- Email: [email protected]
Removal of hydrogen sulfide with steelmaking slag by concurrent reactions of sulfide mineralization and oxidation
Ecological Engineering 63 (2014) 122–126
Contents lists available at ScienceDirect
Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng
Short communication
Removal of hydrogen sulfide with steelmaking slag by concurrent reactions of sulfide mineralization and oxidation Keisuke Okada a,∗ , Tamiji Yamamoto a , Kyung-Hoi Kim c , Satoshi Asaoka d , Shinjiro Hayakawa b , Kazuhiko Takeda a , Tetsuya Watanabe d , Akio Hayashi e , Yasuhito Miyata e a
Graduate School of Biosphere Science, Hiroshima University, 1-4-4 Kagamiyama, Higashi-Hiroshima, Hiroshima 739-8528, Japan Graduate School of Engineering, Hiroshima University, 1-4-1 Kagamiyama, Higashi-Hiroshima, Hiroshima 739-8527, Japan c Research Center for Inland Seas and Graduate School of Maritime Sciences, Kobe University, 5-1-1 Fukaeminami, Higashinada, Kobe 658-0022, Japan d Res. Lab. JFE Mineral Co., 3-8-2 Shiba, Minato-ku, Tokyo 105-0014, Japan e Steel Res. Lab. JFE Steel Co., 2-2-3 Uchisaiwai-tyo, Chiyoda-ku, Tokyo 100-0011, Japan b
a r t i c l e
i n f o
Article history: Received 22 May 2013 Received in revised form 28 October 2013 Accepted 19 December 2013 Available online 20 January 2014 Keywords: Hydrogen sulfide Mineralization Oxidation Steelmaking slag XAFS
a b s t r a c t This study experimentally revealed the entire mechanisms of mineralization and oxidation of H2 S by steelmaking slag. After adding steelmaking slag to an artificially prepared H2 S solution, white precipitates were generated, and were identified as elemental sulfur by X-ray absorption fine structure spectra. Sulfate ion was also detected in the solution, but in small amounts. In addition, FeS was identified on the surface of the steelmaking slag. An increase in the oxidation–reduction potential implies the oxidation of H2 S by the steelmaking slag, resulting in the formation of oxidized sulfur compounds. We confirmed that steelmaking slag can effectively remove H2 S through oxidation as well as through its mineralization to FeS and that these processes are promoted by both Fe and Mn originating from the slag. © 2013 Elsevier B.V. All rights reserved.
1. Introduction The formation of hypoxic water in the bottom layer of enclosed water bodies has been observed worldwide (Gianni et al., 2011; Wu et al., 2013). In addition to the consumption of dissolved oxygen (DO) due to the decomposition of organic matter, hypoxia of the bottom water can occur due to the consumption of DO by reductants such as hydrogen sulfide (H2 S; Endo and Shigematsu, 2010), which is generated by sulfate-reducing bacteria under hypoxic conditions (Ito, 1996). Because of the acute toxicity of H2 S, the survival rate of benthic organisms is decreased (Marumo and Yokota, 2012). Thus, the generation of H2 S should be suppressed in order to improve the quality of the benthic layer of enclosed water bodies. Steelmaking slag is produced when steel is manufactured from pig iron. It is commonly utilized as a raw material in the form of coarse aggregates for concrete and roadbed construction. Recent studies have shown some of the advantages of steel slag, such as the
∗ Corresponding author. Tel.: +81 82 424 7945; fax: +81 82 424 2459. E-mail address: [email protected] (K. Okada). 0925-8574/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ecoleng.2013.12.016
removal of phosphorus (Li et al., 2013) and wastewater treatment (Wang et al., 2010). The present study evaluates the effect of steelmaking slag on improving the sediment quality by reducing H2 S, which binds to iron, forming FeS and FeS2 (Hayashi et al., 2012a,b). Furthermore, H2 S forms not only FeS but also sulfur (S(0) ), MnS, and sulfate (SO4 2− ) when reacting with steelmaking slag (Kim et al., 2012). The objective of the present study is to understand the primary processes and mechanisms involved in H2 S removal, which may be mineralization and oxidation, using steelmaking slag. 2. Materials and methods 2.1. Steelmaking slag The steelmaking slag had a diameter of 0.85–2.0 mm and was provided by JFE Steel Corporation, Ltd. (Tokyo, Japan). It was mainly composed of SiO2 , CaO, T-Fe, MnO, and Al2 O3 (Table 1). The mineral forms were identified as FeO, CaO–SiO2 –Al2 O3 . Glass was not identified because the slag was not quenched.
