- Email: [email protected]
TiO2–Al2O3 catalysts
Sci. Bull. DOI 10.1007/s11434-015-0782-3
www.scibull.com www.springer.com/scp
Article
Chemistry
Selective oxidation of methanol to dimethoxymethane over V2O5/TiO2–Al2O3 catalysts Tuo Wang • Yali Meng • Liang Zeng Jinlong Gong
•
Received: 7 February 2015 / Accepted: 6 March 2015 Ó Science China Press and Springer-Verlag Berlin Heidelberg 2015
Abstract This paper describes the effect of the preparation method of binary oxide supports (TiO2–Al2O3) on catalytic performance of V2O5/TiO2–Al2O3 catalysts for methanol selective oxidation to dimethoxymethane (DMM). The TiO2–Al2O3 supports are synthesized by a number of methods including mechanical mixing, ball milling, precipitation, co-precipitation, and sol–gel method, which is followed by incipient wetness impregnation to produce V2O5/TiO2–Al2O3 catalysts. Among these samples, the V2O5/TiO2–Al2O3 catalyst prepared by the sol–gel method has the best catalytic performance with a maximum methanol conversion of 48.9 % and a high DMM selectivity of 89.9 % at 393 K, showing superior performance than V2O5/TiO2 and V2O5/Al2O3. The excellent catalytic performance of V2O5/TiO2–Al2O3 is attributed to the effective interaction between the active component and the mixed support. Such interaction changes the chemical states of supported active V components, produces an increased amount of V4? species, and facilitates the electron transfer between support and active component. Additionally, the incorporation of titanium cation into the alumina structure could also help produce an appropriate amount of acidic sites, which increases the DMM selectivity. The coordinated environment of the dispersed vanadia on TiO2–Al2O3 mixed support improves the catalytic efficiency on methanol oxidation to DMM. T. Wang Y. Meng L. Zeng J. Gong Key Laboratory for Green Chemical Technology of Ministry of Education, School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China T. Wang Y. Meng L. Zeng J. Gong (&) Collaborative Innovation Center of Chemical Science and Engineering, Tianjin 300072, China e-mail: [email protected]
Keywords Methanol selective oxidation Dimethoxymethane V2O5/TiO2–Al2O3 Binary oxide support
1 Introduction Dimethoxymethane (DMM) has successfully attracted increasing attention as a downstream chemical product of methanol in industrial practice. DMM is a useful intermediate in organic synthesis and an excellent solvent in perfume and pharmaceutical industries. It can also act as an environmentally friendly fuel with high efficiency and low toxicity [1]. Moreover, it has been used as a starting monomer in the synthesis of polyoxymethylene dimethylether (POMM) [2]. The traditional method for DMM synthesis includes two consecutive steps: gas-phase methanol partial oxidation to formaldehyde (FA) and the subsequent acetalization of methanol and formaldehyde over acid catalysts [3]. However, this reaction scheme suffers from high reaction temperatures, complicated procedures, and severe equipment corrosion [2]. In addition to the target product DMM in methanol oxidation, several byproducts could form, including methyl formate (MF), dimethyl ether (DME), FA, and oxycarbide. An alternative synthetic pathway by one-step selective oxidation of methanol to DMM under mild reaction conditions becomes more desirable with both environmental and economic advantages. To obtain a high DMM yield in one step, the design of the catalyst is followed by the principle of appropriately balanced redox ability and acidity [4]. Excessive redox sites of the catalyst could lead to the abundant amount of FA, MF, and oxycarbide, while superfluous acidic sites result in a large amount of DME.
123
Sci. Bull.
