Mechanism of the cathodic process coupled to the oxidation of iron monosulfide by dissolved oxygen

Journal of Colloid and Interface Science 467 (2016) 51–59

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science journal homepage: www.elsevier.com/locate/jcis

Mechanism of the cathodic process coupled to the oxidation of iron monosulfide by dissolved oxygen Ma˘da˘lina I. Duinea a, Andreea Costas b, Mihaela Baibarac b, Paul Chirița˘ a,⇑ a b

University of Craiova, Department of Chemistry, Calea Bucuresti 107I, Craiova 200478, Romania National Institute of Materials Physics, Laboratory of Optical Processes in Nanostructured Materials, P.O. Box MG-7, Bucharest R077125, Romania

g r a p h i c a l a b s t r a c t

a r t i c l e

i n f o

Article history: Received 20 October 2015 Revised 5 January 2016 Accepted 5 January 2016 Available online 6 January 2016 Keywords: FeS oxidation FeS/water interface EIS SEM/EDX Raman spectroscopy

a b s t r a c t This study investigated the mechanism of iron monosulfide (FeS) oxidation by dissolved oxygen (O2(aq)). Synthetic FeS was reacted with O2(aq) for 6 days and at 25 °C. We have characterized the initial and reacted FeS surface using Scanning Electron Microscopy coupled with Energy Dispersive X-ray (SEM/ EDX) analysis, Raman spectroscopy and Fourier transform infrared spectroscopy (FTIR). It was found that during the aqueous oxidation of FeS new solid phases (disulfide, polysulfide, elemental sulfur, ferric oxyhydroxides and Fe3O4) develop on the mineral surface. The results of potentiodynamic polarization experiments show that after 2 days of FeS electrode immersion in oxygen bearing solution (OBS) at initial pH 5.1 and 25 °C the modulus of cathodic Tafel slopes dramatically decreases, from 393 mV/dec to 86 mV/dec. This decrease is ascribed to the change of the mechanism of electron transfer from cathodic sites to O2 (mechanism of cathodic process). The oxidation current densities (jox) indicate that mineral oxidative dissolution is not inhibited by pH increase up to 6.7. Another conclusion, which emerges from the analysis of jox, is that the dissolved Fe3+ does not intermediate the aqueous oxidation of FeS. The results of electrochemical impedance spectroscopy (EIS) show that after 2 days of contact between electrode and OBS the properties of FeS/water interface change. From the analysis of the EIS, FTIR spectroscopy, Raman spectroscopy and SEM/EDX data we can conclude that the change of FeS/water interface properties accompanies the formation of new solid phases on the mineral surface. The new characteristics of the surface layer and FeS/water interface do not cause the inhibition of mineral oxidation. Ó 2016 Elsevier Inc. All rights reserved.

⇑ Corresponding author. E-mail address: [email protected] (P. Chirița˘). http://dx.doi.org/10.1016/j.jcis.2016.01.010 0021-9797/Ó 2016 Elsevier Inc. All rights reserved.

52

M.I. Duinea et al. / Journal of Colloid and Interface Science 467 (2016) 51–59

1. Introduction Iron monosulfide minerals (pyrrhotite, troilite and mackinawite) are present in various geological environments [1]. Also, iron monosulfide phases (FeS) can be present on the iron/steel surface as result of the reaction between H2S (product of sulfate reduction [2]) and Fe2+ (product of iron/steel corrosion [3]). By its products of oxidation (ferric iron and protons), FeS oxidation can produce severe environmental problems [4,5]. In acidic media, the ferric iron is an effective oxidant which dissolves new amounts of FeS and other mineral sulfides [6] releasing toxic species such as Cu, Cd, Hg, Pb or As. Since the oxidation of FeS phases is fast (much faster than that of pyrite [7]), the understanding of the mechanism of aqueous oxidation of FeS is very important. Badica and Chirita [8] have shown that the aqueous oxidation of troilite by dissolved oxygen (O2(aq)) occurs after an electrochemical mechanism. The oxidative dissolution of FeS in the presence of oxygen (the most common oxidant at the Earth’s surface) generates a surface layer which incorporate ferric iron, oxygen, disulfide, polysulfide and elemental sulfur [4,9]. Also, FeS oxidation produces soluble species such as ferrous iron and sulfate:

FeS þ 1=2O2 þ 2Hþ ¼ Fe2þ þ S þ H2 O

ð1Þ

FeS þ 2O2 ¼ Fe2þ þ SO2 4

ð2Þ

The aqueous oxidation of FeS is always accompanied by the non-oxidative dissolution, which is faster [4,10]:

FeS þ 2Hþ ¼ Fe2þ þ H2 S

ð3Þ

The non-oxidative dissolution is an acid-consuming process. The released Fe2+ is oxidized by O2(aq) to Fe(III). If the reaction occurs at acidic pH, the soluble Fe(III) (Fe3+) will oxidize FeS producing Fe2+ [11,12]:

