High catalytic activity for formaldehyde oxidation of an interconnected network structure composed of δ-MnO2 nanosheets and γ-MnOOH nanowires

Among the transition metal oxide catalysts, manganese oxides have great potential for formaldehyde (HCHO) oxidation at ambient temperature because of their high activity, nontoxicity, low cost, and polybasic morphologies. In this work, a MnO2-based catalyst (M-MnO2) with an interconnected network structure was successfully synthesized by a one-step hydrothermal method. The M-MnO2 catalyst was composed of the main catalytic agent, δ-MnO2 nanosheets, dispersed in a nonactive framework material of γ-MnOOH nanowires. The catalytic activity of M-MnO2 for HCHO oxidation at room temperature was much higher than that of the pure δ-MnO2 nanosheets. This is attributed to the special interconnected network structure. The special interconnected network structure has high dispersion and specific surface area, which can provide more surface active oxygen species and higher surface hydroxyl groups to realize rapid decomposition of HCHO.


Introduction
Formaldehyde (HCHO) is a typical indoor air pollutant that is mainly released from furnishings and building materials [1][2][3]. Long-term exposure to HCHO can induce adverse health consequences, such as allergic rhinitis, asthma, respiratory system diseases, hepatitis, pneumonia, nasopharyngeal cancer, and leukemia [4][5][6][7]. In previous studies, several approaches for HCHO oxidation have been developed, such as biological degradation, photocatalytic oxidation, adsorption, and thermal catalytic oxidation. It has been proved that room-temperature catalytic oxidation is one of the most promising and cost-effective methods for high-efficiency conversion of HCHO into harmless species (H 2 O, CO 2 ) over a metal oxide catalyst [8][9][10].
Noble metal catalysts, mainly containing Pt, Au, Pd, and Ag as the active component supported on various metal oxides (TiO 2 , Al 2 O 3 , ZrO 2 , and Fe 2 O 3 ), show high performance for HCHO oxidation at ambient temperature [11][12][13][14]. However, their high cost, poor thermal stabilities, and limited resources have prevented a wide application of noble metal catalysts. It was found that transition metal oxide catalysts had enormous potential for HCHO oxidation at room temperature. In particular, the transition metal oxide catalysts such as MnO x , Co 3 O 4 , and CeO 2 show good thermal stability and are less expensive and more abundant than noble metal catalysts [15][16][17][18].
Among the transition metal oxide catalysts, manganese oxides have been widely used in catalysis because of their high activity, nontoxicity, low cost, polybasic structures, and morphologies [10][11][12][13][14][15][16][17][18][19][20][21][22]. The structures of manganese oxide, including a-MnO 2 , b-MnO 2 , c-MnO 2 , and d-MnO 2 , play an important role in catalytic capacity [23][24][25][26][27]. Zhang et al. [25] found that d-MnO 2 had the highest activity because of its special 2D layer tunnel structure that contained the most active oxygen species and the highest amount of lattice oxygen species on the catalyst surface. Birnessite (d-MnO 2 ) [23,25] is a layered manganese dioxide constructed by an edge-sharing [MnO 6 ] octahedra framework with a certain number of cations (Li ? , Na ? , Ca 2? ) and water molecules located between the manganese oxide layers for charge balance. Wang et al. [17] found that the water content in the birnessite was essential for HCHO oxidation-that was, better activity was obtained for materials with higher water content. They also found that the HCHO oxidation efficiency was influenced by the different types of interlayer cation (K ? , Mg 2? , Ca 2? , Fe 3? ) [18,19]. However, the agglomeration of the nanosheets in d-MnO 2 can reduce the exposed active sites. In the past few years, MnO 2 has usually been supported on carbon materials (graphene hydrogel, carbon foam, carbon nanotubes, and carbon fibers) to solve the problem of agglomeration [28,29]. However, these carbon materials are inert components in catalysts and may cover the active sites for catalytic reactions. To solve this problem, Rong et al. [30,31] developed an organic and template-free approach to prepare 3D MnO 2 by freeze-drying aqueous solutions of the MnO 2 nanosheets and nanowires. Although improved catalytic performance was obtained, this method involved multiple steps and was cumbersome.
In this report, a one-step hydrothermal method for the fabrication of a MnO 2 -based catalyst (M-MnO 2 ) composed of d-MnO 2 nanosheets anchored on a nanowire-shaped c-MnOOH framework is described. It exhibits high performance for HCHO oxidation at room temperature.

