Tuning the K+ Concentration in the Tunnel of OMS-2 Nanorods Leads

Nov 1, 2013 - ABSTRACT: OMS-2 nanorods with tunable K+ concentration were prepared by a facile hydrothermal redox reaction of MnSO4, (NH4)2S2O8, ...
0 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/est

Tuning the K+ Concentration in the Tunnel of OMS‑2 Nanorods Leads to a Significant Enhancement of the Catalytic Activity for Benzene Oxidation Jingtao Hou,† Liangliang Liu,‡ Yuanzhi Li,*,† Mingyang Mao,† Haiqin Lv,† and Xiujian Zhao† †

State Key Laboratory of Silicate Materials for Architectures, Wuhan University of Technology, 122 Luoshi Road, Wuhan 430070, P. R. China ‡ Department of Physics, Wuhan University, Luojia Hill, Wuhan 430072, P. R. China S Supporting Information *

ABSTRACT: OMS-2 nanorods with tunable K+ concentration were prepared by a facile hydrothermal redox reaction of MnSO4, (NH4)2S2O8, and (NH4)2SO4 at 120 °C by adding KNO3 at different KNO3/MnSO4 molar ratios. The OMS-2 nanorod catalysts are characterized by X-ray diffraction, transmission electron microscopy, N2 adsorption and desorption, inductively coupled plasma, and X-ray photoelectron spectrometry. The effect of K+ concentration on the lattice oxygen activity of OMS-2 is theoretically and experimentally studied by density functional theory calculations and CO temperature-programmed reduction. The results show that increasing the K+ concentration leads to a considerable enhancement of the lattice oxygen activity in OMS-2 nanorods. An enormous decrease (ΔT50 = 89 °C; ΔT90 > 160 °C) in reaction temperatures T50 and T90 (corresponding to 50 and 90% benzene conversion, respectively) for benzene oxidation has been achieved by increasing the K+ concentration in the K+-doped OMS-2 nanorods due to the considerable enhancement of the lattice oxygen activity.

■

INTRODUCTION Volatile organic compounds (VOCs) as major air pollutants are not only hazardous to human health but also harmful to the environment. It is highly desirable to control the emissions of VOCs. Among various VOCs, the carcinogenic and recalcitrant benzene, which is one of the most abundant aromatic hydrocarbons found in polluted urban atmospheres, has been regarded as a priority hazardous substance for which efficient treatment technologies are needed.1,2 Until now, a number of methods have been developed for the removal of VOCs. Among the strategies for the elimination of VOCs, catalytic oxidation is believed to be one of the most important technologies. Expensive supported noble metals are conventionally used as the most efficient catalysts in the catalytic oxidation of VOCs.2 Finding economical catalysts that are alternatives to noble metal catalysts would be scientifically and technologically significant. The cryptomelane-type octahedral molecular sieve (OMS-2), which is inexpensive and environmentally benign, is a type of manganese oxide having edge- and corner-shared MnO6 octahedra forming a 2 × 2 tunnel structure.3 Because of its unique structural characteristics such as porous structure, mixed valency (2+, 3+, and 4+) of Mn, easy release of lattice oxygen, etc.,3 OMS-2 has been extensively investigated for diverse applications of catalytic oxidation.3−16 Because the tunnel cavity of OMS-2 is 0.46 nm long, some cations, such as Li+, K+, Na+, Rb+, Cs+, NH4+, etc., can be introduced into the tunnel, which can tailor its chemical and physical properties. Several strategies © 2013 American Chemical Society

have been reported for improving the catalytic activity of OMS2 by the incorporation of metal cations into the OMS-2 tunnel. For example, Suib and co-workers have reported for the first time systematically using alkali metal cations (e.g., Li+, Na+, and Rb+) and NH4+ cations as templates to synthesize 2 × 2 tunnel structures (A-OMS-2; A = Li, Na, Rb, or NH4+).17 They found that the crystallinity, microstructures, and properties of these AOMS-2 materials such as the surface area, thermal stability, and chemical composition greatly depend on the nature of cations, and their catalytic activities for the oxidation of cyclohexanol to cyclohexanone are related to their surface areas. Santos et al. have successfully incorporated Cs and Li cations into the tunnel of OMS-2 using the ion-exchange method and reported that the doping of Li and Cs cations enhanced the catalytic activity of OMS-2 for the total oxidation of ethyl acetate, which is explained in terms of the enhanced reducibility and framework basicity of the doped OMS-2.5 However, they found that the amount of Li and Cs in the doped OMS-2 does not play a significant role in the catalytic performance. Potassium ions have been taken as the ideal cation templates for forming and stabilizing the 2 × 2 tunnel structure in synthetic cryptomelane materials, because the dimensions of the tunnels of OMS materials were believed to be controlled Received: Revised: Accepted: Published: 13730

September 3, 2013 October 25, 2013 November 1, 2013 November 1, 2013 dx.doi.org/10.1021/es403910s | Environ. Sci. Technol. 2013, 47, 13730−13736