FeS þ 2Fe3þ ¼ 3Fe2þ þ S

ð4Þ

þ FeS þ 8Fe3þ þ 4H2 O ¼ 9Fe2þ þ SO2 4 þ 8H

ð5Þ

As in the case of pyrite [13], it can appear a cycle in which the Fe2+ is oxidized to Fe3+, and the Fe3+ subsequently oxidizes the FeS and produces Fe2+. Fe2+/Fe3+ cycle can be interrupted at pH > 4 [4] (in many cases, the pH of oxidizing solutions) by the precipitation of Fe3+

Fe3þ þ 2H2 O ¼ FeOOH þ 3Hþ

ð6Þ

Fe3þ þ 3H2 O ¼ FeðOHÞ3 þ 3Hþ

ð7Þ

The reactions (6) and (7) are acid-producing processes. The precipitation of Fe3+ can affect the oxidation of FeS by, on the one hand, the decrease of Fe3+ concentration [12], and, on the other hand, the hindering of the access of O2(aq) to mineral surface [14]. Because the great majority of the experimental studies were performed in acidic media [4,15], the details of the reaction mechanism (the rate determining step(s), intermediates, reaction products, etc.) of FeS oxidation with O2(aq) at high pH in aqueous media have not yet established. In this paper, the aqueous oxidation of FeS in oxygen bearing solution (OBS) at initial pH 5.1 and 25 °C was studied by various techniques. At pH 5.1 the ferric iron solubility is very low (1011 M) and it continues to decreases when pH increases up to 8 (1012.26 M) [16]. Hence, in this pH range, Fe3+ cannot be the oxidant of FeS. To observe any behavioral changes, the aqueous oxidation of FeS was monitored by means of electrochemical techniques (potentiodynamic polarization and electrochemical impedance spectroscopy (EIS)) over a period of 6 days of FeS electrode immersion in OBS. The reaction products were analyzed using

chemical methods, Scanning Electron Microscopy coupled with Energy Dispersive X-ray (SEM/EDX) analysis, Raman spectroscopy and Fourier transform infrared spectroscopy (FTIR). For comparison, we carried out a series of electrochemical experiments with pyrite as electrode material. 2. Experimental 2.1. Materials The characteristics of synthetic FeS (96 wt.% of troilite and 4 wt. % of elemental iron) and natural FeS2 (pyrite) used in this study have been previously described in detail elsewhere [8,17–19]. All chemicals were of analytical grade and all solutions were prepared with distilled water. 2.2. Electrochemical experiments The electrochemical measurements were performed in a standard three-electrode cell with an electrochemical workstation Zahner Elektrik IM6e. The working electrode (WE) was prepared by encapsulation of FeS (or FeS2) in epoxy resin. Only one side, with the effective area of 0.5 cm2 (or 1 cm2 for the pyrite electrode) was exposed. It has been polished with 600, 2000 and 3000 grade silicon carbide paper. After polishing the electrode surface was rinsed with distilled water and acetone. The counter electrode was platinum foil, while reference electrode was saturated calomel electrode (SCE). The working electrode was immersed in OBS at initial pH 5.10 and 25 °C. The initial pH was adjusted to 5.10 by addition of diluted HCl, without any other background electrolyte. A constant flow of air was bubbled continuously through the solution. The initial pH and pH before each electrochemical measurement was measured with a combined glass electrode (Consort). Before each measurement the pH electrode was calibrated against two commercial pH buffers (pH 4.01 and pH 7.00). Potentiodynamic polarization measurements were carried out with a scan rate of 1 mV/s in the potential range from 250 to +250 mV relative to the open circuit potential (OCP). Before the polarization measurement corresponding to 0 days, the electrode was allowed to equilibrate with the oxygen bearing solution for 40 min. The electrochemical parameters (oxidation potential (Eox), oxidation current density (jox), anodic and cathodic Tafel slopes (ba and bc)) were determined by analyzing the polarization data with Thales 3.16 software. EIS measurements were carried out at OCP over a frequency range of 3 MHz–10 mHz with a signal amplitude perturbation of 10 mV. The impedance data were validated using Z-HIT transform test (Fig. S1 of the Supplementary material) [8,17] and then fitted with an equivalent electric circuit using Thales software. 2.3. Aqueous batch experiment In order to characterize the solid layer developed on the FeS surface during its aqueous oxidation the unreacted FeS powder and the residual solid resulted from an aqueous batch experiment were analyzed by FTIR spectroscopy. FeS powder was prepared by crushing in an agate mortar the same material used for the construction of WE. The specific surface area of the FeS powder was determined using BET method with a Micromeritics TriStar 3000 equipment and found to be 2.02 m2/g. The aqueous batch experiment was performed by the suspension of 1 g FeS powder in 0.5 L oxygen bearing HCl solution (initial pH 5.10) at 25 °C. The experiment lasted for 6 days, and a constant flow of air was bubbled continuously through the solution. At the end of aqueous batch experiments aliquots of suspension were removed with a

53

M.I. Duinea et al. / Journal of Colloid and Interface Science 467 (2016) 51–59

syringe connected to 0.2 lm filter. The filtrates were analyzed for total dissolved iron (Fetotal) and sulfate (SO2 4 ). The concentration of total dissolved iron ([Fetotal]) was determined (after the reduction of aqueous Fe(III) by a solution of 10% hydroxylamine) by spectrophotometry (PG Instruments T70-UV–vis spectrophotometer) using the 2,20 -dipirydyl method at 522 nm [20]. The concentration of sulfate ([SO2 4 ]) was analyzed by turbidimetry at 420 nm [21].

