Simultaneous Oxidation and Sequestration of Arsenic(III) from Aqueous Solution by Copper Aluminate with Peroxymonosulfate: A Fast and Efficient Heterogeneous Process.

The major problem in arsenic (As(III)) removal using adsorbents is that the method is time-consuming and inefficient owing to the fact that most of the adsorbents are more effective for As(V). Herein, we report a new discovery regarding the significant simultaneous oxidation and sequestration of As(III) by a heterogeneous catalytic process of copper aluminate (CuAl2O4) coupled with peroxymonosulfate (PMS). Oxidation and adsorption promote each other. With the help of the active radicals, the As(III) removal efficiency can be increased from 59.4 to 99.2% in the presence of low concentrations of PMS (50 μM) and CuAl2O4 (300 mg/L) in solution. CuAl2O4/PMS can work effectively in a wide pH range (3.0-9.0). Other substances, such as nitrate, sulfate, chloride, carbonate, and humic acid, exert an insignificant effect on As(III) removal. Based on X-ray photoelectron spectroscopy (XPS) analysis, the exposed reductive copper active sites might drive the redox reaction of Cu(II)/Cu(I), which plays a key role in the decomposition of PMS and the oxidation of As(III). The exhausted CuAl2O4 could be refreshed for cycling runs with insignificant capacity loss by the combined regeneration strategy because of the stable spinel structure. According to all results, the CuAl2O4/PMS with favorable oxidation ability and stability could be employed as a promising candidate in real As(III)-contaminated groundwater treatment.

Introduction

Generally, among various heavy metals found in natural groundwater, exposure to arsenic is a major public issue threatening human society, and it has become major suffering for more than 100 countries, especially in South and Southeast Asia.[1,2] For instance, around millions of deaths occur in Bangladesh alone according to estimates, which is caused by arsenic poisoning and the same number of people are living under the threat of arsenic.[3] Given the potential hypertoxicity, bio-accumulation, and environmental persistence of arsenic, the World Health Organization (WHO) and many countries such as America and China have regulated the maximum limit of arsenic in water for safe drinking as 10 μg/L.[2] Reliable and economical methods are urgently needed for its efficient removal from water given the enormous threat of arsenic on human health.

Numerous techniques, such as precipitation, ion exchange, coagulation, membrane separation, and adsorption, have been developed to remove arsenic from contaminated water,[4,5] among which adsorption is a promising method due to its simplicity of operation and low cost.[6,7] Over the past few decades, a variety of adsorbents have been utilized to reduce arsenic in water, especially nanomaterials;[8,9] however, most adsorptions are effective only for As(V) oxyanions (H2AsO4– or HAsO42– and its pKa1, pKa2, and pKa3 are 2.2, 6.97, and 11.53, respectively) but not valid for As(III) because of its neutral charge (H3AsO3 and its pKa1, pKa2, and pKa3 are 9.2, 12.13, and 13.4, respectively).[10] Of note, the dominant species of arsenic in surface water or groundwater environment is arsenite (As(III)),[11] which exists as a neutral species in most aqueous environments and exhibits much higher mobility and toxicity than inorganic arsenate (As(V)).[12] Hence, the peroxidation of As(III) to As(V) is usually regarded as an effective strategy to improve the removal efficiency of As(III).[13−15] To date, the removal of As(III) using various nanocomposite adsorbents have been proposed, which are capable of integrating oxidation and adsorption processes.[15−19] Nevertheless, some disadvantageous factors, such as the complex synthesis routes of materials, prohibitive cost of synthetic reagents, as well as time-consuming process, have limited the scaled-up application of this strategy for arsenic removal. For instance, the preparation of redox polymer-based Fe(III) oxide nanocomposite (HFO@PS-Cl) requires more than eight reagents; moreover, it needs 700 min to reach adsorption equilibrium for the removal experiments of 1 mg/L arsenic.[15]

Nowadays, considerable research studies have been devoted to developing new approaches for As(III) oxidation and sequestration. Several studies indicated that a heterogeneous catalytic process, such as H2O2/Al2O3,[20] PMS/CuFe2O4,[8] and persulfate/nZVI[21] coupling oxidation with adsorption would simplify the overall removal process. Nanosized metal oxide such as Cu(II),[22,23] Fe(III),[24,25] and Zn(II)[26,27] oxides are most widely used in arsenic adsorption owing to their specific affinity toward the target pollutants.[28] Compared to other metal-based materials, the spinel-type particle is very promising taken into account its excellent properties, such as the recoverability and mechanical stability,[29,30] but the coupling reactions have been rarely studied.[8]