Several catalytic systems have been reported to be active for the methanol oxidation to DMM by carefully coupling the redox sites with suitable supports. For example, Re oxide has been considered as a potential active site for methanol oxidation, and the supported Re oxide like Re/c-Fe2O3 catalyst has a DMM selectivity of 91 % and a methanol conversion of 48.4 % at 513 K [5]. But a bulk ReO3 catalyst without support has a rather low methanol conversion, only *12.4 % [5]. The V2O5/TiO2based catalysts by acidic modification have shown superior performance. For instance, V2O5/TiO2/SO42- achieved a 47 % methanol conversion at 393 K, which is much better than the conventional V2O5/TiO2 catalyst with 17 % methanol conversion [4, 6, 7]. The bulk H5PV2Mo10O40 only showed a 41.1 % methanol conversion with a 40.2 % DMM selectivity at 453–513 K, while the supported heteropoly acid catalysts exhibited 60.3 %–68.4 % methanol conversion with 67 %–51 % DMM selectivity. This is benefitted from the Keggin structures of H3?nPVnMo12-nO40 providing both redox and Bronsted acid sites required for producing DMM [8]. Additionally, several other catalytic systems have been reported to be active for the methanol oxidation to DMM, including modified V/TS-1 catalysts [9], amorphous multi-metal oxides [10, 11], Fe2(MoO4)3–MoO3 catalysts [2, 12]. Among these catalysts, the supported V2O5 catalysts have shown excellent catalysis performance in the selective oxidation of methanol to DMM, owing to the outstanding redox properties of V2O5. The catalytic activity and selectivity of the catalyst would be improved by the use of an appropriate support. Due to its high specific surface area, weak acidity, and good thermal stability, c-Al2O3 has been widely used as a support for a number of catalysts. Recently, some interesting results have been obtained for TiO2 as a promoter on high surface area supports. The chemical states of supported metal or metal oxides, adsorption properties of the catalyst and their catalytic activity are influenced by the incorporation of TiO2 with the support. Particularly, TiO2– Al2O3 support has drawn special attention owing to its versatile catalytic properties for various commercially important reactions such as isomerization of alkenes, dehydration of alcohols, and hydrodesulphurization [13–15]. Particularly, the TiO2–Al2O3 supported catalysts show better performance than single Al2O3 supported catalyst in the hydrodesulfurization (HDS) reaction [14]. It is also well established that the supported vanadium oxides are effective catalysts for partial oxidation of hydrocarbons and alcohols [13, 16, 17]. And the structure of the dispersed vanadium species is closely related to the nature of the specific oxide support, the loading amount, and the preparation procedure [18].
123
Considering versatile catalytic properties of TiO2–Al2O3 support and good redox properties of V2O5, we have developed V2O5/TiO2–Al2O3 catalysts for methanol oxidation to DMM. The aim of this study is to explore the influence of preparation method of mixed TiO2–Al2O3 support on the catalytic activity of V2O5. The V2O5/TiO2– Al2O3 catalysts prepared by five different methods have been employed in selective methanol oxidation to DMM, including mechanical mixing, ball milling, precipitation, co-precipitation, and sol–gel methods. The materials are characterized with the N2 adsorption/desorption, X-ray diffraction (XRD), temperature-programmed reduction (H2-TPR), Raman spectroscopy, temperature-programmed desorption of ammonia (NH3-TPD), and X-ray photoelectron spectroscopy (XPS) techniques. A set of activity tests has been conducted in a fixed-bed micro-reactor for collecting and comparing their kinetic data. The relationship between catalytic activities and the behaviors of both redox and acidic sites of V2O5/TiO2–Al2O3 catalysts is analyzed and discussed.
2 Experimental 2.1 Catalyst preparation A series of 15V2O5/TiO2–Al2O3 catalysts were prepared by a wetness impregnation method with the TiO2–Al2O3 complex support. The V2O5 content is all fixed at 15 % for well dispersion and effective comparison reason, as the monolayer coverage of V2O5/TiO2 and V2O5/Al2O3 catalysts is about 12.5 % and 20 %, respectively [7, 19]. The TiO2–Al2O3 complex support were prepared by five different methods—mechanical mixing, ball milling, precipitation, sol–gel, and co-precipitation method—with the uniform constitute of Al2O3 and TiO2 (Al2O3:TiO2 = 2:1 by weight). A c-Al2O3 powder support and an industrial TiO2 (P25) (80 % anatase and 20 % rutile) powder were calcined in air at 573 K for 2 h prior to the catalyst preparation. For the complex support with the mechanical mixing method (denoted as TiO2–Al2O3-mix), the required amount of Al2O3 and P25 were mechanically mixed with each other until smooth powder was formed. For the support with the ball-milling method (denoted as TiO2–Al2O3mil), the required amount of Al2O3 and P25 were spread in the high energy ball-milling pot and milled for 12 h. The complex support with the precipitation method (denoted as TiO2–Al2O3-imp) was prepared by the method of gelsupported precipitation [13]. In the case with the sol–gel method (denoted as TiO2–Al2O3-sol), it was prepared by the procedure adapted from the literature [20]. For the complex support with the co-precipitation method (denoted
Sci. Bull.
as TiO2–Al2O3-cop), the mixed oxide support was prepared by a homogeneous co-precipitation method using urea as hydrolyzing agent [18]. All the oxide supports were calcined in air at 773 K for 4 h before the impregnation of ammonium metavanadate. The V2O5/TiO2–Al2O3 catalysts were obtained by the impregnation of calcined TiO2–Al2O3 support with the mixed ammonium metavanadate and oxalic acid aqueous liquid solution. Subsequently, samples were obtained upon the evaporation of the solvent at 353 K and then dried at 393 K for 12 h and calcined at 773 K for 6 h in furnace.