Environmental Science & Technology

Article

10 °C min−1 to 700 °C in a flow of 5 vol % O2/He at a rate of 24 mL min−1. Density Functional Theory Calculation Method. Density functional theory (DFT) calculations were used to study OMS-2 (KMn8O16). The calculations were performed using the Vienna Ab-initio Simulation Package (VASP).19−21 The valence electronic states are expanded in a set of periodic plane waves, and the interaction between core electrons and the valence electrons is implemented through the projector augmented wave (PAW) approach. The Perdew−Burke− Ernzerhof (PBE) GGA-exchange correlation functional is applied in the calculations.22−24 In relaxation, summations over the Brillouin zone (BZ) are performed with a 2 × 2 × 2 Monkhorst−Pack k-point mesh. The smooth part of the wave functions is expanded in plane waves with a kinetic energy cutoff of 400 eV, and the convergence criteria for the electronic and ionic relaxation are 10−4 eV and 0.02 eV Å−1, respectively. Our calculated lattice parameters of bulk KMn8O16 are as follows: a = 9.755 Å, and c = 2.96 Å (close to those of Cockayne and Li25). To ensure the convergence of the calculations, we repeated all the calculations with a larger super cell size of ∼100 atoms (1 × 1 × 4). The lowest-energy position of K atoms by calculation is at eight oxygen near neighbors. Catalytic Activity. The catalytic activity measurements for the catalytic oxidation of benzene were taken at atmospheric pressure in a continuous flow fixed-bed quartz tubular reactor on a WFS-2015 online gas-phase reaction apparatus; 0.0500 g of the catalyst was loaded in the middle of the quartz reactor (with an inner diameter of 16 mm and a length of 600 mm) supported by quartz wool. A thermocouple placed inside the reactor, which contacted the catalyst bed, monitored the reaction temperature. The reactor was placed in a temperaturecontrolled tubular electrical oven. The gaseous reactant of benzene was generated by causing air to flow into a vapor saturator, which was kept at 6 °C in a temperature-controlled water bath. The air stream saturated with the reactant was diluted with another flow of air to yield a final reactant concentration of 2000 mg m−3. The total flow rate was 40 mL min−1 [SV = 48000 mL (g of catalyst)−1 h−1]. The reactor was connected to a GC9560 gas chromatograph equipped with a flame ionization detector (FID), a methane converter, a Porapak R column, and a PEG20M column through an automatically sampling 10-way valve (VALCO) with an air actuator. The effluent reaction products contained only CO2 and H2O, and no other byproducts were detected by the gas chromatograph. The conversion was calculated on the basis of benzene consumption. It is the same as the conversion calculated by CO2 production, indicating a good carbon balance. The unreacted reactant in the effluent from the reactor was collected by a cooling tank placed in a salt−ice bath.

directly by the sizes of the templates used.17 Luo et al. synthesized the K+-doped OMS-2 samples with different K+ concentrations through HCl/KOH treatments of the precursor α-MnO2 nanotubes and found that their magnetic properties were dramatically influenced by the K+ concentration.18 Although many reported OMS-2 catalysts contain K+, few works have reported the effect of the K+ concentration on the catalytic activity of OMS-2. At the same time, whether and why the incorporation of K+ can promote the catalytic activity of OMS-2 are still open questions. Herein, we develop a facile method for preparing K+-doped OMS-2 nanorod catalysts with different K+ concentrations. The effect of K+ concentration on the lattice oxygen activity of OMS-2 is theoretically and experimentally studied. For the first time, we demonstrate an enormous effect of K+ concentration on the catalytic activity of the K+-doped OMS-2 nanorods for benzene oxidation.

■

EXPERIMENTAL SECTION Catalyst Preparation. OMS-2 nanorod samples with different K+ concentrations were prepared by a facile hydrothermal redox reaction of MnSO4·H2O, (NH4)2S2O8, and (NH4)2SO4 via addition of different amounts of KNO3. The detailed procedure is as follows. To start, 1.3521 g of MnSO4·H2O, 1.8256 g of (NH4)2S2O8, and 1.9821 g of (NH4)2SO4 were dissolved in 40 mL of distilled water in a beaker; 0, 0.4044, 1.6176, or 3.2353 g of KNO3 was added into the mixture solution described above under vigorous magnetic stirring until it was dissolved. Then, the mixture solution was transferred to a 100 mL Teflon bottle, which was sealed tightly in a stainless-steel autoclave. The autoclave was placed in an electrical oven, heated to 120 °C, and kept at that temperature for 20 h. After the autoclave had cooled to room temperature, the precipitate was washed with distilled water and dried under an infrared lamp. Catalyst Characterization. X-ray diffraction (XRD) patterns were observed on a Rigaku Dmax X-ray diffractometer using Cu Kα radiation. Transmission electron microscopy (TEM) images were obtained by using a JEM-100CX electron microscope. The surface area of the OMS-2 samples was measured on an AUTOSDRB-1 instrument using N 2 adsorption at −196 °C. The chemical compositions of the OMS-2 samples were measured by inductively coupled plasma/ optical emission spectroscopy (ICP-OES, PerkinElmer Optima 4300DV). The OMS-2 samples were analyzed by a VG Multilab 2000 X-ray photoelectron spectrometer using Mg Kα radiation. The spectra generated by X-ray photoelectron spectrometry (XPS) of the OMS-2 nanorod samples were calibrated by referencing the binding energy to adventitious carbon (C1s, 284.6 eV) and were deconvoluted using special software. CO temperature-programmed reduction (CO-TPR) and O2 temperature-programmed oxidation (O2-TPO) were conducted on a TP-5080 multifunctional adsorption apparatus equipped with a TCD detector. To start, 0.0100 g of the OMS-2 sample was located in a quartz reactor. A NaOH particle trap was placed in the front of the detector to adsorb the water and CO2 produced. Before the TPR analysis, the OMS-2 samples were pretreated in 5 vol % O2/He at 200 °C for 1 h. TPR was performed by heating the pretreated OMS-2 sample (0.0100 g) at a rate of 10 °C min−1 to 700 °C in a flow of 5 vol % CO/He at a rate of 40 mL min−1. Before the TPO analysis, the OMS-2 samples were reduced in 5 vol % CO/He at 220 °C for 1 h. TPO was performed by heating the OMS-2 sample at a rate of

