A fuel-cell-assisted iron redox process for simultaneous sulfur recovery and electricity production from synthetic sulfide wastewater

Journal of Hazardous Materials 243 (2012) 350–356

Contents lists available at SciVerse ScienceDirect

Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat

A fuel-cell-assisted iron redox process for simultaneous sulfur recovery and electricity production from synthetic sulfide wastewater Lin-Feng Zhai a , Wei Song a , Zhong-Hua Tong b , Min Sun a,∗ a b

Department of Chemical Engineering, Hefei University of Technology, Hefei 230009, China Department of Chemistry, University of Science and Technology of China, Hefei 230026, China

h i g h l i g h t s

g r a p h i c a l

a b s t r a c t

 A novel FC-IR process was reported to recover sulfur and electricity from sulfide.  The fuel cell and iron based liquid redox sulfur recovery techniques were combined.  The sulfide oxidation and Fe(III) regeneration were individually investigated.  Results of coupling operation confirmed the feasibility of the FC-IR process.

a r t i c l e

i n f o

Article history: Received 3 September 2012 Received in revised form 22 October 2012 Accepted 24 October 2012 Available online 2 November 2012 Keywords: Iron redox sulfur recovery Fuel cell Coupling Electricity

a b s t r a c t Sulfide present in wastewaters and waste gases should be removed due to its toxicity, corrosivity, and malodorous property. Development of effective, stable, and feasible methods for sulfur recovery from sulfide attains a double objective of waste minimization and resource recovery. Here we report a novel fuel-cell-assisted iron redox (FC-IR) process for simultaneously recovering sulfur and electricity from synthetic sulfide wastewater. The FC-IR system consists of an oxidizing reactor where sulfide is oxidized to elemental sulfur by Fe(III), and a fuel cell where Fe(III) is regenerated from Fe(II) concomitantly with electricity producing. The oxidation of sulfide by Fe(III) is significantly dependent on solution pH. Increasing the pH from 0.88 to 1.96 accelerates the oxidation of sulfide, however, lowers the purity of the produced elemental sulfur. The performance of fuel cell is also a strong function of solution pH. Fe(II) is completely oxidized to Fe(III) when the fuel cell is operated at a pH above 6.0, whereas only partially oxidized below pH 6.0. At pH 6.0, the highest columbic efficiency of 75.7% is achieved and electricity production maintains for the longest time of 106 h. Coupling operation of the FC-IR system obtains sulfide removal efficiency of 99.90%, sulfur recovery efficiency of 78.6 ± 8.3%, and columbic efficiency of 58.6 ± 1.6%, respectively. These results suggest that the FC-IR process is a promising tool to recover sulfur and energy from sulfide. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Sulfide is ubiquitously present in many industrial wastewaters and gaseous streams and needs to be removed due to its toxicity, corrosivity, and malodorous property. As one of the most

∗ Corresponding author. Fax: +86 551 2901450. E-mail address: [email protected] (M. Sun). 0304-3894/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jhazmat.2012.10.046

important desulfurization processes, the removal of sulfide by converting it to sulfur attains a double objective of waste minimization and resource recovery. Development of effective, stable, and feasible methods for sulfur recovery from sulfide has been the focus of extensive research [1–4]. Due to the excellent electrochemical activity of sulfide, an aircathode fuel cell was proposed to recover sulfur from aqueous sulfide [5,6]. In such a device, the utilization of air-driven cathode resulted in spontaneous oxidization of sulfide at the anode,

