Solvent Effect on the Solvothermal Synthesis of Mesoporous NiO Catalysts for Activation of Peroxymonosulfate to Degrade Organic Dyes.

In this work, we successfully prepared three different mesoporous NiO nanostructures with preferential (111) planes using three different solvents-water, a water-ethanol mixture, and a water-ethylene glycol mixture. The NiO nanosheets prepared from the water-ethylene glycol mixture and denoted as NiO-EG showed a nanosheet morphology thinner than 10 nm, whereas the water-ethanol and water samples were 30-40 nm and above 100 nm thick, respectively. The NiO-EG catalyst was found to exhibit a high catalyzing ability to activate peroxymonosulfate (PMS) for decoloring dyes, by which 94.4% of acid orange 7 (AO7) was degraded under the following reaction conditions: AO7 = 50 mg/L, catalyst = 0.2 g/L, PMS = 0.8 g/L, pH = 7, and 30 min reaction time. The dye degradation rate was investigated as a function of the catalyst dosage, pH, and dye concentration. According to quenching experiments, it was found that SO4 •-, HO•, and O2 •- were the dominant radicals for AO7 degradation, and oxygen vacancies played a significant role in the generation of radicals. High surface area, thin flaky structure, rich oxygen vacancies, fast charge transport, and low diffusion impedance all enhanced the catalytic activity of NiO-EG, which exhibited the highest degradation ability due to its abundant accessible active sites for both adsorption and catalysis.

Introduction

Environmental pollution, especially water pollution, is increasingly becoming a challenging multidisciplinary problem with the rapid development of industry. Water pollution due to recalcitrant organic contaminants could lead to serious ecological impacts.[1,2] Many organic contaminants are toxic and non-biodegradable with lethality and carcinogenicity.[3] Meanwhile, organics are difficult to remove by the conventional biological, physical, and chemical methods of wastewater treatment.[4] Therefore, advanced treatment techniques that offer remarkable treatment efficiency are highly desired.

In recent decades, advanced oxidation processes (AOPs) aiming at treating refractory organic matter in industrial wastewater have attracted increasing attention. AOPs involve highly reactive radicals such as hydroxyl radicals (HO•) as the main oxidation species, which can attack the non-biodegradable organic molecules and oxidize them into biodegradable low-molecular-weight organics or inorganics, such as water and carbon dioxide.[5−7] Recently, besides HO•, a sulfate radical (SO4•–) has become a promising alternative due to several advantages compared with HO•. SO4•– can be generated by activation of peroxymonosulfate (PMS, HSO5–, or persulfate (PS, S2O82–)) via ultrasound,[8] ultraviolet,[9] metal-free heterogeneous catalysis,[10] transition-metal catalysis,[11,12] and miscellaneous activation methods. In addition, SO4•– possesses an oxidation potential E0 = 2.5–3.1 V,[13] which is comparable to or even higher than that of HO• (E0 = 2.8 V),[14] long lifetime (t1/2 = 30–40 μs),[15] and high activity with organic compounds over a wide pH range of 2–8.[16] These properties make SO4•– an ideal oxidant of AOPs in practical applications.

Transition-metal compounds have been a research hotspot in the generation of SO4•– due to their low cost, convenience, and wide variety for selection. Anipsitakis and Dionysiou[17] in 2004 systematically studied several metal ions, Fe2+, Fe3+, Co2+, Mn2+, Ni2+, Ru3+, Ce3+, and Ag+, for the activation of peroxymonosulfate, hydrogen peroxide (H2O2), and peroxydisulfate for degradation of 2,4-dicholorophenol. In recent years, heterogeneous metal oxide compounds have received more attention attributed to their high activity and good recyclability. Up to now, single and bimetallic oxides consisting of active Co, Fe, Mn, and Cu have been reported.[18] Nickel is a transition metal with valence-changing properties. Therefore, nickel compounds, such as NiO, have been widely investigated as photocatalysts or electrocatalysts. However, as for the PMS catalytic activation, although studies concerning bimetallic or trimetallic oxides, such as NiFe2O4[19] and NiCo2O4,[20] have been reported, the single oxide, NiO, has never been investigated for catalyzing PMS.