As one of the emerging spinel nanoparticles, CuAl2O4 has shown promising prospects for catalytic oxidation, such as ethanol dehydrogenation,[31] ozonation,[32] etc. With prospective potential for many other applications, CuAl2O4 would probably be more reactive in the heterogeneous process, in particular the removal of arsenic due to its higher specific surface area than CuFe2O4.[32,33] In this work, to extend the heterogeneous process for arsenic removal, CuAl2O4 nanoparticles were synthesized by a simple method and used to activate PMS for simultaneous oxidation and adsorption of arsenic, particularly in the complex aqueous environment. Moreover, to investigate the active oxygen species (ROS) responsible for the oxidation of As(III), a series of radical quenching experiments and instrument verification were tested. The potential reaction mechanisms for the promising removal properties of the CuAl2O4/PMS system were proposed. Finally, the removal efficiency of As(III) at a low concentration (<1 mg/L) in different water samples were characterized. The results of the experiment suggested that the joint method may provide an option to efficiently remove the low concentration arsenic contaminant from water without adjusting the pH value and peroxidizing natural organic matter (NOM).

Results and Discussion

Characterization of CuAl2O4

The X-ray diffraction (XRD) pattern of CuAl2O4 and CuO are presented in Figure a. The crystalline phases of CuAl2O4 show some amorphous agglomerates because of multiple peaks. The different peaks of CuAl2O4 located at 19.01°(111), 31.29°(220), 36.86°(311), and 44.84°(400) matched well with the spinel phase (CuAl2O4, JCPDS No. 78-1605). Moreover, obvious reflection peaks of CuO located at 35.56 (1̅11), 38.76°(111), 48.74°(2̅02), and 53.53°(020) were observed (CuO, JCPDS No. 80-1268) in the spectra of the CuAl2O4 sample, which is observed in the following scanning electron microscope (SEM) images. Especially, the well-crystallized CuO nanoparticles were obtained and used as a comparison in this work.

Figure 1

(a) XRD spectra of CuAl2O4 and CuO calcined at 400 °C, (b) SEM and (c) energy dispersive X-ray spectroscopy (EDX) images of CuAl2O4.

The morphology and microstructure of CuAl2O4 with a calcination temperature of 400 °C were examined by field emission scanning electron microscope (FESEM). As shown in Figure b, the porous grid-like surface was seen in the SEM images of CuAl2O4, which could provide abundant active sites for PMS catalytic oxidation and As(III) adsorption. Moreover, a spot of sphere-like CuO particles with a smooth and well-defined edge were anchored to the CuAl2O4 surface, and Al2O3 appears like nanowires in this work, which was prepared by sol–gel methods (Figure S1). The as-prepared grid-like CuAl2O4 is agglomerated by small flake-shaped nanoparticles. This nanoscale structure is naturally favorable for the interaction with targeted contaminants and oxidants during the catalysis and adsorption process. Furthermore, the representative elemental composition of the particle surface suggests CuAl2O4 with a metal element atomic ratio of 1:2 was successfully prepared, as seen from the EDX analysis images (Figure c).

The removal efficiency of As(III) by CuAl2O4 at different calcination temperatures was investigated, and the results are depicted in Figure S2. The highest As(III) removal efficiency was obtained by the sample calcined at 400 °C, which may be attributed to the higher Brunauer–Emmett–Teller (BET) and more Lewis acid sites.[33] Thus, the following experiments were examined for a fixed calcination temperature at 400 °C. Of note, most of the organic residuals generated from the sol–gel process could be decomposed during the thermal treatment of 300 °C.[35]

Activity of As-Prepared Adsorbents

Fast and efficient removal of heavy metals is one of the targets for wastewater treatment. The higher specific surface area and pore-volume of CuAl2O4 could provide more binding sites for arsenic removal (Table S1). Herein, the removal tests of the different initial concentrations of As(III) were conducted, and the results are depicted in Figure a. Intriguingly, CuAl2O4 coupled with the PMS system, presented excellent arsenic removal in a relatively short time, even at a high arsenic initial concentration, and fast reaction equilibrium was obtained and As(III) was almost completely removed in 100 min at a 1 mg/L initial concentration. Furthermore, even at 11 mg/L initial arsenic concentration, the expected removal rate of 74.7% was still obtained within 100 min.

