Removal of Gaseous Hydrogen Sulfide by a FeOCl/H2O2 Wet Oxidation System.

The removal of gaseous hydrogen sulfide using FeOCl/H2O2 was studied. The effects of the FeOCl dosage, the H2O2 concentration, the reaction temperature, and the gas flow rate on the removal of H2S were investigated. The reaction products were analyzed, and the characterization of FeOCl was carried out by X-ray diffraction, scanning electron microscopy, X-ray photoelectron spectroscopy, and electron paramagnetic resonance spectroscopy. Furthermore, radical quenching experiments were carried out using butylated hydroxytoluene, isopropanol, and benzoquinone. It was found that the H2S removal rate for a H2S gas concentration of 160 ppm reached 85.6% when bubbling through 100 mL of an aqueous solution containing FeOCl (1 g/L) and H2O2 (0.33 mol/L) at 293 K with a flow rate of 135 mL/min. Although the dissolution of chlorine in FeOCl was found to result in reduced catalytic performance, the activity was restored after soaking the catalyst in concentrated hydrochloric acid (37%) and subsequent calcination. The mechanism of H2S removal was also discussed, and it was found that this process was controlled by H2S diffusion. FeOCl was found to activate H2O2 and produce radicals, such as •OH and •O2 -, resulting in the formation of a water film rich in radicals on the FeOCl surface. Following the diffusion of H2S into the water film, it underwent oxidation by radicals to produce SO4 2-. Overall, the catalyst and the product can be effectively separated.

Introduction

Hydrogen sulfide (H2S) is a colorless, flammable, and toxic gas with the smell of rotten eggs, and it is a common impurity in natural gas and syngas. Upon its combustion, H2S produces sulfur dioxide, which is released into the atmosphere and leads to air pollution.[1] In addition, in industry, the acidity and corrosiveness of H2S can cause serious corrosion to equipment and pipelines, while in daily life, an atmospheric concentration of >500 ppm can lead to a partial loss of human neurological function and can even be fatal.[1−4]

Thus, to reduce H2S emissions, a large number of H2S removal technologies have been developed to date, including adsorption/catalytic desulfurization,[5−17] biodesulfurization,[18] membrane separation,[19,20] and washing desulfurization.[21−31] More specifically, the adsorption desulfurization process is simple and the level of investment required is low; however, the regeneration energy consumption is high, and the adsorbent is easily deactivated.[3,4] In contrast, the biological desulfurization technology can operate at room temperature and pressure, its operating costs are low, and it does not tend to cause secondary pollution; however, in the presence of high concentrations of H2S, treatment is slow, and the process efficiency tends to be poor.[18] In addition, the membrane separation technology is easily operated and boasts both a high separation efficiency and a low level of pollution; however, the membranes exhibit poor corrosion resistances and short service lives, in addition to being particularly expensive.[19,20] In terms of the chemical washing desulfurization technology, the operation is also simple, and this process exhibits strong adaptability and a high desulfurization rate, resulting in it being the most widely used and mature technology at present.[3,21,23] Nonetheless, the most commonly used traditional absorbents for this process, including ethanolamine and carbonate, require high amounts of energy for their regeneration.[3] In contrast, the wet (solution) oxidation desulfurization technology does not suffer from issues related to absorbent regeneration, the process and equipment required are simple, and a high sulfur recovery tends to be achieved. As a result, this technology has attracted growing attention in recent years.[3,22,24−30] Based on these advantages, Liu et al.[26−30,32] reported the application of the wet oxidation desulfurization technology in an advanced oxidation system. More specifically, hydrogen peroxide, persulfate, and Oxone were used as oxidants, while Fe3+, Cu2+, Fe2+, and UV irradiation were used to promote the production of radicals. Hydrogen peroxide is a low-cost, widely used chemical that does not require complex safety measures.[3] Moreover, Fe3+, Cu2+, Fe2+, and hydrogen peroxide form a homogeneous Fenton or Fenton-like oxidation system, which can effectively remove gaseous H2S.[26,27] However, the recovery of Fe3+, Cu2+, and Fe2+ is challenging, and during evaporation crystallization, wherein the final oxidation product (sulfuric acid) is recovered by the addition of ammonia to produce ammonium sulfate,[28,30] the product can become contaminated with these metal elements.

