Fast Degradation of Monochloroacetic Acid by BiOI-Enhanced UV / S(IV) Process: E ﬃ ciency and Mechanism

: Iodide (I − ) could promote ultraviolet-activated S(IV) processes (UV / S(IV)) and degrade aqueous halogenated organic compounds and hazardous oxoanions. With the interest of promoting use of this technology, this study investigated the feasibility of using bismuth oxyiodide (BiOI) as an I − source to enhance UV / S(IV) where monochloroacetic acid (MCAA) was selected as a testing model compound. Degradation of MCAA by UV / S(IV) increased by 50% in presence of BiOI. Results of competitive kinetics indicated that the promotion e ﬀ ect brought by BiOI mainly originated from its sustainable release of I − , and subsequent enhanced generation of hydrated electrons. Electron spin resonance detection and ﬂuorescence characterization proved increased formation of sulﬁte radical, resulting from sulﬁte oxidation by UV-excited BiOI. However, the sulﬁte radical only made a small contribution (9%) to MCAA degradation due to its moderate reactivity toward MCAA (4.2 × 10 5 M − 1 · s − 1 ). UV / S(IV) combined with BiOI signiﬁcantly decreasing the biotoxicity of MCAA solution. BiOI can be regenerated using I − -containing solution. Our ﬁndings provide evidence that BiOI is a promising I − source and photocatalyst, which progresses the I − -assisted UV / S(IV) process towards practical application. The competitive kinetics method, electron spin resonance (ESR) spectroscopy, and ﬂuorescence spectroscopy were used to identify involved reactive species, quantify their contribution, and clarify how BiOI play e ﬀ ectiveness. MCAA degradation products and their biotoxicity were analyzed to evaluate the biocompatibility of treated water. A regeneration test of BiOI was performed to explore its possible repeated use. Validation, L.W.; Formal Analysis, L.W.; Investigation, L.W.; Resources, X.L.; Data Curation, L.W.; Writing—Original Draft Preparation, L.W.; Writing—Review & Editing, X.L.; Visualization, L.W.; Supervision, X.L.; Project Administration, L.W. and X.L.; Funding Acquisition, L.W. and X.L.