Catalyst preparation
All the chemicals used in this study were of analytical grade, and deionized water was used for all preparations. KMnO 4 and (NH 4 ) 2 C 2 O 4 ÁH 2 O were purchased from Sinopharm Chemical Reagent Co., Ltd. Deionized water was obtained from a highly pure water system (Thermo Co., USA).
The catalysts were synthesized by a facile redox reaction through the hydrothermal method. A series of reactions for different MnO 2 structures were carried out by changing the dosage of the reducing agent. Typically, KMnO 4 (6.33 mmol) and (NH 4 ) 2 C 2 O 4 ÁH 2 O (2.81, 5.98 and 11.26 mmol) were mixed with deionized water (40 mL) under vigorous stirring at room temperature. Then, the solution was transferred into a 100-mL Teflon-lined stainless-steel autoclave maintained at 100°C for 10 h at a heating rate of 5°C/min in an electric oven. Then, the autoclave was cooled to room temperature, and the precipitate was collected by centrifugal separation, washed three times with distilled water and ethanol, and dried in vacuum at 80°C for 12 h.

Characterization
The catalysts were characterized using an X-ray diffractometer (D8 ADVANCE, Bruker, Germany) operated at 40 kV and 40 mA with Cu-K a radiation (k = 0.154 2 nm) at a scanning rate of 5°/min in the 2h range of 5°-80°. The morphology and microstructure of the samples were determined using a Magellan 400 field-emission scanning electron microscope (FE-SEM, USA) and a Tecnai G2 F20 transmission electron microscope (TEM, Netherlands). X-ray photoelectron spectroscopy (XPS) measurements were performed using an ESCALAB 250 X-ray photoelectron spectrometer (Thermo Fisher, UK) microprobe with an Al-K a source (k = 1 486.7 eV) at a pass energy of 30 eV. The XPSPEAK41 peak fitting program was applied to fit the Mn 2p 3/2 , Mn 2p 1/2 , and O 1s spectra. The specific surface area and the pore volume of the samples were determined by the multipoint Brunauer-Emmett-Teller method and Barrett-Joyner-Halenda analyses.
Hydrogen temperature-programmed reduction (H 2 -TPR) and oxygen temperature-programmed desorption (O 2 -TPD) were performed using a Chemisorb 2920 instrument (USA) with a thermal conductivity detector (TCD). For H 2 -TPR, the sample (50 mg) was first pretreated with He (30 mL/ min) at 150°C for 1 h and then cooled to room temperature. Next, the temperature was increased from 25°C to 600°C at a heating rate of 5°C/min with a flow of 5% H 2 / Ar at a flow rate of 30 mL/min that passed the samples monitored by the TCD. For O 2 -TPD, the sample (50 mg) was first pretreated with He (30 mL/min) at 300°C for 30 min to remove physically adsorbed and interlayer H 2 O and surface oxygen, and then it was cooled to room temperature with flowing O 2 for 30 min. Next, the sample was purged with He for 1 h to remove the weakly adsorbed O 2 . Finally, the temperature was increased from room temperature to 500°C at a heating rate of 10°C/min under He flow (30 mL/min). In situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) was performed using a Fourier transform infrared spectrometer (Nicolet 6700, Thermo Fisher, USA) to identify the intermediate species during HCHO oxidation. All spectra were measured with a resolution of 4 cm -1 and an accumulation of 32 scans. Prior to exposure to HCHO flowing gas, the samples were pretreated by He (30 mL/min) for 30 min at room temperature. Then, the reactant gas mixture (80 mg/ kg of HCHO/N 2 ? 20% O 2 /N 2 ) was injected into the DRIFTS cell at a flow rate of 60 mL/min. Before the data were recorded, the samples were swept by He (30 mL/min) for 1 min to remove weakly adsorbed species.

Catalytic activity tests
The evaluation of the catalytic activity system hast two parts: static and dynamic experiments.
For static experiments, the catalytic activities of different samples were evaluated in a 5-L glass reactor at room temperature. A small electric fan (0.5 W) was securely fixed on the top of the reactor to ensure the uniform dispersal of the HCHO gas. The temperature of the glass reactor was maintained at 25°C during the reaction. Moreover, the relative humidity inside the reactor was controlled to 60%, which was close to the environmental humidity. Typically, the sample (100 mg) was spread on a watch glass that was placed at the bottom of the glass reactor, and then HCHO solution (38%, 15 lL) was injected into the reactor. The initial concentration of HCHO gas was 180 mg/kg when the HCHO solution was volatilized completely and then was analyzed by the 3-methyl-2-benzothiazolinone hydrazine method.