Compared with the homogeneous Fenton system, the heterogeneous Fenton system allows the more facile recovery of the catalyst. Among the various catalysts reported to date for this process, iron oxychloride (FeOCl) has been widely studied in terms of the degradation of organic pollutants present in wastewater due to its high catalytic activity, low raw material costs, and its simple preparation and regeneration.[31,33−39] However, to date, the removal of gaseous H2S by the FeOCl/H2O2 system has yet to be examined.

Thus, we, herein, report a novel desulfurization method wherein gaseous H2S is removed by bubbling in a FeOCl/H2O2 wet oxidation system. The technical feasibility of this new technology and the effects of various process parameters (e.g., the FeOCl dosage, the H2O2 concentration, the reaction temperature, and the gas flow rate) on the removal of H2S are studied. In addition, the reaction products and the reaction mechanism of H2S removal are studied systematically. It is expected that the results of this study will provide the necessary theoretical basis and data for the development of new H2S removal technologies and will promote the practical application of the advanced oxidation technology in the field of gaseous H2S removal.

Materials and Methods

All analytically pure reagents were provided by Aladdin Industrial Corporation (Shanghai, China), and were used directly as received without any additional treatment.

Catalyst Preparation

FeOCl was prepared by thermal decomposition as described previously.[38,40] More specifically, ferric trichloride (FeCl3·6H2O) was ground in an agate mortar, spread flat on the bottom of a crucible (thickness ≤ 3 mm), and heated to 220 °C at a rate of 10 °C/min in a muffle furnace. After maintaining this temperature for 2.5 h, the crucible was allowed to cool to room temperature, then washed with deionized water, and finally dried overnight at 60 °C in a vacuum drying box to obtain the desired FeOCl as a dark fuchsia powder.

H2S Removal

As outlined in Figure , the experimental device employed herein consists of a raw material gas supply system, a reaction system, a gas absorption system, and a waste gas treatment system. The raw gas supply system is composed of 300 ppm H2S/N2 (1), high-purity N2 (2), and a gas mixing system (3–5). The reaction system is mainly composed of three flasks (7) and a heat-collecting magnetic agitator (10), while the gas absorption system to be tested and the waste gas treatment system were prepared using absorption bottles containing zinc acetate (12, 14) and sodium hydroxide (13, 15), respectively.

Figure 1

Schematic representation of the experimental system employed herein: (1) 300 ppm H2S/N2 gas cylinder, (2) N2 gas cylinder, (3) H2S/N2 gas flowmeter, (4) N2 gas flowmeter, (5) gas mixer, (6) three-way valve, (7) 250 mL three-neck flask, (8) sand core, (9) rotor, (10) heat-collecting magnetic agitator, (11) thermometer, (12, 14) zinc acetate absorption bottles, (13, 15) sodium hydroxide absorption bottles, and (16) bypass (used to determine the concentration of H2S in the gas entering the reactor).

The H2S/N2 gas mixture was prepared by mixing the two individual gases as raw materials. By adjusting the flow rates of H2S/N2 and N2, the flow rate of the mixture was changed. A heat-collecting magnetic agitator was used to stir the system and to control the temperature. After mixing the gas evenly and adjusting the reaction temperature to the required value, a catalyst adsorption experiment was carried out using only the catalyst and water (100 mL) in a three-neck flask. Following saturation of the catalyst, the desired amount of H2O2 was added to the three-port flask to allow the catalytic oxidation process to take place. The H2S concentrations of the inlet and outlet gases were determined by iodometry[41] after being subjected to absorption by zinc acetate. The tail gas was discharged after passing through the waste gas treatment system. The removal rate of H2S was calculated by eq . During the reaction time of 120 min, four samples were taken at each 30 min interval and the average value was calculated.