L.-F. Zhai et al. / Journal of Hazardous Materials 243 (2012) 350–356

concomitant with production of elemental sulfur and electricity. This technology would be cost-effective if the produced sulfur can be successfully harvested, while the electricity generation is an additional advantage. However, there are several operating issues yet to be resolved before it becomes ready for practical application, including passivation of the electrodes by sulfur, difficulty in sulfur separation and handling, and occurrence of undesirable secondary electrochemical reactions [7,8]. Liquid redox sulfur recovery (LRSR) process is among the most promising and thus intensively studied techniques for sulfur recovery from gaseous sulfide. The LRSR process comprises two coupled steps: (1) the gaseous sulfide is absorbed into a solution and then oxidized by an intermediate (in the oxidized form) to solid elemental sulfur, which is thereafter removed by gravity settling; and (2) the reduced form of the intermediate is converted back to the oxidized form. Redox couples including As(V)/As(III), V(V)/V(IV), Co(III)/Co(II), and Fe(III)/Fe(II) can be used as intermediates in the LRSR process, among which the iron redox couple Fe(III)/Fe(II) is predominantly favored due to its specific oxidative potential toward sulfide [9,10]. In the step of Fe(III) regeneration, Fe(II) is typically oxidized to Fe(III) via contact with air or electrolysis [10,11]. Unfortunately, the low solubility of atmospheric oxygen in Fe(II) solution makes the direct oxidation by air at low efficiency, while the requirement of power supply makes electrolysis at high economic cost. An alternative technology based on air-cathode fuel cell was recently developed to recover Fe(III) from Fe(II) [12,13]. In the fuel cell system, ferrous iron (Fe2+ ) is oxidized to ferric iron (Fe3+ ) at the anode, and oxygen in the air is reduced to water at the cathode. Such a process was proved to be effective in treating the synthetic acidmine drainage, with Fe2+ completely oxidized to ferric state which formed precipitates in the anodic chamber [12,13]. The utilization of fuel cell technology offers a preferable approach to regenerate Fe(III) in the LRSR process, which achieves simultaneous Fe(III) regeneration and electricity production. The aim of the present work was thus to develop a novel fuelcell-assisted iron redox (FC-IR) process for simultaneous sulfur recovery and electricity generation from aqueous sulfide, by incorporating the fuel cell technology into the iron based LRSR process. The FC-IR system consisted of one oxidizing reactor for recovering sulfur from sulfide and one fuel cell for regenerating Fe(III) from Fe(II) (Fig. 1). In the oxidizing reactor, sulfide ions (S2− ) react with hydrogen ions (H+ ) to form hydrogen sulfide (H2 S) in acidic solution. The H2 S is then oxidized to elemental sulfur (S0 ) by Fe3+ which is reduced to Fe2+ . In the fuel cell, Fe2+ is oxidized to Fe(III) precipitates such as Fe (OH)3 under neutral conditions at the anode, which is thereafter converted to Fe3+ in the presence of excess acid. The atmospheric oxygen (O2 ) is reduced to water (H2 O) at the cathode, and electricity is simultaneously produced. The related reactions in the whole process can be expressed as follows: Reactions in the oxidizing reactor: S2− + 2H+ → H2 S H2 S + 2Fe

3+

→ S0 ↓ + 2Fe

(1) 2+

+ 2H

+

(2)

Reactions in the fuel cell: Anode :

2Fe2+ + 6H2 O → 2Fe(OH)3 ↓ + 6H+ + 2e−

2Fe(OH)3 ↓ + 6H+ → 2Fe3+ + 6H2 O Cathode :

1/2O2 + 2H+ + 2e− → H2 O

(3) (4) (5)

351

sulfide solution. We demonstrate here that effective sulfur recovery from sulfide can be achieved and that electricity can be produced concurrently. The efficiency of FC-IR process was comprehensively evaluated in terms of sulfide removal, and sulfur recovery and electricity production. 2. Materials and methods 2.1. Oxidation of sulfide The experiment on sulfide oxidation was carried out in an oxidizing reactor, which was a 1000 mL cylindrical glass reactor with liquid volume of 750 mL. The reactor was filled with 200 mM of NaCl as basal solution and Fe2 (SO4 )3 as Fe(III) source, and the pH was adjusted by 1 M HCl. To determine the effects of operating conditions on sulfide oxidation, the concentration of Fe(III) varied from 8.05 to 12.25 mM, and solution pH from 0.88 to 1.96. The reaction was started by adding the 1 M of Na2 S stock solution (iodometrically standardized) into the reactor at a final concentration of 3.5 mM. Experiments were conducted in duplicate at ambient temperature of 25 ◦ C. The sulfur particles were collected by filtration and then washed with deionized water. 2.2. Regeneration of Fe(III) The experiment on Fe(III) regeneration was conducted in an air-cathode single chamber fuel cell operated in fed-batch mode. Fig. S1 (see Supplementary material) gives a laboratory-scale prototype of the fuel cell. The fuel cell was made of glass. The volume of the anodic chamber was 175 mL with a net volume of 150 mL. A 4 cm × 2.5 cm carbon paper (non wet-proofed, 090S, Toray Co., Japan) was located in the chamber as anode. The cathode was a 2 cm × 2 cm carbon paper (090S, wet-proofed, Toray Industries Inc., Japan) with 0.05 mg cm−2 Pt catalyst on one side. The coated side of the cathode was positioned facing the cation exchange membrane (GEFC-10N, GEFC Co., China) and the uncoated side was directly exposed to air. Titanium wires (1 mm in diameter) were used to connect both electrodes and the external load (1 k resistor). The anodic chamber was filled with a solution containing 200 mM NaCl as electrolyte and 50 mM NaHCO3 as pH buffer, and then purged with N2 to remove dissolved oxygen. FeSO4 ·7H2 O was added to the anodic chamber at 3.5 mM as Fe(II) source in an anaerobic glove box, and the pH was adjusted using 1 M HCl or NaOH. Solution pH was varied in the range of 4.0 and 8.0 to determine the effect of pH on Fe(II) oxidation and power generation. Experiments were conducted in duplicate at ambient temperature of 25 ◦ C. 2.3. Coupling operation of the FC-IR system The FC-IR coupled system was operated in fed-batch mode. The 3.5 mM of sulfide was oxidized by Fe(III) in a 1000 mL reactor at Fe(III)/S2− ratio of 2.3 and pH of 0.88. The produced sulfur particles were recovered by filtration after 24 h of reaction, and 1 M of NaOH was added into the residual solution until pH 5.0. NaHCO3 was then added at 50 mM and solution pH was adjusted to 6.0. The 150 mL of solution was moved into the anodic chamber of a fuel cell for Fe(III) regeneration. Batch test was ended when the cell voltage dropped below 1 mV. The recovered Fe(III) precipitates were dissolved by adding 5 M of HCl into the anodic chamber and then the concentration of Fe(III) was measured.