2.2 Catalyst characterization N2 adsorption/desorption analysis was performed at 77 K with a Micromeritics Tristar 300 surface area and porosity analyzer. The specific surface area was calculated by the Bruauer–Emmett–Teller (BET) method. The pore parameters (pore volume and pore diameter) were evaluated from the desorption branch of isotherms based on the Barrett–Joyner–Halenda (BJH) model. XRD patterns were collected on a Rigaku D/Max-2500 diffractometer using C-filtered Cu Ka radiation ˚ ), operated at 40 kV and 200 mA, at a scan (k = 1.54056 A rate of 0.02°/s from 10° to 90°. Raman spectra were collected at ambient temperature using a Thermo Scientific DXRTM Raman Microscope with a green semiconductor laser (532 nm) as the excitation source. Samples were pretreated in a vacuum drier at 393 K for 12 h before any measurements. NH3-TPD spectra were recorded on an Autochem 2920 chemisorption analyzer. The sample was pretreated at 573 K under Ar flow (30 mL min-1) for 1 h and was then cooled down to 333 K. NH3 was subsequently introduced into the flow system for 30 min. The TPD spectra were recorded at a ramp rate of 10 K min-1 from 333 to 973 K under a He flow. H2-TPR measurements were carried out on an Autochem 2920 chemisorption instrument with a thermal conductivity detector (TCD). The sample (100 mg) was pretreated at 573 K under Ar flow (30 mL min-1) for 1 h and cooled to 333 K. The sample was then contacted with a H2–Ar mixture (10 vol% H2) at a flow rate of 30 mL min-1 and heated at a rate of 10 K min-1 from 333 to 950 K. XPS measurements were performed on a Physical Electronics PHI-1600 XPS system equipped with a hemispherical electron analyzer and an Mg Ka X-ray source (1,253.6 eV). The samples were vacuum-dried at 393 K for 4 h prior to any measurements. The energy region of the photoelectrons was scanned at a pass energy of 29.35 eV. The binding energies were referenced to the C 1s signal at 284.6 eV. The data were processed using the PHI Multipack Program (Gaussian peak fitting).
2.3 Catalyst activity measurements Catalyst activity tests were carried out in a fixed-bed micro-reactor at ambient pressure. The reactor tube was made of quartz with an inner diameter of 7 mm. Methanol was introduced into the reaction zone by bubbling a gas mixture of O2 and N2 (99.99 %) through a glass saturator filled with methanol (99.9 %) at 288 K. In each test, 0.3 g of tableted catalysts was loaded. The catalyst was pretreated with a 25 vol% O2 and 75 vol% N2 mixture gas (40 mL min-1) at 673 K for 2 h before reaction test. Upon cooling to the reaction temperature, a mixture of methanol, N2, and O2 was introduced to the catalyst bed with a flow rate of 48 mL min-1. The gas hourly space velocity (GHSV) was kept at 8,000 h-1 with a feed gas of CH3OH:O2:N2 = 1:2.5:7.5 in volume. The reaction temperature ranged from 373 to 453 K. The products were analyzed by an online SP-2100 gas chromatography (GC) using a Propack T column connected with a flame ionization detector (FID) and a TCD detector. The original products like methanol, MF, FA, DME, and DMM were analyzed on the FID line, while the carbon monoxide (CO) and carbon dioxide (CO2) were quantified on the TCD line. The gas lines were kept at 373 K to prevent condensation of the reactants and products. The carbon balance was 99.0 % ± 1 % for all GC measurements.
