince, K.E., St, C., Smart, R., Gerson, A.R., 2001. Secondary ion mass
spectrometry investigation of the interaction of xanthate with galena. Int. J.
occurring and a similar reaction mechanism to Eqs. (8) and (9)
Miner. Process. 61, 121–143.
were proposed. Mixtures of chalcopyrite and galena resulted in a Buckton, G., 1995. Applications of isothermal microcalorimetry in the
lower heat of reaction than for either of the pure minerals and it pharmaceutical sciences. Thermochim. Acta 248, 117–129.
Rao, S.R., 2004. Surface Chemistry of Froth Flotation, second ed. Kluwer Academic
was suggested that the galvanic interaction of chalcopyrite and
Publishers, New York.
galena resulted in the passivation of the surface and a reduction Rong, X.M., Huang, Q.Y., Jiang, D.H., Cai, P., Liang, W., 2007. Isothermal
in the extent of the collector–mineral reaction. Mixtures of collec- microcalorimetry: a review of applications in soil and environmental
tors, SIBDTC and SEX, resulted in reactions with more of the char- sciences. Pedosphere 17 (2), 137–145.
Spiewak, B.E., Dumesic, J.A., 1998. Applications of adsorption microcalorimetry for
acter of the stronger collector (SIBDTC) since they exhibited only a the characterisation of metal-based catalysts. Thermochim. Acta 312 (1–2), 95–
single reaction peak. However, the extent of the reaction was 104.
reduced when a collector mixture was used. Sui, C.C., Brienne, S.H.R., Ramachandra Rao, S., Xu, Z., Finch, J.A., 1995. Metal ion
production and transfer between sulphide mineral. Miner. Eng. 8 (12), 1523–
1539.
Wadso, I., Goldberg, R.N., 2001. Standards in isothermal microcalorimetry (IUPAC
Acknowledgements Technical Report). Pure Appl. Chem. 73 (10), 1625–1639.
Yelloji Rao, M.K., Natarajan, K.A., 1989. Effect of galvanic interaction between
The authors wish to acknowledge funding from Anglo American grinding media and minerals on sphalerite flotation. Int. J. Miner. Process. 27,
95–109.
Platinum, Impala Platinum, Lonmin, Nkomati, the University of
Cape Town and the National Research Foundation.