Table 1 Time dependencies of electrochemical parameters obtained by Tafel polarization for FeS (or FeS2) electrodes in OBS at 25 °C and different pH values. The error for the jox ranged from 2% to 8%, being the lowest for the FeS2. For error evaluation, at least four independent experiments were performed for each electrode, and at each initial pH.

2.4. SEM/EDX analysis The morphology and composition of initial FeS surface (before its immersion in OBS) and FeS surface oxidized for 6 days in OBS were investigated using a Tescan LYRA3 XMU Scanning Electron Microscope (SEM) with Schottky field emission cathode having Energy Dispersive X-ray (EDX) analysis Bruker Quantax 200, Peltier cooled X-ray detector as accessory. 2.5. Raman spectroscopy Raman spectra were recorded using a T64000 Raman spectrophotometer from Horiba Jobin Yvon endowed with a Ar laser. The resolution of Raman spectra was of 1 cm1. 2.6. FTIR measurements The FTIR spectra were recorded with a Bruker Alpha spectrometer using KBr technique. They were collected in the range between 375 and 4000 cm1, with a resolution of 4 cm1, and are the average of 64 scans. In order to limit the sample oxidation, the FTIR spectra were recorded immediately after pellet preparation. 3. Results and discussion 3.1. Potentiodynamic polarization measurements The potentiodynamic polarization curves of FeS electrode at different times of contact between mineral and OBS are presented in Fig. 1. The calculated electrochemical parameters are listed in Table 1. One can be seen that after 2 days of contact between FeS and OBS the Eox shifts to positive direction. Initial Eox is 336.7 mV, and after 2 days of immersion of FeS electrode in OBS it increases up to 197.9 mV. After 6 days of contact between FeS and OBS the Eox becomes 256.8 mV. The increase of the oxidation current densities (jox) after 2 days of FeS electrode immersion in OBS indicates that the FeS oxidation is not inhibited by the prolonged contact of mineral with OBS. The anodic Tafel slopes (ba) show a moderate oscillation between 287 mV/dec and

a b c

Time (days)

Electrode

pH

Eox (V)

jox (lA cm2)

ba (V dec1)

bc (V dec1)

0a 0a 2 4 6 0a 2 4 9

FeS FeS FeS FeS FeS FeS2 FeS2 FeS2 FeS2

5.10b 6.50b 6.41c 6.30c 6.70c 5.10b 5.15c 5.11c 5.17c

0.337 0.316 0.198 0.250 0.257 0.216 0.267 0.270 0.256

16.8 15.8 20.4 24.0 26.8 0.283 0.389 0.298 0.468

0.359 0.287 0.397 0.344 0.396 0.310 0.300 0.277 0.316

0.393 0.332 0.086 0.084 0.080 0.115 0.124 0.113 0.126

Immediately (40 min) after immersion of FeS (or FeS2) electrodes in OBS. Initial pH. pH before the electrochemical measurements.

397 mV/dec. Very interestingly, the modulus of cathodic Tafel slopes (bc) dramatically decreases after 2 days of contact between FeS and OBS, from 393 mV/dec to 86 mV/dec. The trend of Tafel slopes indicates that, on the one hand, the mechanism of anodic process (the oxidation of FeS) is not affected by the 6 days of contact between FeS electrode and OBS and, on the other hand, the mechanism of cathodic process (the reduction of O2) occurring on the FeS electrode changes after 2 days of WE immersion in OBS. The decrease of the modulus of bc indicates that the product acn (the only variable in bc, part of its denominator) increases [3,22]. ac is the cathodic charge transfer coefficient and n is the number of the electrons transferred during the rate determining step of cathodic process. The increase of acn means that after 2 days of reaction either the activated complex adopts predominantly the structure of the reduced species (ac shifts toward 1) [3], the rate determining step of cathodic process involves the transfer of more electrons than immediately after the contact with OBS [22] or both. In order to assess the role of pH increase during the oxidative dissolution experiments (Table 1), we have compared the electrochemical parameters determined for FeS electrode immediately (40 min) after its immersion in aerated solutions (25 °C) at initial pH 5.1 and, respectively, initial pH 6.5 (Table 1). The initial pH was adjusted to 6.50 by addition of diluted HCl and NaOH solutions. Although there are some differences between the cathodic Tafel slopes obtained for polarized FeS electrode immediately after its immersion in OBS at pH 5.1 and 6.5, respectively, these differences are not comparable with those registered between the cathodic Tafel slopes determined for the polarized FeS electrode immediately (40 min) after its immersion in OBS and, respectively, after

Fig. 1. Potentiodynamic polarization curves recorded at (a) initial pH 5.10 and different times of contact between FeS electrode and OBS and (b) initial pH 5.10 and 6.50, respectively, immediately after FeS electrode immersion in OBS. SHE = Standard hydrogen electrode.