In eq , η is the removal rate of H2S, Cin is the H2S inlet concentration, and Cout is the H2S outlet concentration.

Characterization

The crystal structure of FeOCl was determined by X-ray diffractometry (XRD, DX-2700, China) with Cu Kα radiation. The scanning angle (2θ) was 5–80° and the step size was 0.02°. The morphology of the FeOCl sample was analyzed by scanning electron microscopy (SEM, FEI Inspect F50). X-ray photoelectron spectroscopy (XPS, Thermo Escalab 250Xi) was employed to determine the element valences and chemical information of the FeOCl surface. The binding energy data were calibrated with the C 1s signal at 284.6 eV. Paramagnetic signals were collected on an electron paramagnetic resonance spectrometer (EPR, Brooke A300, Germany) using 5,5-dimethyl-1-pyrrolidine N-oxide (DMPO) to give the EPR spectra of the samples. The oxidation products of H2S were analyzed by ion chromatography[32] (IC, ICS-5000).

Results and Discussion

Performance of H2S Removal by the FeOCl/H2O2 System

The effects of the amount of an FeOCl catalyst, the H2O2 concentration, the reaction temperature, the gas flow rate, and the reaction time on the removal of H2S by the FeOCl/H2O2 system were investigated, and the results are presented in Figure .

Figure 2

Effects of different operating parameters on H2S removal by the FeOCl/H2O2 system. (A) Dosage of FeOCl, (B) H2O2 concentration, (C) reaction temperature, (D) gas flow rate, and (E) reaction time. Reaction conditions: H2S concentration, 160 ppm; dosage of FeOCl, 1 g/L; initial H2O2 concentration, 0.33 mol/L; initial pH value, 3.16; reaction temperature, 293 K; gas flow rate, 265 mL/min; reaction time, 120 min; and liquid volume, 100 mL.

As can be seen from Figure A, when the amount of FeOCl is increased from 0 to 1.0 g/L, η increased from 11.1 to 56.6%, indicating that FeOCl has a catalytic effect on the removal of H2S by H2O2, and so η increases with an increase in the FeOCl dosage. This can be accounted for by considering that increasing the amount of FeOCl will provide additional catalytic active centers,[42] which will produce a greater number of active species, and η increases. However, when the amount of FeOCl was increased further to 2 g/L, η increased only from 56.6 to 58.4%. This was likely due to the fact that an excess of the catalyst led to an agglomeration of catalyst particles, thereby hindering the catalytic action of the reaction active sites.[43]

As outlined in Figure B, when the H2O2 concentration was increased from 0 to 0.33 mol/L, η increased from 0 to 56.6%, although a further increase in the H2O2 concentration to 1.65 mol/L gave only a slight increase in η, i.e., to 60.7%. This was considered that when the concentration of H2O2 in the reaction system is low, the solid catalyst can fully interact with the available H2O2, and so the amount of an active material (and thus η) increases with an increase in the H2O2 content.[44] However, on increasing hydrogen peroxide continuously, the amount of hydrogen peroxide was in excess to the catalyst relatively, and the reactive material did not increase with the increase in H2O2; in addition, adding excess H2O2 would result in the self-consumption of reactive substances.[43] Thus, for the purpose of subsequent experiments, an initial H2O2 concentration of 0.33 mol/L was selected.

As shown in Figure C, upon increasing the temperature from 283 to 293 K, η increased from 43.6 to 56.6%, although a further increase in temperature to 313 K resulted in a slight decrease in η (i.e., from 56.6 to 55.4%). In this context, it should be noted that as the temperature increases, the reaction rate also increases initially.[32] However, the solubility of H2S in water is reduced at higher temperatures, and so the reduced concentration of reactants ultimately results in a lower reaction rate.