Overall reactions are shown in Eq. (6) as: S2− + 2H+ + 1/2O2 → S0 ↓ + H2 O + electricity

(6)

To investigate the feasibility of this novel FC-IR process, we therefore examined the treatment efficiency of a synthetic

2.4. Analytical methods The concentration of Fe(II) was measured using the modified phenanthroline method [14]. Fe(III) concentration was determined

352

L.-F. Zhai et al. / Journal of Hazardous Materials 243 (2012) 350–356

Fig. 1. Flow diagram of the FC-IR process.

after being reduced to Fe(II) by ascorbic acid. The concentration of sulfide was measured according to the iodimetric method [15]. The sulfur valence states of recovered sulfur particles were determined by X-ray photoelectron spectroscopy (XPS) performed on an ESCALAB 250 spectrometer (VG Instrument Group Ltd., UK) equipped with a monochromatic Mg KR X-ray source (1253.6 eV). Before analysis, the sulfur particles were washed with deionized water and then dried at 60 ◦ C until constant weight. Survey scans were carried out over 1100 – 0 eV binding energy range with 1.0 eV steps. The residual pressure inside the analytical chamber was below 1 × 10−8 Pa. The C1s electron binding energy corresponding to graphitic carbon was set at 284.6 eV and used as a reference to position the other peaks. Voltage across the 1 k external resistor was recorded at 10 min intervals using a data acquisition system (USB2801, ATD Co., China). The current was calculated using Ohm’s law, and the current density was then normalized by the projected surface areas of two sides of the anode (20 cm2 ). Coulombic efficiency (CE) was calculated as previously described [16], based on the assumption that 1 mol of electrons are produced from the oxidation of 1 mol of Fe(II). To investigate the redox property of Fe(II) in the fuel cell, the initial potential and cyclic voltammetry (CV) of carbon anode was obtained in Fe(II) solution at different pH levels, using a CHI 660D electrochemical workstation (CH Instruments Inc., USA) with an Hg/Hg2 Cl2 reference electrode and a Pt counter electrode. The CV scanning was taken at 0.1 V s−1 and between −0.8 and 0.8 V, with an initial scanning from a low potential to a high one, then followed by a reversed one. The electrolyte solution was purged with nitrogen for 30 min prior to electrochemical measurements and a nitrogen atmosphere was maintained throughout the experiments.

3. Results 3.1. Oxidation of sulfide by Fe(III) and sulfur recovery Fig. 2 shows the pH-dependent efficiency of sulfide removal at different Fe(III)/S2− molar ratios. The sulfide removal efficiencies were generally above 90% after 90 min of reaction, suggesting the high oxidizing rate of sulfide by Fe(III). Sulfide oxidation was accelerated when the solution pH increased from 0.88 to 1.96, but not dramatically so. The theoretical molar ratio of Fe(III) to sulfide was 2–1 according to the stoichiometric equation. Adding Fe(III) in excess of the stoichiometric requirement has very limited effect on