Figure 2

(a) As(III) removal of different initial concentration by the CuAl2O4/PMS system, (b) removal efficiency of As(III) with different adsorbents, (c) the removal efficiency of As(III) with different adsorbent dosages, and (d) the corresponding rate constant of (c). Conditions: [As(III)] = 1.0–11.0 mg/L for (a), [As(III)] = 1920 μg/L for (b), [PMS]0 = 200 μM, Solid dosage = 300 mg/L, pH = 7.0, T = 25 °C.

Given that copper and aluminum could be predominant active sites in CuAl2O4,[33,36] a series of contrastive tests were performed to specify the role of each related process involved in As(III) removal. Figure b shows the variations of As(III) removal efficiency with time in PMS coupled with CuAl2O4, CuO, Al2O3, and CuO + Al2O3 (1:1), respectively. Obviously, CuO and Al2O3 slightly improved As(III) removal compared with CuAl2O4 adsorption alone. In contrast, it was noticed that a rapid decrease of the As(III) and almost complete removal of As(III) was achieved with the PMS/CuAl2O4 mixture. The results revealed that the activity of CuAl2O4 is higher than that of well-crystallized CuO and Al2O3 nanoparticles; this may be related to the unique spinel structure of CuAl2O4. A detailed discussion regarding the efficiency of CuAl2O4 and the possible mechanism for As(III) removal will be presented in the next section.

In this work, it can be observed that the As(III) removal kinetic data under different dosages of PMS/CuAl2O4 or PMS/CuO can be fitted well with the pseudo-second-order model (as in eq ).where q and qe (mg·g–1) represent the adsorption capacity at time t (min) and at equilibrium, respectively. k2 (g·mg–1·min–1) is the adsorption rate constants of the pseudo-second-order process. The reaction rate constant obtained from the model could be used as a more intuitive index to compare reaction activity for As(III) adsorption. As depicted in Figure c,d, the removal efficiency with a remarkable increase from 79.69 to 100% could be observed when the dosage of CuAl2O4 increases from 0.05 to 1.2 g/L. Comparatively, the As(III) removal efficiency by CuO/PMS is far lower than that of CuAl2O4/PMS. The k2 value in the CuAl2O4/PMS system is higher than that in the CuO/PMS system irrespective of the dose of CuO used for the reaction. The results suggested that CuAl2O4 has better adsorptive and catalytic performance than CuO. As the CuAl2O4 dosage increased from 0.05 to 0.9 g/L, the k2 increased progressively, but a further increase in the adsorbent dosage would cause a slight drop in the rate constant. More adsorbent dosages would provide more adsorption sites, whereas excess adsorbent probably induced the diffusion limitation phenomenon in heterogeneous reactions, similar to previous studies.[36−38]

As(III) Removal Isotherm

In addition, the adsorption properties of As(III) by CuAl2O4 were also investigated, and the data were fitted by Langmuir and Freundlich adsorption models, as expressed in eqs and 3, respectively. The obtained adsorption parameters are listed in Table S2.where qe is the As(III) capacity at equilibrium (mg/g) and qmax is the maximum adsorption capacity (mg/g), b is the binding constant (L/mg), KF is roughly an indicator of the adsorption capacity, and n is the heterogeneity factor. As clearly shown in Figure , the adsorption capacity and efficiency of As(III) at a higher PMS dosage (66 μM to 4 mM) was slightly higher than that at a fixed PMS dosage of 200μM. The maximum adsorption capacity from the Langmuir model is 66.25 mg/g, in fact, it can be believed that the qmax should be higher because the highest qe in adsorption experiments does not reach a plateau. A simple review on different spinel-type particles as adsorbents for removing arsenic contaminant are presented in Table S3. Comparatively, the arsenic adsorption capacity in this experiment outperforms that of many other adsorbents, such as MnFe2O4 (27.27 mg/g) and CuFe2O4 (41.2 mg/g), indicating that the CuAl2O4/PMS system is a promising alternative for arsenic removal. The higher correlation coefficients of the Freundlich model under both PMS dosages suggest that the adsorption process of As(III) on CuAl2O4 exhibits irreversible adsorption on the heterogeneous adsorption site.[8]

Figure 3

Adsorption isotherms for As(III) by 300 mg/LCuAl2O4 with different PMS dosage at initial pH 7.0 and 25 °C.