On the other hand, the morphology of catalysts significantly affects their properties.[21] It was found that the shape and crystalline facets and pore microstructure affect the catalytic properties to a great extent.[22,23] NiO nanoparticles with various morphologies have been synthesized by different methods and used in the catalytic field.[24] The effect of the NiO microstructure on their catalytic performance for oxidative dehydrogenation of propane to propene has been reported.[25] Layered structure materials usually have high specific surface area and fast electron transfer, which can accelerate the catalytic reaction. It is well known that the solvothermal route can be widely used for the synthesis of nanomaterials, which is convenient to control the microstructure, facets, and pore size of nanomaterials. It is reported that different microstructures such as nanoparticles,[26,27] nanorods,[28] nanosheets, spheres, nanoflowers consisting of flakes, and nanoflakelets[29] have been prepared.

In this study, three mesoporous NiO with different morphologies have been prepared via a facile solvothermal method only by changing the solvent. On this basis, their crystal structure, morphology, pore size distributions, and surface area were measured. The prepared NiO catalysts were found, for the first time, to activate PMS for degrading the organic dye, acid orange 7 (AO7), with high efficiency. The physicochemical properties, catalytic mechanism, and recyclability of the used catalysts were also reported.

Results and Discussion

Characterization

The X-ray powder diffraction (XRD) patterns of precursors achieved by the solvothermal process are shown in Figure S1. It can be seen that the NiO-EG and NiO-E precursors are both indexed to Ni3(NO3)2(OH)4 (PDF card No. 00-022-0752), whereas that of NiO-W is indexed to Ni(OH)2 (PDF card No. 00-014-0117). These samples after calcination at 350 °C exhibit XRD patterns attributed to NiO (JCPDS card no. 00-047-1049), in which the diffraction peaks at 37.2, 43.3, and 62.9° correspond to the (111), (200), and (220) reflections, respectively. No peaks of other crystalline phases are found, suggesting the high purity of NiO. Compared with NiO-E and NiO-W, the NiO-EG sample demonstrates an obviously higher (111) peak intensity, indicating a more preferential (111) lattice in this sample (Figure ).

Figure 1

XRD patterns of (a) NiO-EG, (b) NiO-E, and (c) NiO-W catalysts.

The difference in the solvent used has a critical influence on the morphology and microstructure of the NiO catalysts, as shown in scanning electron microscope (SEM) images (Figure ). NiO-EG presents as nanoflakes with only 1–10 nm thickness and abundant nanopores on the surface (Figure a,b). As for NiO-E, a flowerlike structure formed by self-assembly of nanosheets is obtained with a few pores on the surface of petals (Figure c,d). The flowerlike spheres have a size ranging from 1 to 8 μm and are formed by nanosheets of 15–30 nm thickness. In the presence of pure water as the solvent, NiO-W exhibits a regular hexagonal platelet shape and a sponge structure with a size of 1 μm and 100 nm thickness.

Figure 2

SEM images of (a, b) NiO-EG; (c, d) NiO-E; and (e, f) NiO-W catalysts.

Because all NiO catalysts were prepared under the same conditions except for solvent, we can infer that the difference in morphology is a result of changing the solvent system (Scheme ). As an organic solvent, ethylene glycol has two hydroxyl groups. Zhu et al. reported that because of the effects of hydrogen bonding between hydroxyl groups, ethylene glycol could serve as a ligand to form a chainlike coordination complex with metal ions, and the as-formed chainlike complexes would congregate and self-assemble into bundles when such chains become sufficiently long.[30] However, due to the presence of water and absence of a surfactant, some effects of ethylene glycol are weakened, such as viscosity and chelation, which prohibits the formation of spherelike superstructures but leads to the growth of flakes.[31] As shown in Figure S2, a significant difference observed between NiO-EG-2 and NiO-EG-6 is the size of nanosheets, which grew bigger after 4 h of the solvothermal process. These two samples have the same microstructure and lattice structure as NiO-EG, confirming the growth mechanism. Li et al. synthesized flowerlike CuS in an ethanol/water system. They found that ethanol has a tendency to self-aggregate in aqueous medium at high temperatures, thus forming micro-heterogeneities, which can improve the microemulsion droplet nucleation rate in the hydrothermal reaction process, meaning that ions can be assembled onto the microemulsion droplet surfaces more easily.[32] In the same way, Ni2+ followed a similar process in an ethanol/water system and self-assembled into a flowerlike structure. As for NiO-W, Ni(OH)2 is formed by the hydrolysis of urea as a precursor at the initial stage of the hydrothermal process.[33] The faster initial nucleation and growth rates and an oriented attachment mechanism result in the formation of a hexagonal platelet structure.[34]