K. Okada et al. / Ecological Engineering 63 (2014) 122–126 Table 1 Chemical composition of the steelmaking slag used in the present study. Substance
Average ± SD (%)
SiO2 CaO T-Fe MnO Al2 O3
32.0 31.5 15.6 6.35 4.88
± ± ± ± ±
1.4 0.4 1.2 0.6 0.6
Substance
Average ± SD (%)
MgO P2 O5 TiO2 S Na2 O
4.33 3.42 1.16 0.12 0.73
± ± ± ± ±
0.8 0.1 0.0 0.0 0.1
123
Maidstone, Kent, UK). The samples for the X-ray absorption fine structure (XAFS) analyses were prepared by drying the samples in a nitrogen atmosphere. Iron, manganese, and sulfur K-edge XAFS spectra measurements and their data processing were conducted according to a previous report (Asaoka et al., 2012) using BL3 at the Ritsumeikan SR Center, Japan, and BL11 at the Hiroshima Synchrotron Research Center (HiSOR), respectively. 3. Results
2.2. H2 S removal experiments Experiments were carried out for 1 week using 150 mL glass vials containing 100 mL H2 S solutions with concentrations of 0, 25, 50, 75, 100, and 150 mg S/L in order to observe the differences in the reactions with steelmaking slag at different H2 S concentrations. The test period (1 week) was selected based on the decreasing trend observed in the preliminary experiments (data not shown). The H2 S solution was prepared as follows: 1 M analytical grade, autoclaved Tris–HCl buffer (Wako Pure Chemical Industries, Osaka, Japan) was added to Milli-Q water that had been deaerated with N2 gas to reach a final concentration of 30 mmol/L to maintain a stable pH in the solution. One aliquot of analytical grade Na2 S·9H2 O (Nacalai Tesque, Kyoto, Japan) was then dissolved in the deaerated water. The H2 S solution was slowly dispensed into the vials and 0.2 g steelmaking slag was added gently. Each vial was plugged with a rubber cork and sealed with an aluminum cap after displacing the air in the head space with N2 gas. The vials were then agitated at 40 rpm at 25 ◦ C in a water bath for 1 week, and the concentrations of H2 S in the solution were determined with a detection tube (200SA or 200SB; Komyo Rikagaku Kogyo, Kanagawa, Japan). The pH and oxidation reduction potential (ORP) were measured with a pH electrode (F-22; Horiba, Ltd., Kyoto, Japan) and an ORP electrode (RM-20P; DKKTOA Corporation, Tokyo, Japan), respectively. The ORP value was converted to the Eh value [Eh = ORP + 206 − 0.7(t − 25); t: water temperature (◦ C)]. Control experiments were also carried out without using the steelmaking slag. All experiments were performed in triplicate. The remaining H2 S was scavenged by adding zinc acetate to form ZnS, as described by Ito (1996), to determine the sulfate (SO4 2− ) concentrations. After the precipitates settled, the supernatant solution was filtered through a hydrophilic polyvinylidene difluoride filter with a pore size of 0.45 m (Millex; EMD Millipore Corporation, Billerica, Massachusetts, USA). The pH was then adjusted to neutral, and the SO4 2− concentrations were determined using an ion chromatograph (ICS-200; Dionex, Sunnyvale, CA, USA). The total dissolved Fe and Mn concentrations were determined using inductively coupled plasma-atomic emission spectroscopy (ICP-AES; Optima 7300DV; PerkinElmer, Waltham, MA, USA) for the filtered samples, to which HNO3 was added to obtain a final concentration of 3%. Because two types of precipitates (white and black) were formed in the solution, and color changes in the slag surface were also observed, and additional experiments and analyses were conducted as described below. 2.3. X-ray absorption fine structure analyses and data processing In order to analyze the precipitates and substances formed on the slag surface, which become black, additional experiments were carried out under the same conditions (1 week in a 50 mg S/L solution). The white precipitates formed in the solution were collected through a glass fiber filter (Whatman GF/F; Whatman Plc,
3.1. Results of H2 S removal experiments The concentration of H2 S decreased under all the experimental conditions compared to control (Fig. 1a). Sulfate was detected in small amounts after the experiment (ca. 1/20 of the H2 S concentration), and initial H2 S concentrations of 0, 100, and 150 mg S/L in the solutions, SO4 2− concentrations were higher compared to control (Fig. 1d). During the experiment, the pH values increased by 0.2–0.5 compared to control (Fig. 1b). The Eh values of the slags increased up to +148 mV from −109 mV and +238 mV from −126 mV at initial concentrations of 25 and 50 mg S/L, respectively, while the Eh increase in the control was not statistically significant (Fig. 1c). The Fe and Mn concentrations in the solution determined by ICP-AES were 0.03–0.60 mg/L and 0.25–0.65 mg/L, respectively (Fig. 2). Both the elements behaved similarly, showing high concentrations under low (<50 mg S/L) H2 S concentrations and low concentrations under high (>75 mg S/L) H2 S concentrations. White precipitates were generated in the 25 and 50 mg S/L solutions. In these solutions, no H2 S was detected in either the gas or liquid phases. White precipitates were not observed in the control. Furthermore, a yellowish color was observed in all the experimental solutions, but this color disappeared upon the formation of white precipitates under the conditions with initial H2 S concentrations of 25 and 50 mg S/L. The yellowish color did not disappear under higher (>75 mg S/L) H2 S concentrations. Black substances were also observed on the surfaces of several slag particles. They were easily removed from the slag surface by shaking. 3.2. Results of X-ray absorption fine structure analyses The sulfur K-edge XAFS spectra of the standard substances and samples are shown in Fig. 3a the slag after the experiment showed a peak at 2470 eV, which is identified as that of FeS from the spectra of the standards (Fig. 3a). On the other hand, a peak at 2472 eV was detected for the precipitates, which was identified as S(0) . Furthermore, S(0) and FeS were not observed on the surface of the initial steelmaking slag (Fig. 3a). The iron K-edge XAFS spectra of the standard substances and samples are shown in Fig. 3b. The iron K-edge spectra of the initial slag can be successfully fitted by a combination of FeO 48%, iron (III) hydroxide 18%, and iron (II) hydroxide 34%. On the other hand, the iron K-edge spectra of the slag after the experiment can be successfully fitted by a combination of FeO 74% and iron (III) hydroxide 26%. The manganese K-edge XAFS spectra of the standard substances and samples are shown in Fig. 3c. The manganese K-edge spectra of the initial slag can be successfully fitted by a combination of MnSO4 58%, MnO 24%, and MnO2 18%. On the other hand, the manganese K-edge spectra of the slag after the experiment can successfully be fitted by a combination of initial slag 32%, MnSO4 48%, and MnO 20% with no MnO2 that was observed on the initial slag.