■

RESULTS AND DISSCUSSION Characterization. OMS-2 nanorod samples with different potassium concentrations were prepared through a facile hydrothermal redox reaction of MnSO4·H2O, (NH4)2S2O8, and (NH4)2SO4 at 120 °C by adding KNO3 with different KNO3/MnSO4 molar ratios of 0, 0.5, 2, and 4. The obtained samples are denoted KMn0, KMn0.5, KMn2, and KMn4, respectively. XRD analysis (Figure 1) reveals that all the assynthesized samples are indexed to the pure tetragonal cryptomelane structures of OMS-2 (KMn8O16) (JCPDS-2913731

dx.doi.org/10.1021/es403910s | Environ. Sci. Technol. 2013, 47, 13730−13736

Environmental Science & Technology

Article

V /V0 = exp{−1/(E0β)2 [RT ln(P0/P)2 ]}

where V is the volume adsorbed, V0 the micropore volume, E0 the characteristic energy dependent on the pore structure, and β the affinity coefficient that is characteristic of the adsorber. As shown in Table 1, the OMS-2 nanorod samples exhibit low DR method micropore volumes because of the diffusion limitations of N2 molecules with a kinetic diameter of 3.60 Å in accessing the narrow channel of OMS-2 with K+ or H3O+ cations inside the tunnel (4.6 Å × 4.6 Å).26 TEM images reveal that all of the KMn0, KMn0.5, and KMn2 samples are characterized by the morphology of nanorods (Figure 2). High-resolution TEM (HRTEM) images indicate that all of the K+-doped OMS-2 samples exhibit singlecrystal structure with an exposed {200} facet.

Figure 1. XRD patterns of OMS-2 nanorod samples.

1020). The average crystal sizes of KMn0, KMn0.5, KMn2, and KMn4 determined by using the Scherrer formula (L = 0.89λ/β cos θ) at 2θ = 18.1° (corresponding to the {200} plane) are 18.2, 12.9, 17.2, and 17.7 nm, respectively. The bulk contents of potassium and manganese of the samples were determined by ICP (Table 1). For the KMn0.5 sample prepared with the Table 1. Surface Areas and Atomic Ratios of the OMS-2 Nanorod Samples K/Mn atomic ratio

sample

initial K/Mn atomic ratio

ICP

XPS

KMn0 KMn0.5 KMn2 KMn4

0 0.5 2 4

0 0.03 0.07 0.08

0 0.04 0.08 −

surface Mn3−/Mn4− area atomic ratio (m2 g−1) 0.05 0.09 0.16 −

65.1 89.2 109.7 −

Vmicropore (cm3 g−1) 0.025 0.031 0.040 −

initial KNO3/MnSO4 reactant molar ratio of 0.5, the K/Mn molar ratio of the obtained KMn0.5 sample measured by ICP is 0.03; increasing the initial KNO3/MnSO4 reactant molar ratio leads to an increase in the potassium content of OMS-2 nanorods. When the initial KNO3/MnSO4 reactant molar ratio increases from 0.5 to 2, the K/Mn molar ratio increases from 0.03 to 0.07 (KMn2). However, further increasing the initial KNO3/MnSO4 reactant molar ratio to 4 does not lead to an obvious increase in the K/Mn atomic ratio. The K/Mn atomic ratio (0.08) of KMn4 is very close to that of KMn2 (0.07). Therefore, we focus on only the KMn0, KMn0.5, and KMn2 samples below. The lattice parameters of the OMS-2 samples with tetragonal cryptomelane structure are calculated by the refinement of their XRD patterns. The lattice parameters of KMn0 without K+ inside the channel of OMS-2 are as follows: a = b = 9.8152 Å, c = 2.8469 Å, and α = β = γ = 90° [which are the almost same as those in the literature (JCPDS-29-1020; a = b = 9.815 Å, c = 2.847 Å, and α = β = γ = 90°)]. After the addition of K+ to the channel of OMS-2, the XRD peak positions of KMn0.5 and KMn2 remain unchanged as compared to that of KMn0 as shown in Figure 1. This observation suggests that the addition of K+ cannot change their lattice parameters. N2 adsorption or desorption reveals that the BET surface areas of KMn0, KMn0.5, and KMn2 are 65.1, 89.2, and 109.7 m2 g−1, respectively (Table 1 and Figure S1 of the Supporting Information). The micropore volumes of the samples are obtained using the Dubinin−Radushkevich equation:16,26

Figure 2. TEM and high-resolution TEM images of KMn0 (a and b), KMn0.5 (c and d), and KMn2 (e and f).