Effect of PMS on As(III) Removal

The removal kinetics of As(III) by CuAl2O4 with different amount of PMS are illustrated in Figure . Obviously, the addition of PMS can remarkably enhance the kinetics of As(III) adsorption regardless of the concentration of PMS. Nearly 59.4% As(III) removal was observed during CuAl2O4 adsorption alone in 100 min, but the removal efficiency reached 99.2% in the presence of PMS at concentrations as low as 50 μM. Residual As(III) can drop dramatically from the 960 μg/L initial concentration to <10 μg/L in 100 min (Figure S3) in the presence of PMS. This result proves that PMS did promote As(III) adsorption on CuAl2O4. No obvious difference in the removal efficiency can be observed when the PMS concentration further increased to 500 μM. However, the k2 and qe fitted to the pseudo-second-order model increased from 0.2504 to 0.4231 g·mg–1·min–1 and from 1.932 to 3.224 mg·g–1, respectively (Table S4). The results revealed the potential environmental application of the CuAl2O4/PMS system to remove even high concentrations of organic and inorganic contaminants.

Figure 4

As(III) removal by CuAl2O4 with different PMS dosages. Conditions: [As(III)]0 = 960 μg/L, solid dosage = 300 mg/L, pH = 7.0, T = 25 °C.

As(III) Adsorption Mechanism

In general, the removal of As(III) by adsorbents is realized via two pathways: one is the direct adsorption of As(III), and the other is the oxidation of As(III) into As(V) followed by As(V) adsorption. Such enhancement in As(III) removal by the CuAl2O4/PMS system possibly results from the oxidation of As(III) to As(V) followed by adsorption. To further verify this assumption, the adsorption kinetics of As(V) by CuAl2O4 was also investigated, and the results are illustrated in Figure . The adsorption efficiency of As(III) and As(V) by CuAl2O4 alone are 44.8 and 84.5%, respectively, indicating that As(V) is more easily adsorbed by CuAl2O4. However, after the addition of PMS, the adsorption efficiency of As(III) was rapidly increased to 94.8%. Meanwhile, the adsorption efficiency of As(V) was also increased to 99.1% from 84.5%, as shown in Figure a. The results indicated that the addition of PMS plays a key role in promoting the adsorption capacity of both As(III) and As(V) on CuAl2O4. More efficient As(III) removal could be ascribed to the in situ oxidation of As(III) to As(V) under the existence of PMS.

Figure 5

(a) Adsorption efficiency of As(III) and As(V) with CuAl2O4 with or without PMS. Conditions: [As(III)]0 = 1450 μg/L, [As(V)]0 = 1480 μg/L,[PMS]0 = 200 μM, [CuAl2O4]300 mg/L. (b) XPS spectra of As 3d and (c) O 1s on the surface of CuAl2O4 after As adsorption.

After fulfilling the arsenic adsorption experiments with or without PMS, the arsenic species on the surface of CuAl2O4 were identified by XPS analysis, and the results are presented in Figure b. Obviously, the As(III) adsorbed on CuAl2O4/PMS was in the form of As(V), that is, the adsorbed As(III) was completely oxidized to As(V). Interestingly, the adsorbed As(III) by CuAl2O4 alone was also completely transformed to As(V) rather than partial oxidation. No obvious As(III) species can be observed in the XPS spectra.

In general, three possible mechanisms, monodentate mononuclear, bidentate mononuclear, and bidentate binuclear complexes could depict the inner-sphere arsenic complexes.[39,40] The surface hydroxyl groups (OH–) on metal-based oxide are generally supposed to have a great influence on arsenic adsorption. After the uptake of As(III) on CuAl2O4, the speciation of the O element on the prepared samples was probed with the aid of XPS analysis. The sample of fresh CuAl2O4 was also measured for reference, and the obtained results are shown in Figure c. Obviously, the content ratio of fresh CuAl2O4 and CuAl2O4 saturated with arsenic was nearly 0.65, revealing that the monodentate complexes could inform in adsorbent surface. Furthermore, the ratio of OH–/O2– on the surface of CuAl2O4 increased from 34.11 to 56.38 and 58.28% for fresh CuAl2O4, CuAl2O4 adsorbing alone, and CuAl2O4 adsorbing under PMS, respectively, indicating the formation of inner-sphere monodentate mononuclear species.[8,15,39,40] Conclusively, the fast and efficient As(III) removal might occur through the following pathways: in situ oxidation of As(III) followed by the direct sequestration on CuAl2O4. Afterward, the inner-sphere monodentate mononuclear complexes could produce between arsenic and CuAl2O4.