For the dynamic experiments, HCHO oxidation was performed in a fixed bed reactor under atmospheric pressure within a temperature range of 25-180°C. The sample (100 mg) was placed in a quartz tube with a diameter of 6 mm that was used as the reactor. The temperature of the reactor was controlled using a constant temperature controller. HCHO gas was generated by flowing compressed air (21% O 2 ) over the paraformaldehyde in a water bath at 25°C at a flow rate of 100 mL/min. The inlet concentration of HCHO was 100 mg/kg, corresponding to a gas hourly space velocity (GHSV = 60 000 mL/(g cat Áh).
For kinetics measurements, the HCHO conversion was controlled to be \ 15%. The reaction rate (m) was calculated from the dynamic experiments according to where C HCHO is the HCHO concentration of the feed gas (mol/mL), F gas the total flow rate during HCHO oxidation, g the stable HCHO conversion, and m cat the mass of the catalyst in the reactor bed.
To calculate the apparent activation energy (E a ), it is assumed that HCHO oxidation follows the first-order reaction kinetics. According to the Arrhenius formula, the obtained kinetic model can be described by where R is the molar gas constant, T the catalytic temperature, k 0 the reaction constant. 88-0649). After more reductants (11.26 mmol) were added, the birnessite phase disappeared, and there was only manganite; therefore, the sample was named ''c-MnOOH'' [32,33]. The wide peak widths of birnessite in d-MnO 2 and M-MnO 2 indicate poor crystallinity, which may be caused by the damage to the ordered structure. The morphology and microstructure of these three samples were observed using SEM and TEM, as shown in  and c-MnOOH, respectively. It was concluded that the sample with this special morphology was a mixture of d-MnO 2 and c-MnOOH. This is also supported by the XRD results. The morphology of the as-synthesized products can be regulated by modulating the ratio of (NH 4 ) 2 C 2 O 4 / KMnO 4 during the hydrothermal process. With the increase of the ratio of (NH 4 ) 2 C 2 O 4 /KMnO 4 , the low-valent manganese compound c-MnOOH appeared. The d-MnO 2 nanosheets tended to anchor on a nanowire-shaped c-MnOOH, which increased the specific surface area and the pore volume compared with the pure d-MnO 2 , as shown in Table 1.   In addition, as shown in Fig. 3b, the O 1s spectra of the three different samples were deconvoluted into two peaks located at 529.9 eV and 531.4 eV that were associated with lattice oxygen (O latt ) and surface adsorbed oxygen (O ads ) species, respectively. The O ads /O latt ratios of the three samples are calculated from the XPS data using their peak area ratios and are presented in Table 1 and ÁO - [34]. Previous studies have shown that the catalytic oxidation activity at low temperatures is closely related to the content of surface oxygen species, and most of the O ads is generated from the adsorption and activation of adsorbed oxygen at oxygen vacancies [31]. The highest O ads /O latt ratios are ascribed to the high-specific-surface area and can expose more active sites, which can be deduced from the SEM images and higher-surface-area data. H 2 -TPR experiments were performed to investigate the reducibility of the three samples. Figure 3c shows the H 2 -TPR profiles of c-MnOOH, d-MnO 2 , and M-MnO 2 catalysts with increasing temperature that can be deconvoluted into four peaks, named peaks 1, 2, 3, and 4. Among these, the peak area ratio of peaks 2, 3, and 4 is approximately 3:1:2 for all samples. This can be associated with the sequential reduction of MnO 2 to Mn 2 O 3 , Mn 2 O 3 to Mn 3 O 4 , and Mn 3 O 4 to MnO [28]. However, the first reduction peaks for c-MnOOH, d-MnO 2 , and M-MnO 2 are located at 240°C, 228°C, and 197°C, respectively. M-MnO 2 showed the lowest initial reduction temperature among all of the samples, suggesting that its surface oxygen species showed the highest reducibility and activity. It is clear that the highest oxidation potential is in the order of M-MnO 2 [ d-MnO 2 [ c-MnOOH, which is consistent with the results for the catalytic activity of HCHO oxidation by these samples.