XPS full spectra indicate that only manganese and oxygen are in the KMn0 sample and manganese, oxygen, and potassium are in the KMn0.5 and KMn2 samples (Figure S2 of the Supporting Information). The surface K/Mn atomic ratios of KMn0, KMn0.5, and KMn2 obtained by XPS are 0, 0.04, and 0.08, respectively (Figure 3A and Table 1), which are very close to their corresponding bulk K/Mn atomic ratios measured by ICP, indicating that potassium is well dispersed in the K+doped OMS-2 nanorods. The oxidation state of Mn and the relative levels of the Mn3+/Mn4+ atomic ratio are analyzed by 13732

dx.doi.org/10.1021/es403910s | Environ. Sci. Technol. 2013, 47, 13730−13736

Environmental Science & Technology

Article

Figure 4. Calculated super cell of OMS-2 (K0Mn32O64) without K+ (a), OMS-2 (K2Mn32O64) with 2 atom % K+ (b), and OMS-2 (K4Mn32O64) with 4 atom % K+ (c): purple for K, gray for Mn, red for O, and blue for the oxygen to be removed. Figure 3. K 2p and Mn 2p3/2 XPS spectra of the OMS-2 nanorod samples.

adjacent to the K+ is 2.41 eV, which is lower than that of the OMS-2 super cell without K+. Further increasing the concentration of K+ from 2 to 4 atom % leads to a decrease in EVO to 2.32 eV (Figure 4c). This result suggests that the presence of K+ located in the tunnels of OMS-2 can improve its lattice oxygen activity and the higher the concentration of K+, the lower the removal energy of lattice oxygen. CO-TPR and O2-TPO. The theoretical calculation result is experimentally confirmed by CO temperature-programmed reduction (CO-TPR). Figure 5A illustrates the CO-TPR

studying the Mn 2p3/2 spectra.13,27,28 The Mn 2p3/2 spectra of the OMS-2 nanorod samples are decomposed into two peaks around 641.8 and 642.5 eV (Figure 3B),27,28 which are attributed to Mn3+ and Mn4+, respectively. No Mn2+ species are detected by XPS. As shown in Table 1, the higher the K/ Mn atomic ratio, the higher the Mn3+/Mn4+ atomic ratio. KMn0 without K+ possesses the lowest Mn3+/Mn4+ atomic ratio (0.05). With an increase in the initial reactant KNO3/ MnSO4 molar ratio from 0 to 0.5, the Mn3+/Mn4+ atomic ratio of the OMS-2 nanorod sample increases from 0.05 to 0.09 (KMn0.5). Further increasing the initial KNO3/MnSO4 molar ratio to 2 leads to an increase in the Mn3+/Mn4+ atomic ratio to 0.16 (KMn2). DFT Calculation. It is commonly accepted that the catalytic oxidation on OMS-2 proceeds via the Mars-van Krevelen mechanism.12−14 Organic molecules adsorbed on the catalyst surface are oxidized by surface lattice oxygen, and the resultant oxygen vacancies are subsequently replenished by gas-phase O2. According to this mechanism, the fundamental and significant thing for understanding and improving the catalytic activity of OMS-2 is to reveal the impact factors of its lattice oxygen reactivity. Thus, the effect of K+ located at the tunnel on the lattice oxygen activity of OMS-2 is theoretically studied by calculating the energy of removal of lattice oxygen (oxygen vacancy formation energy) in the absence or presence of different K+ concentrations using DFT calculations. The oxygen vacancy formation energy in a calculated super cell (Figure 4) can be defined as16 E VO = Edef − E bulk + 1/2EO2

Figure 5. CO-TPR (A) and O2-TPO (B) profiles of OMS-2 nanorods.

where Edef is the system energy with the loss of one oxygen atom (O), Ebulk the energy of a slab without the loss of an oxygen atom, and EO2 the energy of an O2 molecule in the gas phase. For the OMS-2 super cell without K+ (Figure 4a), the removal energy of one lattice oxygen atom (EVO) is 2.61 eV. Interestingly, when the OMS-2 super cell contains 2 atom % K (Figure 4b), the removal energy of one lattice oxygen atom

profiles of the as-prepared OMS-2 nanorod samples. For the KMn0 sample without doping K+, maximal CO consumption occurs around 287 and 551 °C. These major peaks indicate a two-step reduction for OMS-2: KxMn8O16 to Mn3O4 and Mn3O4 to MnO (in agreement with the literature).11,29 Interestingly, compared to the corresponding TPR peaks of 13733