the sulfide oxidation rate, evidenced by the slightly improved sulfide removal efficiency when the Fe(III)/S2− molar ratios increased from 2.3 to 3.5. The purity of produced sulfur was analyzed by XPS. Fig. 3 shows the S2p spectra of sulfur particles obtained under different operating conditions. The S2p3/2–1/2 doublet at 163.9 and 164.9 eV was characteristic of S(0) in elemental sulfur [17,18]. The broad peak centered at 168.5 eV was attributed to the high valences like S(+IV) and S(+VI), and the irregular peak at ca. 161.4 was assigned to S(−II) [8,19]. The detection of S(+IV)/S(+VI) in all sulfur samples implied the presence of sulfur oxyanions, such as thiosulfate, sulfite, and sulfate. Comparison of the relative areas of corresponding peaks indicated the fraction of sulfur oxyanions in sulfur particles increased with pH. Besides S(+IV) and S(+VI), S(−II) was also detected in the sulfur samples obtained at pH 1.96, suggesting more impurities were formed at higher pH. Totally speaking, the operation of the sulfide oxidizing reactor at Fe(III)/S2− ratio of 2.3 and pH 0.88 seems advantageous as less byproducts occurs while sulfide removal is still reasonably good.

3.2. Fe(III) regeneration and electricity production in fuel cells Fig. 4 illustrates the electricity generation of fuel cells operated at different pH. The current densities of all the fuel cells shared a similar evolution tendency. Upon addition of 3.5 mM FeSO4 , an instant current was obtained, immediately followed by a drop in the current density which gradually fell to 0 mA over the following several hours. The electricity generation was strongly dependent on the solution pH. At pH 4.0, the initial current density was no more than 5 mA m−2 , which decreased to null within 15 h. When solution pH increased, current density was correspondingly promoted, with a maximum initial value of 180 mA m−2 observed at pH 8.0. The electricity generation maintained for as long as 106 h at pH 6.0, whereas lower pH led to an obvious shortening in the duration time and higher pH slightly shortened it. For the fuel cells operated at pH above 6.0, no Fe(II) was detected in the anodic solution at the end of one batch cycle, indicating Fe(II) was completely oxidized to Fe(III). Only 13.9% and 34.4% of Fe(III) was recovered at pH 4.0 and 5.0, respectively (Fig. 5A). The highest CE of 75.7% was achieved at pH 6.0, which dropped to 0.8% at pH 4.0 and to 63.3% at pH 8.0 (Fig. 5B). In order to explain the low CEs of fuel cells, one fuel cell was operated in open circuit mode that the circuit was disconnected

L.-F. Zhai et al. / Journal of Hazardous Materials 243 (2012) 350–356

Sulfide removal (%)

100

C

B

A

353

80

60

pH 0.88

pH 0.88

pH 0.88

pH 1.27

pH 1.27

pH 1.27

pH 1.96

pH 1.96

pH 1.96

Fe(III):S2-= 2.3:1

Fe(III):S2-= 3:1

Fe(III):S2-= 3.5:1

40 0

20

40

60

80

0

20

Time (min)

40

60

80

0

20

40

60

80

Time (min)

Time (min)

Fig. 2. Sulfide removal efficiencies in the sulfide oxidizing reactor at different pH and Fe(III)/S2− ratios.

the anodic potential declined drastically from 57 to −533 mV vs Hg/Hg2 Cl2 . To deeply investigate the effect of pH on the redox activity of Fe(II), cyclic voltamogramms of the carbon anodes at different solution pH were compared. As anticipated, the redox activities of Fe(II) were strongly dependent on the solution pH, evidenced by the distinctive anodic peaks attributed to Fe(II) oxidation (shown in Fig. 6). At pH 4.0 and 5.0, the peak intensities were quite weak with correspondence to the poor performance of fuel cells. Significant increases of peak intensity were achieved when solution pH went up to 6.0 and 7.0, indicating enhanced activity of Fe(II) to be oxidized. Further rise of pH to 8.0 depressed Fe(II) oxidation, evidenced by the weakened peak intensity. In order to fully understand the role of Fe(II) speciation on its oxidation, one fuel cell was operated at solution pH of 6.0 while without NaHCO3 buffer. As shown in Fig. 6, very negligible

and no current flow was obtained in the operation. As shown in Fig. S2 (see Supplementary material), Fe(II) was gradually oxidized to Fe(III) in the fuel cell with open circuit, while the oxidation rate of Fe(II) in such a reactor was relatively lower than that in the fuel cell with close circuit. This observation suggested that some oxidants, such as oxygen diffused from the cation exchange membrane [20], existed in the anodic chamber of fuel cell that could oxidize Fe(II). It is thus presumed the loss of electrons from Fe(II) to other competitive oxidation processes was the most likely reason for the low CEs of fuel cells. 3.3. Electrochemical analyses of Fe(II) oxidation activity Fig. 5C shows the initial anodic potentials of the fuel cells under different pH conditions. When the pH increased from 4.0 to 8.0,