Scheme 1

Formation Mechanism of NiO Catalysts

The NiO catalysts were further characterized by transmission electron microscope (TEM) and high-resolution TEM (HRTEM) (Figure ). All NiO catalysts exhibit a porous structure. In Figure b, the HRTEM image of NiO-EG nanoflakes demonstrates clear lattice fringes. The interplanar spacing is 0.147 nm, corresponding to the (220) planes of the isometric system of NiO. The inset of Figure b shows the selected-area electron diffraction (SAED) pattern of the NiO-EG nanoflakes, revealing that the nanoflakes consist of crystalline NiO. The set of diffraction spots can be indexed as the [1̅11] zone axis of cubic NiO. The SAED pattern of NiO-E (inset of Figure d) consists of three diffraction rings, which represent the (111), (200), and (220) planes, respectively, indicating that the petals of flowerlike NiO-E are polycrystalline. The TEM image of NiO-W (Figure e) shows a single hexagonal platelet shape, which agrees with the SEM result discussed above. The 0.147 nm intervals of the lattice observed in the HRTEM image (Figure f) agree well with the spacing of the (220) plane. This indicates that the NiO nanoplatelets are perfect single-crystalline sheets and the (111) planes form the main surfaces of the nanosheets.[35] The much thinner nanosheets of NiO-EG means it contains a much larger percentage of the (111) planes than the other two samples, which is consistent with the XRD results. It has been previously reported that the (111) planes of NiO have more unsaturated surface O atoms and higher surface energy than the commonly exposed (220) facets, which results in an enhanced catalytic activity in the activation of PMS.[36] Therefore, the NiO-EG catalyst is expected to possess higher catalytic ability for PMS decomposition and further dye degradation in this study.

Figure 3

TEM and HRTEM images of (a, b) NiO-EG; (c, d) NiO-E; and (e, f) NiO-W. The insets are the corresponding SAED patterns of NiO-EG, NiO-E, and NiO-W.

N2 adsorption and desorption measurements were performed to investigate the specific surface area, the average pore size, and its distribution in different NiO catalysts, and the results are depicted in Figure . The Brunauer–Emmett–Teller (BET) specific surface areas of the NiO-EG, NiO-E, and NiO-W catalysts are 87.36, 50.93, and 94.84 m2 g–1, respectively. The average pore sizes calculated by Barrett–Joyner–Halenda (BJH) methods are 3.1, 15.5, and 7.8 nm, respectively. The isotherm of NiO-EG is of type II, representing the presence of a macroporous structure, whereas the isotherms of the NiO-E and NiO-W catalysts are all of type IV, representing predominantly mesoporous structure characteristics. However, the isotherm of NiO-EG has a hysteresis loop of type H3, which confirms the existence of slitlike open pores, whereas a hysteresis loop of type H2 is observed in the isotherms of both NiO-E and NiO-W, which is characteristic of ink-bottle-type pores on the surface of catalysts. The pore is formed by the release of water or NO2 in the thermal transformation of Ni(OH)2 or Ni3(NO3)2(OH)4 into NiO. The molecular size of AO7 and PMS is 1.36[37] and 0.2 nm,[38] respectively, which is much smaller than the pore size. Compared with ink-bottle-type pores, the slitlike open pores on the surface of NiO-EG provide more accessible active sites and accelerate the catalytic reaction. Due to the narrow neck of the ink bottle, it is difficult for the dye molecules and PMS molecules to be adsorbed and attach to the active sites inside the catalysts and the reaction products to desorb, which is unfavorable for mass transfer. The pore structure is consistent with that in the thick NiO-E and NiO-W, as observed from the SEM and TEM images. The large pore size of NiO-E causes the loss of surface area and active surface sites. Although the NiO-W has a much larger pore size and an even higher surface area than NiO-EG, its thickness as high as 100 nm may hinder the dye molecule diffusion into and the product out of the pores.

Figure 4

N2 adsorption–desorption isotherms and the pore size distributions (inset) of (a) NiO-EG, (b) NiO-E, and (c) NiO-W catalysts.