Results and discussion
O 2 -TPD experiments were performed to investigate the activity of the surface adsorbed oxygen in different samples. Several O 2 desorption peaks were observed from in the 40-500°C, as shown in Fig. 3d. The first desorption High catalytic activity for formaldehyde oxidation of an interconnected network structure… 433 peak was assigned to the surface adsorbed oxygen (ÁO 2 -/ ÁO -), while the rest of the peaks were related to the release of lattice oxygen [35,36]. The first surface adsorbed oxygen desorption occurred at temperatures of 113°C, 100°C, and 72°C for c-MnOOH, d-MnO 2 , and M-MnO 2 , respectively. M-MnO 2 showed the lowest surface adsorbed oxygen desorption temperature, indicating that the surface adsorbed oxygen of M-MnO 2 exhibited the highest mobility and activity. Furthermore, the special interconnected network structure of M-MnO 2 favors mass transfer and exposes more active sites.
The specific surface area and pore volume of c-MnOOH, d-MnO 2 , and M-MnO 2 are shown in Table 1. The specific surface area and the pore volume of c-MnOOH were higher than those of the other two species, and those of d-MnO 2 were the lowest. This was because of the high dispersion of c-MnOOH nanowires and the severe agglomeration of d-MnO 2 nanosheets. M-MnO 2 has a special structure composed of d-MnO 2 nanoparticles anchored on nanowireshaped c-MnOOH, which avoids the severe agglomeration of d-MnO 2 .
The catalytic activities of c-MnOOH, d-MnO 2 , and M-MnO 2 for HCHO removal at room temperature are compared in Fig. 4. For the static experiment, shown in Fig. 4a, the concentration of HCHO was decreased from 180 to 142, 80 and 34 mg/kg within 1 h by c-MnOOH, d-MnO 2 , and M-MnO 2 , respectively. In the static catalytic activity test, M-MnO 2 showed significantly higher HCHO removal ability than the other samples, and the catalytic efficiency of c-MnOOH was quite low. The test demonstrated that d-MnO 2 was the main active component for HCHO oxidation in the M-MnO 2 catalyst.
The dynamic experiment results further illustrate the catalytic activities of the different samples, as shown in Fig. 4b. The catalytic efficiencies of c-MnOOH, d-MnO 2 , and M-MnO 2 were 4%, 36%, and 53% at room temperature, respectively. Meanwhile, T 100 refers to the temperature of complete HCHO conversion. The T 100 of c-MnOOH, d-MnO 2 , and M-MnO 2 were 160°C, 120°C, and 90°C, respectively. This shows that the c-MnOOH catalyst needs a tremendous amount of energy (160°C) to achieve complete HCHO conversion. The three catalysts' variation tendency in Fig. 4b is similar to that of the static The catalytic stability and HCHO removal efficiency are important for the practical applications of the catalysts [34]. Therefore, the catalytic HCHO removal efficiencies of the three samples were tested in a series of relatively long-term experiments, as shown in Fig. 4c. The catalytic efficiency of M-MnO 2 was greater than 50% and remained stable during the entire 10 h test. Meanwhile, the HCHO removal efficiency of pure d-MnO 2 decreased sharply from 36% to 27% within 10 h, while that of pure c-MnOOH was initially 6% and then rapidly decreased to zero. These results further indicate that the catalytic efficiency of M-MnO 2 is higher than that of the other two catalysts and can be maintained for a long time at room temperature.
The apparent activation energies (E a ) of the three samples are given in Fig. 4d and were calculated according to the Arrhenius plots of the catalysts for HCHO oxidation. The E a values for c-MnOOH, d-MnO 2 , and M-MnO 2 were 34.1, 18.2 and 15.1 kJ/mol, respectively. It is clear that HCHO is oxidized more easily by M-MnO 2 because it has a lower E a value than c-MnOOH and d-MnO 2 .
Thermogravimetry (TG) analysis was used for evaluating the weight loss and qualitative analysis for detachable species in different samples with increasing temperature (see Fig. 5). Mass spectral analysis was used to confirm the detachable species according to mass charge ratio (m/z). It was found that the m/z of all detachable species was close to 18, which was consistent with the water molecule. The content of the largest surface adsorbed water loss (less than 100°C) was 9.21% for M-MnO 2 , followed by d-MnO 2 and c-MnOOH, which were only 3.68% and 2.71%, respectively. According to the literature, the surface adsorbed water can provide a rich surface hydroxyl group, which promotes the efficiency of HCHO oxidation at room temperature [17].