Fe(III):S2-= 2.3:1

pH 0.88 Fe(III):S2-= 2.3:1

S0

S0

Counts Per Second

S+IV/S+VI

pH 1.27 Fe(III):S2-= 2.3:1

pH 1.96

S0 S+IV/S+VI

S+IV/S+VI

S-II

Fe(III):S2-= 3:1

2pH 0.88 Fe(III):S = 3:1

pH 1.27Fe(III):S2-=3:1

pH 1.96

Fe(III):S2-= 3.5:1

pH 0.88 Fe(III):S2-=3.5:1

pH 1.27 Fe(III):S2-= 3.5:1

pH 1.96

172

168

164

160 172

168

164

160 172

168

164

Binding Energy (eV) Fig. 3. XPS analysis of sulfur particles recovered at different pH and Fe(III)/S2− ratios.

160

354

L.-F. Zhai et al. / Journal of Hazardous Materials 243 (2012) 350–356

pH8.0

150 100 50 0

pH7.0

150 50

-2

Current density (mA m )

100 0 pH6.0

150 100 50 0

pH5.0

50 25 0 5

pH4.0

0 0

25

50

75

100

Time (h) Fig. 4. Electricity generation from Fe(II) oxidation in the fuel cell at different pHs.

oxidation occurred during potential scanning in the CV measurement, and accordingly, the fuel cell failed to produce electricity, suggesting the importance of carbonate in facilitating Fe(II) oxidation.

20

Fig. 5. (A) Coulombic efficiencies; (B) Fe(III) recoveries; and (C) initial anodic potentials of the fuel cells operated at different pH.

pH 4

pH 5

pH 6

pH 7

10 0

Current (mA)

-10 20 10 0 -10 20

pH 8

NaCl/Fe(II)

10 0 -10 -800

-400

0

400

800 -800

-400

0

400

Potential (mV vs. Hg/Hg2Cl2) Fig. 6. Cyclic voltamogramms of 3.5 M FeSO4 on carbon electrode at different pHs.

800

L.-F. Zhai et al. / Journal of Hazardous Materials 243 (2012) 350–356

3.4. Performance of the FC-IR coupled system The processes of sulfide oxidation and Fe(III) regeneration were coupled to evaluate the feasibility of the FC-IR system. The sulfide removal efficiency was 99.90%, and 0.066 ± 0.007 g of sulfur were recovered with sulfur recovery efficiency of 78.6 ± 8.3%. Electricity generation maintained for 104 h in the fuel cell at average current density of 79.2 ± 2.1 mA m−2 . The CE was calculated as 58.6 ± 1.6%, based on the assumption that 2 mol of electrons were produced from 1 mol of sulfide removal. Notably, the Fe(III) was recovered at an efficiency of approximately 100%, suggesting the iron redox couple can be recycled in the FC-IR operation.