dx.doi.org/10.1021/es403910s | Environ. Sci. Technol. 2013, 47, 13730−13736

Environmental Science & Technology

Article

the KMn0 sample, the incorporation of K+ into the tunnel of OMS-2 nanorods leads to a significant shift of the TPR peaks to lower temperatures. When the initial KNO3/MnSO4 molar ratio increases from 0 to 0.5 (KMn0.5), the lowest-temperature peak (TL) and the highest-temperature peak (TH) decrease from 287 and 551 °C to 245 and 370 °C, respectively. This result indicates that the incorporation of K+ into the tunnel of OMS-2 nanorods significantly improves the lattice oxygen activity. Further increasing the initial KNO3/MnSO4 molar ratio to 2 (KMn2) leads to a decrease in TL from 245 to 212 °C while TH remains almost unchanged. This result indicates that increasing the K+ concentration of the K+-doped OMS-2 nanorods results in the improvement of the lattice oxygen activity. In addition, two lower-temperature TPR peaks below 300 °C are observed for KMn0.5 and KMn2 samples, suggesting the removal of two kinds of lattice oxygen in KMn0.5 and KMn2 by CO. We quantitatively measure the amounts of CO consumed for the OMS-2 samples by calculating the area of their TPR profiles and the profile for a known amount of CO. The total amounts of CO consumed for KMn0, KMn0.5, and KMn2 are 9.53, 9.83, and 10.32 mmol g−1, respectively. The result indicates that the three samples consume a similar amount of CO, which is close to their amount of theoretical CO consumption (11.50 mmol g−1) corresponding to the reduction of KxMn8O16 to MnO. The amounts of CO consumed corresponding to the first-step reduction of OMS-2 to Mn3O4 for KMn0, KMn0.5, and KMn2, which is calculated by fitting their TPR profiles, are 5.46, 7.17, and 7.04 mmol g−1, respectively. O2 temperature-programmed oxidation (O2-TPO) of the OMS-2 nanorod samples prereduced at 220 °C for 1 h in the flow of 5 vol % CO/He was performed (Figure 5B). For the prereduced KMn0 sample, it starts to be reoxidized at ∼120 °C, and maximal O2 consumption occurs around 284 and 421 °C. According to the assignment of the CO-TPR peaks as discussed above, the lower- and higher-temperature peaks are assigned to two-step reoxidation for the OMS-2 nanorod samples prereduced by CO: Mn3O4 to KxMn8O16 and MnO to Mn3O4, respectively. The negative peak at ∼560 °C is attributed to the evolution of the lattice oxygen of the OMS2 nanorod samples and the transformation of cryptomelane to Mn2O3.13,16 Compared to the prereduced KMn0 sample without doping K+, increasing the concentration of K+ in the OMS-2 nanorod samples leads to a downshift of the reoxidation peak to the lower temperature. The prereduced KMn2 sample with the highest K+ concentration exhibits the lowest lower-temperature peak (176 °C), follwed by KMn0.5 (246 °C). We also quantitatively measure the amounts of O2 consumed for the prereduced OMS-2 samples by calculating the area of their TPO profiles and the profile for a known amount of O2. The total amounts of O2 consumed for the prereduced samples of KMn0, KMn0.5, and KMn2 are 1.33, 2.21, and 3.41 mmol g−1, respectively. The maximal reoxidation peak temperature of the KMn0, KMn0.5, and KMn2 samples prereduced by CO is lower than the corresponding maximal reduction peak temperature of KMn0, KMn0.5, and KMn2 samples prereduced by CO (Figure 5). These results reveal that the reoxidation of the OMS-2 nanorod samples prereduced by CO is faster than the reduction of the OMS-2 nanorod samples, suggesting that the reducibility or activity of lattice oxygen of OMS-2 plays a decisive role in its catalytic activity according to the Mars-van Krevelen mechanism.

Catalytic Performance. The effect of the K+ concentration on the catalytic activity of the OMS-2 nanorod samples is investigated by comparing the T50 and T90 reaction temperatures (corresponding to 50 and 90% benzene conversion, respectively). Figure 6A exhibits the benzene conversion over

Figure 6. Benzene conversion vs reaction temperature (A) and reaction time over KMn2 for the oxidation of benzene at 280 °C (B) at a benzene concentration of 2000 mg m−3 and an SV of 48000 mL (g of catalyst)−1 h−1.

the catalysts as a function of reaction temperature at a benzene concentration of 2000 mg m−3 and an SV of 48000 mL (g of catalyst)−1 h−1. As shown in Figure 6A and Table 2, the KMn0 Table 2. T50 and T90 Values and Specific Rates of Benzene Oxidation over the Catalysts catalytic activity (°C)

sample

T50

T90

KMn0 KMn0.5 KMn2

289 226 203

− 258 240

specific rate at 200 °C (μmol of catalyst m−2 rate at 200 °C min−1) (μmol min−1) 0.045 0.229 0.456