124
K. Okada et al. / Ecological Engineering 63 (2014) 122–126
Fig. 1. Experimental results for hydrogen sulfide removal using steelmaking slag. (䊉) with steelmaking slag, () control (without steelmaking slag). Concentrations of (a) dissolved hydrogen sulfide, (b) pH, (c) Eh, and (d) SO4 2− in the solution after 1 week.
Fig. 2. Experimental results for hydrogen sulfide removal using steelmaking slag. Concentrations of (a) iron and (b) manganese in the solution after 1 week.
4. Discussion In the present study, the amount of H2 S was reduced by the addition of steelmaking slag, resulting in an increase in Eh. The following two processes are assumed to contribute to the increasing Eh: (1) the dissolution of ions from oxides contained in the steelmaking slag, and (2) the removal of H2 S because of its chemical bonding to FeS. Because H2 S itself is a potent reducer, the decreasing H2 S concentrations led to an increase in Eh.
The present sulfur K-edge XAFS analyses revealed that the black substance coating the slag after the experiment was FeS. This is consistent with a previous report that FeS is black (Pelffer et al., 1992). The iron K-edge spectra of the initial slag were successfully fitted with a combination of FeO, iron (III) hydroxide, and iron (II) hydroxide, while those of the slag after the experiment were fitted with a combination of FeO and iron (III) hydroxide (Fig. 3b). In addition, the presence of iron in the solution was confirmed by ICP-AES
K. Okada et al. / Ecological Engineering 63 (2014) 122–126
125
analyses (Fig. 2a). It was assumed iron elution from steelmaking slag was occurred. Iron (II) hydroxide reacts with HS− to afford FeS (Zhang et al., 2009). Therefore, it is possible that FeS was generated in the present study through a reaction between iron (II) hydroxide and HS− . Fe2+ + HS− → FeS ↓ + H+
(1)
Under the initial H2 S concentrations of 25 and 50 mg S/L, the iron ion concentration was three times higher than under high (>75 mg S/L) H2 S concentrations (Fig. 2a). Because iron ion reacts with H2 S to afford FeS, the low iron ion concentration in the vials with high H2 S concentrations is thought to result from consumption through the reaction with H2 S. On the other hand, at lower (<50 mg S/L) H2 S concentrations, dissolved H2 S was completely removed from the solution. Hence, the shortage of sulfides in the solution would have led to an increase in the iron ion concentration. The increasing Eh because of the decreasing content of H2 S by oxidation is assumed to have caused the generation of white precipitates, which were observed only in the solutions with initial H2 S concentrations of 25 and 50 mg S/L. Before applying the XAFS analyses, we conducted X-ray fluorescence (XRF; XRF-1700, Shimadzu, Kyoto, Japan) and scanning electron microscopy-energy dispersive X-ray spectroscopy (SEM-EDX; S-4300, Hitachi, Tokyo, Japan and Apollo XL, Ametek, Paoli, PA, USA) analyses of the precipitates. The precipitates were identified as being composed of 25% S and 26–45% Fe. These precipitates were confirmed to be S(0) by XAFS analysis. Although the Fe compounds were not examined by iron K-edge XAFS, those identified by the XRF and SEM-EDX analyses could be the fragments of iron oxide compounds of the slag and/or FeS formed during the experiments. In the present study, the yellowish color was observed under all experimental conditions, but disappeared upon the formation of the white precipitates when the initial H2 S concentrations were 25 and 50 mg S/L. Polysulfide, which is formed by the oxidation of H2 S (Steudel, 2003), is yellow (Takamatsu et al., 2010). Therefore, it is considered that these results were caused by the oxidation of H2 S into polysulfide and then into S(0) . However, under high (>75 mg S/L) H2 S concentrations, the yellow color of the solutions persisted until the end of the experiments. It is considered that when the initial H2 S concentration is high, not all of the sulfide can be oxidized into sulfur. The manganese K-edge spectra of the initial intact slag showed the existence of MnO2 on the slag surface (Fig. 3c); however, no such MnO2 peak was detected at the end of the experiments. Furthermore, the presence of manganese in the solution was confirmed by the ICP-AES analyses (Fig. 2b). Because MnO2 is an oxide, it can contribute to the increasing Eh; it can also oxidize FeS into S(0) (Schippers and Jorgensen, 2001). Therefore, it is possible that in the present study, the MnO2 contained in the slag oxidized FeS into S(0) , as shown in the following equation: FeS + 1.5MnO2 + 3H+ → 3Fe(OH)3 + S(0) + 1.5Mn2+
(2)
The iron K-edge spectra of the initial intact slag showed the existence of iron (III) hydroxide on the slag surface (Fig. 3b). Therefore, it is assumed that the iron (III) hydroxide was formed because of the low Eh value. It has been reported that Fe3+ reacts with HS− to generate S(0) (Zhang et al., 2009). Thus, S(0) was considered to have been generated by the oxidation of S2− and the concomitant reduction of Fe3+ to Fe2+ (Eq. (3)). 2Fe3+ + S2− → 2Fe2+ + S(0) 2−
Fig. 3. (a) Sulfur K-edge, (b) iron K-edge and (c) manganese K-edge XAFS spectra for steelmaking slag submerged in 50 mg-S/L hydrogen sulfide solution for 1 week. Sulfur K-edge XAFS spectra for the white precipitates were also collected on a glass fiber filter in the solution.