As can be seen from Figure D, when the gas flow rate was increased from 135 to 370 mL/min, η decreased from 85.6 to 20.5%. It should be noted that the initial pH of the reaction system is 3.16, and so H2S exists mainly in its molecular form in the aqueous environment (Ka1(H2S) = 1.3 × 10–7, [HS–]/[H2S] = 1 × 10–3.56). For their effective removal, the H2S molecules must diffuse to the catalyst surface, and so as the gas flow rate is increased, the residence time of H2S in the reactor decreases, greater volumes of H2S are discharged from the reactor prior to diffusion to the catalyst surface, and η is reduced. These results clearly indicate that the removal of H2S is controlled by the diffusion of H2S.

As outlined in Figure E, when the H2S gas was continuously introduced into the reaction system over 540 min, η decreased gradually from 56.6 to 40.1%. This can be attributed to a reduced concentration of H2O2 over time and/or a loss in the catalytic performance of FeOCl. Generally, the oxidation products of H2S could be sulfur[32] and/or sulfur oxides (e.g., SO32– and SO42–4); in addition, the amount of ineffective H2O2 decomposition is small under acidic conditions (i.e., pH ∼ 3).[45] Assuming that the consumption of H2O2 is mainly attributed to the oxidation of H2S, the oxidation of H2S to S0 consumes the least amount of H2O2 (eq ),[32] while the generation of SO42– consumes the most amount of H2O2 (eq ).[32] Under the experimental conditions employed herein, a total of 0.353 mmol H2S enters the reaction system during the initial 180 min. Thus, if the removal rate of H2S is 60%, the consumption of H2O2 during the oxidation of H2S to SO42– is 0.848 mmol. However, it was found that the maximum consumption of H2O2 reached only 2.5% during this time, thereby indicating that its concentration remained relatively constant. Respectively, 33 mmol H2O2 was added at 180 and 270 min to ensure that the concentration was maintained at >0.33 mol/L. These results indicate that the decrease in the H2S removal rate could be attributed to the reduced catalytic performance of FeOCl.

Analysis of the Reduced Catalytic Performance of FeOCl

In general, the decreased performances of heterogeneous catalysts can be attributed to the reaction products covering or blocking the catalyst active centers or pores or changes in the composition and the structure of the catalyst.[46] As noted above, the products of H2S oxidation by H2O2 are sulfur[32] and sulfur oxides.[4] If sulfur is produced, it can cover the active center of the catalyst or block the catalyst pores of the catalyst, whereas if SO32– and SO42– are produced, the pH of the reaction system may decrease. In addition, the interaction between FeOCl and H2S may produce FeS. Thus, after the completion of the reaction (i.e., 540 min), the reaction mixture was separated by centrifugation. In the case where SO32– and SO42– are formed, these components should be present in the supernatant, whereas if S0 and FeS are formed, they would exist in the solid. The liquid phase (or supernatant) was analyzed by ion chromatography and its pH was determined (Table ); the solid phase (i.e., “used FeOCl”) was dried under vacuum at 333 K and characterized by XRD, SEM, and XPS. As listed in Table , after 540 min, the pH of the liquid phase was 2.88, and both SO42– and Cl– were detected. Since only FeOCl, H2O2, H2O, and H2S/N2 were added to the reaction system, Cl– present in the liquid phase must originate from FeOCl. The oxidation of H2S to SO42– and the dissolution of chlorine from FeOCl to Cl– resulted in the reduced pH of the reaction system (i.e., from 3.16 to 2.88).

Table 1

Analysis of the Liquid Phase (Supernatant) of the FeOCl/H2O2 System after the Reactiona

parameter	SO42– (mmol/L)	SO32–	Cl– (mmol/L)	pH
value	5.1	NDb	0.3	2.88

After the reaction of 540 min under the abovementioned reaction conditions.

ND—not detected.