0.014 0.051 0.083

turnover frequency at 200 °C (min−1) 0.00076 0.0032 0.0070

sample without doping K+ exhibits the lowest catalytic activity. Its T50 is 289 °C, while its benzene conversion is only 72.7% even at a reaction temperature of 400 °C. Increasing the K+ concentration in the tunnel of OMS-2 nanorod samples leads to significant decreases in T50 and T90. When the K/Mn ratio in the OMS-2 nanorod samples increases from 0 to 0.03 (KMn0.5), T50 and T90 decrease to 226 and 258 °C, respectively. KMn2 with the highest K/Mn ratio (0.07) exhibits the highest catalytic activity; its T50 and T90 decrease to 203 and 240 °C, respectively. These results show that an enormous decrease (ΔT50 = 86 °C; ΔT90 > 160 °C) in the T50 and T90 reaction temperatures is achieved by the incorporation of different K+ concentrations into the tunnel of the OMS-2 nanorods. For comparison, we also tested the catalytic activity 13734

dx.doi.org/10.1021/es403910s | Environ. Sci. Technol. 2013, 47, 13730−13736

Environmental Science & Technology

Article

oxidation.14 (4) Via the strategy of increasing the oxygen vacancy defect concentration of OMS-2 nanorod samples, a maximal decrease (ΔT50 = 61 °C; ΔT90 = 102 °C) in the T50 and T90 reaction temperatures is reported in our previrous work for benzene oxidation.16 The current strategy of improving catalytic activity by the incorperation of K+ into the tunnel of OMS-2 nanorod samples is one of the most efficient approaches, evidenced by an enormous decrease (ΔT50 = 86 °C; ΔT90 > 160 °C) in the T50 and T90 reaction temperatures, and a 6.0-fold enhancement in the specific catalytic activity for benzene oxidation. In summary, OMS-2 nanorods with a tunable K + concentration were prepared by a facile hydrothermal redox reaction of MnSO4·H2O, (NH4)2S2O8, and (NH4)2SO4 at 120 °C by adding KNO3 at different KNO3/MnSO4 molar ratios. We demonstrate an enormous effect of K+ concentration on the catalytic activity of OMS-2. Increasing the K+ concentration leads to a considerable enhancement of the lattice oxygen activity in the K+-doped OMS-2 nanorod catalysts, thus significantly increasing the catalytic activity. Via the strategy of tuning the K+ concentration, we successfully prepared a lowcost, environmentally benign, and highly efficient oxidation catalyst that is comparable to the expensive noble metal catalyst. These results should aid in the rational design and facile preparation of highly efficient oxidation catalysts of OMS2 by the incorporation of other different cation concentrations into the tunnel of OMS-2, which are widely used in both the partial oxidation of petrochemical materials and the complete oxidation of organic compounds in environmental purification and fuel cells.

of a commercial supported noble metal catalyst (0.5% Pt/ Al2O3, 320.4 m2 g−1). Both T50 (204 °C) and T90 (260 °C) of 0.5% Pt/Al2O3 are higher than the corresponding values of KMn2. By tuning the K+ concentration in the OMS-2 strategy, we successfully prepared a low-cost, environmentally benign, and highly efficient oxidation catalyst, which is comparable to expensive supported noble metal catalysts. The long-term catalytic stability of KMn2 was evaluated by using the benzene oxidation reaction test at 280 °C (Figure 6B). At the initial 2 h, the level of benzene conversion on OMS-70 was 95.0%. After 60 h, the level of benzene conversion remained as high as 92.8%, suggesting that KMn2 with the highest K+ concentration exhibits good catalytic stability. To clarify whether the significant enhancement of the catalytic activity of OMS-2 nanorod samples by the incorporation of K+ mainly arises from the different surface areas (Table 1), their specific benzene reaction rates (per unit surface area of catalyst) at 200 °C, which represents the intrinsic catalytic efficiency of the OMS-2 nanorod samples, are compared (Table 2). For the KMn0 sample without doping K+, its specific benzene reaction rate is 0.014 μmol of catalyst m−2 min−1. Compared to KMn0, when the K/Mn ratio (bulk phase) in the OMS-2 nanorod sample increases from 0 to 0.03 (KMn0.5), its specific benzene reaction rate increases to 0.051 μmol of catalyst m−2 min−1. The KMn2 sample with the highest K+ concentration exhibits the highest specific benzene reaction rate (0.083 μmol of catalyst m−2 min−1), which is 1.6 and 6.0 times higher than those of KMn0.5 and KMn0, respectively. The turnover frequencies (TOFs) of the catalysts at 200 °C, defined as the number of benzene molecules reacting per active site in unit time, are also compared. As the benzene oxidation on OMS-2 proceeds via the Mars-van Krevelen mechanism, the surface lattice oxygen of OMS-2 acts as an active site. OMS-2 experiences a two-step reduction for OMS-2: OMS-2 to Mn3O4 and Mn3O4 to MnO as discussed for CO-TPR. The secondstep reduction takes place at a temperature higher than the reaction temperature for the catalytic oxidation of benzene, so the lattice oxygen of OMS-2 corresponding to the first-step reduction of OMS-2 to Mn3O4 is used as the active sites for calculating TOFs. As shown in Table 2, the KMn2 sample with the highest K+ concentration exhibits the highest TOF (0.0070 min−1), which is 2.2 and 9.2 times higher than those of KMn0.5 and KMn0, respectively. In the past several decades, several strategies have been reported for improving the catalytic activity of OMS-2. (1) Via the strategy of decreasing the particle size or increasing the surface area, the level of conversion increases from 21 to 52% for selective oxidation of benzyl alcohol, from 63 to 95% for the oxidation of anisyl alcohol to anisyl aldehyde, and from 10 to 31% for the oxidation of furfuryl alcohol.9,30,31 However, the strategy of decreasing the particle size or increasing the surface area did not result in a considerable evolution of the intrinsic property (specific reaction rate: per unit surface area of catalyst) of OMS-2. (2) Via the strategy of controlling the morphology of OMS-2, a maximal decrease (ΔT50 ∼ 13 °C; ΔT90 ∼ 13 °C) in the T50 and T90 reaction temperatures is reported for toluene oxidation.27 (3) Via the strategy of ion exchange of K+ in the channel of OMS-2 with Li+ and Cs+, a maximal decrease (ΔT50 ∼ 17 °C; ΔT90 ∼ 27 °C) in the T50 and T90 reaction temperatures is reported for the total oxidation of ethyl acetate.5 (4) Via the strategy of multiple substitutions of Mo, V, Cu, and Fe ions into the OMS-2 structure, the level of conversion increases from 35 to 96% for diphenylmethanol