54

M.I. Duinea et al. / Journal of Colloid and Interface Science 467 (2016) 51–59

obtained for FeS electrode after 2 days of contact with OBS (Table 1).

3.2. EIS measurements

Fig. 2. Potentiodynamic polarization curves recorded at initial pH 5.10 and different times of contact between FeS2 and OBS. SHE = Standard hydrogen electrode.

at least 2 days of immersion in OBS. These findings indicate that the pH does not cause the change of the mechanism of cathodic process occurring on the mineral surface. It is important to mention that, the behavior of FeS electrode was not observed for pyrite electrode. Although the cathodic current densities measured for FeS2 electrode increase after 2 days of immersion in OBS (Fig. 2) the anodic and cathodic Tafel slopes show only a minor variation during the 9 days of reaction and indicate that the mechanism of FeS2 interaction with O2(aq) does not change. Also, if we take into consideration the jox values, our results confirm that the reactivity of FeS is higher than that of pyrite. It is interesting to stress that the bc values obtained for FeS2 electrode are close to the values

The Nyquist plots of the FeS electrode in aerated HCl solution at OCP and 25 °C are shown in Fig. 3. As we can easily observe the impedance behavior of FeS electrode immediately after its immersion in OBS (Fig. 3a) is very different from the impedance behavior registered after 2, 4 and 6 days of contact between FeS and OBS (Fig. 3b–d). The observed differences show that the interfacial properties of FeS electrode after 2, 4 and 6 days of contact with OBS are changed relative to interfacial properties of the FeS electrode immediately (40 min) after its immersion in OBS. These differences can be an explanation for the observed change of the mechanism of electron transfer from cathodic sites on FeS surface to O2. The electrical circuit used to fit the EIS data for the oxidation process of FeS electrode immediately (40 min) after its immersion in OBS is shown in Fig. 4a. This circuit is a slight variation of the one used in a previous study by members of our group [8] and was chosen because it is in accordance with the general mechanism of FeS oxidative dissolution (which involves a mix of surface reaction and diffusion [8,17,23]) and give the best statistically fit of the experimental data. The equivalent circuit shown in Fig. 4b was used to fit the impedance data for the oxidation process of FeS electrode after 2 days of immersion in OBS. This circuit is also a modification of the one used by Badica and Chirita [8] to characterize FeS/water interface. It gives the best statistically fits of the experimental data obtained after 2 days of FeS immersion in OBS. The differences between equivalent electric circuit used by Badica and Chirita [8] and the circuits of Fig. 4 can be explained by the

Fig. 3. Nyquist plots (10 mHz–3 MHz) for FeS electrode (a) immediately (40 min) after contact with OBS and, respectively, after (b) 2, (c) 4 and (d) 6 days in contact with OBS. More details regarding the fitting procedure can be found in the Supplementary material. h Measure samples and s fitting (simulated data).

55

M.I. Duinea et al. / Journal of Colloid and Interface Science 467 (2016) 51–59

trolled by diffusion process across the surface layer [8]. After 2 days of reaction of FeS with OBS the rate determining step of the mineral oxidation becomes the charge transfer process (Rct is higher with approximately one order of magnitude than Rsl, and a Warburg diffusion element is associated to Rct (Fig. 4b)). The Warburg element can be correlated with low frequencies diffusion process of electroactive species to the mineral interface from solution [26,27]. In comparison with Cdl, Qdl decreases with approximately two orders of magnitude (by between 46 and 67 times) after 2 days of FeS electrode immersion in OBS. The situation is reversed in the case of Qsl. It increases with approximately one order of magnitude (by between 4 and 12 times) in comparison with Csl. These findings indicate important modifications of the characteristics (area of the plates, distance between plates and/or the composition of the dielectric medium) of the corresponding capacitors. 3.3. Reaction products Fig. 4. Equivalent electric circuits used to fit EIS data of FeS electrode (a) immediately (40 min) after its immersion in OBS and (b) after at least 2 days of immersion in oxidizing solution.