It was then considered whether a reduction in the pH of the reaction system would lead to a η decrease. Thus, dilute sulfuric acid was used to regulate the initial pH of the reaction system, and the effect of the initial pH on the removal of H2S was investigated (Figure ). As shown, when the initial pH values were 2.00, 2.88, and 3.16, the η values were 61.5, 57.2, and 56.6%, respectively, indicating that between values of 2.88 and 3.16, the pH had little effect on the removal of H2S. However, when the pH value was 2, η was slightly higher, and this was attributed to the strong acidity of the solution, leading to the dissolution of Fe from the FeOCl, and the amount of H2S removed by the homogeneous Fenton system cannot be ignored.[47] These results indicate that a reduction in the system pH during the reaction process does not result in a decrease in η.

Figure 3

Effect of the initial reaction system pH on the removal rate of H2S. Reaction conditions: H2S concentration, 160 ppm; dosage of FeOCl, 1 g/L; initial H2O2 concentration, 0.33 mol/L; reaction temperature, 293 K; gas flow rate, 265 mL/min; reaction time, 120 min; and liquid volume, 100 mL.

Figure shows the XRD patterns and SEM images of FeOCl and used FeOCl. As can be seen from Figure A, the intensity of the XRD diffraction peaks of used FeOCl decreased significantly compared to FeOCl. In addition, the two-dimensional flake morphology of FeOCl can be clearly seen in Figure B,[35] while in the case of used FeOCl, this structure was destroyed to a certain extent. Overall, these observations confirm that the crystal structure of FeOCl was partly destroyed during the reaction.

Figure 4

(A) XRD patterns and (B) SEM images of FeOCl. (a) FeOCl, and (b) used FeOCl.

Figure shows the XPS spectra of FeOCl, which allowed the elemental composition and valence distribution of the elements on the catalyst surface to be calculated, as listed in Table . From Figure A, it is apparent that sulfur was not present in used FeOCl, indicating that little or no S0 and FeS were produced during H2S removal. This observation, therefore, indicates that blockage of the catalyst pores or covering of the catalyst active centers was not responsible for the decreased catalytic activity of FeOCl.

Figure 5

XPS spectra of FeOCl. (A) Full survey spectrum of FeOCl, (B) high-resolution O 1s spectra, and (C) high-resolution Fe 2p spectra. (a) FeOCl and (b) used FeOCl.

Table 2

Elemental Composition on the FeOCl Surface

	atomic %			ratio
sample	Fe	O	Cl	FeII/FeIII	Fe–O/H2O
FeOCl	23.59	49.42	26.99	0.26	1.06
used FeOCl	27.43	63.84	8.37	0.42	0.47

As outlined in Table , the contents of chlorine in FeOCl and used FeOCl were 26.99 and 8.37% respectively, indicating the loss of chlorine during the reaction. It can be seen from Figure B that the O 1s peaks at 530.35 and 531.72 eV can be attributed to Fe–O and O–H of H2O.[48] Compared with FeOCl, the intensity of the Fe–O peak was lower for used FeOCl, while the intensity of the H2O peak increased. As can be seen from Table , the Fe–O/H2O ratios of FeOCl and used FeOCl were 1.06 and 0.47, respectively, and of oxygen were 49.42 and 63.84%, respectively. Fe–Cl in FeOCl is weak, and so chlorine is easily replaced by O–H from H2O,[49] resulting in a decrease in the chlorine content, an increase in the oxygen content, and a higher Fe–O/H2O ratio. It is, therefore, possible that the decreased catalytic activity may be caused by the dissolution of chlorine from FeOCl. It was, therefore, expected that the addition of chlorine to used FeOCl could restore its catalytic activity.[50]

Thus, used FeOCl was supplemented with chlorine by soaking in a 37% HCl solution (used FeOCl/HCl = 1 g:10 mL) for 30 min, then calcined at 220 °C for 1 h, and washed and dried to obtain the regenerated catalyst, i.e., regenerated FeOCl.[50] As shown in Figure A, the XRD peak intensity of the regenerated FeOCl is similar to that of FeOCl. Figure B shows that η reached 55.4% (average after 120 min) for regenerated FeOCl, which is similar to FeOCl. These results indicate that the crystal structure of FeOCl can be restored by the addition of chlorine, ultimately resulting in the restoration of the catalytic activity.