■

ASSOCIATED CONTENT

S Supporting Information *

N2 adsorption−desorption isotherm and XPS full spectra of the OMS-2 nanorod samples. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Telephone: +86-027-87651856. Fax: +86-027-87883743. Email: [email protected] or [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (21273169), the National Basic Research Program of China (2009CB939704), the Innovative Research Team Project of Hubei Province (2010CDA070), and the Fundamental Research Funds for the Central Universities. We are grateful to Prof. Chunxu Pan for his help in the DFT calculation.

■

REFERENCES

(1) Huang, H. J.; Li, D. Z.; Lin, Q.; Zhang, W. J.; Shao, Y.; Chen, Y. B.; Sun, M.; Fu, X. Z. Efficient degradation of benzene over LaVO4/ TiO2 nanocrystalline heterojunction photocatalyst under visible light irradiation. Environ. Sci. Technol. 2009, 43, 4164−4168. (2) Liotta, L. F. Catalytic oxidation of volatile organic compounds on supported noble metals. Appl. Catal., B 2010, 100, 403−412. (3) Suib, S. L. Porous manganese oxide octahedral molecular sieves and octahedral layered materials. Acc. Chem. Res. 2008, 41, 479−487. 13735

dx.doi.org/10.1021/es403910s | Environ. Sci. Technol. 2013, 47, 13730−13736

Environmental Science & Technology

Article

(22) Blochl, P. E. Projector augmented-wave method. Phys. Rev. B 1994, 50, 17953−17979. (23) Kresse, G.; Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave method. Phys. Rev. B 1999, 59, 1758−1775. (24) Kresse, G.; Hafner, J. Ab initio molecular dynamics for openshell transition metals. Phys. Rev. B 1993, 48, 13115−13118. (25) Cockayne, E.; Li, L. First-principles DFT + U studies of the atomic, electronic, and magnetic structure of α-MnO2(cryptomelane). Chem. Phys. Lett. 2012, 544, 53−58. (26) Fairen-Jimenez, D.; Moggach, S. A.; Wharmby, M. T.; Wright, P. A.; Parsons, S.; Duren, T. Opening the gate: Framework flexibility in ZIF-8 explored by experiments and simulations. J. Am. Chem. Soc. 2011, 133, 8900−8902. (27) Wang, F.; Dai, H. X.; Deng, J. G.; Bai, G. M.; Ji, K. M.; Liu, Y. X. Manganese oxides with rod-, wire-, tube-, and flower-like morphologies: Highly effective catalysts for the removal of toluene. Environ. Sci. Technol. 2012, 46, 4034−4041. (28) Wei, Y. J.; Yan, L. Y.; Wang, C. Z.; Xu, X. G.; Wu, F.; Chen, G. Effects of Ni doping on [MnO6] octahedron in LiMn2O4. J. Phys. Chem. B 2004, 108, 18547−8551. (29) Xu, R.; Wang, X.; Wang, D.; Zhou, K.; Li, Y. Surface structure effects in nanocrystal MnO2 and Ag/MnO2 catalytic oxidation of CO. J. Catal. 2006, 237, 426−430. (30) Nyutu, E. K.; Chen, C.-H.; Sithambaram, S.; Crisostomo, V. M. B.; Suib, S. L. Systematic control of particle size in rapid open-vessel microwave synthesis of K-OMS-2 nanofibers. J. Phys. Chem. C 2008, 112, 6786−6793. (31) Ding, Y. S.; Shen, X. F.; Sithambaram, S.; Gomez, S.; Kumar, R.; Crisostomo, V. M. B.; Suib, S. L.; Aindow, M. Synthesis and catalytic activity of cryptomelane-type manganese dioxide nanomaterials produced by a novel solvent-free method. Chem. Mater. 2005, 17, 5382−5389.