low ionic strength of the initial solution with pH 5.1 or by the oxidation of the electrode surface after 2 days of contact with OBS. Rs represents the solution resistance, Rsl is the resistive component of the mass transport through the surface layer, Rct is the charge transfer resistance, Csl is the capacitance exerted by the surface layer and Cdl is the double layer capacitance. For better fitting results, in the case of the experimental data for the oxidation process of FeS electrode after 2 days of immersion in OBS Csl and Cdl were substituted with the constant phase elements CPEsl and CPEdl, respectively. Cdl and CPEdl reveal the existence of a capacitive loop at low frequencies while Csl and CPEsl can be associated to a capacitive loop at high frequencies. Both equivalent circuits are completed with an inductive component L. This inductance can be assigned either to the increase of mineral dissolution rate as result of surface layer dissolution or to adsorption processes produced on the electrode surface [8]. The equivalent circuit in Fig. 4b also includes a Warburg element. The impedance parameters obtained from the fitting of the experimental data using the equivalent electric circuits of Fig. 4 are summarized in Table 2. In Table 2 Qsl and Qdl are the parameters of CPEsl and CPEdl, and nsl and ndl are the corresponding constant phase element exponents [24]. nsl and ndl could be attributed to non-uniform current distribution and surface roughness [25]. The high value of Rsl indicates that a surface layer is initially present on FeS surface and is formed during the polishing of the electrode. Because the initial Rsl is higher with approximately two orders of magnitude than Rct it results that immediately (40 min) after immersion of FeS electrode in OBS the overall oxidative dissolution of FeS is con-

3.3.1. SEM/EDX analysis Another explanation for the observed change of the cathodic slopes after 2 days of FeS electrode immersion in OBS may reside in the formation of new solid phases on the mineral surface. Analyzing the SEM/EDX data of initial FeS surface (before its immersion in OBS) (Figs. 5a and S2a of the Supplementary material) and FeS surface oxidized for 6 days in OBS (Figs. 5b, c and S2b of the Supplementary material) one can observe that the main components of initial surface are iron and sulfur, and the outermost layer developed on the FeS surface mainly incorporates oxygen and iron. This layer formed on the electrode surface is heterogeneous (Figs. 5b and c). The acicular morphology which can be seen in Figs. 5b and c is typical of goethite (a-FeOOH) [28]. 3.3.2. Raman spectroscopy Additional information concerning the layer developed on the FeS surface are shown in Fig. 6. Raman spectrum of the FeS electrode shows: (i) in the low frequencies spectral range two lines of very low intensity situated at 212 and 276 cm1, that are assigned to the asymmetric and symmetric stretching vibrational modes of FeS, respectively [29], and (ii) in the high frequencies spectral range a Raman line of very high intensity peaked at 3079 cm1, which corresponds to the OH stretching vibrational mode of water and hydroxyl groups [30]. A characteristic of the FeS electrode in the initial state is the value of the ratio between the relative intensities of the Raman lines situated in the high frequencies spectral range and low frequencies spectral range, equal with 12. According with Fig. 6, a new Raman line in the low frequencies spectral range is observed at 661 cm1, which is not very long situated of Raman line of Fe3O4 (665 cm1) [31]. In this last case, the value of the ratio between the relative intensities of the

Table 2 Time dependencies of impedance parameters for FeS immersed in OBS at 25 °C and different pH values. The error is lower than 8% for the impedance parameters obtained after 2 days of immersion in OBS and 33% for the parameters derived immediately (40 min) after FeS electrode immersion in OBS.

a b c

Time (days)

pH

0a 2 4 6

5.10b 6.41c 6.30c 6.70c

Rct (kX cm2)

Cdl (mF cm2)

Rsl (kX cm2)

Csl (pF cm2)

CPEdl Qdl (lF cm2)

ndl

0.07 0.66 0.49 0.44

22.4 – – –

4.0 – – –

367.8 – – –

– 334.8 484 428.8

– 0.719 0.728 0.726

Immediately (40 min) after immersion of FeS electrode in OBS. Initial pH. pH before the electrochemical measurements.

DW (cm2)

L (lH cm2)

Rsl (X cm2)

CPEsl Qsl (nF cm2)

nsl

– 568 414 514.8

3400 32 14.9 10

– 56.5 54 41.5

– 4.52 1.68 1.91

– 0.864 0.992 0.997

56

M.I. Duinea et al. / Journal of Colloid and Interface Science 467 (2016) 51–59

Fig. 5. SEM images obtained (a) before and (b and c) after 6 days of immersion in OBS and their corresponding EDX analysis.

Raman lines situated in the high frequencies spectral range and low frequencies spectral range is equal with 1. 3.3.3. FTIR spectroscopy The chemical analysis of the solution resulted from aqueous batch experiment indicates that iron (Fetotal) was preferentially released relative to sulfur (as SO2 4 ) from the mineral surface

([Fetotal]:[SO2 4 ] = 145:22 lM:lM). This finding is in line with the results of previous studies [4,32–34] and indicates that under the new formed Fe(III)/Fe(II) bearing phases (outermost layer) there is a sulfur rich layer (SRL) which may incorporate species such as disulfide, polysulfide and elemental sulfur. It should be noted that details regarding the speciation of aqueous iron and sulfur can be found in Table S1 of the Supplementary material.

M.I. Duinea et al. / Journal of Colloid and Interface Science 467 (2016) 51–59

57

Scheme 1. Proposed electron transfer mechanism from the FeS to O2(aq) via the surface layer (i.e., Fe(III)/Fe(II) and/or S–S bonds). Aqueous Fe(III) (i.e., Fe3+) does not intermediate the mineral oxidation. It is formed by Fe(II) oxidation and is in balance with Fe(III) species present on FeS surface.