Removal of AO7 with NiO Catalysts by PMS Activation

A typical degradation profile of AO7 by NiO catalysts is given in Figure . For comparing the capacity of different NiO catalysts in AO7 removal, a blank test was also conducted. According to the results, NiO-EG possesses the highest catalytic capacity in removing AO7 in the presence of PMS compared to NiO-E and NiO-W. For example, 94.4% of AO7 was removed in 30 min by NiO-EG, whereas 91.0% and 89.7% of AO7 were removed by NiO-E and NiO-W under the same conditions, respectively. Only 16% of AO7 was removed after 30 min under the same conditions without catalysts. Meanwhile, 42% of AO7 was absorbed by NiO-EG in 30 min without PMS and the AO7 concentration showed no change in the next 30 min, indicating that the NiO itself cannot degrade AO7 but has strong adsorption ability toward it. This suggests that all three NiO catalysts possess the capacity of catalyzing PMS to oxidize AO7, while different microstructures result in different catalytic abilities. NiO-W has the highest BET specific surface area and high adsorbance of AO7 but the lowest removal of AO7, owing to its microstructure. According to the SEM images, NiO-W has a spongelike structure with size ranging from 0.9 to 1.5 μm and 100 nm in thickness, which makes it more difficult for AO7 and PMS molecules to spread into the NiO-W nanoplatelets and attach to more active sites. However, NiO-EG displays a flaky structure with a thickness of a few nanometers and open porous surface, which provides more available active sites and accelerates the activation of PMS, resulting in the excellent performance of the NiO-EG catalyst.

Figure 5

AO7 removal with time in adsorption and catalytic oxidation. Reaction conditions: [AO7] = 50 mg/L, catalyst = 0.2 g/L, PMS = 0.8 g/L, and pH = 7.

The results of the catalytic degradation of AO7 and some other organic contaminants by catalyst/PMS systems reported previously are listed and compared with our study in Table . In comparison to other research studies, the NiO-EG prepared in this study exhibited higher catalytic efficiency even at an increased dye concentration and lower catalyst dosage. We explored the possible reasons besides the surface area and the (111) planes of the catalyst in our study, which was further investigated later.

Table 1

Comparison with Reported Works of Other Catalyst/PMS Systems or Targeted Contaminants

	dosage
system	catalyst (g/L)	oxidant (g/L)	target (mg/L)	time (min)	removal efficiency (%)	ref
NiO/PMS	0.2	0.8	50 (AO7)	30	94.4	this article
Co3O4/PMS	0.1	∼0.57	50 (2,4-DCP)	120	∼47	(39)
GAC/PMS	1.0	3.51	20 (AO7)	300	85	(40)
Fe3O4@C/PMS	0.6	0.1	20 (AO7)	60	77	(41)
Mn3O4/PMS	0.1	0.3	10 (RhB)	60	90	(42)
Co3O4/PMS	0.67	2.0	20 (Phenol)	30	∼60	(43)
GO-Fe3O4/H2O2	0.5	0.3	35 (AO7)	180	76	(44)
NiFe2O4/PMS	0.1	0.61	1.22 (BA)	60	82.5	(19)
NiCo2O4/PMS	0.2	0.3	10 (humic acid)	120	∼90	(20)

Electrochemical Impedance Spectroscopy (EIS) Measurements

EIS measurements were employed to investigate the charge transfer ability of the NiO catalysts. The results are shown in Figure . These spectra consist of a typical semicircle in the high–medium-frequency region and a linear portion in the low-frequency range, which can be ascribed to the charge-transfer resistance and diffusion impedance, respectively. The radius of the semicircle of NiO-EG is smaller than that of the other two NiO catalysts, indicating a decrease in the charge-transfer resistance and enhancement of the charge transfer in the NiO-EG catalyst. At low frequencies, NiO-EG presents a smaller slope than NiO-E and NiO-W, suggesting that the NiO-EG exhibits a low diffusion resistance inside the catalyst. This should be attributed to the flaky open-porous structure of NiO-EG and is propitious to the charge transfer for PMS activation (Table ).

Figure 6

Nyquist plots of NiO catalysts.

Table 2

Physicochemical Properties and AO7 Removal of the Three Different NiO Catalysts

samples	shape	SBET (m2 g–1)	pore size (nm)	volume of pores (cm3 g–1)	AO7 removal in 30 min (%)
NiO-EG	flake	87.36	3.1	0.521	94.4
NiO-E	flower	50.93	15.5	0.452	91.0
NiO-W	hexagonal platelet	94.84	7.8	0.329	89.7

Effect of Catalyst Dosage, Initial AO7 Concentration, and Solution Initial pH

The influence of catalyst dosage on AO7 degradation is shown in Figure a. With the increase in NiO-EG dosage from 0.1 to 0.3 g/L, the AO7 degradation efficiency was enhanced from 68.0 to 98.3% in 20 min. As the catalyst dosage increased, more active sites were provided for PMS activation, which caused faster formation of active radicals, and more AO7 molecules were absorbed, resulting in the remarkable enhancement of AO7 degradation efficiency.