Introduction
Recently, efficient degradation of some halogenated organic compounds (HOCs) and hazardous oxoanions was achieved through UV-activated (bi)sulfite (S(IV)) reduction (abbreviated as UV/S(IV)) [1][2][3], which is classified as a hydrated electron (e − aq )-based advanced reduction process (e − aq -ARP). However, besides the valid concentration of S(IV) (main precursor of e − aq ), UV photon absorbance by S(IV) dominates the production efficiency of e − aq for this reduction process [4]. The UV sources currently used in water treatment are the low-pressure mercury lamp (UV-L) and medium-pressure mercury lamp (UV-M). UV-L produces light that is almost entirely at 254 nm, and UV-M mainly emits light in the range of 300 to 600 nm [5]. Hence, a low production efficiency of e − aq and a low energy utilization rate are expected when UV-L and UV-M are used to activate S(IV) due to its relatively low absorptivity at wavelength >250 nm ( Figure 1). As a result, high S(IV) dosage, high pH (to decrease the proportion of bisulfite), and high power of UV lamp are usually needed to increase e − aq generation and/or energy utilization for UV/S(IV) process. In previous works, Yu et al. [1] and our group [7] tried to introduce iodide (I ) into UV/S(IV) system (UV/S(IV)/I ) and observed an enhancement of e aq generation. I was found to play a homogeneous catalyst-like role, and its recycling relies on S(IV) reduction. Nevertheless, as aqueous I solution also absorbs UV light at wavelength >250 nm very weakly (Figure 1), UV/S(IV)/I don't perform well enough from the angle of energy utilization and still need to operate under alkaline conditions to guarantee e aq yield. Moreover, aqueous I is easy to be oxidized if exposed to air. A possible solution is to replace I with a photocatalyst that can simultaneously release I and absorb wide-range UV light to generate reactive species. Based on this consideration, bismuth oxyiodide (BiOI) may be an ideal candidate. Bismuth oxyiodide, as a semiconductor photocatalyst, has a bandgap energy of ~1.7 eV [8], and thus can be photoexcited by UV/visible light of 200 to 600 nm [9]. The photocatalytic ability of BiOI was reported to be comparable to or superior to that of commercial TiO2 (P25) [10]. Photoreduction enhancement on the surface of BiOI is expected in the presence of S(IV) (as hole scavengers). Furthermore, BiOI is stable and can release I slowly in aqueous solution [8], eliminating the concern about deterioration of the I working solution. Given these facts, we hypothesize that UV/S(IV) combined with BiOI (UV/S(IV)/BiOI) is a promising ARP, which can make full use of UV energy input and meanwhile keep high e aq yield. However, the efficiency and the underlying mechanisms of UV/S(IV)/BiOI for contaminant degradation are totally unknown.
Thus, this study aims to investigate the efficiency and mechanism of UV/S(IV)/BiOI for the decomposition of micropollutants. Monochloroacetic acid (MCAA), due to its simple chemical structure and its distinguishing reactivity toward various radicals, was selected as the model compound to probe the availability of reactive species in the proposed process. The competitive kinetics method, electron spin resonance (ESR) spectroscopy, and fluorescence spectroscopy were used to identify involved reactive species, quantify their contribution, and clarify how BiOI play effectiveness. MCAA degradation products and their biotoxicity were analyzed to evaluate the biocompatibility of treated water. A regeneration test of BiOI was performed to explore its possible repeated use.  In previous works, Yu et al. [1] and our group [7] tried to introduce iodide (I − ) into UV/S(IV) system (UV/S(IV)/I − ) and observed an enhancement of e − aq generation. I − was found to play a homogeneous catalyst-like role, and its recycling relies on S(IV) reduction. Nevertheless, as aqueous I − solution also absorbs UV light at wavelength >250 nm very weakly (Figure 1), UV/S(IV)/I − don't perform well enough from the angle of energy utilization and still need to operate under alkaline conditions to guarantee e − aq yield. Moreover, aqueous I − is easy to be oxidized if exposed to air. A possible solution is to replace I − with a photocatalyst that can simultaneously release I − and absorb wide-range UV light to generate reactive species. Based on this consideration, bismuth oxyiodide (BiOI) may be an ideal candidate. Bismuth oxyiodide, as a semiconductor photocatalyst, has a bandgap energy of~1.7 eV [8], and thus can be photoexcited by UV/visible light of 200 to 600 nm [9]. The photocatalytic ability of BiOI was reported to be comparable to or superior to that of commercial TiO 2 (P25) [10]. Photoreduction enhancement on the surface of BiOI is expected in the presence of S(IV) (as hole scavengers). Furthermore, BiOI is stable and can release I − slowly in aqueous solution [8], eliminating the concern about deterioration of the I − working solution. Given these facts, we hypothesize that UV/S(IV) combined with BiOI (UV/S(IV)/BiOI) is a promising ARP, which can make full use of UV energy input and meanwhile keep high e − aq yield. However, the efficiency and the underlying mechanisms of UV/S(IV)/BiOI for contaminant degradation are totally unknown.