Figure 6

XRD patterns and the H2S removal performances of FeOCl, used FeOCl, and regenerated FeOCl. (A) XRD patterns and (B) H2S removal performances. (a) FeOCl, (b) used FeOCl, and (c) regenerated FeOCl. Reaction conditions: H2S concentration, 160 ppm; dosage of FeOCl, 1 g/L; initial H2O2 concentration, 0.33 mol/L; initial pH value, 3.16; reaction temperature, 293 K; gas flow rate, 265 mL/min; and liquid volume, 100 mL.

Analysis of Radicals in the FeOCl/H2O2 System

The polarized plan of FeOCl is filled with a large number of unsaturated iron atoms,[43] and it is known that H2O2 can approach this polarization plane more easily than water.[31] Thus, the H2O2 molecules adsorbed on the catalyst active sites can be activated to form hydroxyl radicals (•OH) and superoxide radicals (•O2–).[32,40,51−54] To confirm the presence of such radicals and their possible contribution to the reaction mechanism, EPR spectroscopy (Figure ) and radical quenching experiments (Figure ) were carried out on the FeOCl/H2O2 system. In the EPR experiment, the DMPO trap reacts with •OH and •O2– in solution to form adducts, resulting in the production of characteristic signals.[55] As can be seen from Figure , no characteristic signals corresponding to •OH and •O2– were detected in the H2O2 system and the FeOCl/H2O system. However, characteristic signals corresponding to both radicals were observed for the FeOCl/H2O2 system, thereby indicating that •OH and •O2– were produced from H2O2 upon catalysis by FeOCl.

Figure 7

EPR spectra of •OH and •O2– captured by DMPO. (A) H2O2 solution, (B) FeOCl/H2O system, and (C) FeOCl/H2O2 system. Reaction conditions: dosage of FeOCl, 1 g/L; initial H2O2 concentration, 0.33 mol/L; initial pH value, 3.16; and liquid volume, 100 mL.

Figure 8

Influence of different radical quenchers on the degradation of H2S. Conditions: H2S concentration, 160 ppm; gas flow rate, 265 mL/min; initial H2O2 concentration, 0.33 mol/L; dose of FeOCl, 1 g/L; reaction temperature, 283 K; initial pH, 3.16; IPA concentration, 10 mM; BQ concentration, 1 mM; BHT concentration (in isopropanol), 10 mM; and liquid volume, 100 mL.

As can be seen from Figure C, two large peaks are visible for the Fe 2p orbitals at 711.05 and 724.46 eV, which correspond to the 2p3/2 and 2p1/2 binding energies of FeIII, respectively,[56] while two smaller peaks at 712.9 and 726.12 eV correspond to the 2p3/2 and 2p1/2 binding energies of FeII, respectively,[42] indicating that both Fe3+ and Fe2+ are present in FeOCl, where Fe3+ is the main valence state. In addition, the data presented in Table show that the FeII/FeIII ratios of FeOCl and used FeOCl catalysts are 0.26 and 0.42, respectively, which can be attributed to the progress of•O2– (eqs –6)[53] and •OH (eq )[37,38] that were produced from H2O2 upon catalysis by FeOCl; FeIII is easily converted to FeII, while the reverse process is challenging.[43]