Mechanism of In Situ Oxidation of As(III)

Extensive research studies presented that PMS can be induced by spinel-type materials to generate powerful reactive oxygen species (ROS), such as radicals, for the degradation of organic matter.[36,37] Thereby, we can infer that PMS could release a large number of involved reactive species for the oxidation of As(III) in solution. To clarify the major ROS in the CuAl2O4/PMS/As system, a series of radical scavenging experiments were carried out. As depicted in Figure S4, a significant decrease in the removal efficiency of As(III) was observed when EtOH or benzoquinone (BQ) was added into the reaction system and even presented a higher negative effect when the dosages of EtOH or BQ were further increased. In contrast, the removal efficiency of As(III) slightly increased from 97.3% (without quencher) to 97.8% in 40 min in the presence of another radical quencher, tert-butyl alcohol (TBA). It is well-known that both •OH and SO4•– can be rapidly scavenged by EtOH.[41] BQ and TBA are widely used as a standard scavenger for O2•– and •OH, respectively.[42−44] This particular phenomenon could be attributed to •OH generated on the CuAl2O4 surface, and it may preferentially react with the target adsorbed onto the catalyst. In contrast, TBA with hydrophilic properties predominantly scavenged the radicals in solution rather than those on the surface of the catalysts. Similar results were also found by others.[33,45]

To further confirm the existence of ROS, electron paramagnetic resonance (EPR) experiments were conducted with 5,5-dimethyl-1-pyrroline N-oxide (DMPO) as the spin trapping agent. The dosage of PMS was fixed at 3 mM to clearly identify the radical species, and the results are shown in Figure . No apparent signal was detected in the presence of PMS alone, suggesting that there are nearly no radicals in the bare PMS solution at neutral pH. After adding CuAl2O4, four-line peaks with the relative intensities of 1:2:2:1 was detected, which matched well with the spectra of the DMPO/•OH adduct.[46] The characteristic spectra of the DMPO/SO4•– adduct generated from the sulfate radical were also identified clearly,[47] as shown in Figure a. Meanwhile, the signal for the DMPO/O2•– adduct was also clearly observed (as shown in Figure b). The EPR results revealed that both •OH, SO4•–, and O2•– contributed to As(III) oxidation in the CuAl2O4/PMS system.

Figure 6

EPR spectra of (a) DMPO-SO4•–/•OH and (b) DMPO-O2•–.

In this experimental process, the solution of the CuAl2O4/PMS system was in equilibrium with the atmosphere in terms of oxygen. To investigate the effect of dissolved oxygen on ROS, control experiments were carried out. The results confirm that there was not much difference between the air-saturated and N2-saturated systems (as shown in Figure S5), implying that the generation of ROS is not related to dissolved oxygen. Accordingly, it is reasonable to believe that the generation process of active species could be related to the redox reaction between Cu(II)/Cu(I) based on previous reports on CuFe2O4.[36,38,48,49] The possible generation mechanism of ROS in the CuAl2O4/PMS system can be depicted as followsTo determine the reactive sites and verify the above hypothesis, the surface characteristics of the used CuAl2O4 were characterized by XPS analysis, and the results are presented in Figure . In the high-resolution XPS spectra of Cu 2p, the signal could be divided into six peaks at around 932.8, 934.5, 941.6, 943.7, 952.6, and 954.2 eV. The main peaks at binding energies of 934.5, 954.2, and 941.6 eV are attributed to the Cu(II) oxide species,[33,50] and the peaks appearing at the binding energies of 932.8, 943.7, and 952.6 eV are assigned to the Cu 2p3/2 and Cu 2p1/2 characteristic peaks of Cu(I), respectively.[51−53] No Cu(I) could be detected on the surface of the fresh CuAl2O4 because only Cu(II) can be formed on the surface of particles under the conditions of high-temperature calcination.[33] The results of XPS suggest that Cu(II) was transferred to Cu(I) partially by eq (29) in the CuAl2O4/PMS system, and the cycles of Cu(II)/Cu(I) could emerge through eqs and 6, meanwhile generating SO4•– and •OH.[29,38] The generation of •OH could also be attributed to the oxidation of SO4•– (eqs and 8),[8,29,36] and O2•– might be produced from the interaction of •OH with H2O2 in the solution, as described by eqs –11.[33,50] Likewise, the conversion of Cu(II)/Cu(I) was observed even in the absence of PMS, suggesting that the copper might play a key role in the adsorption of arsenic.