Fig. 6. Raman spectra of the FeS electrode (a) before and (b) after 6 days of immersion in OBS.

In order to obtain additional information regarding the nature of solid phases formed on FeS electrode, the solid residue collected after 6 days of contact between FeS powder (the same material like that used for the electrode construction) and an aerated solution with initial pH 5.1 was analyzed by FTIR spectroscopy. The FTIR spectra of initial and reacted FeS powder are presented in Fig. 7. The spectrum of initial FeS sample shows peaks at 1021, 1117, 1632 and 3444 cm1. The peaks at 1021 and 1117 can be assigned to sulfate ions formed during the preparation of FeS powder [8,23]. The peak at 1632 cm1 can be assigned to H–O–H bending modes [23]. The broad band centered at about 3444 cm1 results from stretching modes of surface water or iron hydroxo groups [23,35]. It is plausible that similar chemical species appear on the FeS electrode during polishing procedure. The FTIR spectrum of reacted FeS contains all the four signals present in the FTIR spectrum of initial FeS (but they are more intense) and additional signals at 473, 569, 801, 884, 2923 and 3223 cm1. The peak at 473 cm1 indicates the presence of elemental sulfur, polysulfide

and disulfide which are incorporated in the sulfur rich layer formed on FeS surface during its oxidation [23,35]. The peak at 569 cm1 can be ascribed to stretching vibrations of disulfide groups formed on the oxidized FeS surface [36,37]. The other peaks (801, 884, 2923 and 3223 cm1) can be associated to Fe(III) oxyhydroxides. The first two peaks result from the Fe–O–H bending modes of Fe(OH)3 and goethite [35,38], while the last two are produced by the stretching modes of Fe(OH)3 and goethite [23,35]. The small peaks observed in the range of 1350–1500 cm1 can be associated with carbonate species [23].

3.4. Mechanism of cathodic process If we take into account the experimental data, it is reasonable to assume that the variation of the characteristics of surface layer and FeS/water interface is largely responsible for the change of the mechanism of cathodic process. Since the rate of aqueous oxidation of FeS (which is directly proportional to jox [8]) does not decrease, although the pH increases, it results that aqueous oxidation of FeS minerals at high pH values is not intermediated by the dissolved ferric iron (Fe3+ (aq)) (whose concentration decreases, when the pH increases) [16]. Fe(III) oxyhydroxides and Fe3O4 (a mixed valence Fe(III)/Fe(II) oxide), which are incorporated in the surface layer and separates the O2(aq) and cathodic sites, may enable the transfer of electrons [39] between unreacted FeS and solution (Scheme 1). Also, the transfer of electrons between cathodic sites and oxygen may be facilitated by the S–S bonds in polysulfide and disulfide species [8]. For the latter mechanism of electron transfer (through S–S bonds) advocates the bc values obtained for pyrite elec-

Fig. 7. FTIR spectra of the initial FeS powder and FeS powder reacted for 6 days in OBS. The experimental conditions were initial pH 5.10 and a temperature of 25 °C.