3 Results 3.1 Structural properties Physical properties of the catalysts with different preparation methods are presented in Table 1. The 15V2O5/TiO2– Al2O3 catalyst with the impregnation method has the largest BET surface area, which is almost the same as pure Al2O3 support (221 m2 g-1 BET surface area, 0.34 m3 g-1 pore volume, and 6.0 nm average pore diameter). The BET surface areas of the catalysts with mechanical mixing and ball-milling methods are similar, about 110 m2 g-1, whereas the average pore radius of the ball-milling catalyst is larger than the mechanical mixing catalyst, about 9.3 nm, indicating the ball milling enhances the interaction between TiO2 and Al2O3 [21]. The BET surface areas of the catalysts with sol–gel and co-precipitation methods are lower than samples with other methods. The pore volumes of sol–gel and co-precipitation catalysts decrease substantially compared with other catalysts, while the average pore diameters are much greater than other catalysts. XRD patterns of the 15V2O5/TiO2–Al2O3 catalysts with different preparation methods are shown in Fig. 1. It is apparent that vanadium oxide species are well dispersed on the mixed TiO2–Al2O3 support, except for the sample of
123
Sci. Bull. Table 1 Porous properties of V2O5/TiO2–Al2O3 catalysts with different preparation methods Sample
SBET (m2 g-1)
Pore volume (m3 g-1)
15V2O5/TiO2–Al2O3-mix
110.7
0.24
15V2O5/TiO2–Al2O3-imp
214.0
0.37
6.8
32.7
15V2O5/TiO2–Al2O3-cop
45.3
0.10
23.4
64.3
15V2O5/TiO2–Al2O3-mil
109.1
0.31
9.3
15.2
15V2O5/TiO2–Al2O3-sol
61.3
0.13
14.4
18.1
a
Average pore diameter (nm) 7.4
Crystallite size (nm)a 26.4
Calculated by the TiO2 peak of XRD data
Fig. 1 (Color online) XRD patterns of the catalysts with different preparation methods
Fig. 2 (Color online) Raman spectra of catalysts with different preparation methods
15V2O5/TiO2–Al2O3-mix. Compared with 15V2O5/TiO2– Al2O3-mix, the Al2O3 crystal phase (characteristic peaks appear at 45.8°, 67°, and 37.6° (JCPDS No. 10-425)) is not presented on the 15V2O5/TiO2–Al2O3-mil, which only possesses TiO2 phases (sharp peaks at 22°, 23°, 28°, 31°, 34°) (JCPDS No. 21-1272). And there are no sharp V2O5 phase peaks at 20° and 31° (JCPDS No. 45-1074) on the 15V2O5/TiO2–Al2O3-mil. It indicates that the ball-milling process might promote the dispersion of active species and decrease the crystallinity of the Al2O3 in the mixed support [22]. For the sol–gel and co-precipitation catalysts, there are a large amount of TiO2 crystal phases and traces of V2Ti7O17 on the surface [23], and the intensity of the lines due to TiO2 phase on the co-precipitation catalyst is sharper than that of the sol–gel method. The crystallite size of the catalysts is also calculated from the XRD data, and the results are listed in Table 1; it is apparent that the ballmilling and sol–gel catalysts have smaller crystallite sizes than other samples. Raman spectra of V2O5/TiO2–Al2O3 catalysts with different preparation methods are presented in Fig. 2 for more detailed structural information. The pure Al2O3 support exhibits almost no absorption peak, and the band at 1,003 cm-1 is assigned to V=O stretches in monovanadates [19]. The sharp intense Raman bands at 639, 513, 292, and
152 cm-1 are attributed to crystallized anatase TiO2, and the bands at 705, 413, and 223 cm-1 are corresponded to rutile TiO2 [16]. For the 15V2O5/TiO2–Al2O3-imp sample, the degree of TiO2 crystallization is the lowest, which indicates that most of the Ti species are highly dispersed on the Al2O3 surface. Compared with 15V2O5/TiO2–Al2O3mix, it is apparent that the intensity of TiO2 phase for 15V2O5/TiO2–Al2O3-mil sample is decreased significantly and the sharpest bands at 639 cm-1 gradually disappeared, implying that the ball milling promotes the non-equilibrium phase transition of TiO2 phase and formation of lattice defects [24]. For the 15V2O5/TiO2–Al2O3-sol sample, there is no apparent peak at 1,003 cm-1, illustrating that vanadium oxide species are well dispersed on the TiO2–Al2O3 support. Moreover, the bands at 152 cm-1 shift to higher wave number for about 3 cm-1, declaring that the interaction of Ti and Al species is enhanced. For the 15V2O5/ TiO2–Al2O3-cop sample, in addition to the sharper intensity of crystallized TiO2 phase, the shift of band at 152 cm-1 is more distinct; the interaction of Ti and Al species on the mixed TiO2–Al2O3-cop support is much stronger than that of samples prepared by other methods. The Raman bands at 248 cm-1 of 15V2O5/TiO2–Al2O3cop is assigned to the stretching and bending modes of V– O–Ti bridging bonds as the result of the interaction
123
Sci. Bull.