4. Discussion The results of this study provide a proof that the FC-IR process is a promising tool in treating sulfide in wastewater, with concomitant sulfur and energy recovery. The features of the FC-LR process are summarized as follows: (1) sulfide can be removed at high efficiency and sulfur can be recovered; (2) part of the chemical energy stored in sulfide can be utilized as electricity; and (3) the process is performed at ambient temperature. Notably, the FC-IR process is not limited in treating aqueous sulfide, but also can be potentially applied to remove hydrogen sulfide in gaseous streams. As is well known, iron-redox LRSR has developed into a very versatile processing scheme to treat hydrogen sulfide [10]. As a novel process based on LRSR, the FC-IR process has the characteristic of spontaneously producing electricity without energy input, which makes it economically advantageous over traditional LRSR processes. Electricity was produced from the oxidation of Fe(II) in the single-chamber fuel cell, but it was found a large percentage of Fe(II) was oxidized without electricity generation. To increase the power density of the system, approaches should be pursued to increase the fraction of Fe(II) that contributes to electricity generation. It appears that substantial losses of Fe(II) here resulted from passive oxygen diffusion into the anodic chamber across the cation exchange membrane [20]. Note that ca. 36% of the Fe(II) was oxidized in the open-circuit fuel cell (see Fig. S2 in Supplementary material), oxidation of Fe(II) by diffused oxygen could account for a considerable portion of the electron loss. This problem can be solved by employing the two-chamber fuel cell architecture with separate compartments for the anode and cathode [12]. However, the two-compartment design may lead to operational difficulty in large-scale systems. The capital cost of the two-chamber architectures is also substantially higher than the single-chamber ones. From practical point of view, Fe(II) oxidation due to passive oxygen transfer may actually be economical when compared to the forcedair systems in conventional LRSR process. Nevertheless, additional work will be needed to optimize operational parameters for an improved power production. In the conventional LRSR processes, oxygen is not strictly excluded from the system when sulfide is oxidized [11]. However, oxygen in the sulfide oxidizing reactor is of adverse influence on the power generation in the fuel cell. Since part of sulfide is oxidized by oxygen, the amount of Fe(III) participating in the reaction is reduced, caused less Fe(II) to be oxidized in the fuel cell. Consequently, the CE of FC-IR coupled system calculated based on the sulfide consumption is considerably lower than that of the fuel cell based on the Fe(II) consumption. Solution pH is a key variable to control the sulfide oxidation, which can affect key chemical equilibrium and the rates of reactions. Since the solubility of iron species increases with decreased pH, solution pH should be low enough to avoid formation of Fe precipitates [11]. Although the oxidation of sulfide is accelerated with increased pH, the occurrence of secondary reactions is

355

facilitated as well. In a sulfide oxidizing reactor which did not rigidly eliminate oxygen, such as that in this study as well as those in typical LRSR processes [11], oxidation of sulfide over the state of elemental sulfur results in the formation of sulfur oxyanions [21]. Efficient production of separable solid sulfur with high purity is important for an economic consideration. However, the produced sulfur oxyanions, like sulfate and sulfite, are absorbed on the sulfur particles and difficult to be removed, thus lowering the purity of sulfur. Therefore, a compromise operating pH should be carefully selected not only to avoid Fe precipitation and control by-products, but also to give acceptable sulfide-to-sulfur conversion rate. In this study, the sulfide oxidation is better operated below pH 1.0, because no Fe precipitation occurs and the by-products are less while sulfide removal is still favorably good. The oxidation of Fe(II) in fuel cell is also a strong function of pH. The accelerated oxidation of Fe(II) with increased pH was ever attributed to the participation of H+ /OH− in cell reactions [12]. However, the Fe(II) experiences more complicated processes to be oxidized in the anodic solution buffered with carbonate, with a number of iron species involved in the Fe(II) oxidation, including hydroxyl–Fe(II) species FeOH+ and Fe(OH)2 , carbonate–Fe(II) species FeCO3 , FeHCO3 + , Fe(CO3 )2− and Fe(OH)CO3 − , species FeCl+ and FeSO4 , and free hydrated Fe2+ [22,23]. Individual iron species are oxidized at different rates, and the overall Fe(II) oxidation rate can be interpreted as the sum of the oxidation rates of individual Fe(II) species. The FeCl+ and FeSO4 are expected to be kinetically unfavorable and usually ignored in the analysis of Fe(II) oxidation rate [22]. Accordingly, a failure of electricity generation was observed in the fuel cell without carbonate buffer. In the presence of carbonate, the Fe(OH) and Fe(CO3 ) species contribute to the fast kinetics of Fe(II) oxidation [24–26]. Therefore, a noticeable enhancement of fuel cell performance was obtained upon addition of NaHCO3 . At pH below 5.0, the concentrations of active Fe(II) species were so low as to be negligible, whereas the kinetically less reactive species Fe2+ dominated the solution [21,26]. As a result, very limited Fe(II) oxidation was observed for the fuel cells operated at pH 4.0 and 5.0. In the experimental pH range (pH 4.0–8.0), the concentrations of Fe(OH) and Fe(CO3 ) species, especially Fe(OH)2 and Fe(CO3 )2 2− , rose steeply with pH increase [25,27]. Since these species were far more readily to be oxidized, accelerated Fe(II) oxidation with higher power output as well as longer maintenance of electricity production was achieved for the fuel cells operated above pH 6.0. Notably, a slightly decrease of power output was observed when pH increased from 6.0 to 8.0, probably attributed to the decreased concentration of soluble Fe(II) species [27]. Moreover, the Fe(II) loss due to oxygen diffusion from membrane was another possible reason for the reduced power output [20]. The accelerated Fe(II) oxidation rate made Fe(II) more easily oxidized by diffused oxygen at higher pH, thus causing decreased CE and power output. The system used here is a very basic type of reactor architecture to demonstrate the feasibility of this new FC-IR process. Further developments are expected to make this technique more effective and applicable. Improvements are pursued for high efficiencies of fuel cell, such as modification of electrodes to facilitate Fe(II) oxidation, design of new fuel cell architectures to restrain air leakage, and optimization of operational parameters. Due to the very low solubility of iron in aqueous solutions, the sulfide oxidation should be performed in acidic solution below pH 2.0 to avoid iron precipitation. However, the fuel cell for Fe(III) regeneration is better operated at pH 6.0–8.0, which causes iron species to precipitate and leads to operational troubles. The introduction of chelating agents may bring a possible solution to such a problem. The chelating agents, such as EDTA and NTA, have been suggested to prevent Fe(III) species from precipitating while at the same time