To explain why the M-MnO 2 catalyst shows excellent catalytic activity and to observe the reaction process and mechanism, in situ DRIFTS spectra of the three samples exposed in the 80 mg/kg HCHO/N 2 ? 20% O 2 /N 2 for HCHO oxidation at room temperature were obtained (see Fig. 6). There was no obvious material change on the surface of c-MnOOH, which indicated that the catalytic reaction was weak. The material change on the surface of d-MnO 2 and M-MnO 2 was obvious, which also confirmed that d-MnO 2 was the main catalytic agent, and the catalyst activity of c-MnOOH could be ignored at room temperature. The band at 1 465 cm -1 represents the dioxymethylene (DOM) species (d(CH 2 )), suggesting that HCHO first absorbed on the samples to form DOM because the carbonyl carbon in HCHO is electrophilic and tends to adsorb on the nucleophilic surface oxygen atom. The weak intensity of the DOM band indicates that DOM can be rapidly transformed into formate species with surface oxygen. Moreover, the band at 1 465 cm -1 is much lower in the M-MnO 2 sample, suggesting that the DOM on M-MnO 2 can be decomposed faster than that on a pure d- Formate species are the well-known intermediates during the catalytic oxidation of HCHO and are detected in the experiments. Four intense bands located at 2 914 (m(CH)), 1 564 (m as (COO -)), 1 415(d(CH)), and 1 340 (m s (COO -)) cm -1 can be ascribed to the formate species [30,31]. The intensity of the formate species bands on d-MnO 2 and M-MnO 2 increased with the HCHO exposure time, suggesting that formate species accumulated gradually on the catalyst surface. Meanwhile, the pure d-MnO 2 sample showed a much higher relative intensity of the formate species than M-MnO 2 . According to the previous studies [28,29], the catalytic performance of the catalysts decreased, with the formate species occupying the active sites of the catalysts. This can explain the gradual decrease in the catalytic performance of d-MnO 2 over time. The decomposition of formate species is the governing process for HCHO oxidation at room temperature, and the formate species can be transformed to carbonate species by the surface hydroxyl groups [17]. The intensity of the negative and large band at 3533 cm -1 (m(OH)) decreased gradually with the exposure time, indicating that the hydroxyl groups in the structure were continuously consumed during the reaction. Moreover, the carbonate species, located at 1 685 (m as (CO 3 )) cm -1 and 1 201 (m s (CO 3 )) cm -1 , were only found on M-MnO 2 , as shown in Fig. 6c. These indicate that the decomposition efficiency of the formate species on M-MnO 2 is higher than on pure d-MnO 2 . It has been reported that hydroxyl groups play a key role in MnO 2based catalysts and the surface hydroxyl groups can be regenerated by the reaction between surface active oxygen (ÁO -/ÁO 2 -) and surface adsorbed water through the Based on the previous analysis, the mechanism of HCHO oxidation over M-MnO 2 at room temperature can be described as shown in Fig. 7. Normally, HCHO is oxidized by active oxygen (ÁO -/ÁO 2 -) into DOM and formate species after adsorbing on the catalyst surface, while the formate species tend to deposit on the catalyst surface, which requires more energy or surface hydroxyl groups to decompose. However, M-MnO 2 is composed of d-MnO 2 nanosheets dispersed on c-MnOOH nanowires, and this structure can be beneficial for the surface active oxygen compared with d-MnO 2 . According to the above- This special morphology can provide more surface active oxygen species, more surface adsorbed water, and thus, higher surface hydroxyl groups. Therefore, the formate species can be rapidly transformed to carbonate species and then decomposed to CO 2 and H 2 O.

Conclusions
A special interconnected network structure, M-MnO 2 , composed of d-MnO 2 nanosheets and c-MnOOH nanowires for HCHO oxidation was prepared by a facile onestep hydrothermal method. The obtained M-MnO 2 exhibited high catalytic activity, which reached 50% for 100 mg/ kg HCHO oxidation for 10 h under a gas hourly space velocity (GHSV = 60 000 mL/(g cat Áh) higher than that of pure d-MnO 2 nanosheets. It was found that the interconnected network structure provided sufficient surface active oxygen species and surface adsorbed water, resulting in excellent catalytic activity and stability for HCHO oxidation at room temperature. Moreover, the synthetic method is simple and can be easily implemented for large-scale industrial application. Hong-Jie Luo is a professor at shanghai university. He received his doctor's degree from shanghai institute of ceramics in 1991. He was supported by China National Science Fund for Distinguished Young Scholars and the onehundred-talent program of CAS. He is mainly engaged in the research of cultural relics conservation and energy-saving materials.
High catalytic activity for formaldehyde oxidation of an interconnected network structure… 439