(3)
The formation of SO4 was not the main pathway for removing H2 S, as indicated by the ion chromatography measurements (Fig. 1d). The major pathways were considered to be the formation of S(0) and FeS, although the amounts of these compounds could not
126
K. Okada et al. / Ecological Engineering 63 (2014) 122–126
be quantified because the precipitates are a mixture of these compounds and fragmented slag. The SO4 2− detected in the vials with initial H2 S concentrations of 0 mg S/L probably originated from the steelmaking slag, because it contains sulfate. The present study is similar to that from Kim et al. (2012) with respect to the methodology, and we confirmed many of their results. Although they concluded that H2 S was removed by manganese. On the other hand, the present our study points out as following: (1) Fe also contributed to the chemical reaction on removal of H2 S with formation of FeS, and (2) the generations of polysulfide and sulfur were induced by increasing Eh attributed to the oxidation by Fe and Mn. Therefore, this is the valuable study that shows the entire mechanism of sulfide mineralization and oxidation by both Fe and Mn originating from the steelmaking slag. 5. Conclusions The present study revealed that steelmaking slag can effectively remove H2 S through oxidation into S(0) and mineralization (chemical bonding) with Fe, forming FeS. Based on these results, steelmaking slag can be applied to remediate organically enriched sediments containing large amounts of H2 S in eutrophicated regions. Acknowledgments This work was carried out as part of Joint Research under the Agreement on Comprehensive Study by JFE Steel Corporation and Hiroshima University. Experiments at HiSOR were carried out under the approval of the HSRC Program Advisory Committee (11B-11). XAFS analyses were carried out at the Ritsumeikan SR Center under the approval of the “Nanotechnology Network Japan Program” (S23-15), and we are deeply grateful to Dr. Misaki Katayama. 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.ecoleng. 2013.12.016.
References Asaoka, S., Hayakawa, S., Kim, K., Takeda, K., Katayama, M., Yamamoto, T., 2012. Combined adsorption and oxidation mechanisms of hydrogen sulfide on granulated coal ash. J. Colloid Interface Sci. 377, 284–290. Endo, T., Shigematsu, T., 2010. An experimental study on effect of oxygen supply to the bottom of hypoxic sea on oxygen consumption characteristic and sea bottom environment. J. Japan Soc. Civil Eng. Ser. B2 (Coastal Engineering) 66, 1196–1200. Gianni, A., Kehayias, G., Zacharias, I., 2011. Geomorphology modification and its impact to anoxic lagoons. Ecol. Eng. 37, 1869–1877. Hayashi, A., Asaoka, S., Watanabe, T., Kaneko, R., Takahashi, K., Miyata, Y., Kim, K., Yamamoto, T., Inoue, R., Ariyama, T., 2012a. Mechanism of suppression of sulfide ion in seawater using steelmaking slag. Tetsu-to-Hagane 98, 624–631. Hayashi, A., Watanabe, T., Kaneko, R., Takano, A., Takahashi, K., Miyata, Y., Matsuo, S., Yamamoto, T., Inoue, R., Ariyama, T., 2012b. Decrease of sulfide in enclosed coastal sea by using steelmaking slag. Tetsu-to-Hagane 98, 207–214. Ito, K., 1996. Analytical methods for sulfur compounds and its dynamics in the sediment. In: Samukawa, S., Hiiro, K. (Eds.), The Latest Sediment Analysis Methods and Environmental Dynamics. Gihodo, Tokyo, pp. 77–116. Kim, K., Asaoka, S., Yamamoto, T., Hayakawa, S., Takeda, K., Katayama, M., Onoue, T., 2012. Mechanisms of hydrogen sulfide removal with steel making slag. Environ. Sci. Technol. 46, 10169–10174. Li, H., Li, Y., Gong, Z., Li, X., 2013. Performance study of vertical flow constructed wetlands for phosphorus removal with water quenched slag as a substrate. Ecol. Eng. 53, 39–45. Marumo, K., Yokota, M., 2012. Review on aoshio and biological effects of hydrogen sulfide. Rep. Mar. Ecol. Res. Inst. 15, 23–40. Pelffer, S., dos Santos Afonso, M., Wehrli, B., Gachter, R., 1992. Kinetics and mechanism of the reaction of H2 S with lepidocrocite. Environ. Sci. Technol. 26, 2408–2413. Schippers, A., Jorgensen, B.B., 2001. Oxidation of pyrite and iron sulfide by manganese dioxide in marine sediments. Geochim. Cosmochim. Acta 65, 915–922. Steudel, R., 2003. Inorganic polysulfides Sn2− and radical anions Sn• − . Top. Curr. Chem. 231, 127–152. Takamatsu, N., Nishioka, M., Fukushima, H., Kuwahara, N., 2010. Coloration mechanism of green colored hot spring waters. Balneol. Soc. 60, 119–133. Wang, S., Yang, J., Lou, S., Yang, J., 2010. Wastewater treatment performance of a vermifilter enhancement by a converter slag–coal cinder filter. Ecol. Eng. 36, 489–494. Wu, J., Cui, N., Cheng, S., 2013. Effects of sediment anoxia on growth and root respiratory metabolism of Iris pseudacorus: implications for vegetation restoration of eutrophic waters in China. Ecol. Eng. 53, 194–199. Zhang, L., Keller, J., Yuan, Z., 2009. Inhibition of sulfate-reducing and methanogenic activities of anaerobic sewer biofilms by ferric iron dosing. Water Res. 43, 4123–4132.
