Promoted VOC oxidation over homogeneous porous Co_xNiAlO composite oxides derived from hydrotalcites: effect of preparation method and doping

Abstract

Homogeneous porous and curve plated Co_xNiAlO composite metal oxide catalysts are obtained from the thermal decomposition of Co_xNiAl-layered double hydroxide (LDH) precursors, which are prepared by urea co-precipitation with surfactant, followed by a hydrothermal treatment. The as-prepared samples were characterized by XRD, BET, SEM, TEM, H₂-TPR and XPS. The Co₃AlO sample shows 90% benzene conversion (T₉₀) at 236 °C at a high space velocity (SV = 60 000 mL g⁻¹ h⁻¹), and possesses much higher activity than Co₃AlO prepared with NaOH co-precipitation without surfactant, with T₉₀ = 288 °C. This is mainly correlated with the narrower pore size (2.9 vs. 17.2 nm) and lower temperature reducibility (319 vs. 360 °C). The Co₂NiAlO sample exhibits enhanced activity at T₉₀ = 227 °C with the low activation energy of 39.0 kJ mol⁻¹, and its lower temperature reducibility is ascribed to the larger amount of surface accessible Co³⁺. The Co₂NiAlO sample owns good reproducibility and superior reversibility and long stability with prolonged time on benzene stream in the presence of 3.5% water vapor. Moreover, a monolithic Co₂NiAlO film catalyst is fabricated by the thermal decomposition of an LDH film precursor through an in situ growth methodology, with a high reaction rate of 1.21 mmol g⁻¹ h⁻¹ under T₉₀ = 275 °C.

---

1. Introduction

The release of anthropogenic toxic pollutants such as volatile organic compounds (VOCs) into the atmosphere is a worldwide threat of growing concern. VOCs may be found in non-industrial indoor air environments, and can act as irritants to the human organism and have negative health effects, and cause bad indoor air quality.¹ Catalytic oxidation^2,3 with high efficiency is recognized as an effective method for air purification compared with adsorption, UV photocatalytic oxidation, non-thermal plasma, plasma-assisted catalysis and catalytic incineration, which have been observed with potentially harmful intermediates and/or by-products.^4,5 A rational control of the size and reactivity of the pore walls of metal–organic frameworks to maximize the performance of capture and catalytic degradation of harmful gases has been summarized.⁶ The design of high-performance catalysts in terms of activity and resistance to deactivation is of great importance. Noble metal catalysts over Pd/Co₃AlO,⁷ Pt/mesoporous silica⁸ and Ag–Au alloy⁹ exhibit high activity for the reduction of VOCs, however, they are reported for their high cost, their ease of sintering and coking, and are susceptible to poisoning.¹⁰ Non precise metal oxides as a potential alternative have attracted much attention for their low cost, high resistance to poisoning, and good reducibility.¹¹ With smart design, transition metallic composite oxides own equivalent or even better properties than precious metals because of the synergistic effect.^12–14 For example, a Co-nanocasting synthesis of mesoporous Cu–Mn composite oxides using a siliceous template SBA-15 exhibited promoted catalytic activity towards benzene removal.¹³ Exerting the synergistic effect to the maximum degree for mixed metal oxides with high catalytic activity is still filled with challenges and opportunities.

Besides, the design of catalysts with small particles or pore sizes and hierarchical morphologies fabricated with thinner two-dimensional sheets possessing more active sites and better dispersion is highly beneficial for outstanding catalytic activity, but is still a challenge.¹⁵ Zhao reported ultrathin ZnAl-LDH nanosheets with coordinatively unsaturated Zn ions with lateral sizes less than 30 nm that were synthesized using a facile bottom-up strategy, and exhibited extraordinarily high activities for the photo-reduction of CO₂ to CO.¹⁶ Li fabricated flower structured CoZnAl–MMO/Al₂O₃ catalysts, which exhibited a high dispersion of cobalt species due to the well-developed three-dimensional flower-like CoZnAl–MMO platelets, which showed much higher catalytic activity and selectivity for the oxidation of ethylbenzene compared with a conventional supported Co-based catalyst prepared by incipient impregnation.¹⁷ Recently, many efforts proved that the calcination of layered double hydroxide (LDH) precursors may obtain well dispersed and stable catalysts with the above advantages.^18,19

Layered double hydroxides (LDHs) are a class of anion clay materials generally expressed by the formula [M²⁺_1−xM³⁺_x(OH)₂](Aⁿ⁻)_x/n·mH₂O, in which M^II and M^III cations disperse in an ordered and uniform manner in brucite-like layers, and Aⁿ⁻ is a charge compensating anion.^20,21 The specific ordered cation arrangements in the hydroxide layers of LDHs, revealed by multinuclear NMR spectroscopy, could enhance the functional optimization of these materials in catalysis oxidation, photocatalysis and environmental remediation.^20,22–24 Therefore, the resulting mixed oxide catalysts from the topotactic transformation of LDHs give finely dispersed mixed oxides of M^II and M^III metals, with the promotion of a synergistic effect.²⁵ Fan reviewed recent advances in the applications of layered double hydroxides (LDHs) with smart design in heterogeneous catalysis, both as directly prepared or after thermal treatment and/or reduction.²⁶ Besides, LDH particles can now be synthesized smaller and thinner with enhanced surface area ranging from several micrometers to several tens of nanometers by choosing a proper preparation method, which will be notably favorable towards the exposure of active sites.²⁷ Moreover, LDH lamellate plates growing into a film through in situ crystallization can assemble to yield flower, core–shell or comb-like structures, leading to a large surface area, and the separate plates inhibit the agglomeration of active components, giving high activity and long-term stability.^25,28,29 However, to our knowledge, there is limited information on the possible cooperative effects that can exist for Co_xNiAlO catalysts synthesized by different methods or with varying dopant amounts.