356

L.-F. Zhai et al. / Journal of Hazardous Materials 243 (2012) 350–356

attaining high Fe(II) oxidation rates at neutral to slightly caustic pH [21,27]. 5. Conclusions A novel FC-IR system that incorporates fuel cell technology into iron based LRSR process was developed to simultaneously recover elemental sulfur and electricity from aqueous sulfide. In such a system, sulfide is oxidized to sulfur by Fe(III) which is reduced to Fe(II), and then Fe(II) is reoxidized to Fe(III) and electricity is concomitantly generated in the air-cathode fuel cell. The processes of sulfide oxidation and Fe(III) regeneration were individually investigated under different operating conditions. Increasing the solution pH slightly accelerated the oxidation of sulfide, however, lowered the purity of produced elemental sulfur. The preferable operating condition for sulfide oxidation was determined as pH of 0.88 and Fe(III)/S2− ratio of 2.3. The performance of fuel cell was also strongly dependent on solution pH. Fe(II) is completely oxidized to Fe(III) when the fuel cell is operated at a pH above 6.0, whereas only partially oxidized below pH 6.0. The highest columbic efficiency of 75.7% is achieved at pH 6.0 with electricity production maintaining for as long as 106 h, whereas lower pH led to noticeably reduction in power output and higher pH lightly lowered it. Coupling operation of the FC-IR system obtains sulfide removal efficiency of 99.90%, sulfur recovery efficiency of 78.6 ± 8.3%, and columbic efficiency of 58.6 ± 1.6%, respectively. In summary, the FC-IR process described in this work is able to simultaneously recover sulfur and energy in the sulfide treatment. In addition to the environmental benefits available, the significant economic potential makes this technology commercially prospective. Successful development of this technology should focus on the operational issues to enable it applicable and scaleable. Acknowledgments The authors wish to thank the NSFC (51008108) and the NSFCRGC Joint Project (21021140001) for the partial support of this work. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jhazmat. 2012.10.046. References [1] J.S. Eow, Recovery of sulfur from sour acid gas: a review of the technology, Environ. Prog. 21 (2002) 143–162. [2] L. Zhang, P.D. Schryver, B.D. Gusseme, W.D. Muynck, N. Boon, W. Verstraete, Chemical and biological technologies for hydrogen sulfide emission control in sewer systems: a review, Water Res. 42 (2008) 1–12.