Figure 7

Effects of (a) catalyst dosage, (b) initial AO7 concentration, and (c) initial pH on the catalytic degradation of AO7 by the NiO/PMS system.

The influence of initial AO7 concentration on its degradation efficiency via the activation of PMS is shown in Figure b. At the initial AO7 concentrations of 25, 40 and 50 mg/L, the AO7 degradation efficiency achieved were 98.3, 98.3, and 94.4% within 30 min, respectively. But only 88.7 and 72.1% of AO7 were decomposed at the initial concentrations of 60 and 75 mg/L in the same reaction time, respectively. In general, with the increase in AO7 concentration, the degradation efficiency decreased. As the AO7 concentration increased, more sulfate radicals were required. However, considering that the same concentration of catalysts and oxidants only produced the same amount of sulfate radicals, the degradation efficiency inevitably decreased. In addition, massive AO7 molecules may reduce the contact probability between NiO and PMS, which is not conducive to the formation of sulfate radicals.

The influence of the initial pH on the AO7 degradation efficiency with PMS by NiO-EG is shown in Figure c. The removal efficiency of AO7 at 30 min could reach over 90% in a wide pH range from 5 to 8. With the increase in solution initial pH, the removal efficiency increased from 82.9 to 98.8%. In general, pH plays a significant role in the activation of PMS and surface charge of catalysts. In acidic medium, NiO-EG exhibited increasing AO7 removal from 82.9 to 95.5% as the pH rose from 4 to 7. At pH = 4, only 82.9% of AO7 could be removed within 30 min, which might be ascribed to the fact that excessive H+ would scavenge the SO4•– and HO• radicals (eqs and 2).[45] The advantages of the NiO/PMS system in degrading AO7 were obviously presented in both acid and alkaline medium (pH 5–8).

Recycle Tests

Recyclability is an important factor that evaluates the performance of catalysts in practical applications.[46] Duan et al. synthesized cobalt-based perovskites for activating PMS to degrade phenol with a removal efficiency of 100% in 30 min, which decreased to 82% after 3 cycles.[47] Zhang et al. used granular activated carbon to catalyze PMS to degrade AO7 with a degradation efficiency of 85% within 300 min, and 78.5% of AO7 removal was achieved in the fourth run.[40]

Four cycling runs were performed under the same experimental conditions to evaluate the reusability of the prepared NiO-EG catalyst. The related results are shown in Figure . It is apparent that the degradation removal of AO7 decreases only slightly from 95.4 to 94.8% at 30 min after four cycles, which shows the excellent stability of NiO-EG during the degradation. These results indicate that NiO-EG processes excellent stability, which may provide a fundamental basis for its practical applications.

Figure 8

Consecutive runs of the catalytic activities of NiO-EG. Reaction conditions: [AO7] = 50 mg/L, catalyst = 0.2 g/L, PMS = 0.8 g/L, and pH = 7.

Exploration of the Catalytic Mechanism

Quenching experiments were conducted to explore the efficient radicals in the NiO/PMS system. As reported, activation of PMS via heterogeneous transition-metal oxides usually generates several critical active species, namely, sulfate (SO4•–), hydroxyl (HO•), superoxide radical (O2•–), and peroxymonosulfate (SO5•–). Among them, SO5•– with a redox potential of 1.1 V at pH 7, which is much lower than that of SO4•– and HO•, makes negligible contribution to the degradation of AO7. Therefore, only superoxide, sulfate, and hydroxyl radicals are considered. Ethanol is an efficient scavenger of both SO4•– and HO• (kHO = 1.2–2.8 × 109 M–1 s–1; kSO = 1.6–7.7 × 107 M–1 s–1), while the reaction rate between tert-butyl alcohol (TBA) and HO• (3.8–7.6 × 108 M–1 s–1) is about 1000-fold faster than that with SO4•– (4.0–9.1 × 105 M–1 s–1).[48] Thus, ethanol was chosen as a scavenger for both SO4•– and HO•, whereas TBA was employed as a scavenger for HO•. In addition, benzoquinone was used as a quencher to examine the role of O2•–.[49][49]