Herein, we fabricated Co_xNiAlO powdered and film composite metal oxide catalysts through the calcination of Co_xNiAlLDH precursors by a urea co-precipitation method, followed with hydrothermal treatment. The effect of the preparation method on the total catalytic degradation of benzene was evaluated through the Co₃AlO samples obtained with urea and NaOH co-precipitation methods. The effect of Co (or Ni) doping for the Co_xNiAlO samples on their textural properties and catalytic activities was further investigated.

2. Experimental section

2.1 CoNiAlLDH powder and film precursor preparation

All chemicals used in this study were commercially available (from Merck) and used without further purification. A pure aluminum substrate (purity > 99.99%; thickness 0.2 mm) was purchased from Beijing General Research Institute. Co_xNiAlLDH powder precursors with varying Co contents were prepared by a hydrothermal method (x = the molar ratio percentage of Co to Al). Briefly, a 75 mL aqueous solution containing Co(NO₃)₂·6H₂O (15, 10, 7.5, 5, 0 mmol), Ni(NO₃)₂·6H₂O (molar number of Ni = 15 mmol – molar number of Co), Al(NO₃)₃·9H₂O (5 mmol), urea (75 mmol) and sodium dodecyl sulfonate (SDS, 2.2 mmol) was delivered into a 100 mL Teflon autoclave after being stirred thoroughly at 120 °C for 6 h. A Co₃AlLDH precursor was prepared through a co-precipitation method for comparison.³⁰ Briefly, a 20 mL mixed salt solution with 15 mmol Co²⁺ and 5 mmol Al³⁺, and a 30 mL 45 mmol NaOH solution were added dropwise simultaneously to a 180 mL 10 mmol Na₂CO₃ solution, maintaining a constant pH of 9.5 at 65 °C, under constant stirring for 18 h. The resulting solution was denoted as Co₃AlLDH-Ref. The resulting suspension was centrifuged with ethanol and deionized water several times until the surfactant was thoroughly removed, and then dried at 80 °C for overnight.

The CoNiAlLDH film precursor was prepared by in situ crystallization on an aluminum substrate, which was cleaned with acetone, ethanol and deionized water in sequence before use. The Al substrate (4 × 15 cm) was rolled into a tubular shape and immersed vertically in the same solution as for the preparation of Co₂NiAlLDH (Co²⁺ 10 mmol), with the exception of NH₃·H₂O, which was replaced by urea to adjust the pH to 9, then the Co₂NiAlLDH film (Co₂NiAlLDH-F) was grown along the ab-orientation vertically to the Al substrate. Afterwards, the substrate was withdrawn, rinsed with deionized water and dried at room temperature.

2.2 CoNiAlO powder and film catalysts preparation

The metal oxide powder and film catalysts were obtained by calcinating Co_xNiAlLDH powder and film precursors in air at 400 °C for 4 h with a heating rate of 2 °C min⁻¹, with the resulting materials denoted as Co_xNiAlO, and Co₂NiAlO-F. The detailed textural properties of Co₃NiAlO-Ref are shown in our previous report.³⁰

2.3 Characterization of the catalysts

X-ray diffraction (XRD) patterns of the catalysts were measured on a Rikagu SmartLab system using Cu-Kα radiation in the diffraction angle (2θ) range of 8° to 90°. The specific surface areas and pore size distributions of all catalysts were obtained according to the Brunauer–Emmett–Teller (BET) and Barrett–Joyner–Halenda (BJH) methods, respectively, using the N₂ adsorption–desorption method on an automatic surface analyser (AS-1-C TCD, Quantachrome Cor., USA) at 77 K. Before measurement, each sample was degassed at 200 °C for 3 h. The morphology of the samples was recorded on a scanning electron microscope (SEM, JEOL JSM-6700F, Japan, 15 kV, 10 mA). The microstructures of samples were obtained using transmission electron microscopy (TEM, JEOL JEM-2010F) with an accelerating voltage of 200 kV. Surface species of the as-prepared catalysts were determined by X-ray photoelectron spectroscopy (XPS) and X-ray induced Auger electron spectroscopy (XAES) using a XLESCALAB 250Xi electron spectrometer from VG Scientific with monochromatic Al Kα radiation (1486.6 eV). Hydrogen temperature programmed reduction (H₂-TPR) with 30 mg catalyst (40–60 mesh) was carried out in a U-shaped quartz reactor under gas flow (5% H₂ balanced with Ar, 25 mL min⁻¹). In each procedure, the temperature was raised to 800 °C from room temperature at a constant rate of 10 °C min⁻¹.