(4) Wang, R. H.; Li, J. H. Effects of precursor and sulfation on OMS2 catalyst for oxidation of ethanol and acetaldehyde at low temperatures. Environ. Sci. Technol. 2010, 44, 4282−4287. (5) Santos, V. P.; Soares, O. S. G. P.; Bakker, J. J. W.; Pereira, M. F. R.; Orfao, J. J. M.; Gascon, J.; Kapteijn, F.; Figueiredo, J. L. Structural and chemical disorder of cryptomelane promoted by alkali doping: Influence on catalytic properties. J. Catal. 2012, 293, 165−174. (6) Hernández, W. Y.; Centeno, M. A.; Ivanova, S.; Eloy, P.; Gaigneaux, E. M.; Odriozola, J. A. Cu-modified cryptomelane oxide as active catalyst for CO oxidation reactions. Appl. Catal., B 2012, 123− 124, 27−35. (7) Chen, X.; Shen, Y. F.; Suib, S. L.; O’Young, C. L. Catalytic decomposition of 2-propanol over different metal-cation-doped OMS2 materials. J. Catal. 2001, 197, 292−302. (8) Abecassis-Wolfovich, M.; Jothiramalingam, R.; Landau, M. V.; Herskowitz, M.; Viswanathan, B.; Varadarajan, T. K. Cerium incorporated ordered manganese oxide OMS-2 materials: Improved catalysts for wet oxidation of phenol compounds. Appl. Catal., B 2005, 59, 91−98. (9) Dharmarathna, S.; Kingondu, C. K.; Pedrick, W.; Pahalagedara, L.; Suib, S. L. Direct sonochemical synthesis of manganese octahedral molecular sieve (OMS-2) nanomaterials using cosolvent systems, their characterization, and catalytic applications. Chem. Mater. 2012, 24, 705−712. (10) Tian, H.; He, J. H.; Zhang, X. D.; Zhou, L.; Wang, D. H. Facile synthesis of porous manganese oxide K-OMS-2 materials and their catalytic activity for formaldehyde oxidation. Microporous Mesoporous Mater. 2011, 138, 118−122. (11) Genuino, H. C.; Dharmarathna, S.; Njagi, E. C.; Mei, M. C.; Suib, S. L. Gas-phase total oxidation of benzene, toluene, ethylbenzene, and xylenes using shape-selective manganese oxide and copper manganese oxide catalysts. J. Phys. Chem. C 2012, 116, 12066− 12078. (12) Luo, J.; Zhang, Q.; Garcia-Martinez, J.; Suib, S. L. Adsorptive and Acidic Properties, Reversible Lattice Oxygen Evolution, and Catalytic Mechanism of Cryptomelane-Type Manganese Oxides as Oxidation Catalysts. J. Am. Chem. Soc. 2008, 130, 3198−3207. (13) Santos, V. P.; Pereira, M. F. R.; Orfao, J. J. M.; Figueiredo, J. L. The role of lattice oxygen on the activity of manganese oxides towards the oxidation of volatile organic compounds. Appl. Catal., B 2010, 99, 353−363. (14) Kingondu, C. K.; Opembe, N.; Chen, C. H.; Ngala, K.; Huang, H.; Iyer, A.; Garces, H. F.; Suib, S. L. Manganese oxide octahedral molecular sieves (OMS-2) multiple framework substitutions: A new route to OMS-2 particle size and morphology control. Adv. Funct. Mater. 2011, 21, 312−323. (15) Peluso, M. A.; Gambaro, L. A.; Pronsato, E.; Gazzoli, D.; Thomas, H. J.; Sambeth, J. E. Synthesis and catalytic activity of manganese dioxide (type OMS-2) for the abatement of oxygenated VOCs. Catal. Today 2008, 133−135, 487−492. (16) Hou, J. T.; Li, Y. Z.; Liu, L. L.; Ren, L.; Zhao, X. J. Effect of giant oxygen vacancy defects on the catalytic oxidation of OMS-2 nanorods. J. Mater. Chem. A 2013, 1, 6736−6741. (17) Liu, J.; Makwana, V.; Cai, J.; Suib, S. L.; Aindow, M. Effects of Alkali Metal and Ammonium Cation Templates on Nanofibrous Cryptomelane-type Manganese Oxide Octahedral Molecular Sieves (OMS-2). J. Phys. Chem. B 2003, 107, 9185−9194. (18) Luo, J.; Zhu, H. T.; Liang, J. K.; Rao, G. H.; Li, J. B.; Du, Z. M. Tuning magnetic properties of α-MnO2 nanotubes by K+ doping. J. Phys. Chem. C 2010, 114, 8782−8786. (19) Kresse, G.; Hafner, J. Ab initio molecular dynamics for liquid metals. Phys. Rev. B 1993, 47, 558−561. (20) Kresse, G.; Hafner, J. Ab initio molecular-dynamics simulation of the liquid-metal-amorphous-semiconductor transition in germanium. Phys. Rev. B 1994, 49, 14251−14269. (21) Kresse, G.; Furthmuller, J. Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set. Comput. Mater. Sci. 1996, 6, 15−50. 13736

dx.doi.org/10.1021/es403910s | Environ. Sci. Technol. 2013, 47, 13730−13736