between the active component and the support [25]. The Raman analysis is consistent with the XRD results. 3.2 Acidic properties NH3-TPD profiles of the V2O5/TiO2–Al2O3 catalysts with different preparation methods are shown in Fig. 3. In the NH3-TPD profiles, there are three acidic sites, which are weak acidic sites (350–600 K), medium strong acidic sites (600–750 K), and strong acidic sites (750–900 K), respectively. The profiles show that the strength and distribution of acidic sites vary significantly with different preparation methods. The strong acidic sites could be observed distinctly on the 15V2O5/TiO2–Al2O3-mix sample, which is similar with the c-Al2O3 support profiles, indicating the invariance of the Al2O3 structure. For the catalysts prepared by the ball milling, impregnation, and sol– gel methods, there are mainly weak acidic sites on these catalysts. The induction of weaker acid sites is related to the lowered intensity of Al–O–Al bonds, which is a consequence of the incorporation of TiO2 into Al2O3 with these methods, changing the original Al2O3 structure [20]. The middle strong acidic sites of the 15V2O5/TiO2–Al2O3cop catalysts tend to increase compared with the other samples, which might be related to the stronger interaction of Ti and Al species by the formation of intensely bridged Ti–O–Al bonds as well as the higher electron acceptability by the close interaction between vanadium oxide on the TiO2–Al2O3 supports [20]. 3.3 Redox properties The chemical states of Ti 2p, Al 2p, and O 1s on the catalyst support surface significantly influence the oxidation activity of methanol to DMM, which also could reflect the degree of the mutual function between the active
Fig. 3 (Color online) NH3-TPD profiles of the catalysts with different preparation methods
component and the support [15]. From the Ti 2p, Al 2p, and O 1s XPS analysis shown in Fig. 4, it is evident that the binding energies of Ti 2p and Al 2p shift to lower values, especially on the V2O5/TiO2–Al2O3-sol catalyst, indicating there is lower positive charge density on the alumina and titanium centers, due to the mutual interaction between the active component and the support [15, 18]. Such enhanced interaction could induce more oxygen vacancies on the V2O5/TiO2–Al2O3 catalyst [24]. From binding energies results of O 1s shown in Fig. 4, the binding energy at around 530.0–530.4 eV is the characteristic of lattice oxygen of Ti–O–Ti in TiO2, while the binding energy at around 531.1–531.4 eV is the lattice oxygen of Al–O–Al in Al2O3 and 529.6–529.9 eV is the lattice oxygen of V–O or V=O in V2O5. The O 1s binding energies of V2O5/TiO2– Al2O3-sol sample shift to lower values, implying that the amount of active lattice oxygen species of the catalyst is increased, which could account for better performance in methanol oxidation [18, 26]. XPS analysis also reveals the chemical states of active V species on the support surface, which has significant influence on the redox capability of the catalyst. Binding energies and peak fitting results of V 2p3/2 obtained from XPS spectra of V2O5/TiO2–Al2O3 catalysts are shown in Fig. 5 and Table 2. From the V 2p3/2 deconvolution results, it is clearly seen that a greater amount of V4? species is presented on the V2O5/TiO2–Al2O3-mil and V2O5/TiO2– Al2O3-sol surface than other samples, indicating that the degree of reduction of V2O5 is enhanced and the formation of more oxygen vacancies with the ball-milling and sol–gel methods. The reducibility of the V2O5/TiO2–Al2O3 catalysts with different preparation methods was also characterized by H2-TPR (Fig. 6). The H2-TPR profiles for the V2O5/TiO2– Al2O3 samples are presented as a peak in the 650–850 K temperature range, which is assigned to reduction of V5?– V3? in various polyvanadates species. It is apparent that the temperature of the first peak (Tmax) for V2O5/TiO2– Al2O3-sol is the lowest, which demonstrates that its redox capability is the strongest among all the samples. Additionally, 15V2O5/TiO2–Al2O3-sol sample has an additional reduction peak in the range from 850 to 920 K, which is attributed to reduction of Ti4? to Ti3? [27]. The V2O5/ TiO2–Al2O3-mil sample also has strong redox ability that is preceded only by the V2O5/TiO2–Al2O3-sol sample, while the redox capability of V2O5/TiO2–Al2O3-mix is the poorest one. Most interestingly, for the 15V2O5/TiO2– Al2O3-cop sample, there is large peak presented at about 869 K, which is the overlapped reduction peaks of V5? to V3? and Ti4? to Ti3? in V–O–Ti structures, as a result of the strong interaction between active species and interior supports.
123
Sci. Bull.