2.4 Catalytic activity tests

The performance of the catalysts was evaluated in a continuous-fixed-bed quartz microreactor (i.d. 6 mm) at a space velocity (SV) of 60 000 mL g⁻¹ h⁻¹. The temperature of the reactor was controlled using a temperature controller. Catalysts (100 mg, 40–60 mesh), mixed with 400 mg quartz sand (40–60 mesh) were loaded in the quartz reactor with quartz wool packed at both ends of the catalyst bed. 100 ppm benzene balanced with air was purged into the reactor at a continuous flow rate of 100 mL min⁻¹ using mass flow controllers. For consideration of effect of water vapour, mixed gas, composed of a relative 3.5% water vapour and 50 ppm benzene generated by mixing an 50 mL min⁻¹ air flow used for bubbling water and 50 mL min⁻¹ of air flow with benzene, was purged into the reactor for catalytic tests. The concentration of benzene in the effluent gas was analysed by a gas chromatograph (Shimadzu GC-2014) equipped with a flame ionization detector (FID) and the concentration of CO₂ in the outlet gas was detected by another FID with a methanizer furnace for converting CO₂ to CH₄. To evaluate the benzene catalytic degradation, the gas produced from the reaction was extracted by dissolving it in hexane solvent at a different reaction temperature, followed by analysis using GC-MS (Shimadzu QP2010 Ultra, equipped with an Rxi-5Sil column) with the scanning mode of the mass spectrometer under EI temperature at 200 °C. Catalytic evaluation of the Co₂NiAlO-F catalyst was performed by replacing the Co_xNiAlO powder samples with a film catalyst which was pre-tailored to very small fragments. The complete conversion of benzene (W_benzene) and the yield of CO₂ (η_CO2) were calculated as follows:

where C_benzene,in (ppm), C_benzene,out (ppm) and C_CO2,out (ppm) are the concentrations of benzene in the inlet and outlet gas, and CO₂ in the outlet gas, respectively.

3. Results and discussion

3.1 Textural study of the precursors and catalysts

The XRD patterns of the Co_xNiAlLDH precursors are shown in Fig. 1. In each case, the XRD pattern exhibits the characteristic reflections of LDH materials with a series of (00l) peaks appearing as narrow symmetric lines at a low angle with 3R packing of layers. With the increased doping of Ni, the characteristic diffraction peaks (00l) became broader than those of pure Co₃AlLDH, from which it could be deduced that the size of LDH nanoparticles became smaller with the effect of Ni incorporation. After calcination in air at 400 °C, the XRD diffraction patterns of Co_xNiAlO catalysts are given in Fig. 1. The strongest reflection at 2θ = 36.8° (311) for the Co₃AlO sample, together with peaks around 19.1°, 31.4°, 45.1°, 59.5°, and 65.5°, correspond to a Co₃O₄ spinel phase in Fig. 1 (PDF card no. 34-0425), which are the similar to the peaks also reported in other spinel oxide phases (Co₂AlO₄, PDF card no. 38-0814; CoAl₂O₄, PDF card no. 44-0160). They are hard to distinguish by X-ray diffraction. Diffraction reflections at 2θ = 37.5°, 43.6°, 63.2°, 75.8° and 79.5°, corresponding to the 111, 200, 220, 311 and 222 planes, respectively, indicate the formation of an NiO phase in Fig. 1 for the Ni₃AlO sample (PDF card no. 47-1049). As the Ni content is increased, the Co₃O₄ peaks decrease in intensity, become broader and are shifted to lower angles. The absence of NiO diffraction peaks reveals the formation of a solid solution for the Co_xNiAlO samples, which will inhibit crystal growth and is expected to bring about a synergistic effect to alter their physicochemical properties. No peak attributed to Al₂O₃ can be found from the XRD patterns in all samples, revealing that the amorphous Al₂O₃ matrix in mixed oxides can play a role of support to stabilize and disperse Co₃O₄ spinel phases,^31,32 which are also favorable towards high catalytic activity.

Fig. 1 XRD patterns of Co_xNiAlLDH and Co_xNiAlO.

The morphologies of Co_xNiAlLDH and their composite metal oxides revealed by SEM are shown in Fig. 2. They all show a uniform and randomly oriented hexagonal lamellar structure. The particle size of Co₃AlLDH-Ref is ca. 300 nm.³⁰ The Co₃AlLDH particle size is ca. 200 nm in the ab plane with 20 nm thickness in the c-orientation, while they decreased to 120–180 nm with 10–14 nm thickness for Co₂NiAl, Co_1.5NiAl and Co₁NiAlLDH, which is coincident with the gradually broader XRD peak at 2θ = 11.6° with enhanced Ni content. The result also indicated the formation of a solid solution, which was indeed inferred from the XRD data. As commonly reported, this can decrease the size of crystals. The smaller sized particles will generate more crystal defects which will be active in catalytic reactions.³³ Besides, for all the Ni-rich samples, the hexagonal lamellar structures tend to curl and gradually aggregate together to irregular flower petal. With the increased doping of Ni, the extent of curl markedly enhanced. Ni₃AlLDH shows a petal morphology with even thinner platelets around 6–8 nm. It can be deduced that many stack holes exist through the self assembly of LDH platelets to form an analogous petal structure, which may be leading the augmentation of surface area and pore volume. The Co_xNiAlO samples exhibit similar morphologies to their corresponding LDH samples, which shows that the metal oxide catalysts derived from the calcination of LDH precursors can maintain the LDH structure, keeping the higher dispersion of Co₃O₄ phases. The tendency of the Co/Ni atomic ratios in the catalysts, detected from SEM-EDX and summarized in Table 1, was reasonably coincident with the initial input ratios with lower deviation. The Co/Ni ratios of the catalysts surveyed by XPS in Table 1 exhibit lower ratios than those from EDX. The results show that the Co/Ni ratios of the catalysts from the bulk are higher than at their surface. The direct element dispersion information, together with the various element peaks and their related proportions for the Co₂NiAlO sample are shown in Fig. 3. There is very little Na and S detected from the EDX data for Co₂NiAlO, which manifested that the surfactant stuck on the surface of LDHs was almost removed through washing with ethanol, following calcination. The EDX mapping demonstrates a uniform distribution of various elements (O, Al, Ni or Co) in the particle region. The amorphous Al₂O₃ makes the active elements maintain their good dispersion. The well dispersed Co and Ni may promote the catalytic activity.