Results and Discussion
Thus, this study aims to investigate the efficiency and mechanism of UV/S(IV)/BiOI for the decomposition of micropollutants. Monochloroacetic acid (MCAA), due to its simple chemical structure and its distinguishing reactivity toward various radicals, was selected as the model compound to probe the availability of reactive species in the proposed process. The competitive kinetics method, electron spin resonance (ESR) spectroscopy, and fluorescence spectroscopy were used to identify involved reactive species, quantify their contribution, and clarify how BiOI play effectiveness. MCAA degradation products and their biotoxicity were analyzed to evaluate the biocompatibility of treated water. A regeneration test of BiOI was performed to explore its possible repeated use. The results of MCAA elimination by BiOI adsorption, BiOI/S(IV) adsorption/reduction, UV photolysis, UV/BiOI photocatalytic degradation, and UV/S(IV) reduction are also displayed for comparison. Under current experimental conditions, processes including BiOI adsorption, BiOI/S(IV) adsorption/reduction, UV photolysis, and UV/BiOI, display weak MCAA depletion (<10%). Better degradation performance (50%) was exhibited by UV/S(IV). As expected, the best degradation performance was obtained by UV/S(IV)/BiOI. These results indicate that BiOI adsorption, S(IV) reduction, and UV photolysis played minor roles for MCAA degradation in the UV/S(IV)/BiOI system. The results of MCAA elimination by BiOI adsorption, BiOI/S(IV) adsorption/reduction, UV photolysis, UV/BiOI photocatalytic degradation, and UV/S(IV) reduction are also displayed for comparison. Under current experimental conditions, processes including BiOI adsorption, BiOI/S(IV) adsorption/reduction, UV photolysis, and UV/BiOI, display weak MCAA depletion (<10%). Better degradation performance (50%) was exhibited by UV/S(IV). As expected, the best degradation performance was obtained by UV/S(IV)/BiOI. These results indicate that BiOI adsorption, S(IV) reduction, and UV photolysis played minor roles for MCAA degradation in the UV/S(IV)/BiOI system.

Identification of Main Reactive Species for MCAA Degradation by UV/S(IV)/BiOI
It has been established that MCAA reduction by UV/S(IV) and UV/S(IV)/I at pH 9.0 was induced by e aq [11]. At neutral pH, H • may show increased contribution due to the conversion of e aq to H • (pKa(H • )=9.6 [12]). Recently, several works reported that sulfite radical (SO • ) also played an important role during degradation of organic pollutants by sulfite-mediated processes [13][14][15]. Addition to unsaturated bond and H-abstraction are the two main reaction types for SO • with organics [16]. In the UV/S(IV)/BiOI system, SO • can be produced from UV photolysis of S(IV) (including sulfite (SO ) and bisulfite (HSO )) and reaction of S(IV) with e aq /H • , h /HO•, and reactive iodine species (RISs) (Equations (1)- (15) [7,14]). Additionally, as no deoxygenation step is taken, photogenerated hole/electron (h /e ), superoxide radical (O • ), and hydroxyl radical (HO•) are considered to occur in the UV/S(IV)/BiOI system [17]. Therefore, contributions of h /e 、O • 、 HO•, SO • , and e aq /H • are necessary to clarify.

Role of BiOI in the UV/S(IV)/BiOI System
As stated above, introduction of BiOI into the UV/S(IV) system greatly enhanced MCAA degradation efficiency (50% increase), and aq was the main contributor. There were two possible pathways for BiOI to enhance the UV/S(IV) process: (1) BiOI was excited by UV irradiation to generate ℎ , , and ℎ oxidized S(IV) to generate SO • , and (2) releasing from BiOI enhanced the formation of aq . As most of the photogenerated ℎ / pairs unavoidably recombine at the surface or at bulk trapping sites along with release of photons and heat, leading to