Figure 7

XPS spectra of Cu 2p for CuAl2O4 used with or without PMS.

Effect of Solution Chemistry

Effect of pH

The pH of the medium is a significant parameter for the removal of contaminants because the surface charges of adsorbents and the particular species of As(III)/As(V) are significantly related to solution pH.[51] In this study, the role of the initial solution pH ranging from 3 to 11 in the removal of As(III) in the CuAl2O4/PMS system was investigated, and the results are shown in Figure a. Interestingly, the CuAl2O4/PMS system presented appreciable As(III) removal efficiency under acidic, neutral, and weakly basic (pH 9) conditions. Under all pH conditions (pH 3–9), the residual arsenic can dramatically reduce to <10 μg/L when equilibrium was reached. However, upon further increase of pH to 11, the removal efficiency of As(III) obviously decreases, and the residual As(III) increased to 270 μg/L. Such suppression of As(III) removal at pH 11 probably results from the electrostatic repulsion between arsenate oxyanions (H2AsO4– and H2AsO42–) and adsorbents because the pHpzc of the CuAl2O4 was around 8.1 (Figure S6). Similar phenomena have been reported for other adsorbents.[8,19,54]

Figure 8

Influence of As(III) removal by CuAl2O4/PMS system as a function of (a) pH, (b) coexisting anions, and (c) humic acid (HA). Conditions: [As(III)]0 = 960 μg/L, [PMS]0 = 200 μM, Solid dosage = 300 mg/L, pH = 7.0 (for b, c), T = 25 °C.

Moreover, we tested the stability of the adsorbent in various pHs, and the results present the lower ion leaching of CuAl2O4. The concentrations of the leached Cu2+ and Al3+ are lower than 50 and 120 μg/L, respectively, when reached equilibrium for the entire pH range 3.0–11.0, suggesting the high stability of CuAl2O4 in potential applications (Table S5). Generally, the leaching of metal ions is serious under acidic conditions because of the vulnerable nature of adsorbents containing a metal oxide, but that was not observed in this work. The desirable acidic resistance of prepared CuAl2O4 is probably ascribed to the unique structure containing copper/aluminum oxides.[55] The high As(III) removal efficiency and low metal leaching of CuAl2O4 in the presence of PMS in the pH range of 3–9 indicates that this As(III) removal system can be extensively used in practical arsenic removal for general contaminated water without adjusting the pH value.

Effect of Inorganic Anions

Herein, the effect of some inorganic anions on the removal efficiency of As(III) by the CuAl2O4/PMS system was examined. The increase in the ionic strength is known to significantly influence outer-sphere interactions between the solute and the particle’s surface due to the reduced ζ-potential of the particle surface in water, whereas inner-sphere complexation is not affected.[38] The As(III) removal was nearly unaffected when the concentration of NaCl, which was used to adjust the ionic strength, increased from 20 to 200 mM in the reaction solution (Figure S7). This reveals again that the As(III) and the generated As(V) anions adsorbed onto CuAl2O4 by means of the formation of strong inner-sphere complexes. As seen from Figure b, nitrate, sulfate, and carbonate at a concentration of 50 mg/L, even 100 mg/L in aqueous, exhibited negligible effect on As(III) removal because these anions are adsorbed were mainly via electrostatic attraction.[15] Comparatively, phosphate suppresses the As(III) adsorption slightly, and the removal efficiency of As(III) decreased from 96.6% (without phosphate) to 82.6% (with 100 mg/L phosphates) in 40 min. The difference in removal performance was probably attributed to the strong competition for binding sites of the CuAl2O4 particle between phosphate and arsenic. Some reports revealed that both of these can form inner-sphere complexes with the hydroxyl groups at the surface of adsorbents.[10,56] However, the high arsenic removal efficiency was observed in this work even in the presence of higher phosphate concentration, compared with other research studies.[8,18,51]

Effect of NOM

The effect of humic acid (HA, representing the natural organic matter (NOM)) on the removal of As(III) was tested, and the corresponding results are presented in Figure c. Interestingly, compared to the adsorption in the absence of HA, no significant influence on the removal efficiency of As(III) can be observed even at a HA concentration of 20 mg/L. These results might arise from the nanoporous structure of the prepared adsorbent. The molecules of HA with relatively large size could not be allowed to diffuse into such narrow pores; thereby, a higher concentration of HA exert a negligible effect on As(III) removal. Such favorable properties against HA is crucial for the practical application of the CuAl2O4/PMS system.