Fig. 2 SEM images of Co_xNiAlLDH and Co_xNiAlO; all the scale bars are 200 nm.

Table 1 Atomic ratio of Co : Ni for Co_xNiAlO catalysts and their BET data

catalysts	Co/Ni atomic ratio			BET surface areas (m^2 g^−1)	Pore volume (cm^3 g^−1)	Pore size (nm)
	Initial	EDX	XPS
Co3AlO-Ref				153.5	0.34	17.2
Co3AlO				148.6	0.34	3.0
Co2NiAlO	2	1.47	1.28	172.7	0.58	3.1
Co1.5NiAlO	1	1.02	0.66	152.4	0.60	3.6
Co1NiAlO	0.5	0.65	0.39	178.9	0.78	3.8
Ni3AlO				193.2	0.87	4.2
Co3AlO-Ref-after test				127.9	0.31	23.6
Co2NiAlO-after test				119.3	0.21	5.2

---

Fig. 3 SEM image, element distribution and EDX data for the Co₁NiAlO sample.

The TEM images of the Co_xNiAlO catalysts shown in Fig. 4 show that the hexagonal plate-like morphology of the original LDH precursor was maintained. Additionally, the Co₃AlO, Co₂NiAlO and Ni₃AlO samples in Fig. 4a, c and e all present well dispersed uniform pores embedded in the plate matrix. The pore sizes are around 2.2 nm for Co₃AlO, 2.6 nm for the Ni₃AlO samples, which facilitates catalysis by augmenting contact between the reactant and the catalysts. The component of the Co_xNiAlO catalysts is revealed by HRTEM (Fig. 4b, d and f). The identified fringes, with a lattice spacing of 0.30 and 0.21 nm, are indexed to the (220) plane of the Co₃O₄ spinel phase for Co₃AlO and the (200) plane of the NiO phase for Ni₃AlO with good crystallization, as shown in Fig. 4b and f, respectively. The lattice fringes for the Co₃O₄ spinel phase are clearly found, but are obscured by the NiO phase for Co₂NiAlO in Fig. 4d. Elemental mapping of the single plate in the HRTEM image for the Co₂NiAlO sample is shown in Fig. 4g, showing the good dispersion of Co and Ni metals. The SAED pattern shown in Fig. 4h illustrates the single crystalline Co₂NiAlO nanostructure with hexagonally arranged spots.³⁴

Fig. 4 (a–f) TEM and HRTEM images of the catalysts, (g and h) elemental mapping of the single plate and SAED pattern of Co₂NiAlO.

Fig. 5 displays the N₂ adsorption–desorption isotherm and the corresponding pore size distribution curve for the Co_xNiAlO catalysts, and their BET and pore volume results are collated in Table 1. All the samples exhibit a typical IV isotherm with an H₃-type hysteresis loop (P/P₀ > 0.4). It is noticeable that the two samples Co₃AlO-Ref and Co₃AlO exhibit similar surface areas and pore volumes; however, their pore size distribution is quite different, being 17.2 nm for Co₃AlO-Ref and 2.9 nm for the Co₃AlO samples. These results indicate that the preparation method with the incorporation of surfactant mainly leads to changes in the pore size and can make a great contribution to the catalytic activity for benzene oxidation. The surface properties upon doping of Ni into the CoAlO samples were also examined, and show a gradually increased specific surface area from 148.5 to 193.1 m² g⁻¹ with the exception of the Co_1.5NiAlO sample, which displayed increased pore volume and pore size from 0.34 to 0.87 cm³ g⁻¹, and 2.9 to 4.1 nm respectively, as a result of the enhanced Ni content (Table 1), in accordance with the different morphology upon Ni doping, which will be beneficial for the disparity of the catalytic activity.

Fig. 5 N₂ adsorption–desorption isotherm curves and pore size distributions calculated from the desorption branch of the as-synthesized Co_xNiAl–MMO catalysts.

3.2 X-ray photoelectron spectra

To obtain further insight into the structure–function correlations, X-ray photoelectron spectra (XPS) for Co 2p, Ni 2p and O 1s were obtained for the fresh Co_xNiAlO samples and are displayed in Fig. 6, with the reference being the residual carbon at a binding energy (BE) of 284.6 eV. The Co 2p XPS spectra was fitted assuming a theoretical ratio of 2 between the Co 2p_3/2 and 2p_1/2 states, with a spin–orbit splitting of 15.2 eV. Two broad and asymmetric peaks located around 780 and 796 eV (Fig. 6A) correspond to Co 2p_3/2 and Co 2p_1/2 from the spin–orbit doublet of Co₃O₄, respectively. The predominant peak at 780.1 eV and the small peak at 781.4 eV are attributed to Co³⁺ in octahedral sites and the Co²⁺ 2p_3/2 configuration in tetrahedral sites, respectively. The Co³⁺ and Co²⁺ 2p_1/2 components appear at 795.4 and 796.6 eV.³⁵ The Co 2p_3/2 shifted from 780.5 to a higher energy of 779.8 eV after Ni incorporation, indicating the existence of an interaction between Co and Ni. This phenomenon was also observed for Ce doping in Co₃O₄, with a high energy at 774.4 eV observed.³⁶ The surface Co³⁺/Co²⁺ ratio calculated from the areas by quantitatively analyzing the XPS spectra of the samples tended to decrease with decreasing cobalt concentration. The Co₃AlO and Co₂NiAlO samples presents higher ratios of 1.50 and 1.33, which deviate from the value of 2 corresponding to the formula Co²⁺(Co³⁺)₂O₄, while the Co₃AlO-Ref sample presents a ratio of 2.13, which is close to Co²⁺(Co³⁺)₂O₄. The structural defects of the spinel Co₃O₄ could play a key role in creating oxygen vacancies on the surface, corresponding to the higher intrinsic catalytic activity. The Ni 2p spectra comprise two regions representing the Ni 2p_3/2 (850–866 eV) and Ni 2p_1/2 (870–885 eV) spin–orbit levels. The Ni 2p_3/2 region consists of a main peak at 854.9 eV with a shoulder peak (∼1 eV above the main peak) and a satellite peak at 861.5 eV. The Ni 2p_1/2 appears at 873.1 eV with a satellite peak (∼7 eV above the main peak), which is consistent with a previous report on NiO.³⁷ Similar features are observed for the Ni 2p region for Co-containing samples shown in Fig. 6B.