58

M.I. Duinea et al. / Journal of Colloid and Interface Science 467 (2016) 51–59

trode. EIS results indicate that the initial surface layer formed during the preparation (polishing) of the electrode surface does not intermediate the transfer of the electrons and it acts as a physical barrier for the diffusion of the soluble reactants/reaction products from solution/surface to surface/solution. 4. Conclusions In this study, SEM/EDX analysis, Raman spectroscopy and FTIR spectroscopy were used to investigate the morphology and speciation of initial and oxidized surface of FeS at pH P 5.1 and 25 °C. It was found that during the aqueous oxidation of FeS new solid species (disulfide, polysulfide, elemental sulfur, ferric oxyhydroxides and Fe3O4) develop on the mineral surface. The results of potentiodynamic polarization experiments indicate that after 2 days of immersion of FeS electrode in OBS the mechanism of electron transfer from cathodic sites to O2 (mechanism of cathodic process) changes. The new formed solid phases on FeS surface at pH P 5.1 change the mechanism of cathodic process, but do not inhibit the mineral oxidation. Impedance data demonstrate that the formation of solid oxidation products is accompanied by the change of FeS/ water interface properties. The mechanism of anodic process (oxidation of FeS) remains practically unchanged over the 6 days of FeS immersion in OBS. Our findings have important implications for understanding FeS behavior at pH P 5.1. The results of potentiodynamic polarization experiments show that at pH P 5.1 the FeS oxidation is not intermediated by Fe3+ (aq). It is likely that after 2 days of FeS immersion in OBS the new solid phases developed on the mineral surface intermediate the electron transfer from cathodic sites to O2. A practical implication of the finding that Fe3+ does not intermediate FeS oxidation is that the mineral oxidative dissolution cannot be inhibited by Fe3+ sequestration with inorganic and/or organic ligands. Future studies should be conducted to evaluate the aqueous oxidation of pyrite at high pH and over a large time scale. Taking into account the differences between FeS2 and FeS reactivity, a substantial surface layer on pyrite should develop slower than in the case of FeS. Knowledge of the variation of the FeS2 reactivity with time will certainly help to have a better image on the mechanism of aqueous oxidation of pyrite. Acknowledgements This work was supported by a grant of the Romanian National Authority for Scientific Research, CNDI-UEFISCDI, project number 51/2012. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2016.01.010. References [1] M.A. Giaveno, G. Pettinari, E. Gonzalez Toril, A. Aguilera, M.S. Urbieta, E. Donati, The influence of two thermophilic consortia on troilite (FeS) dissolution, Hydrometallurgy 106 (2011) 19–25. [2] W. Su, L. Zhang, Y. Tao, G. Zhan, D. Li, D. Li, Sulfate reduction with electrons directly derived from electrodes in bioelectrochemical systems, Electrochem. Commun. 22 (2012) 37–40. [3] C.M.A. Brett, A.M. Oliveira Brett, Electrochemistry, Principles, Methods, and Applications, Oxford University Press Inc., New York, 1993. [4] N. Belzile, Y.W. Chen, M.F. Chai, Y. Li, A review on pyrrhotite oxidation, J. Geochem. Explor. 84 (2004) 65–76. [5] P.T. Behum, L. Lefticariu, K.S. Bender, Y.T. Segid, A.S. Burns, C.W. Pugh, Remediation of coal-mine drainage by a sulfate-reducing bioreactor: a case study from the Illinois coal basin, USA, Appl. Geochem. 26 (2011) S162–S166.