Contents lists available at ScienceDirect
Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng
Short communication
Removal of hydrogen sulfide with steelmaking slag by concurrent reactions of sulfide mineralization and oxidation Keisuke Okada a,∗ , Tamiji Yamamoto a , Kyung-Hoi Kim c , Satoshi Asaoka d , Shinjiro Hayakawa b , Kazuhiko Takeda a , Tetsuya Watanabe d , Akio Hayashi e , Yasuhito Miyata e a
Graduate School of Biosphere Science, Hiroshima University, 1-4-4 Kagamiyama, Higashi-Hiroshima, Hiroshima 739-8528, Japan Graduate School of Engineering, Hiroshima University, 1-4-1 Kagamiyama, Higashi-Hiroshima, Hiroshima 739-8527, Japan c Research Center for Inland Seas and Graduate School of Maritime Sciences, Kobe University, 5-1-1 Fukaeminami, Higashinada, Kobe 658-0022, Japan d Res. Lab. JFE Mineral Co., 3-8-2 Shiba, Minato-ku, Tokyo 105-0014, Japan e Steel Res. Lab. JFE Steel Co., 2-2-3 Uchisaiwai-tyo, Chiyoda-ku, Tokyo 100-0011, Japan b
a r t i c l e
i n f o
Article history: Received 22 May 2013 Received in revised form 28 October 2013 Accepted 19 December 2013 Available online 20 January 2014 Keywords: Hydrogen sulfide Mineralization Oxidation Steelmaking slag XAFS
a b s t r a c t This study experimentally revealed the entire mechanisms of mineralization and oxidation of H2 S by steelmaking slag. After adding steelmaking slag to an artificially prepared H2 S solution, white precipitates were generated, and were identified as elemental sulfur by X-ray absorption fine structure spectra. Sulfate ion was also detected in the solution, but in small amounts. In addition, FeS was identified on the surface of the steelmaking slag. An increase in the oxidation–reduction potential implies the oxidation of H2 S by the steelmaking slag, resulting in the formation of oxidized sulfur compounds. We confirmed that steelmaking slag can effectively remove H2 S through oxidation as well as through its mineralization to FeS and that these processes are promoted by both Fe and Mn originating from the slag. © 2013 Elsevier B.V. All rights reserved.
1. Introduction The formation of hypoxic water in the bottom layer of enclosed water bodies has been observed worldwide (Gianni et al., 2011; Wu et al., 2013). In addition to the consumption of dissolved oxygen (DO) due to the decomposition of organic matter, hypoxia of the bottom water can occur due to the consumption of DO by reductants such as hydrogen sulfide (H2 S; Endo and Shigematsu, 2010), which is generated by sulfate-reducing bacteria under hypoxic conditions (Ito, 1996). Because of the acute toxicity of H2 S, the survival rate of benthic organisms is decreased (Marumo and Yokota, 2012). Thus, the generation of H2 S should be suppressed in order to improve the quality of the benthic layer of enclosed water bodies. Steelmaking slag is produced when steel is manufactured from pig iron. It is commonly utilized as a raw material in the form of coarse aggregates for concrete and roadbed construction. Recent studies have shown some of the advantages of steel slag, such as the
∗ Corresponding author. Tel.: +81 82 424 7945; fax: +81 82 424 2459. E-mail address: [email protected] (K. Okada). 0925-8574/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ecoleng.2013.12.016
removal of phosphorus (Li et al., 2013) and wastewater treatment (Wang et al., 2010). The present study evaluates the effect of steelmaking slag on improving the sediment quality by reducing H2 S, which binds to iron, forming FeS and FeS2 (Hayashi et al., 2012a,b). Furthermore, H2 S forms not only FeS but also sulfur (S(0) ), MnS, and sulfate (SO4 2− ) when reacting with steelmaking slag (Kim et al., 2012). The objective of the present study is to understand the primary processes and mechanisms involved in H2 S removal, which may be mineralization and oxidation, using steelmaking slag. 2. Materials and methods 2.1. Steelmaking slag The steelmaking slag had a diameter of 0.85–2.0 mm and was provided by JFE Steel Corporation, Ltd. (Tokyo, Japan). It was mainly composed of SiO2 , CaO, T-Fe, MnO, and Al2 O3 (Table 1). The mineral forms were identified as FeO, CaO–SiO2 –Al2 O3 . Glass was not identified because the slag was not quenched.