Fig. 6 XPS spectra of (A) Co 2p, (B) Ni 2p and (C) O 1s of the Co_xNiAlO samples.

The asymmetric O 1s peaks ranging between 528–534 eV for Co_xNiAlO samples can be deconvoluted into two contributions with the peaks at 529.8 and 531.6 eV attributed to the lattice oxygen (O_latt) and surface adsorbed oxygen (O_ads) respectively, as shown in Fig. 6C and summarized in Table 2. These have a great impact on catalytic performance. The main peak of O 1s shifted from 530.4 eV for the Co₃AlO sample to 529.8 eV for Ni doped samples in an indication of the interaction between them. The surface O_latt/O_ads molar ratio decreases faintly from 0.96 to 0.67 with increasing Ni incorporation. This may be due to the narrower pore size creating more vacancies and unsaturated chemical bonds, which will lead to an increase in the surface lattice oxygen ratio.

Table 2 Benzene catalytic activity, E_a and XPS results for Co_xNiAlO samples

Catalysts	Benzene conversion			Ea (kJ mol^−1)	Co^3+	Co^2+	Co^3+/Co^2+ molar ratio	Olatt	Oads	Olatt/Oads
	T10	T50	T90
Co3AlO-Ref	215	254	288	67.1	0.68	0.32	2.13
Co3AlO	160	208	236	50.5	0.60	0.40	1.50	0.49	0.51	0.96
Co2NiAlO	152	204	227	39.0	0.57	0.43	1.33	0.47	0.53	0.89
Co1.5NiAlO	167	213	243	59.8	0.56	0.44	1.27	0.45	0.55	0.82
Co1NiAlO	167	223	264	71.7	0.52	0.48	1.08	0.42	0.58	0.72
Ni3AlO	186	292	>300	72.6				0.41	0.59	0.69

---

3.3 Temperature-programmed reduction

H₂-TPR was carried out to check the redox properties of Co_xNiAlO catalysts, as shown in Fig. 7. The Co₃AlO-Ref sample exhibits two separate peaks, one broad low temperature peak centered at 360 °C due to the two-step reduction of relatively large crystalline Co₃O₄ particles via the sequence of Co₃O₄ → CoO → Co⁰,³⁸ and another high temperature peak at 720 °C which is ascribed to the reduction of Co₃O₄ supported on Al₂O₃ or CoAl₂O₄.⁷ For the Co₃AlO sample, four separate reduction peaks are observed with the maximum rate of H₂ consumption occurring at 319, 395, 553 and 650 °C, respectively. The low temperature region contained two peaks, corresponding to the two step reduction of Co(III) to Co(II) and Co(II) to Co⁰.³⁹ The peaks at 553 and 650 °C are ascribed to the reduction of bulk Co₃O₄ embedded into the Al₂O₃ support. The Co₃O₄ reduction peak markedly shifted to lower temperatures compared to those of Co₃AlO-Ref, which may be due to the quite narrower pore size (2.9 nm vs. 17.2 nm), facilitating the diffusion of Co²⁺ ions. The incorporation of Ni into the Co–Al mixed oxides presents particular effects on the H₂-TPR profiles. It can be clearly seen that the low temperature reduction region between 250 and 420 °C shifts gradually toward a lower temperature with the incorporation of Ni into Co_xAlO systems, together with the indiscernible reduction around 650 °C, implying that the enhanced substitution of Ni²⁺ favors the reducibility of Co₃O₄. The Co₂NiAlO sample has a lower reduction temperature for the conversion of Co³⁺ to Co²⁺, together with the high concentration of Co³⁺ observed from the maximum H₂ consumption occurring at 302 °C for the largest TPR area with a strong signal among Co_xNiAlO samples. The lower reduction temperatures of the Co₂NiAlO sample indicate an increase of the lattice oxygen mobility on the catalyst, which will be an important factor in the catalytic oxidation of benzene. The high concentration of Co³⁺ leads to structural defects of the spinel Co₃O₄ lattice, which is beneficial for a larger amount of surface oxygen vacancies, improving catalytic activity for the combustion of VOCs. The Co₂NiAlO sample, possessing a larger pore volume and BET surface area than Co₃AlO, together with the incorporation of Ni, could be favourable for the formation of low temperature reducibility, which is quite useful to promote catalytic oxidation. It should be noticed that the Ni²⁺ reduction peak centered at 492 °C for the Ni₃AlO sample exhibited a little change with the doping of Co. The reduction temperatures are shifted toward lower values, further showing the synergetic interaction between Co and Ni species.

Fig. 7 Temperature-programmed reduction profiles of Co_xNiAlO catalysts.