[6] W. Sand, T. Gehrke, P.G. Jozsa, A. Schippers, (Bio)chemistry of bacterial leaching-direct vs. indirect bioleaching, Hydrometallurgy 59 (2001) 159–175. [7] Y.W. Chen, Y. Li, M.F. Cai, N. Belzile, Z. Dang, Preventing oxidation of iron sulfide minerals by polyethylene polyamines, Miner. Eng. 19 (2006) 19–27. [8] C.E. Badica, P. Chirita, An electrochemical study of the oxidative dissolution of iron monosulfide (FeS) in air-equilibrated solutions, Electrochim. Acta 178 (2015) 786–796. [9] Y. Mikhlin, V. Varnek, I. Asanov, Y. Tomashevich, A. Okotrub, A. Livshits, G. Selyutin, G. Pashkov, Reactivity of pyrrhotite (Fe9S10) surfaces: spectroscopic studies, Phys. Chem. Chem. Phys. 2 (2000) 4393–4398. [10] J.E. Thomas, C.F. Jones, W.M. Skinner, R.St.C. Smart, The role of surface sulfur species in the inhibition of pyrrhotite dissolution in acid conditions, Geochim. Cosmochim. Acta 62 (1998) 1555–1565. [11] J.E. Thomas, R.St.C. Smart, W.M. Skinner, Kinetic factors for oxidative and nonoxidative dissolution of iron sulfides, Miner. Eng. 13 (2000) 10–11. [12] M.P. Janzen, R.V. Nicholson, J.M. Scharer, Pyrrhotite reaction kinetics: reaction rates for oxidation by oxygen, ferric iron, and for nonoxidative dissolution, Geochim. Cosmochim. Acta 64 (2000) 1511–1522. [13] P.C. Singer, W. Stumm, Acidic mine drainage: the rate determining step, Science 167 (1970) 1121–1123. [14] R. De Marco, S. Bailey, Z.T. Jiang, J. Morton, R. Chester, An in situ chronoamperometry/synchrotron radiation grazing incidence X-ray diffraction study of the electrochemical oxidation of pyrite in chloride media, Electrochem. Commun. 8 (2006) 1661–1664. [15] P. Chirita, J.D. Rimstidt, Pyrrhotite dissolution in acidic media, Appl. Geochem. 41 (2014) 1–10. [16] D. Neff, P. Dillmann, M. Descostes, G. Beranger, Corrosion of iron archaeological artefacts in soil: estimation of the average corrosion rates involving analytical techniques and thermodynamic calculations, Corros. Sci. 48 (2006) 2947–2970. [17] P. Chirita, C.A. Constantin, E.C. Badica, I.M. Duinea, M.L. Birsa, E. Matei, I. Baltog, Inhibition of troilite (FeS) oxidative dissolution in air-saturated acidic solutions by O-ethyl-S-2-(2-hydroxy-3,5-diiodophenyl)-2oxoethylxantogenate, Mater. Chem. Phys. 157 (2015) 101–107. [18] P. Chirita, A kinetic study of hydrogen peroxide decomposition in presence of pyrite, Chem. Biochem. Eng. Quart. 21 (2007) 257–264. [19] P. Chirita, Hydrogen peroxide decomposition by pyrite in the presence of Fe (III)-ligands, Chem. Biochem. Eng. Quart. 23 (2009) 259–265. [20] A. Nacu, R. Mocanu, T. Onofrei, M. Gaburici, C. Simion, F. Matei, D. Balba, V. Dulman, E. Doniga, Chimie analitica si analiza instrumentala, Editura Institutului Politehnic Iasi, Iasi, 1988. [21] S. Manescu, M. Cucu, M.L. Diaconescu, Chimia sanitara a mediului, Editura Medicala, Bucuresti, 1994. [22] G. Yue, E. Asselin, Kinetics of ferric ion reduction on chalcopyrite and its influence on leaching up to 150 °C, Electrochim. Acta 146 (2014) 307–321. [23] P. Chirita, M. Descostes, M.L. Schlegel, Oxidation of FeS by oxygen-bearing acidic solutions, J. Colloid Interface Sci. 321 (2008) 84–95. [24] P. Cordoba-Torres, Characterization of frequency dispersion in the impedance response of a distributed model from the mathematical properties of the distribution function of relaxation times, Electrochim. Acta 180 (2015) 591– 603. [25] K. Shimizu, A. Lasia, J.F. Boily, Electrochemical impedance study of the hematite/water interface, Langmuir 28 (2012) 7914–7920. [26] S. Lehner, M. Ciobanu, K. Savage, D.E. Cliffel, Electrochemical impedance spectroscopy of synthetic pyrite doped with As Co, and Ni, J. Electrochem. Soc. 155 (2008) P61–P70. [27] C.A. Constantin, P. Chirita, Oxidative dissolution of pyrite in acidic media, J. Appl. Electrochem. 43 (2013) 659–666. [28] J.P. Jolivet, M. Henry, J. Livage, De la solution a l’oxyde, Condensation des cations en solutions aqueuse, Chimie de surfaces des oxides. InterEditions et CNRS Editions, 1994. [29] E.B. Hansson, M.S. Odziemkowski, R.W. Gillham, Formation of poorly crystalline iron monosulfides: surface redox reactions on high purity iron, spectroelectrochemical studies, Corros. Sci. 48 (2006) 3767–3783. [30] R.L. Frost, R. Scholz, J. Jirasek, F.M. Belotti, An SEM–EDX and Raman spectroscopic study of the fibrous arsenate mineral liskeardite and in comparison with other arsenates kanˇkite, scorodite and yvonite, Spectrochim. Acta A 151 (2015) 566–575. [31] T. Ohtsuka, S. Tanaka, Monitoring the development of rust layers on weathering steel using in situ Raman spectroscopy under wet-and-dry cyclic conditions, J. Solid State Electrochem. 19 (2015) 3559–3566. [32] A.R. Pratt, I.J. Muir, H.W. Nesbitt, X-ray photoelectron and Auger electron spectroscopic studies of pyrrhotite and mechanism of air oxidation, Geochim. Cosmochim. Acta 58 (1994) 827–841. [33] A.R. Pratt, I.J. Muir, H.W. Nesbitt, Generation of acids from mine waste: oxidative leaching of pyrrhotite in dilute H2SO4 solutions at pH 3.0, Geochim. Cosmochim. Acta 58 (1994) 5147–5159. [34] A.R. Pratt, H.W. Nesbitt, Pyrrhotite leaching in acid mixtures of HCl and H2SO4, Am. J. Sci. 297 (1997) 807–820. [35] Y.L. Mikhlin, A.V. Kuklinskiy, N.I. Pavlenko, V.A. Varnek, I.P. Asanov, A.V. Okotrub, G.E. Selyutin, L.A. Solovyev, Spectroscopic and XRD studies of the air degradation of acid-reacted pyrrhotites, Geochim. Cosmochim. Acta 66 (2002) 4057–4067.

M.I. Duinea et al. / Journal of Colloid and Interface Science 467 (2016) 51–59 [36] R. Weerasooriya, M. Makehelwala, A. Bandara, Probing reactivity sites on pyrite-oxidative interactions with 4-chlorophenol, Colloids Surf. A: Physicochem. Eng. Aspects 367 (2010) 65–69. [37] P. Chirita, M.L. Schlegel, Oxidative dissolution of iron monosulfide (FeS) in acidic conditions: the effect of solid pretreatment, Int. J. Min. Proc. 135 (2015) 57–64. [38] M. Mohapatra, S. Gupta, B. Satpati, S. Anand, B.K. Mishra, PH and temperature dependent facile precipitation of nano-goethite particles in Fe(NO3)3–NaOH–

59

NH3NH2HSO4–H2O medium, Colloids Surf. A: Physicochem. Eng. Aspects 355 (2010) 53–60. [39] J.E. Katz, X. Zhang, K. Attenkofer, K.W. Chapman, C. Frandsen, P. Zarzycki, K.M. Rosso, R.W. Falcone, G.A. Waychunas, B. Gilbert, Electron small polarons and their mobility in iron (oxyhydr)oxide nanoparticles, Science 337 (2012) 1200– 1203.