K. Okada et al. / Ecological Engineering 63 (2014) 122–126 Table 1 Chemical composition of the steelmaking slag used in the present study. Substance
Average ± SD (%)
SiO2 CaO T-Fe MnO Al2 O3
32.0 31.5 15.6 6.35 4.88
± ± ± ± ±
1.4 0.4 1.2 0.6 0.6
Substance
Average ± SD (%)
MgO P2 O5 TiO2 S Na2 O
4.33 3.42 1.16 0.12 0.73
± ± ± ± ±
0.8 0.1 0.0 0.0 0.1
123
Maidstone, Kent, UK). The samples for the X-ray absorption fine structure (XAFS) analyses were prepared by drying the samples in a nitrogen atmosphere. Iron, manganese, and sulfur K-edge XAFS spectra measurements and their data processing were conducted according to a previous report (Asaoka et al., 2012) using BL3 at the Ritsumeikan SR Center, Japan, and BL11 at the Hiroshima Synchrotron Research Center (HiSOR), respectively. 3. Results
2.2. H2 S removal experiments Experiments were carried out for 1 week using 150 mL glass vials containing 100 mL H2 S solutions with concentrations of 0, 25, 50, 75, 100, and 150 mg S/L in order to observe the differences in the reactions with steelmaking slag at different H2 S concentrations. The test period (1 week) was selected based on the decreasing trend observed in the preliminary experiments (data not shown). The H2 S solution was prepared as follows: 1 M analytical grade, autoclaved Tris–HCl buffer (Wako Pure Chemical Industries, Osaka, Japan) was added to Milli-Q water that had been deaerated with N2 gas to reach a final concentration of 30 mmol/L to maintain a stable pH in the solution. One aliquot of analytical grade Na2 S·9H2 O (Nacalai Tesque, Kyoto, Japan) was then dissolved in the deaerated water. The H2 S solution was slowly dispensed into the vials and 0.2 g steelmaking slag was added gently. Each vial was plugged with a rubber cork and sealed with an aluminum cap after displacing the air in the head space with N2 gas. The vials were then agitated at 40 rpm at 25 ◦ C in a water bath for 1 week, and the concentrations of H2 S in the solution were determined with a detection tube (200SA or 200SB; Komyo Rikagaku Kogyo, Kanagawa, Japan). The pH and oxidation reduction potential (ORP) were measured with a pH electrode (F-22; Horiba, Ltd., Kyoto, Japan) and an ORP electrode (RM-20P; DKKTOA Corporation, Tokyo, Japan), respectively. The ORP value was converted to the Eh value [Eh = ORP + 206 − 0.7(t − 25); t: water temperature (◦ C)]. Control experiments were also carried out without using the steelmaking slag. All experiments were performed in triplicate. The remaining H2 S was scavenged by adding zinc acetate to form ZnS, as described by Ito (1996), to determine the sulfate (SO4 2− ) concentrations. After the precipitates settled, the supernatant solution was filtered through a hydrophilic polyvinylidene difluoride filter with a pore size of 0.45 m (Millex; EMD Millipore Corporation, Billerica, Massachusetts, USA). The pH was then adjusted to neutral, and the SO4 2− concentrations were determined using an ion chromatograph (ICS-200; Dionex, Sunnyvale, CA, USA). The total dissolved Fe and Mn concentrations were determined using inductively coupled plasma-atomic emission spectroscopy (ICP-AES; Optima 7300DV; PerkinElmer, Waltham, MA, USA) for the filtered samples, to which HNO3 was added to obtain a final concentration of 3%. Because two types of precipitates (white and black) were formed in the solution, and color changes in the slag surface were also observed, and additional experiments and analyses were conducted as described below. 2.3. X-ray absorption fine structure analyses and data processing In order to analyze the precipitates and substances formed on the slag surface, which become black, additional experiments were carried out under the same conditions (1 week in a 50 mg S/L solution). The white precipitates formed in the solution were collected through a glass fiber filter (Whatman GF/F; Whatman Plc,
3.1. Results of H2 S removal experiments The concentration of H2 S decreased under all the experimental conditions compared to control (Fig. 1a). Sulfate was detected in small amounts after the experiment (ca. 1/20 of the H2 S concentration), and initial H2 S concentrations of 0, 100, and 150 mg S/L in the solutions, SO4 2− concentrations were higher compared to control (Fig. 1d). During the experiment, the pH values increased by 0.2–0.5 compared to control (Fig. 1b). The Eh values of the slags increased up to +148 mV from −109 mV and +238 mV from −126 mV at initial concentrations of 25 and 50 mg S/L, respectively, while the Eh increase in the control was not statistically significant (Fig. 1c). The Fe and Mn concentrations in the solution determined by ICP-AES were 0.03–0.60 mg/L and 0.25–0.65 mg/L, respectively (Fig. 2). Both the elements behaved similarly, showing high concentrations under low (<50 mg S/L) H2 S concentrations and low concentrations under high (>75 mg S/L) H2 S concentrations. White precipitates were generated in the 25 and 50 mg S/L solutions. In these solutions, no H2 S was detected in either the gas or liquid phases. White precipitates were not observed in the control. Furthermore, a yellowish color was observed in all the experimental solutions, but this color disappeared upon the formation of white precipitates under the conditions with initial H2 S concentrations of 25 and 50 mg S/L. The yellowish color did not disappear under higher (>75 mg S/L) H2 S concentrations. Black substances were also observed on the surfaces of several slag particles. They were easily removed from the slag surface by shaking. 3.2. Results of X-ray absorption fine structure analyses The sulfur K-edge XAFS spectra of the standard substances and samples are shown in Fig. 3a the slag after the experiment showed a peak at 2470 eV, which is identified as that of FeS from the spectra of the standards (Fig. 3a). On the other hand, a peak at 2472 eV was detected for the precipitates, which was identified as S(0) . Furthermore, S(0) and FeS were not observed on the surface of the initial steelmaking slag (Fig. 3a). The iron K-edge XAFS spectra of the standard substances and samples are shown in Fig. 3b. The iron K-edge spectra of the initial slag can be successfully fitted by a combination of FeO 48%, iron (III) hydroxide 18%, and iron (II) hydroxide 34%. On the other hand, the iron K-edge spectra of the slag after the experiment can be successfully fitted by a combination of FeO 74% and iron (III) hydroxide 26%. The manganese K-edge XAFS spectra of the standard substances and samples are shown in Fig. 3c. The manganese K-edge spectra of the initial slag can be successfully fitted by a combination of MnSO4 58%, MnO 24%, and MnO2 18%. On the other hand, the manganese K-edge spectra of the slag after the experiment can successfully be fitted by a combination of initial slag 32%, MnSO4 48%, and MnO 20% with no MnO2 that was observed on the initial slag.