3.4 Evaluation of the catalytic behavior

The catalytic performances of Co_xNiAlO samples in the total oxidation of benzene were evaluated as a function of the temperature, as shown in Fig. 8A and B. The error bars represent the standard deviation to the average of two experiments with two fresh catalysts from the same batch. The temperatures of 10% (T₁₀), 50% (T₅₀) and 90% (T₉₀) benzene conversion are summarized in Table 2. The benzene conversion and CO₂ yield all increased with increasing reaction temperature over these catalysts, while the CO₂ yield is close to the direct conversion of benzene with a carbon balance of about 98.5%, indicating that most benzene molecules are converted to CO₂ and H₂O completely. To further evaluate the possibility of by-products, the flux gas was abstracted with hexane for two varied reaction temperatures at 200 and 240 °C and analysed by GCMS (Fig. 8C). At T = 200 °C, only benzene was present except for the signal of hexane, while there was only hexane detected at T = 240 °C. The results demonstrated that there are no by-products formed and above 240 °C, benzene has been totally degraded. The Co₃AlO sample obtained through the surfactant method exhibited much higher activity with T₉₀ at 236 °C, compared to that of the Co₃AlO-Ref sample at 288 °C. The excellent activity of Co₃AlO can be attributed to the smaller pore size and lower temperature reduction than those of Co₃AlO-Ref, proved by BET and TPR characterization. Upon increasing the addition of Ni to Co₃AlO, the activity firstly increases, then starts to decrease, to become even worse than that of Co₃AlO. Among the catalysts, Co₂NiAlO exhibited the best benzene oxidation performance with T₉₀ = 227 °C and a reaction rate of about 0.24 mmol g_cat⁻¹ h⁻¹, compared to the Co₃AlO catalyst with 0.19 mmol g_cat⁻¹ h⁻¹ and Ni₃AlO catalyst with 0.02 mmol g_cat⁻¹ h⁻¹. The outstanding activity of Co₂NiAlO may be associated with the lower temperature reducibility from the larger amount of surface accessible Co³⁺, together with the synergistic effect of the incorporation of Ni, inducing a high number of surface oxygen species. The BET surface area for the Co₃AlO-Ref and Co₂NiAlO catalysts after the catalytic test were reduced by ca. 30 m² g⁻¹, and their pore volumes were a little increased, together with increased pore sizes in comparison with fresh catalysts (Table 1). The activities exceeded some other reported Co/Ni-based catalysts, like the Co₃O₄ nanocrystal growing from a restrictive nanoreactor with T₉₀ = 253 °C for benzene degradation reported by Tang,⁴⁰ Pd, Pt, and Ag–Co based spinel oxides for toluene combustion with T₉₀ at 226, 289 and 319 °C, respectively, synthesised by Zhao,⁴¹ and the optimal as-prepared Co_xAlO catalysts obtained by Li using a co-precipitation method with T₉₀ at 300 °C.⁴²

Fig. 8 (A) Benzene and (B) CO₂ conversion as a function of reaction temperature over Co_xNiAlO catalysts under a benzene concentration of 100 ppm in synthetic air, SV = 60 000 mL g_cat⁻¹ h⁻¹, (C) GC-MS data of hexane solvent, the gaseous products of the reaction collected by dissolving it in hexane at reaction temperatures of 200 °C and 240 °C, respectively, mass spectrometer signal of benzene from the gas production collected at T = 200 °C, (D) Arrhenius plots for the oxidation of benzene over the Co_xNiAlO catalysts.

The catalytic performance can also be evaluated by comparing the activation energy (E_a) values of different catalysts, and normally the sample with the lower E_a value will possess preferable catalytic activity. The E_a values (Table 2 and Fig. 8D) could be obtained from the Arrhenius equation from the detailed description in our previous publication.¹³ The plots of ln r versus 1/T exhibit excellent linear relationships (the correlation coefficients (R²) were rather close to 1). The E_a value of the Co₃AlO catalyst, 50.5 kJ mol⁻¹, was lower than that (67.1 kJ mol⁻¹) of Co₃AlO-Ref, which confirms the advantage of the urea co-precipitation method with surfactant. It is also observed that the Co₂NiAlO catalyst possessed the lowest E_a value (39.0 kJ mol⁻¹) among the Co_xNiAlO catalysts, which confirms the promoting effect of Ni.

To investigate the reproducibility and reversibility of the catalysts, the Co₂NiAlO catalyst with the best benzene activity was selected as a model system. Identical samples were mounted and tested three separate times with the same temperature programming system, and the data are shown in Fig. 9A. Satisfactorily, one batch sample essentially shows the same benzene conversion within an error margin of <±5 °C, indicating a sufficiently high degree of reproducibility of the catalytic tests. The reversibility and stability towards a higher temperature for benzene conversion were monitored at three temperatures of 200, 300 and 400 °C in sequence, followed by repetition. The catalytic activities were maintained at 100% benzene conversion at 400 °C over two cycles of tests from 200 to 400 °C lasting 30 h, indicating that the catalysts for benzene degradation are completely reversible and can stand up to higher temperature conditions, as shown in Fig. 9B. The effect of water vapor on benzene conversion for the Co₂NiAlO catalyst was examined at different temperatures (240 and 260 °C) in the absence and presence of 3.5% water vapor, shown in Fig. 9C. The benzene catalytic activity maintained at 97.2% and 98.8% at the two temperatures, with inconspicuous variation for a long time-on-stream reaction. There was significant drop in the benzene conversion from 97.2% to 88.8% when 3.5% H₂O was introduced to the stream at 240 °C, while it can be seen that water vapor had a negligible negative effect on the catalytic activity with 98.8% under 260 °C. These results suggest that water vapor may competitively cover some active sites to restrain the adsorption, mobility and activation of surface oxygen species at low temperature, leading to poor benzene oxidation activity, while surface active oxygen species, particularly O_latt and O_ads, more easily adsorb and migrate on active sites under high temperature, leading to excellent catalytic activity and tolerance to the water vapor.⁴³

Fig. 9 (A) Reproducibility and (B) reversibility for benzene conversion as a function of reaction temperature and time over Co_xNiAlO catalysts under a benzene concentration of 100 ppm in synthetic air. The symbols in A represent the same sample measured three times, (C) the effect of 3.5% water vapor over the Co₂NiAlO catalyst at 240 and 260 °C with and without water vapor, SV = 60 000 mL g_cat⁻¹ h⁻¹.