Fig. 4 (Color online) XPS spectra of Al 2p3/2 (a), Ti 2p3/2 (b), and O 1s (c) signals of catalysts
Fig. 5 (Color online) XPS spectra and peak fitting results of V 2p3/2 signals of catalysts
3.4 Catalytic activities Catalytic activities of V2O5/TiO2–Al2O3 catalysts with different preparation methods are shown in Table 3. It is apparent that most V2O5/TiO2–Al2O3 catalysts have higher DMM yield than the 15V2O5/TiO2 and 15V2O5/Al2O3 catalysts, implying the positive effects of the mixed TiO2– Al2O3 support. Meanwhile, the activity of V2O5/TiO2– Al2O3-sol catalyst is significantly higher than other V2O5/ TiO2–Al2O3 samples, which reaches a maximum conversion of 48.9 % with a DMM selectivity of 89.9 % at 393 K. The 15V2O5/TiO2–Al2O3-mil and 15V2O5/TiO2– Al2O3-cop samples also have relatively higher methanol conversions, about 43.0 % and 42.7 %, respectively. However, the DMM selectivity of the 15V2O5/TiO2–
123
Al2O3-mil sample is 90.7 %, which is much higher than the DMM selectivity of 15V2O5/TiO2–Al2O3-cop catalyst (77.6 %). The 15V2O5/TiO2–Al2O3-mix sample has much lower methanol conversion (25.1 %) and DMM selectivity (79.7 %) than 15V2O5/TiO2–Al2O3-mil sample, which illustrates the important role of the ball-milling process. The selectivity of DME and FA is low since the reaction temperature is as low as 373 K and is unfavorable for the formation of DME and FA [7]. It is acknowledged that the support prepared via the ball-milling method and the sol– gel method could promote the dispersion of active species and decrease the average particle size of catalyst, which is beneficial for increasing the catalytic activity [28, 29]. Additionally, both methods strengthen the mutual function between TiO2 and Al2O3, increasing weaker acid sites by incorporating TiO2 into Al2O3 and improving the redox capability of vanadates by producing more oxygen vacancies [24]. As the temperature has great influence on the activation of reactants and the performance of catalysts, the reactivity of the 15V2O5/TiO2–Al2O3-sol sample is evaluated as a function of temperature, and the results are listed in Table 4. The conversion increases gradually with the rise of reaction temperature from 373 to 453 K. However, accompanied with the increased conversion, the DMM selectivity presents a tendency of rapid decline. When the temperature reaches 393 K, methanol conversion increases to 48.9 % with a high DMM selectivity of 89.9 %. However, the selectivity of MF and DME increases rapidly with rising temperature, especially for MF, which indicates that high reaction temperatures facilitate the formation of MF and DME instead of DMM. Since methanol oxidation to DMM is an exothermic reaction, low temperature is favorable for the DMM production. The formation of FA and COx is also thermodynamically unfavorable at low temperatures. From our results and previous studies [3, 4, 7], it is not difficult to find that the reaction temperature has
Sci. Bull. Table 2 Peak-fitting results of V 2p3/2 XPS spectra for V2O5/TiO2/Al2O3 catalysts V 2p3/2 BE (eV)
Sample
V(IV)/(V(IV)?V(V))
V(IV) 2p3/2
V(V) 2p3/2
(Area %)
15V2O5/TiO2/Al2O3-mix
515.8
517.3
15.9
15V2O5/TiO2/Al2O3-imp
515.8
517.1
17.2
15V2O5/TiO2–Al2O3-cop
515.6
517.1
24.6
15V2O5/TiO2/Al2O3-mil
515.5
517.0
31.2
15V2O5/TiO2–Al2O3-sol
516.6
517.0
26.3
significant influence on the product distribution. Between 343 and 413 K, DMM is the major product with a selectivity of 68 %–95 %. Between 423 and 453 K, MF becomes the main product with a selectivity of 47 %–86 %. The formation of FA, DME, CO, and CO2 is not observed until the temperature is higher than 473 K.