124
K. Okada et al. / Ecological Engineering 63 (2014) 122–126
Fig. 1. Experimental results for hydrogen sulfide removal using steelmaking slag. (䊉) with steelmaking slag, () control (without steelmaking slag). Concentrations of (a) dissolved hydrogen sulfide, (b) pH, (c) Eh, and (d) SO4 2− in the solution after 1 week.
Fig. 2. Experimental results for hydrogen sulfide removal using steelmaking slag. Concentrations of (a) iron and (b) manganese in the solution after 1 week.
4. Discussion In the present study, the amount of H2 S was reduced by the addition of steelmaking slag, resulting in an increase in Eh. The following two processes are assumed to contribute to the increasing Eh: (1) the dissolution of ions from oxides contained in the steelmaking slag, and (2) the removal of H2 S because of its chemical bonding to FeS. Because H2 S itself is a potent reducer, the decreasing H2 S concentrations led to an increase in Eh.
The present sulfur K-edge XAFS analyses revealed that the black substance coating the slag after the experiment was FeS. This is consistent with a previous report that FeS is black (Pelffer et al., 1992). The iron K-edge spectra of the initial slag were successfully fitted with a combination of FeO, iron (III) hydroxide, and iron (II) hydroxide, while those of the slag after the experiment were fitted with a combination of FeO and iron (III) hydroxide (Fig. 3b). In addition, the presence of iron in the solution was confirmed by ICP-AES
K. Okada et al. / Ecological Engineering 63 (2014) 122–126
125
analyses (Fig. 2a). It was assumed iron elution from steelmaking slag was occurred. Iron (II) hydroxide reacts with HS− to afford FeS (Zhang et al., 2009). Therefore, it is possible that FeS was generated in the present study through a reaction between iron (II) hydroxide and HS− . Fe2+ + HS− → FeS ↓ + H+
(1)
Under the initial H2 S concentrations of 25 and 50 mg S/L, the iron ion concentration was three times higher than under high (>75 mg S/L) H2 S concentrations (Fig. 2a). Because iron ion reacts with H2 S to afford FeS, the low iron ion concentration in the vials with high H2 S concentrations is thought to result from consumption through the reaction with H2 S. On the other hand, at lower (<50 mg S/L) H2 S concentrations, dissolved H2 S was completely removed from the solution. Hence, the shortage of sulfides in the solution would have led to an increase in the iron ion concentration. The increasing Eh because of the decreasing content of H2 S by oxidation is assumed to have caused the generation of white precipitates, which were observed only in the solutions with initial H2 S concentrations of 25 and 50 mg S/L. Before applying the XAFS analyses, we conducted X-ray fluorescence (XRF; XRF-1700, Shimadzu, Kyoto, Japan) and scanning electron microscopy-energy dispersive X-ray spectroscopy (SEM-EDX; S-4300, Hitachi, Tokyo, Japan and Apollo XL, Ametek, Paoli, PA, USA) analyses of the precipitates. The precipitates were identified as being composed of 25% S and 26–45% Fe. These precipitates were confirmed to be S(0) by XAFS analysis. Although the Fe compounds were not examined by iron K-edge XAFS, those identified by the XRF and SEM-EDX analyses could be the fragments of iron oxide compounds of the slag and/or FeS formed during the experiments. In the present study, the yellowish color was observed under all experimental conditions, but disappeared upon the formation of the white precipitates when the initial H2 S concentrations were 25 and 50 mg S/L. Polysulfide, which is formed by the oxidation of H2 S (Steudel, 2003), is yellow (Takamatsu et al., 2010). Therefore, it is considered that these results were caused by the oxidation of H2 S into polysulfide and then into S(0) . However, under high (>75 mg S/L) H2 S concentrations, the yellow color of the solutions persisted until the end of the experiments. It is considered that when the initial H2 S concentration is high, not all of the sulfide can be oxidized into sulfur. The manganese K-edge spectra of the initial intact slag showed the existence of MnO2 on the slag surface (Fig. 3c); however, no such MnO2 peak was detected at the end of the experiments. Furthermore, the presence of manganese in the solution was confirmed by the ICP-AES analyses (Fig. 2b). Because MnO2 is an oxide, it can contribute to the increasing Eh; it can also oxidize FeS into S(0) (Schippers and Jorgensen, 2001). Therefore, it is possible that in the present study, the MnO2 contained in the slag oxidized FeS into S(0) , as shown in the following equation: FeS + 1.5MnO2 + 3H+ → 3Fe(OH)3 + S(0) + 1.5Mn2+
(2)
The iron K-edge spectra of the initial intact slag showed the existence of iron (III) hydroxide on the slag surface (Fig. 3b). Therefore, it is assumed that the iron (III) hydroxide was formed because of the low Eh value. It has been reported that Fe3+ reacts with HS− to generate S(0) (Zhang et al., 2009). Thus, S(0) was considered to have been generated by the oxidation of S2− and the concomitant reduction of Fe3+ to Fe2+ (Eq. (3)). 2Fe3+ + S2− → 2Fe2+ + S(0) 2−
Fig. 3. (a) Sulfur K-edge, (b) iron K-edge and (c) manganese K-edge XAFS spectra for steelmaking slag submerged in 50 mg-S/L hydrogen sulfide solution for 1 week. Sulfur K-edge XAFS spectra for the white precipitates were also collected on a glass fiber filter in the solution.