3.5 The structure and catalytic performance of the monolithic Co₂NiAlO-F catalyst

For better practical application, a monolithic Co₂NiAlO-F catalyst has been obtained by the calcination of a Co₂NiAlLDH film precursor on aluminum substrates through in situ growth. Film catalysts have recently attracted a great deal of attention due to their advantages of having sufficient exposure of active sites, and excellent heat conduction. The Co₂NiAlLDH-F precursor (Fig. 10A) exhibited a series of typical hydrotalcite reflections at 2θ of 11.6°, 23.2°, 34.8°and 60.1°, which can be attributed to the (003), (006), (009) and (110) planes, respectively. After calcination, the LDH phase transforms to the Co₃O₄ spinel phase, similarly to Co₂NiAlO in Fig. 1. The SEM image in Fig. 10B shows that the crystallized Co₂NiAlO-F possesses thin curved hydrotalcite-like platelets with their ab plane nearly perpendicular in orientation to the support. The Co₂NiAlLDH-F on an Al substrate, photographed by a camera, turns to a black color from a light green color after calcination (Fig. 10C) with a plain surface and strong adhesion to the substrate. The catalyst growing on the Al substrate was nearly 1.32 mg cm⁻² by weight. The benzene conversion evaluated in Fig. 10D manifested that the T₉₀ (275 °C) for Co₂NiAlO-F was reduced in comparison with the corresponding Co₂NiAlO powdered catalyst at 232 °C, while the reaction rate of Co₂NiAlO-F, with 0.31 mmol g_cat⁻¹ h⁻¹, is a little higher than that of 0.24 mmol g_cat⁻¹ h⁻¹ for Co₂NiAlO at 232 °C. The structural and functional superiority of the film catalyst makes it an exciting efficient candidate in a benzene oxidation system.

Fig. 10 (A) XRD pattern of Co₂NiAlLDH film precursor on the Al substrate (* represents the Al substrate) and Co₂NiAlO-F, (B) SEM images of Co₂NiAlO-F, (C) the photograph of the monolithic Co₂NiAlLDH and Co₂NiAlO-F catalysts on Al substrates, (D) comparison of benzene conversion over Co₂NiAlO and Co₂NiAlO-F catalysts.

4. Conclusions

A series of mesoporous and well-dispersed Co_xNiAlO catalysts were fabricated via the LDH precursor approach. The preparation method remarkably affects their benzene catalytic activities. The Co₃AlO sample shows outstanding activity compared to Co₃AlO-Ref due to the small pore size and good reducibility of the effect of surfactant during preparation. Furthermore, for different Ni doped catalysts, the Co₂NiAlO sample exhibits excellent catalytic behavior with T₉₀ = 227 °C and a low activation energy of 39.0 kJ mol⁻¹, which is ascribed to the lower temperature reducibility from the larger amount of surface accessible Co³⁺, together with the synergistic effect between the Co–Ni metallic oxide solid solution. The Co₂NiAlO sample displays good reproducibility and reversibility and resistance to 3.5% water vapor. The Co₂NiAlO-F film catalyst shows a higher reaction rate of 0.31 mmol g_cat⁻¹ h⁻¹ than that of 0.24 mmol g_cat⁻¹ h⁻¹ for Co₂NiAlO at 227 °C.

Acknowledgements

This research described above was supported by the National Natural Science Foundation of China (No. 21401200), the strategic project of science and technology of Chinese Academy of Sciences (No. XDB05050000), and the National Key Technology Support Program (No. 2014BAC21B).