As shown in Figure , an observed inhibition on the removal efficiency of AO7 occurred within 30 min with the addition of the scavenger. The degradation efficiency was decreased by 6.8% and 15.4% when 5 and 10 mM ethanol were added, respectively. In contrast, a much weaker inhibition effect appeared even when the same amount of TBA (10 mM) was added, indicating that both SO4•– and HO• were produced in the NiO/PMS system and participated in the degradation of AO7. However, this quenching efficiency is quite far from that in other reports.[50,51] Fan et al. quenched SO4•– and HO• with EtOH in an FeOOH/PMS system, obtaining a decrease from 91.7 to 35.9% in AO7 removal efficiency.[52] It has been reported that as hydrophilic compounds, EtOH and TBA had low affinity for the catalyst surface and could only consume SO4•– and HO• in the liquid phase.[40] Thus, if the generated radicals are bound or caged on the surface of the catalyst during the catalytic oxidation process, the quenching effect of EtOH and TBA would be inconspicuous. Meanwhile, other radicals, such as O2•–, generated in the PMS system are also reported. With the addition of benzoquinone, 77.6% of AO7 was removed in 30 min, revealing that O2•– was also involved in the degradation process. Therefore, three radicals, SO4•–, HO•, and O2•–, were generated in the NiO/PMS system. A similar mechanism is also reported in the heterogeneous activation of a CuFe2O4/PMS system.[53]

Figure 9

Inhibition of TBA and methanol on AO7 degradation by the NiO/PMS system. Reaction conditions: [AO7] = 50 mg/L, catalyst = 0.2 g/L, PMS = 0.8 g/L, and pH = 7.

To further explore the catalytic mechanism, X-ray photoelectron spectroscopy (XPS) was employed to analyze the chemical valence change of nickel and oxygen atoms on the surface of the fresh and used NiO catalysts. From Figure S3a, the full-range XPS spectrum of NiO-EG indicates that Ni and O are the main elements in the samples with a peak of C 1s (284.8 eV), which is used as an internal reference. The high-resolution scans of the fresh and used NiO-EG catalyst for Ni 2p3/2 are shown in Figure a. Before the reaction, the peaks at 853.4, 854.9, and 860.2 eV are assigned to the Ni2+ species in NiO, whereas the peaks at 855.9 and 860.9 eV are assigned to the Ni3+ species on the NiO surface that formed due to defects. These values suggest the coexistence of Ni2+ and Ni3+.[54] After the reaction, the peaks of Ni 2p3/2 for Ni2+ and Ni3+ are slightly shifted to 853.6, 855.1, 860.3, 856.2, and 861.2 eV, respectively, indicating that during the reaction, part of Ni2+ were transformed into Ni3+. The extent of the variation of the Ni3+ ions is surveyed by calculating the peak area ratio of Ni3+ to Ni2+. Results show that the percentage of Ni2+ and Ni3+ changed from 70.3 and 29.7% to 68.4 and 31.6% after use, respectively, which is much lower than that reported by other works. It is reported by Wang et al. that Ni2+ in a NiFe2O4 catalyst decreased from 64.6 to 46.3% after degradation of benzoic acid.[19] Zhao et al. synthesized Co–Mn LDH and found that after degradation of dyes, Mn3+ decreased from 71.1 to 55.2%.[50] The increase in Ni3+ percentage in the used sample confirms that part of Ni2+ were transformed into Ni3+ during the degradation reaction. The XPS spectra demonstrate the coexistence of Ni2+ and Ni3+ and the increase of Ni3+ in the used NiO-EG catalyst, which suggest that the Ni2+/Ni3+ redox cycle may be the reaction pathway for the activation of PMS and this redox cycle also makes a contribution to the stability of NiO-EG. The XPS O 1s spectrum and its deconvolution for fresh and used NiO-EG are shown in Figure S4a. Two clearly separated peaks at 529.0 and 530.6 eV can be detected, which are attributed to lattice oxygen and O2– ions in the oxygen-deficient regions, respectively.[55] A shift occurs after the reaction, and the peaks are slightly shifted to 529.2 and 530.8 eV. The peaks at 532.0 and 532.2 eV correspond to the hydroxyl-related species or the loosely bound oxygen on the surface.[56] Compared with NiO-EG, the XPS spectra of NiO-W show similar results (Figures b, Figure S3b and S4b). But the NiO-W catalyst contains fewer Ni3+ (17.9%) on the surface, corresponding to fewer oxide vacancies, which we consider as one of the reasons for the lower degradation efficiency of NiO-W than that of NiO-EG. Further research was performed as given below.

Figure 10

Ni 2p XPS spectra of the fresh and used (a) NiO-EG and (b) NiO-W.