To confirm the participation of radicals in the H2S removal process, radical quenching experiments were carried out using butylated hydroxytoluene (BHT).[57] As shown in Figure C, upon the addition of BHT, η was reduced from 56.6 to 7.8%, which is a similar value to that obtained in the absence of FeOCl (i.e., 10%). These results indicate that the H2S removal process did indeed involve participation from radicals. Subsequently, quenching experiments were carried out using isopropanol (IPA) and benzoquinone (BQ).[54] As shown in Figure D,E, upon the addition of IPA and BQ, η decreased from 56.6 to 35.4 and 38.5%, respectively, which further indicated that •OH and •O2– were involved in the process of H2S removal by the FeOCl/H2O2 system. The chemical reactions involved in this process are given by eqs –10,[32]11, and 12.[58]

Upon the simultaneous addition of both IPA and BQ (Figure F), η was 24.2%, which is higher than that obtained when BHT was added (i.e., 7.8% η), indicating that other radicals were also involved in the H2S removal process. As can be seen from eq , HO2• is also produced during the catalytic process, but this species is further decomposed to produce •O2– (eq ), and the oxidizability of HO2• (oxidation potential 1.51) is significantly weaker than that of •O2– (oxidation potential 2.21) and •OH (oxidation potential 2.80).[59] In addition to •OH, •O2–, and HO2•, there are other forms of radicals involved in the reaction.

Wang[60] found that the radicals produced by H2O2 activation can be divided into two states, namely adsorbed and free states. IPA is a hydrophilic radical scavenger, which cannot effectively quench adsorbed radicals;[61] BQ is also hydrophilic and it was therefore considered that BQ can only quench free superoxide radicals. Radical quenching experiments show that free radicals are involved in the removal of H2S. According to the process of the catalytic reaction, eqs and 13–15 can be given as follows

In eqs –15, the subscript “ads” denotes radicals in the adsorbed state and “free” denotes radicals in the free state.

According to eqs and 13–15, there are adsorbed radicals in the system. Qian[55] reported that the adsorbed radicals are easy to oxidize the pollutants adsorbed on the catalyst. If H2S in the solution can be adsorbed on FeOCl, it shows that the adsorbed radicals are likely to participate in the oxidation of H2S. Thus, the adsorption of H2S by FeOCl in the aqueous phase was investigated and the adsorption kinetics was analyzed, as shown in Figure and Table .

Figure 9

Adsorption of H2S on FeOCl in the aqueous phase. (A) Adsorption curve of H2S on FeOCl, (B) pseudo-first-order fitting, and (C) pseudo-second-order fitting. Conditions: H2S concentration, 160 ppm; dosage of FeOCl, 1 g/L; initial pH value, 3.16; gas flow rate, 265 mL/min; and liquid volume, 100 mL.

Table 3

Analysis of the Adsorption Kinetics of H2S on FeOCla

	pseudo-first order			pseudo-second order
T (K)	qe1 (mg/g)	k1 (min–1)	R2	qe2 (mg/g)	k2 (g/(mg·min))	R2
283	16.35	50.58	0.9851	14.5	0.0029	0.9979
293	32.93	127.34	0.9803	21.0	0.00075	0.9936
303	35.41	86.56	0.9941	26.5	0.00052	0.9969

k1 and k2 are the apparent rate constants of the pseudo-first-order and pseudo-second-order kinetics, respectively, while qe1 and qe2 are the equilibrium adsorption capacities of the quasi-first-order and quasi-second-order kinetics, respectively.

It can be seen from Figure A that with an increase in temperature, the solubility of H2S in solution decreases, but the adsorption capacity of H2S increases. Obviously, the adsorption of H2S by FeOCl in solution accords with the characteristics of chemical adsorption.[62] As indicated in the fitting results shown in Figure B,C and Table , the R2 of the pseudo-second-order adsorption kinetics of H2S on FeOCl is larger than that of the pseudo-first-order adsorption kinetics at 283, 293, and 303 K, indicating that the pseudo-second-order model can better describe the adsorption process.[2,62]

According to the results of kinetic analysis, enthalpy (ΔH) of H2S adsorption by FeOCl can be calculated by eqs –18 to be 21.5 kJ/mol. Generally, ΔH of the chemisorption is >40 kJ/mol,[63] which indicates that the interaction between FeOCl and H2S is relatively weak, and H2S could be activated by adsorbing. Therefore, we speculate that the adsorbed radicals were involved in the removal of H2S. This may also be the reason that FeS was not generated during the catalysis and sulfur was not detected in the XPS.