Role of BiOI in the UV/S(IV)/BiOI System
As stated above, introduction of BiOI into the UV/S(IV) system greatly enhanced MCAA degradation efficiency (50% increase), and aq was the main contributor. There were two possible pathways for BiOI to enhance the UV/S(IV) process: (1) BiOI was excited by UV irradiation to generate ℎ , , and ℎ oxidized S(IV) to generate SO • , and (2) releasing from BiOI enhanced the formation of aq . As most of the photogenerated ℎ / pairs unavoidably recombine at the surface or at bulk trapping sites along with release of photons and heat, leading to   Figure 5. For the UV/BiOI system, the spectra exhibited a broad emission band in the range of 350 to 600 nm, which was attributed to luminescence emission from recombination of h + VB and e − CB [27]. The presence of S(IV) led to a significant decrease in the emission intensity of the fluorescence spectra, demonstrating that the presence of S(IV) can effectively reduce recombination rate of photoinduced h + VB /e − CB and lead to an increased charge carrier separation. In addition, we monitored increased spin adduct of SO •− 3 with 5,5-Dimethyl-1-pyrroline N-oxide (SO •− 3 -DMPO, Figure 6). The capture of fluorescence emission and increase of SO •− 3 formation indicated that the first enhancement pathway is important. Note that introduction of BiOI into UV/S(IV) system promoted formation of SO •− 3 via the reaction of photogenerated h + VB with S(IV). However, enhanced formation of SO •− 3 did not strenthen MCAA degradation due to the low reactivity of SO •− 3 towards to MCAA (4.2 × 10 5 M −1 ·s −1 ). fluorescence emission, the fluorescence spectra was used to investigate the reaction of ℎ with S(IV) on the surface of BiOI. The fluorescence spectra of the BiOI suspension in the presence or absence of S(IV) are shown in Figure 5. For the UV/BiOI system, the spectra exhibited a broad emission band in the range of 350 to 600 nm, which was attributed to luminescence emission from recombination of ℎ and [27]. The presence of S(IV) led to a significant decrease in the emission intensity of the fluorescence spectra, demonstrating that the presence of S(IV) can effectively reduce recombination rate of photoinduced ℎ / and lead to an increased charge carrier separation. In addition, we monitored increased spin adduct of SO • with 5,5-Dimethyl-1-pyrroline N-oxide (SO • -DMPO, Figure 6). The capture of fluorescence emission and increase of SO • formation indicated that the first enhancement pathway is important. Note that introduction of BiOI into UV/S(IV) system promoted formation of SO • via the reaction of photogenerated ℎ with S(IV). However, enhanced formation of SO • did not strenthen MCAA degradation due to the low reactivity of SO • towards to MCAA (4.2 × 10 5 M −1 ·s −1 ).  To verify the role of sustainable release of I − from BiOI, we simulated the release process of BiOI through a peristaltic pump dosing system, where I − f was continuousely added into UV/S(IV) system at a rate nearly equal to that of I − releasing from BiOI. It can be seen from Figure 8 that the degradation curves for the two processes (UV/S(IV)/BiOI and UV/S(IV) with continuous I − f dosing) were very similar (~10% difference). Thus, we can conclude that I − releasing from BiOI is the key enhancement contributor for the UV/S(IV) process. Considering the similar photochemistry of I − in water and in organic solvent along with excellent solubility of (bi)sulfite in mixture of water and organic solvent [28,29], UV/S(IV)/BiOI is believed to efficiently to degrade hazardous HOCs, which are usually treated in organic solvent or mixture of water and organic solvent. For the second enhancement pathway, we first monitored release in the UV/S(IV)/BiOI process. As shown in Figure 7, the concentration of in the reaction solution (free , ) incresed with time. Meanwhile, we observed that iodine contained in BiOI particles ( ) decreased, and the amount of decrease was equal to that of increase. This means there was no loss of releasing . This phenomena is consistent with that observed in UV/S(IV)/ [7]. The recycling mechanism of in UV/S(IV)/ is that RISs, such as • , , and , are generated through photoactivation of and subsequent chain reactions, and are reduced backward to by S(IV). It has been established that presence of can greatly enhance the generration of aq for UV/S(IV) process. To verify the role of sustainable release of from BiOI, we simulated the release process of BiOI through a peristaltic pump dosing system, where was continuousely added into UV/S(IV) system at a rate nearly equal to that of releasing from BiOI. It can be seen from Figure 8 that the degradation curves for the two processes (UV/S(IV)/BiOI and UV/S(IV) with continuous dosing) were very similar (~10% difference). Thus, we can conclude that releasing from BiOI is the key enhancement contributor for the UV/S(IV) process. Considering the similar photochemistry of in water and in organic solvent along with excellent solubility of (bi)sulfite in mixture of water and organic solvent [28,29], UV/S(IV)/BiOI is believed to efficiently to degrade hazardous HOCs, which are usually treated in organic solvent or mixture of water and organic solvent.