Oxygen vacancies are important factors influencing the catalytic reactions, such as catalytic oxidation.[57] On this basis, O2-TPD measurements were conducted to identify the oxide species on the surface of the NiO catalysts (Figure ). The desorption peaks below 150 °C usually belong to the physically adsorbed oxygen species, whereas the desorption peaks between 300 and 700 °C are attributed to the O2– or O– species, formed by the oxygen adsorbed on the surface oxide vacancies.[58] The peaks above 700 °C are associated with the surface lattice oxygen species (O2–). The NiO-W physically adsorbed more oxygen, according to its higher specific surface area, than NiO-EG. However, the amount of adsorbed O2– or O– species for NiO-EG was much higher compared with that for NiO-W, suggesting that there were more oxygen vacancies on the surface of NiO-EG. Oxide vacancies might function as active sites for dissociative adsorption of water on metal oxide surfaces at which hydroxyl groups are formed.[59] In the neutral state, these surface hydroxyl groups would be a major part responsible for the generation of radicals.[60] In addition, oxygen vacancies promote electronic transfer and participate in the redox cycle of Ni2+/Ni3+, which is feasible for the activation of PMS.[61]

Figure 11

O2-TPD spectra of (a) NiO-EG and (b) NiO-W.

Oxygen vacancies play an important role in the high reactivity of NiO-EG. The generation of oxygen vacancies should be attributed to the unique property of ethylene glycol and the surface characteristics of the (111) plane. It was reported that ethylene glycol could easily react with oxygen-terminated oxide surfaces, during which the oxidation of the alcohol groups would produce aldehydes and acids. Then, some surface oxygen atoms are removed, resulting in the formation of oxygen vacancies.[62] Thus, we propose that the highest degradation efficiency shown by NiO-EG is mainly attributed to the combination effect of its flaky porous structure and abundant oxygen vacancies on its surface, which enhance charge transfer and provide easy access of high surface areas and highly reactive surface sites. The preparation methods are simple and versatile, and this work may shed light on the metal oxide catalyst design and synthesis for water treatment.

Based on the above results, we proposed the mechanism of PMS catalytic activation and dye degradation by the NiO/PMS system as followsFirst, SO4•– was generated by the reaction between Ni2+ on the surface of the catalyst and PMS (eq ). HO• was also produced by the reaction of SO4•– with water molecules or hydroxyls in the solution subsequently (eqs and 5). Ni3+ then oxidized PMS to SO5•– that combined with each other to generate SO4•– and O2 and transformed into the original Ni2+ (eqs and 7).[19] Herein, the Ni3+/Ni2+ redox couples formed, enhancing the recyclability of the NiO catalysts, which was also evidenced by XPS. By means of one-electron transfer, O2•– was formed through the oxygen absorbed on the oxygen vacancies (eq ), which could also react with water molecules to produce HO• (eq ).[63] The generated radical species, SO4•–, O2•–, and HO• attacked the dye molecules and degraded the target pollutants into small molecular inorganic compounds (eq ). And the reversible oxidation and reduction between Ni3+/Ni2+ maintained the structure of the catalyst and high activation activity for PMS.

Conclusions

Several mesoporous NiO catalysts with thin flaky, flowerlike, and thick spongelike plate structures were prepared by a simple solvothermal method, and their shape-control activity was evaluated in catalytic activation of PMS for AO7 degradation. The NiO prepared using ethylene glycol as the solvent with a high surface area and thin flaky structure exhibited the highest degradation ability due to its abundant accessible active sites for both adsorption and catalysis. Rich oxygen vacancies, active oxygen species, fast charge transport, and low diffusion impedance all enhanced the catalytic activity of NiO-EG. In addition, the degradation efficiency improved with the increase in NiO-EG dosage but was inhibited with the increase in the initial concentration of AO7. The pH tests showed that the NiO/PMS system had a wide application range of pH (5–8). NiO-EG also exhibited high catalytic stability with little deactivation after four runs. Radical quenching studies suggested that SO4•–, HO•, and O2•– were generated during the PMS activation. The high activity and stable performance of NiO-EG makes it a promising catalyst for activating PMS to oxidize organic pollutants.

Experimental Section

Materials

Nickel nitrate (Ni(NO3)2·6H2O), urea, ethylene glycol, acid orange 7 (AO7), tert-butyl alcohol (TBA), ethanol, sodium nitrite (NaNO2), and peroxymonosulfate (PMS), available as a triple salt of sulfate commercially known as oxone (2KHSO5·KHSO4·K2SO4), were obtained from Sinopharm Chemical Reagent Co. Ltd. (China). Para-benzoquinone was purchased from Aladdin (Shanghai, China). All chemicals were of analytical grade and used without further purification. The experimental solutions were prepared with deionized water. The solution pH was adjusted using 1 M NaOH.