In eqs –18, K is the adsorption equilibrium constant and qe is the equilibrium adsorption capacity of H2S (saturated adsorption capacity, mg/g). According to the results of kinetic analysis, qe = qe2 and ce is the concentration of H2S in the aqueous solution at the point of adsorption equilibrium (mg/g). Since the concentration of H2S in the gas phase remains constant (160 ppm) during the adsorption experiment, the value of ce can be obtained according to Henry’s law. ΔG is the free energy (kJ/mol), ΔH is the enthalpy (kJ/mol), ΔS is the entropy (J/(mol·K)), R is the gas constant (8.314 J/(mol·K)), and T is the adsorption temperature (K).

Mechanistic Analysis

In general, the lifetimes of radicals are extremely short, and so they are mainly concentrated in the water film close to the surface of the solid catalyst.[34] In heterogeneous catalytic reactions, the reaction process is generally composed of several processes occurring in series, such as diffusion of the H2S to the surface of the catalyst, the adsorption of H2S onto the catalyst surface, and conversion of the adsorbed H2S. Due to the fact that a water film rich in radicals is present on the surface of FeOCl, and the free radicals can react directly with H2S,[32] it is possible that two reaction paths exist for H2S: (I) the adsorbed H2S is oxidized by adsorbed radicals and (II) the not adsorbed H2S is oxidized by free radicals. Based on the abovementioned analysis, a mechanism was established for H2S removal by the FeOCl/H2O2 system, as shown in Figure . The oxidation product is mainly SO42–, and the diffusion of H2S controls the rate of H2S removal.

Figure 10

Schematic representation of the heterogeneous degradation of H2S by the FeOCl/H2O2 system. Path I, the adsorbed H2S is oxidized by adsorbed radicals, and path II, the unadsorbed H2S is oxidized by free radicals.

The resulting SO42– can be recovered by the addition of ammonia to produce ammonium sulfate,[32] thereby allowing the effective recovery of the reaction products to achieve zero discharge of liquid waste from our proposed system.

Conclusions

This paper reports the study on the performance of gaseous H2S removal by the FeOCl/H2O2 system. It was found that upon increasing the amount of a catalyst, the H2O2 concentration, and the reaction temperature, the removal efficiency of H2S initially increased prior to becoming stable. In contrast, upon increasing the gas flow rate, the removal efficiency of H2S decreased, and so it was apparent that the process of H2S removal by FeOCl/H2O2 is a diffusion-controlled process. The removal rate of 160 ppm H2S gas reached 85.6% when bubbling through 100 mL of an aqueous solution containing FeOCl (1 g/L) and H2O2 (0.33 mol/L) at 293 K with a flow rate of 135 mL/min. Although the dissolution of chlorine in FeOCl caused a reduction in the catalytic performance, the catalytic activity could be restored by soaking in hydrochloric acid. The main active species involved in the removal of H2S were found to be radicals, such as •OH and •O2–, which were produced from H2O2 upon catalysis by FeOCl. These radicals were likely present in the adsorbed and free states, and the free radicals were mainly distributed in the water film close to the FeOCl surface. In terms of the process mechanism, H2S initially diffuses toward the vicinity of FeOCl or adsorbs onto the catalyst surface. Subsequently, H2S is oxidized by free and adsorbed radicals, respectively, to give SO42–. Importantly, the catalyst and the product of this system can be separated effectively. To enhance the removal rate of H2S further, it is necessary to increase the residence time of H2S in the reactor, such as replacing the three-neck flask with a tower reactor. Overall, this process was found to be a simple and efficient method for H2S purification, and following scale-up, could be suitable for industrial applications.