Degradation Products of MCAA and Biotoxicity of Treated Water
Based on the results of contribution analysis, aq reduction (79.5%), SO • attack (9.0%), and UV photolysis (9.2%) were confirmed to be involved in the transformation of MCAA. The enhanced formation of SO • raises a concern about the generation of organic sulfur-containing products. To shed light on this concern, we checked the loss of inorganic sulfur by determining the concentration variation of inorganic sulfur species during degradation of MCAA ( Figure 9). Sulfate (SO ) was found to be the only product of S(IV) (including SO and HSO ). This result indirectly proved that there no organic sulfur-containing products generated. To gain more insight, other degradation

Degradation Products of MCAA and Biotoxicity of Treated Water
Based on the results of contribution analysis, e − aq reduction (79.5%), SO •− 3 attack (9.0%), and UV photolysis (9.2%) were confirmed to be involved in the transformation of MCAA. The enhanced formation of SO •− 3 raises a concern about the generation of organic sulfur-containing products. To shed light on this concern, we checked the loss of inorganic sulfur by determining the concentration variation of inorganic sulfur species during degradation of MCAA ( Figure 9). Sulfate (SO 2− 4 ) was found to be the only product of S(IV) (including SO 2− 3 and HSO − 3 ). This result indirectly proved that there no organic sulfur-containing products generated. To gain more insight, other degradation products of MCAA by UV/S(IV)/BiOI were analyzed. Acetate, formate, succinate, and chloride (Cl − ) were successfully detected ( Figure 10). Reports regarding the toxicity evaluation of MCAA solution treated by UV/S(IV)/BiOI demonstrated increasing biocompatibility can be achieved with prolonged treatment time (insert in Figure 10).  Figure 10).

Regeneration of BiOI
From the discussion above we have learned that BiOI can enhance UV/S(IV) to degrade MCAA through releasing I at a certain rate. Thus, BiOI faces the problem of iodine loss as a solid carrier of I with increasing time of use, which will compromise its enhancement effect. So BiOI regeneration test was conducted ( Figure 11). Three recycling of BiOI made the 20 min degradation rate of MCAA decrease by 37.4%. However, after regeneration using KI solution,

Regeneration of BiOI
From the discussion above we have learned that BiOI can enhance UV/S(IV) to degrade MCAA through releasing I − at a certain rate. Thus, BiOI faces the problem of iodine loss as a solid carrier of I − with increasing time of use, which will compromise its enhancement effect. So BiOI regeneration test was conducted ( Figure 11). Three recycling of BiOI made the 20 min degradation rate of MCAA decrease by 37.4%. However, after regeneration using KI solution, the activity of such a batch of used BiOI was completely restored. Hence, the activity of BiOI can be renewed through generation at regular intervals when it was used to enhance UV/S(IV) to degrade pollutants. the activity of such a batch of used BiOI was completely restored. Hence, the activity of BiOI can be renewed through generation at regular intervals when it was used to enhance UV/S(IV) to degrade pollutants.

Synthesis and Characterization of BiOX Microstructures
The synthesis of BiOI was carried out through a solvothermal procedure [23]. The scanning electron micrograph of sample scanning using a SU-8010 Scanning electron microscope (Hitachi, Japan) indicate that this process yielded BiOI microspheres ( Figure 12). X-ray diffraction (XRD) analysis of prepared BiOI specimen was conducted using a Rigaku Dmax-2000 diffractometer using Cu Kα irradiation at 40 kV and 40Ma and XRD patterns from 10 • to 80 • 2θ were recorded. XRD patterns of the fabricated material were consistent with the six characteristic peaks observed in the standard JCPD card (Figure 13). The specific surface area (64.3 m 2 ·g −1 ) was measured using an ASAP2020HD88 analyzer (Micromeritics, USA).