References

1. E. Gallego, F. J. Roca, J. F. Perales and X. Guardino, Build. Environ., 2013, 67, 14–25.
2. L. Y. Lin and H. Bai, RSC Adv., 2016, 6, 24304–24313.
3. H. Chen, Y. Yan, Y. Shao and H. Zhang, RSC Adv., 2014, 4, 55202–55209.
4. M. S. Johnson, E. J. K. Nilsson, E. A. Svensson and S. Langer, Environ. Sci. Technol., 2014, 48, 8768–8776.
5. L. Sivachandiran, F. Thevenet, P. Gravejat and A. Rousseau, Chem. Eng. J., 2013, 214, 17–26.
6. E. Barea, C. Montoro and J. A. R. Navarro, Chem. Soc. Rev., 2014, 43, 5419–5430.
7. P. Li, C. He, J. Cheng, C. Y. Ma, B. J. Dou and Z. P. Hao, Appl. Catal., B, 2011, 101, 570–579.
8. S. H. Joo, J. Y. Park, C.-K. Tsung, Y. Yamada, P. Yang and G. A. Somorjai, Nat. Mater., 2009, 8, 126–131.
9. A. Q. Wang, C. M. Chang and C. Y. Mou, J. Phys. Chem. B, 2005, 109, 18860–18867.
10. P. O. Thevenin, A. G. Ersson, H. M. J. Kusar, P. G. Menon and S. G. Jaras, Appl. Catal., A, 2001, 212, 189–197.
11. J. Deng, S. He, S. Xie, H. Yang, Y. Liu, G. Guo and H. Dai, Environ. Sci. Technol., 2015, 49, 11089–11095.
12. S. Xia, L. Zhang, X. Zhou, G. Pan and Z. Ni, Appl. Clay Sci., 2015, 114, 577–585.
13. W. Tang, X. Wu, S. Li, X. Shan, G. Liu and Y. Chen, Appl. Catal., B, 2015, 162, 110–121.
14. G. Liu, J. Li, K. Yang, W. Tang, H. Liu, J. Yang, R. Yue and Y. Chen, Particuology, 2015, 19, 60–68.
15. Y. Sun, S. Gao, F. Lei and Y. Xie, Chem. Soc. Rev., 2015, 44, 623–636.
16. Y. Zhao, G. Chen, T. Bian, C. Zhou, G. I. Waterhouse, L. Z. Wu, C. H. Tung, L. J. Smith, D. O’Hare and T. Zhang, Adv. Mater., 2015, 27, 7824–7831.
17. R. Xie, G. Fan, L. Yang and F. Li, Catal. Sci. Technol., 2015, 5, 540–548.
18. X. Mei, J. Wang, R. Yang, Q. Yan and Q. Wang, RSC Adv., 2015, 5, 78061–78070.
19. R. Han, C. Nan, L. Yang, G. Fan and F. Li, RSC Adv., 2015, 5, 33199–33207.
20. J. Feng, Y. He, Y. Liu, Y. Du and D. Li, Chem. Soc. Rev., 2015, 44, 5291–5319.
21. R. Ma, J. Liang, K. Takada and T. Sasaki, J. Am. Chem. Soc., 2011, 133, 613–620.
22. P. J. Sideris, U. G. Nielsen, Z. Gan and C. P. Grey, Science, 2008, 321, 113–117.
23. Y. Zhao, Q. Wang, T. Bian, H. Yu, H. Fan, C. Zhou, L. Z. Wu, C. H. Tung, D. O’Hare and T. Zhang, Nanoscale, 2015, 7, 7168–7173.
24. Y. Zhao, X. Jia, G. I. N. Waterhouse, L. Z. Wu, C. H. Tung, D. O’Hare and T. Zhang, Adv. Energy Mater., 2016, 6, 1–20.
25. S. He, C. Li, H. Chen, D. Su, B. Zhang, X. Cao, B. Wang, M. Wei, D. G. Evans and X. Duan, Chem. Mater., 2013, 25, 1040–1046.
26. G. Fan, F. Li, D. G. Evans and X. Duan, Chem. Soc. Rev., 2014, 43, 7040–7066.
27. Q. Wang and D. O’Hare, Chem. Rev., 2012, 112, 4124–4155.
28. S. He, Z. An, M. Wei, D. G. Evans and X. Duan, Chem. Commun., 2013, 49, 5912–5920.
29. M. Gabrovska, R. Edreva-Kardjieva, K. Tenchev, P. Tzvetkov, A. Spojakina and L. Petrov, Appl. Catal., A, 2011, 399, 242–251.
30. S. Li, H. Wang, W. Li, X. Wu, W. Tang and Y. Chen, Appl. Catal., B, 2015, 166–167, 260–269.
31. L. He, Y. Huang, A. Wang, X. Wang, X. Chen, J. J. Delgado and T. Zhang, Angew. Chem., Int. Ed., 2012, 51, 6191–6194.
32. M. Q. Zhao, Q. Zhang, W. Zhan, J. Q. Huang, Y. Zhang, D. S. Su and F. Wei, J. Am. Chem. Soc., 2010, 132, 14739–14741.
33. W. Tang, X. Wu, S. Li, W. Li and Y. Chen, Catal. Commun., 2014, 56, 134–138.
34. Y. Han, Z.-H. Liu, Z. Yang, Z. Wang, X. Tang, T. Wang, L. Fan and K. Ooi, Chem. Mater., 2008, 20, 360–363.
35. T. Garcia, S. Agouram, J. F. Sanchez-Royo, R. Murillo, A. Maria Mastral, A. Aranda, I. Vazquez, A. Dejoz and B. Solsona, Appl. Catal., A, 2010, 386, 16–27.
36. G. A. Babu, G. Ravi, T. Mahalingam, M. Navaneethan, M. Arivanandhan and Y. Hayakawa, J. Phys. Chem. C, 2014, 118, 23335–23348.
37. V. Binni, M. V. Reddy, Z. Yanwu, C. S. Lit, T. C. Hoong, G. V. S. Rao, B. V. R. Chowdari, A. T. S. Wee, C. T. Lim and C.-H. Sow, Chem. Mater., 2008, 20, 3360–3367.
38. G. Bai, H. Dai, J. Deng, Y. Liu, F. Wang, Z. Zhao, W. Qiu and C. T. Au, Appl. Catal., A, 2013, 450, 42–49.
39. B. de Rivas, R. López-Fonseca, C. Jiménez-González and J. I. Gutiérrez-Ortiz, J. Catal., 2011, 281, 88–97.
40. W. Tang, Y. Deng, W. Li, S. Li, X. Wu and Y. Chen, Catal. Commun., 2015, 72, 165–169.
41. S. Zhao, K. Li, S. Jiang and J. Li, Appl. Catal., B, 2016, 181, 236–248.
42. D. Li, Y. Ding, X. Wei, Y. Xiao and L. Jiang, Appl. Catal., A, 2015, 507, 130–138.
43. Q. Liu, L.-C. Wang, M. Chen, Y. Cao, H.-Y. He and K.-N. Fan, J. Catal., 2009, 263, 104–113.

---