Chemical Science Research Frontiers 2021 90 Mercury (Hg) is a global pollutant, and the significant adverse impacts caused by Hg and its compounds are of great concern. The Minamata Convention came into effect in August 2017 with the aim of protecting human health and the environment from the anthropogenic emission and release of Hg. Activated carbon (AC) is often used to remove Hg from various emission and release sources, but the capacity of virgin AC (AC-virgin) is limited. Because of the high cost of AC, it is often modified to enhance its capacity and for regeneration. AC impregnated with halogens exhibits excellent Hg 0 removal performance, and iodine-impregnated AC (I-AC) has high Hg 0 removal efficiency, low activation energy, and low volatility of the adsorption product, HgI 2 , which increases the stability of the spent sorbent, compared with AC impregnated with other halogens such as Br and Cl. The mechanism of the adsorption of Hg 0 by I-ACs, such as potassium iodide (KI)-impregnated AC, is via I 2 , the formation of which is the key step in the chemisorption of Hg 0 , as it reacts with Hg 0 to produce HgI 2 . I 2 formation has been proposed to occur via the following reaction: 2 KI + 1/2 O 2 = K 2 O + I 2 . However, starch–iodine tests showed that negligible I 2 is formed at temperatures lower than 300ºC, and the Gibbs free energy of the proposed reaction has a positive value even at 1,000ºC. Thus, whether I 2 can actually be formed in I-ACs remains unclear. To elucidate the mechanism of Hg 0 adsorption by I-ACs, especially with regard to the occurrence and formation of I 2 , we prepared a series of I-ACs with different iodine precursors including KI, NH 4 I, and KIO 3 via one-step impregnation, denoted as AC-KI, AC- NH 4 I, and AC-KIO 3 , respectively [1]. Comparing AC-KI with AC-NH 4 I and AC-KIO 3 , we analyzed the effects of the cation element and valence state of iodine on Hg 0 removal. In an attempt to impregnate the AC with I 2 by the reaction 2 CuSO 4 (aq) + 4 KI(aq) = 2 CuI(s) + I 2 (s) + 2 K 2 SO 4 (aq), we conducted the co-impregnation of KI followed by CuSO 4 . The product is denoted as AC-KI + CuSO 4 . AC-virgin without any precursors was used as the reference material for evaluating the Hg 0 removal performance. We conducted X-ray absorption near-edge structure analysis (XANES) for iodine and Hg at SPring-8 BL01B1 . The iodine XANES spectra (Fig. 1) of AC-KI, AC-NH 4 I, and AC-KI + CuSO 4 before and after the adsorption test were very similar. No similarity was detected between the sorbents and their precursors, KI, NH 4 I, and CuI. As the spectrum of I 2 contained no distinct features (Fig. 1), we could not determine whether I 2 was present in the I-AC samples from only the iodine K -edge spectra. Thus, we conducted the linear combination fitting (LCF) of the XANES data for I-ACs prior to the adsorption test, applying principal component analysis (PCA) to exclude irrelevant reference materials from the spectra. I 2 accounted for approximately 9 – 16% of total iodine in these sorbents. Notably, the ranking of I 2 fractions among the I-ACs was consistent with the order of Hg 0 removal efficiency, which was as follows: AC -KI + CuSO 4 ≈ AC-NH 4 I > AC-KIO 3 > AC-KI. Therefore, I 2 was confirmed to play an important role in Hg 0 removal by I-ACs. The formation of I 2 may occur during impregnation, drying (before the adsorption test), and adsorption. However, our results indicated no I 2 Fig. 1. Iodine K -edge X-ray absorption near- edge structure (XANES) spectra for reference materials (solid KI, solid NH 4 I, solid CuI, solid KIO 3, solid I 2 , I 3 – ion, solid HgI, and solid HgI 2 ) and impregnated ACs (AC-KI, AC-NH 4 I, AC-KI+CuSO 4 , and AC-KIO 3 ) before and after adsorption test (E 0 = 33,172 – 33,174 eV). Identification of chemical species of iodine and mercury on iodine-impregnated activated carbon using X-ray absorption near-edge structure analysis 33150 33170 33190 33210 33230 33250 Photon Energy (eV) Normalized Absorption (–) HgI 2 I 3 – I 2 KIO 3 KI NH 4 I AC-KI+ CuSO 4 (before) AC-KI (before) AC-KI (after) AC-NH 4 I (before) AC-NH 4 I (after) AC-KI+ CuSO 4 (after) AC-KIO 3 (before) AC-KIO 3 (after) CuI HgI Research Frontiers 2021 91 in AC-KI; instead, I 3 – and KI were detected. The LCF results revealed larger proportions of I 3 – in all I-ACs. Notably, sorbents with greater Hg 0 removal efficiency, namely, AC-KI + CuSO 4 and AC-NH 4 I, contained larger fractions of I 3 – than did the other sorbents. Thus, in addition to I 2 , I 3 – may serve as an I 2 donor and play an essential role in Hg 0 removal by I-ACs. Moreover, as the reaction of I 3 – with starch produces a black-blue color, I 3 – formation may have improved the Hg 0 removal performance in previous studies, in which KI was used as the impregnation precursor and the starch test was employed for I 2 detection. We investigated the Hg species present after I-AC adsorption, which was assumed to be HgI 2 in previous research. The Hg L III -edge XANES analysis of sorbents was conducted after the Hg 0 adsorption test to determine the chemical species of Hg (Fig. 2). Two references, HgCl 2 and HgSO 4 , produced a marked shoulder peak at 12,282 eV; this shoulder peak was also produced by AC-virgin, which contains Cl and S, but not by the I-ACs. Therefore, HgCl 2 and HgSO 4 may have been present in AC-virgin, not adsorption products of I-ACs. LCF was conducted after PCA and target transformation to confirm the species of Hg in the spent sorbents and their ratios. The main species of Hg in I-ACs in this study were HgI 2 and HgI, with a higher proportion of HgI 2 than of HgI. The mechanism of HgI 2 generation in I-ACs, according to the adsorption test results, can be explained as follows. In I-ACs, HgI is more readily formed than HgCl because of its lower activation energy. This HgI can be further oxidized into HgI 2 by I 2 , which is released from I 3 – . HgI 2 can also be formed via the direct oxidation of Hg 0 by I 2 . Furthermore, HgI 2 is more stable than HgCl 2 , which is in accord with our finding that HgI 2 , rather than HgCl 2 , dominates the I-AC samples. In this study, the main Hg removal mechanism was clear from both the I- K and Hg- L III -edge XANES spectra. The results were supported by those of other analyses such as X-ray diffraction and X-ray photoelectron spectroscopy. Future research should focus on the stability and regeneration of the spent sorbents. Yingchao Cheng a,b , Kenji Shiota b and Masaki Takaoka b, * a National Institute for Environmental Studies b Department of Environmental Engineering, Kyoto University *Email: takaoka.masaki.4w@kyoto-u.ac.jp References [1] Y. Cheng, K. Shiota, T. Kusakabe, K. Oshita, M. Takaoka: Chem. Eng. J. 402 (2020)126225. Fig. 2. Mercury L III -edge XANES spectra for reference materials (metallic mercury, solid Hg 2Cl 2, solid HgCl 2 , solid HgI 2 , solid HgI, solid HgO [yellow], solid HgS [black], solid Hg 3 S 2 Cl 2 , and solid HgSO 4 ), AC-virgin, and impregnated ACs (AC-KI, AC-NH 4 I, AC-KIO 3 , and AC-KI + CuSO 4 ) after adsorption test (E 0 = 12,284 eV). 12240 12260 12280 12300 12320 12340 Photon Energy (eV) Normalized Absorption (–) HgSO 4 HgS (black) HgO (yellow) HgI AC-virgin (after) AC-KIO 3 (after) Hg (metal) AC-NH 4 I (after) AC-KI (after) AC-KI+ CuSO 4 (after) HgI 2 HgCI 2 Hg 2 CI2 Hg 3 S 2 Cl 2
home
- JP
- EN