Cycles and Regeneration

The regeneration of the used adsorbents plays an important role in the improvement of the removal capacity of adsorbents.[2] As for the arsenic adsorbed on the adsorbents, numerous studies proved that it can be desorbed effectively by soaking in alkaline solution.[15,57] To confirm the adsorptive properties and stability of CuAl2O4, the used adsorbents were recovered (e.g., centrifugation) and allowed to react with 5% NaOH for 6 h, then dried at 80 °C. Under otherwise identical conditions, the As(III) removal efficiency with CuO/PMS decreased evidently with the number of reaction cycles, whereas that with CuAl2O4/PMS showed only a slight decrease up to six recycles, as depicted in Figure . The removal efficiency of As(III) still reached 86.3% within 40 min after six successive adsorption–desorption cyclic runs; the detailed removal dynamics and the corresponding k2 are displayed in Figure S8. The potential change of the CuAl2O4 morphology and microstructure after six cycles were characterized by SEM, XRD, and EDX. Generally, the lower diffraction intensity possibly ascribes to the poor crystallinity of the oxide.[58,59] As shown in Figure S9, the results clearly show that no apparent changes in the diffraction intensity of the used CuAl2O4 can be observed compared with the fresh one. Not much difference was observed in the SEM and EDX images (Figures S10 and S11). The readily recyclable and stable structure implies that CuAl2O4 could facilitate the large scale application and could reduce the treatment cost.

Figure 9

Removal efficiency of As(III) by the CuAl2O4/PMS system as a function of cyclic regeneration. Conditions: [As(III)]0 = 1450 μg/L, pH = 7.0, T = 25 °C, [PMS]0 = 200 μM, solid dosage = 300 mg/L.

Practical Application

To assess the removal capacity of the CuAl2O4/PMS system in practical applications, the removal of As(III) from practical water samples were conducted for tap water, river water, and self-made water in lab (containing CO3–: 20 mg/L; SO42–: 20 mg/L, NO3–: 20 mg/L and Cl–: 20 mg/L). Water samples were prepared by spiking with As(III) to prepare a simulated wastewater sample for investigating As(III) removal. Considering the relatively low arsenic content in natural groundwater, the initial arsenic concentration (C0) in removal experiments was specified as 100–700 μg/L. As presented in Figure S12, the removal efficiency of As(III) (C0 ≤ 500 μg/L) is 100% within 40 min irrespective of the kind of water sample. Even for C0 = 700 μg/L, the Ce values (the equilibrium concentration of arsenic (μg/L)) are less than the maximum contaminant level (10 μg/L) when the reaction reached equilibrium. The expected removal efficiency could be obtained despite the fact that existing background organic matters in real water partly hindered the removal of arsenic, which can be avoided by properly increasing adsorbent dosage.

Conclusions

A simple, nonhazardous, low energy input, and highly efficient oxidation and adsorption process is always desirable for the removal of heavy metals with high toxicity in wastewater treatment. The results of this study suggested that spinel CuAl2O4 coupled with PMS might be a feasible choice to meet this requirement. The fast and efficient oxidation and sequestration of arsenic from an aqueous solution could be attributed to the promising reactivity of the heterogeneous process. More accessible reactive sites in CuAl2O4 were supposed to be for the oxidant and target contaminant due to the special grid-like surface with a high specific surface area and larger pore-volume of CuAl2O4. Some ubiquitous anions showed negligible effect on As(III) removal in the CuAl2O4/PMS system, even high concentrations of phosphate. In this work, the advantages of this heterogeneous process of removing arsenic from water with CuAl2O4 can be summarized as (i) integration of oxidation and sequestration, (ii) negligible ion leaching of adsorbent because of the unique spinel structure, (iii) without by-products generation and energy input, and (iv) economical, efficient, and simple operation. The experimental data clearly state that the CuAl2O4/PMS system is an ideal bifunctional water purifier capable of simultaneous oxidation of As(III) to As(V) and effective sequestration of both of them; as such, it is believed to have great potential for environmental applications.