4 Discussion The catalysts supported on the TiO2–Al2O3 support result in better catalysis performance in methanol oxidation to DMM reaction systems with respect to catalysts supported
over single oxides. As shown in the previous section, the 15V2O5/TiO2–Al2O3-sol sample has a 48.9 % methanol conversion with a DMM selectivity of 89.9 % at 393 K, but the 15V2O5/TiO2 only exhibits a 23.2 % methanol conversion with a DMM selectivity of 86 %, and the 15V2O5/Al2O3 has a 34.6 % methanol conversion with a DMM selectivity of 67.9 %. The comparison of the catalytic performance implies the significant role of mixed TiO2–Al2O3 oxide support for the enhanced activity in the methanol oxidation. Due to the coexistence of TiO2 and Al2O3, the chemical properties of V species on the catalyst are effectively influenced by modifying the interaction between active component and mixed support. Meanwhile, the acid properties of the catalyst have been changed by the interaction of TiO2 with Al2O3, and the catalysts of sol–gel and ball-milling methods have generated more weaker acidic sites that have notable positive effect on DMM selectivity, which is presented in Table 3 and Fig. 3. The mixed TiO2–Al2O3 oxide support makes an important role in the activity of methanol oxidation to DMM through its effect on the acidity of the catalyst and the chemical properties of V species as follows:
4.1 The effect of mixed support on the acidity of the catalysts Fig. 6 (Color online) H2-TPR profiles of catalysts with different preparation methods
The NH3-TPD profiles show that the acidity of mixed oxides is quite different from that of individual component
Table 3 Catalytic activities of V2O5/TiO2–Al2O3 catalysts with different preparation methods Sample
Methanol conversion (%)
Selectivity (%) DMM
MF
DME
FA 0.0
15V2O5/TiO2–Al2O3-mix
25.1
79.7
11.6
8.7
15V2O5/TiO2–Al2O3-imp
32.0
84.9
9.1
6.0
0.0
15V2O5/TiO2–Al2O3-cop
42.7
77.6
13.9
8.2
0.3
15V2O5/TiO2–Al2O3-mil
43.0
90.7
3.4
5.6
0.3
15V2O5/TiO2–Al2O3-sol
48.9
89.9
4.3
5.7
0.1
15V2O5/TiO2
23.2
86.0
10.3
3.7
0.0
15V2O5/Al2O3
34.6
67.9
16.5
14.6
1.0
-1
T = 393 K; feed gas CH3OH/O2/N2 = 1/2.5/7.5 (v:v); gas hourly space velocity (GHSV) = 8,000 h ; reaction duration = 8 h
123
Sci. Bull. Table 4 Catalytic performance at different temperatures for 15V2O5/TiO2–Al2O3-sol Temperature (K)
Methanol conversion (%)
Selectivity (%) DMM
DME
MF
FA
COx
373
39.3
90.4
2.8
6.8
0.0
0.0
393
48.9
89.9
4.3
5.7
0.1
0.0
413
55.5
68.3
6.8
23.8
1.1
0.0
433
59.9
39.4
8.9
46.9
4.8
0.0
453
64.3
18.0
14.3
63.3
3.7
0.7
metal oxide, and such difference could be caused by the excess charges from the formation of Ti–O–Al bonds [20]. The extent of the formation of the bridged hetero metal– oxygen bonds depends on the mixing degree between Ti and Al, which varies with the preparation methods. If proper preparation methods are used, the bonds of Al–O– Al become weaken with the incorporation of Ti into Al2O3 structure, and weaker acidic sites can thus be produced. For example, the catalysts with sol–gel, ball milling, and precipitation methods possess more weaker acidic sites. But the acidity might tend to increase if the interaction between Ti and Al species is so intense that highly bridged Ti–O–Al bonds are formed [13]. As a result, the catalyst with the co-precipitation method has a large amount of middle strong acidic sites. The resulting acidic site strength and distribution significantly affect the product selectivity. The catalytic test results show that the catalysts with more weaker acidic sites, including 15V2O5/ TiO2–Al2O3-sol, 15V2O5/TiO2–Al2O3-mil, and 15V2O5/ TiO2–Al2O3-imp, favor the formation of DMM with a selectivity higher than 85 %. The low DMM selectivity of 15V2O5/TiO2–Al2O3-cop might be due to the large amount of middle strong acid sites, which favors the formation of MF and DME [9]. Although the 15V2O5/ TiO2–Al2O3-imp sample has a comparatively lower methanol conversion (32.0 %) than 15V2O5/TiO2–Al2O3cop sample (42.7 %), the DMM selectivity of 15V2O5/ TiO2–Al2O3-imp (84.9 %) is higher than 15V2O5/TiO2– Al2O3-cop (77.6 %), attributing to the weaker acid sites on the catalyst surface [10]. 4.2 The effect of mixed support on the chemical properties of V species In this work, it is apparent that the V2O5 supported on mixed TiO2–Al2O3 support with different preparation methods show huge differences on the catalytic conversion of methanol to DMM. To some extent, such differences reflect the chemical properties of V species affected by the TiO2–Al2O3 mixed support. In many catalytic systems, the support not only allows for a high dispersion of the active
123