Experimental Procedures
The experiments were performed in a thermostated photoreactor, which has already been reported in our previous work [7]. UV irradiation was generated by a 10W Heraeus low-pressure UV lamp (ozone-free, light centered at 254 nm, Hanau, Germany), and the photoreactor was surrounded by aluminum foil to prevent UV leak and interference of light from environment. The MCAA concentration was set at 10 μM and the initial concentration of sulfite (Na2SO3) was kept at 1 mM in the reaction solution. Subsequently, the pH was adjusted with 10 mM borate buffer (to pH 7.4), and 0.10 g of BiOI microspheres were dispersed into the 1 L reaction solution. During the

Experimental Procedures
The experiments were performed in a thermostated photoreactor, which has already been reported in our previous work [7]. UV irradiation was generated by a 10W Heraeus low-pressure UV lamp (ozone-free, light centered at 254 nm, Hanau, Germany), and the photoreactor was surrounded by aluminum foil to prevent UV leak and interference of light from environment. The MCAA concentration was set at 10 μM and the initial concentration of sulfite (Na2SO3) was kept at 1 mM in the reaction solution. Subsequently, the pH was adjusted with 10 mM borate buffer (to pH 7.4), and 0.10 g of BiOI microspheres were dispersed into the 1 L reaction solution. During the Figure 13. X-ray diffraction patterns of prepared BiOI powder.

Experimental Procedures
The experiments were performed in a thermostated photoreactor, which has already been reported in our previous work [7]. UV irradiation was generated by a 10W Heraeus low-pressure UV lamp (ozone-free, light centered at 254 nm, Hanau, Germany), and the photoreactor was surrounded by aluminum foil to prevent UV leak and interference of light from environment. The MCAA concentration was set at 10 µM and the initial concentration of sulfite (Na 2 SO 3 ) was kept at 1 mM in the reaction solution. Subsequently, the pH was adjusted with 10 mM borate buffer (to pH 7.4), and 0.10 g of BiOI microspheres were dispersed into the 1 L reaction solution. During the reaction, samples were collected using a syringe at predetermined time intervals and filtered immediately with 0.22 µm membrane for analysis of the MCAA concentration. To investigate iodine distribution in aqueous and solid phase during the degradation process, four group experiments were performed, and reactions were quenched (turning off UV lamp and separating BiOI immediately) at 5, 10, 15, and 20 min, respectively. All separated BiOI particles from the whole 1 L reaction solution were used to analyze I − s , and concentration of I − in the bulk solution (I − f ) was recorded meanwhile.
Parameter settings: field sweep, 80 G; microwave frequency, 9.78 GHz; microwave power, 25 mW; modulation amplitude, 0.5 G; conversion time, 164 ms; time constant, 164 ms; and receiver gain, 5 × 10 4 . Photoluminescence spectroscopy analysis of BiOI was performed in a Shimadzu RF-6000 fluorophotometer (Shanghai, China). The toxicity tests were performed on a Microtox model 500 and followed the standard Microtox procedure outlined in the Microtox system operating manual. Water samples were pretreated with membrane filtration to remove BiOI particles. The measurements of light output were carried out after 15 minutes using the 81.9% screening test method. The toxicity was recorded as the light inhibition rate, which was calculated using Equation (16).
Detoxification rate (%) = T 0 − T t T 0 (16) where T 0 and T t are the toxicity of water sample at reaction time 0 min and t min, respectively.

Determination of Reaction
The ultrasonic irradiation was provided by a Biosafer650-92 apparatus (650 W, 20 kHz, Shanghai, China). Ultrasonic irradiation experiments were performed in a cylindrical jacketed glass reactor (inside diameter: 8.2 cm; height: 16.5 cm) with magnetic stirring. The second-order rate constant of W·L -1 was inserted into the solution at a depth of 5.8 cm and the solution was irradiated with the mode "work 5s-stop 2s". The solution temperature was kept at 25 ± 1 ℃ by cooling water. k(SO • + MCAA) can be calculated using Equation (20). The second-order rate constant of SO • with phenol (k(SO • + Phenol)) was reported to be 6.