Materials and Methods

Materials

As (III) and As (V) stock solutions (1.0 g/L), peroxymonosulfate (KHSO5·0.5 KHSO4·0.5 K2SO4), and 5,5-dimethyl-1-pyrroline N-oxide (DMPO) were purchased from J&K Chemical Company. 1,4-Benzoquinone (C6H4O2, BQ), tert-butyl alcohol (C4H10O, TBA), ethanol (C2H6O, EtOH), humic acid (HA), copper nitrate (Cu(NO3)2·3H2O), aluminum nitrate (Al2(NO3)3·9H2O), and citric acid (C6H8O7·H2O) were purchased from Macklin Biochemical Co., Ltd (Shanghai, China). Details of all reagents mentioned above are listed in Table S6. Stock solutions were always prepared in ultrapure water produced by a Milli-Q system.

Preparation of Copper Aluminate

The CuAl2O4 nanoparticle was synthesized by a modified citrate combustion method according to the previous reports.[33] Briefly, the precursors with an atomic ratio of copper and aluminum of 1:2 were dissolved in 100 mL of ultrapure water. Then, the specific amount of citric acid (0.03 M) was dissolved in 100 mL of ultrapure water and was drop-wise added to the solution containing metal precursors under magnetic stirring for 2 h to form a mixed solution. The obtained homogeneous solution was constantly stirred at 90 °C until the formation of a blue sticky gel, and then the gel was calcined at 400 °C for 4 h. Finally, the resultant brown particles were ground, washed with ultrapure water until the pH reached neutral, and dried at 80 °C for 24 h.

Pure CuO and Al2O3 nanoparticles were used as a comparison in this work and were prepared according to the aforementioned methods used for CuAl2O4, but without aluminum nitrate and copper nitrate, respectively.

Characterization and Analysis

The crystal phases of the prepared materials were determined by X-ray diffraction (XRD, Bruker D8 Advance, Germany) with a Cu Kα X-ray source (40 kV, 40 mA). The surface morphology was visualized using a field emission scanning electron microscope operating at 5 kV (FESEM, Hitachi SU-8010, Japan). The content of metal element was examined by Energy Dispersive X-Ray Spectroscopy (Shimadzu EDX-720, Japan). The BET surface area and average pore size of the oxides were determined on a JW-BK 112 analyzer. The surface chemistry properties of the adsorbent were determined by X-ray photoelectron spectroscopy (XPS, Thermo Escalab 250XI, American). The generated reactive oxygen species from aqueous were confirmed using an electron paramagnetic resonance spectrometer operating at a resonance frequency of 9.77 GHz, microwave power of 19.97 mW, and modulation amplitude of 1.0 G (EPR, Bruker Biospin GmbH E500-9.5/12, Germany).

Batch Experiments

A common stock solution of each reactant was prepared first. The above stock solutions were aliquoted and then blended to achieve the initial experimental conditions. All reactions were carried out in 100 mL brown glass vials. Unless otherwise specified, the initial pH values of aqueous solution were adjusted to 7.0 ± 0.5 using 0.1 M HNO3 or NaOH solution, with the catalyst dosage of 300 mg/L. Specific amounts of CuAl2O4 and PMS solution were initially dispersed in 80 mL of ultrapure water. After mixing for 1 min, a certain dosage of As(III) stock solution was added to start the reaction. The suspension was stirred at room temperature (25 ± 2 °C) at a rate of 120 rpm under air, and samples were withdrawn through 0.45 μm filters to remove the catalyst at predetermined time intervals. To accurately analyze the concentration of arsenic, excess sodium nitrite was immediately introduced into the filtrate to quench residual PMS. All of the As(III) removal experiments were performed in triplicate, and the average results were reported, with error bars in figures representing one standard deviation.

Analytical Methods

The concentration of As(III) was determined using an atomic fluorescence spectrophotometer (AFS-9700) (Shimadzu, Japan), and the concentration of As(V) was characterized using the modified molybdate-based method.[34] Briefly, 2 mL of the reaction solution was acidified with 0.3 mL of 1% HCl rather than HNO3 immediately after withdrawal and mixed with specified molybdate agent in a 4 mL cuvette. The absorbance of the mixed solution was measured at 880 nm with a UV–vis spectrophotometer (TU-1901, China) after 20 min. The leaching metal ions from the catalysts in the solution were analyzed with an atomic absorption spectrophotometer (AAS) (TAS-990, Purkinje, China) after the completion of the removal experiment.