(3)
The formation of SO4 was not the main pathway for removing H2 S, as indicated by the ion chromatography measurements (Fig. 1d). The major pathways were considered to be the formation of S(0) and FeS, although the amounts of these compounds could not
126
K. Okada et al. / Ecological Engineering 63 (2014) 122–126
be quantified because the precipitates are a mixture of these compounds and fragmented slag. The SO4 2− detected in the vials with initial H2 S concentrations of 0 mg S/L probably originated from the steelmaking slag, because it contains sulfate. The present study is similar to that from Kim et al. (2012) with respect to the methodology, and we confirmed many of their results. Although they concluded that H2 S was removed by manganese. On the other hand, the present our study points out as following: (1) Fe also contributed to the chemical reaction on removal of H2 S with formation of FeS, and (2) the generations of polysulfide and sulfur were induced by increasing Eh attributed to the oxidation by Fe and Mn. Therefore, this is the valuable study that shows the entire mechanism of sulfide mineralization and oxidation by both Fe and Mn originating from the steelmaking slag. 5. Conclusions The present study revealed that steelmaking slag can effectively remove H2 S through oxidation into S(0) and mineralization (chemical bonding) with Fe, forming FeS. Based on these results, steelmaking slag can be applied to remediate organically enriched sediments containing large amounts of H2 S in eutrophicated regions. Acknowledgments This work was carried out as part of Joint Research under the Agreement on Comprehensive Study by JFE Steel Corporation and Hiroshima University. Experiments at HiSOR were carried out under the approval of the HSRC Program Advisory Committee (11B-11). XAFS analyses were carried out at the Ritsumeikan SR Center under the approval of the “Nanotechnology Network Japan Program” (S23-15), and we are deeply grateful to Dr. Misaki Katayama. 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.ecoleng. 2013.12.016.
References Asaoka, S., Hayakawa, S., Kim, K., Takeda, K., Katayama, M., Yamamoto, T., 2012. Combined adsorption and oxidation mechanisms of hydrogen sulfide on granulated coal ash. J. Colloid Interface Sci. 377, 284–290. Endo, T., Shigematsu, T., 2010. An experimental study on effect of oxygen supply to the bottom of hypoxic sea on oxygen consumption characteristic and sea bottom environment. J. Japan Soc. Civil Eng. Ser. B2 (Coastal Engineering) 66, 1196–1200. Gianni, A., Kehayias, G., Zacharias, I., 2011. Geomorphology modification and its impact to anoxic lagoons. Ecol. Eng. 37, 1869–1877. Hayashi, A., Asaoka, S., Watanabe, T., Kaneko, R., Takahashi, K., Miyata, Y., Kim, K., Yamamoto, T., Inoue, R., Ariyama, T., 2012a. Mechanism of suppression of sulfide ion in seawater using steelmaking slag. Tetsu-to-Hagane 98, 624–631. Hayashi, A., Watanabe, T., Kaneko, R., Takano, A., Takahashi, K., Miyata, Y., Matsuo, S., Yamamoto, T., Inoue, R., Ariyama, T., 2012b. Decrease of sulfide in enclosed coastal sea by using steelmaking slag. Tetsu-to-Hagane 98, 207–214. Ito, K., 1996. Analytical methods for sulfur compounds and its dynamics in the sediment. In: Samukawa, S., Hiiro, K. (Eds.), The Latest Sediment Analysis Methods and Environmental Dynamics. Gihodo, Tokyo, pp. 77–116. Kim, K., Asaoka, S., Yamamoto, T., Hayakawa, S., Takeda, K., Katayama, M., Onoue, T., 2012. Mechanisms of hydrogen sulfide removal with steel making slag. Environ. Sci. Technol. 46, 10169–10174. Li, H., Li, Y., Gong, Z., Li, X., 2013. Performance study of vertical flow constructed wetlands for phosphorus removal with water quenched slag as a substrate. Ecol. Eng. 53, 39–45. Marumo, K., Yokota, M., 2012. Review on aoshio and biological effects of hydrogen sulfide. Rep. Mar. Ecol. Res. Inst. 15, 23–40. Pelffer, S., dos Santos Afonso, M., Wehrli, B., Gachter, R., 1992. Kinetics and mechanism of the reaction of H2 S with lepidocrocite. Environ. Sci. Technol. 26, 2408–2413. Schippers, A., Jorgensen, B.B., 2001. Oxidation of pyrite and iron sulfide by manganese dioxide in marine sediments. Geochim. Cosmochim. Acta 65, 915–922. Steudel, R., 2003. Inorganic polysulfides Sn2− and radical anions Sn• − . Top. Curr. Chem. 231, 127–152. Takamatsu, N., Nishioka, M., Fukushima, H., Kuwahara, N., 2010. Coloration mechanism of green colored hot spring waters. Balneol. Soc. 60, 119–133. Wang, S., Yang, J., Lou, S., Yang, J., 2010. Wastewater treatment performance of a vermifilter enhancement by a converter slag–coal cinder filter. Ecol. Eng. 36, 489–494. Wu, J., Cui, N., Cheng, S., 2013. Effects of sediment anoxia on growth and root respiratory metabolism of Iris pseudacorus: implications for vegetation restoration of eutrophic waters in China. Ecol. Eng. 53, 194–199. Zhang, L., Keller, J., Yuan, Z., 2009. Inhibition of sulfate-reducing and methanogenic activities of anaerobic sewer biofilms by ferric iron dosing. Water Res. 43, 4123–4132.