[3] H. Huang, Y. Yu, K.H. Chung, Recovery of hydrogen and sulfur by indirect electrolysis of hydrogen sulfide, Energy Fuels 23 (2009) 4420–4425. [4] E. Sahinkaya, H. Hasar, A.H. Kaksonen, B.E. Rittmann, Performance of a sulfide-oxidizing, sulfur-producing membrane biofilm reactor treating sulfidecontaining bioreactor effluent, Environ. Sci. Technol. 45 (2011) 4080– 4087. [5] K. Rabaey, K. Vandesompel, L. Maignien, N. Boon, P. Aelterman, P. Clauwaert, L. DeSchamphelaire, H.T. Pham, J. Vermeulen, M. Verhaege, P. Lens, W. Verstraete, Microbial fuel cells for sulfide removal, Environ. Sci. Technol. 40 (2006) 5218–5224. [6] P.K. Dutta, K. Rabaey, Z. Yuan, J. Keller, Spontaneous electrochemical removal of aqueous sulfide, Water Res. 42 (2008) 4965–4975. [7] P.K. Dutta, R.A. Rozendal, Z. Yuan, K. Rabaey, J. Keller, Electrochemical regeneration of sulfur loaded electrodes, Electrochem. Commun. 11 (2009) 1437–1440. [8] M. Sun, Z.X. Mu, Y.P. Chen, G.P. Sheng, X.W. Liu, Y.Z. Chen, Y. Zhao, H.L. Wang, H.Q. Yu, L. Wei, F. Ma, Microbial-assisted sulfide oxidation in the anode of a microbial fuel cell, Environ. Sci. Technol. 43 (2009) 3372–3377. [9] G.X. Hua, D. McManus, J.D. Woollins, The evolution, chemistry and applications of homogeneous liquid redox sulfur recovery techniques, Comment. Inorg. Chem. 22 (2001) 327–351. [10] D.L. Heguy, G.J. Nagl, Consider optimized iron-redox processes to remove sulfur, Hydrocarb. Process. 82 (2003) 53–57. [11] Y. Gendel, N. Levi, O. Lahav, H2 S (g) removal using a modified, low-pH liquid redox sulfur recovery (LRSR) process with electrochemical regeneration of the Fe catalyst couple, Environ. Sci. Technol. 43 (2009) 8315–8319. [12] S. Cheng, B.A. Dempsey, B.E. Logan, Electricity generation from synthetic acidmine drainage (AMD) water using fuel cell technologies, Environ. Sci. Technol. 41 (2007) 8149–8153. [13] S. Cheng, J.H. Jang, B.A. Dempsey, B.E. Logan, Efficient recovery of nano-sized iron oxide particles from synthetic acid-mine drainage (AMD) water using fuel cell technologies, Water Res. 45 (2011) 303–307. [14] L. Herrera, P. Ruiz, J.C. Aguillon, A. Fehrmann, A new spectrophotometric method for the determination of ferrous iron in the presence of ferric iron, J. Chem. Technol. Biotechnol. 44 (1989) 171–181. [15] American Public Health Association (APHA), Standard Methods for Examination of Water and Wastewater, 20th ed., Washington, DC, 1998. [16] M. Sun, G.P. Sheng, L. Zhang, C.R. Xia, Z.X. Mu, X.W. Liu, H.L. Wang, H.Q. Yu, R. Qi, T. Yu, M. Yang, An MEC-MFC-coupled system for biohydrogen production from acetate, Environ. Sci. Technol. 42 (2008) 8095–8100. [17] I. Kartio, G. Wittstock, K. Laajalehto, D. Hirsch, J. Simola, T. Laiho, R. Szargan, E. Suoninen, Detection of elemental sulphur on galena oxidized in acidic solution, Int. J. Miner. Process. 51 (1997) 293–301. [18] M.A. Hampton, C. Plackowski, A.V. Nguyen, Physical and chemical analysis of elemental sulfur formation during galena surface oxidation, Langmuir 27 (2011) 4190–4201. [19] E. Tomaszewicz, M. Kurzawa, Use of XPS method in determination of chemical environment and oxidation state of sulfur and silver atoms in Ag6 S3 O4 and Ag8 S4 O4 compounds, J. Mater. Sci. 39 (2004) 2183–2185. [20] H. Liu, B.E. Logan, Electricity generation using an air-cathode single chamber microbial fuel cell in the presence and absence of a proton exchange membrane, Environ. Sci. Technol. 38 (2004) 4040–4046. [21] D. DeBerry, Chemical evolution of liquid redox processes, Environ. Prog. 16 (1997) 193–199. [22] D.W. King, Role of carbonate speciation on the oxidation rate of Fe(II) in aquatic systems, Environ. Sci. Technol. 32 (1998) 2997–3003. [23] J.M. Santana-Casiano, M. Gonzalez-Davila, F.J. Millero, Oxidation of nanomolar levels of Fe(II) with oxygen in natural waters, Environ. Sci. Technol. 39 (2005) 2073–2079. [24] F.J. Millero, W. Yao, J. Archer, The speciation of Fe(II) and Fe(III) in natural waters, Mar. Chem. 50 (1995) 21–39. [25] D.W. King, R. Farlow, Role of carbonate speciation on the oxidation of Fe(II) by H2 O2 , Mar. Chem. 70 (2000) 201–209. [26] J.M. Santana-Casiano, M. Gonzalez-Davila, F.J. Millero, The role of Fe(II) species on the oxidation of Fe(II) in natural waters in the presence of O2 and H2 O2 , Mar. Chem. 99 (2006) 70–82. [27] B. Morgan, O. Lahav, The effect of pH on the kinetics of spontaneous Fe(II) oxidation by O2 in aqueous solution-basic principles and a simple heuristic description, Chemosphere 68 (2007) 2080–2084.