Synthesis of NiO Catalysts

Three NiO catalysts were synthesized via solvothermal methods with different solvents. Typically, 5 mmol nickel nitrate and urea were dissolved in 25 mL of solvent and 5 mL of deionized water to form a green solution. The sample using ethylene glycol as the solvent was referred to as NiO-EG, that using ethanol as the solvent was referred to as NiO-E, and that using deionized water as the solvent was referred to as NiO-W. The mixed solution was transferred into a Teflon-lined autoclave. The autoclave was sealed and maintained at 180 °C for 10 h and was then cooled to room temperature naturally. The solvothermal product was filtered, washed with deionized water and ethanol, and dried in air at 50 °C overnight. Finally, all three samples were calcined at 350 °C under air for 4 h to obtain the catalysts. To investigate the formation of the NiO-EG catalyst, two samples were synthesized in the same way as NiO-EG, except that the reaction time in the solvothermal process was shortened to 2 and 6 h, respectively. These samples were referred to as NiO-EG-2 and NiO-EG-6, respectively.

Characterization of Catalysts

X-ray diffraction (XRD) patterns were obtained on a Bruker D8 Discover (Bruker-AXS, Karlsruhe, Germany) diffractometer using a filtered Cu Kα radiation source (λ = 1.54178 Å), scanned at 2θ from 10 to 80°. The specific surface area of catalysts was investigated by N2 adsorption at 77 K, using the multipoint Brunauer–Emmett–Teller (BET) method (Autosorb iQ, Quantachrome). Prior to the measurements, the samples were dehydrated in vacuum at 175 °C for 6 h. The pore size distributions were calculated from desorption branches of the isotherms by the Barrett–Joyner–Halenda (BJH) method. The morphology of the catalyst was observed on a TF20 JEOL 2100F transmission electron microscope (TEM) and a FEI Magellan 400 field-emission scanning electron microscope. O2 temperature-programmed desorption (O2-TPD) was conducted on a gas flow system (AutoChem II 2920, Micromeritics). Typically, the catalyst sample (40 mg) was placed in a U-shaped quartz reactor and pretreated in flowing O2 (15 mL/min) for 2 h at 350 °C, followed by cooling at room temperature. Then the temperature was raised from room temperature to 900 °C at a rate of 15 °C/min in a He flow (30 mL/min). The X-ray photoelectron spectroscopy (XPS) data were taken on a Thermo Scientific instrument, ESCALAB 250XI using X-ray source Al Kα radiation (hv = 1486.6 eV).

Electrochemical Measurements

The electrochemical measurements were performed using a Solartron Analytical 1470E CellTest System in a standard three-electrode cell. An Ag/AgCl (3 M KCl) electrode and a platinum wire were used as the reference electrode and counter electrode, respectively. A glassy carbon electrode modified with a homogeneous catalyst ink, which was prepared via sonication of 10 mg of catalyst powder, 40 μL of Nafion solution (5 wt %, Sigma-Aldrich) and 1 mL of absolute ethanol, was used as the working electrode, and the electrolyte was 0.1 M Na2SO4. Electrochemical impedance spectra (EIS) were recorded within a frequency range from 105 to 10–1 Hz.

Catalytic Activity Tests

Catalytic activity tests were carried out in a 150 mL conical flask containing 100 mL of aqueous solution with a certain concentration of AO7. First, the catalyst was added into the solution. The mixture was stirred for 30 min to reach the balance between the catalyst and AO7. Subsequently, a fixed amount of PMS was added to start the oxidation reaction. Then, 1 M NaOH solution was also added into the solution to adjust the pH to 7. Samples were obtained at regular intervals, withdrawn from the mixture using a syringe filter of 0.22 μm and mixed with NaNO2 immediately to quench the reaction. The concentration of AO7 was analyzed by a 725N UV–vis spectrophotometer at 484 nm. For comparison, the catalytic activity tests were conducted without adding the catalyst or PMS under the same conditions. During the recycling experiment, the catalyst was collected by vacuum filtration and thoroughly washed with distilled water and ethanol after each recycle. Then the catalyst was dried in a vacuum oven at 50 °C for 24 h to remove water and ethanol.