phase, but also plays a role in determining the catalytic activity by interacting with the active phase. From the literature [13, 16, 19], both the V5? and V4? species are the active sites in the oxidation process. However, as V4? could provide more oxygen vacancies for improving the electron-transfer capability, more V4? species lead to better catalytic activity. The redox process of V species between high oxidation state and low oxidation state allows the reactant to be oxidized with the release of lattice oxygen in V species. From the XPS results in Table 2, the sol–gel, ball milling, and co-precipitation methods generate more V4? species for methanol oxidation. The generation of lower valence state of V species is related to the interaction between the active component and the mixed support, which is confirmed by the appearance of traces of V2Ti7O17 phases on these catalysts in Fig. 1 and the shift of Al 2p and Ti 2p binding energy in Fig. 4. The definite degree of interaction of active component and the support is in favor of producing more oxygen vacancies to facilitate the electron transfer between support and active components [23], which leads to stronger redox capability. The catalysts with high redox capability, including 15V2O5/ TiO2–Al2O3-sol, 15V2O5/TiO2–Al2O3-mil, and 15V2O5/ TiO2–Al2O3-cop, favor the methanol oxidation with conversions higher than 43 %. 4.3 The nature of the V2O5/TiO2–Al2O3 catalyst with different preparation methods The metal oxide–support interaction cannot behave well without specific support structure. Different support structure is generated from the variation in preparation methods of the mixed TiO2–Al2O3 support. The substance of the V2O5/TiO2–Al2O3 catalyst with different preparation method could reflect the metal oxide–support interaction to some extend. The theoretical monolayer capacity of supported TiO2 species on the Al2O3 is about 0.66 mmol/ 100 m-2 Al2O3 (about 0.168 g/g c-Al2O3). Thus, with the given high TiO2 mass content of 0.33 g/g c-Al2O3 (TiO2:Al2O3 = 1:2 wt%), TiO2 might have been crystallized into phases on the Al2O3 support surface or interacted
Sci. Bull.
with Al2O3 [20]. For example, the sol–gel method allows a portion of Ti species to incorporate into Al2O3 lattice to change the electronic properties, as shown in Fig. 5, which further enhances the interaction between the active component and the support. The ball-milling method facilitates the non-equilibrium phase transition of Al2O3 crystal and promotes the interface interaction of TiO2 with Al2O3 companied with the decrease in crystallite size [24]. Additionally, the interaction between the support and active components of sol–gel and ball-milling samples is stronger than other samples, which is helpful to promote the electron-transfer properties, which could improve the redox properties for methanol oxidation. However, the precipitation and mechanical mixing method have little influence on the improvement of the interaction between the support and active component, not to mention the enhancement of redox ability, as confirmed by the XPS profiles in Fig. 4 and the structure of catalysts in Fig. 1. In brief, a proper preparation method would allow one species to incorporate into the lattice of another species, enhancing the interaction between support and active component, changing the electronic properties, and facilitating the electron transfer between support and active component, thereby improving the redox and acidic properties of the V2O5/TiO2–Al2O3 catalyst for methanol oxidation.
species. The V4? could offer more oxygen vacancies to facilitate the electron transfer between the support and active component, thereby improving the redox properties of the V2O5/TiO2–Al2O3 catalyst for increased methanol conversion. Providing weaker acidic sites is enabled by the incorporation of titanium cation into the alumina structure, weakening the Al–O–Al bonds. The resulted weaker acidic sites are favorable for improving the DMM selectivity. In summary, the study emphasizes the role of the mixed support on the chemical properties of active vanadium oxide species and the acidic properties. The coordinated environment of the V chemical state and the mixed TiO2–Al2O3 support is significant for improving the catalytic efficiency on selective methanol oxidation to DMM. Acknowledgments This work was supported by the National Natural Science Foundation of China (21006068, 21222604), the Program for New Century Excellent Talents in University (NCET-100611), Specialized Research Fund for the Doctoral Program of Higher Education (20120032110024), the Scientific Research Foundation for the Returned Overseas Chinese Scholars (MOE), and the Program of Introducing Talents of Discipline to Universities (B06006). Conflict of interest of interest.
The authors declare that they have no conflict
References 5 Conclusions We have studied the catalytic properties of V2O5/TiO2– Al2O3 catalysts for the selective methanol oxidation to DMM. In order to investigate the mechanism of V2O5/TiO2– Al2O3 catalysts in methanol oxidation to DMM, we have explored five different preparation methods and obtained diversified catalytic performance on the V2O5/TiO2–Al2O3 samples. The V2O5/TiO2–Al2O3 samples have superior catalytic properties than the V2O5/TiO2 and V2O5/Al2O3 samples. The catalyst prepared by the sol–gel method has the best catalytic performance with a maximum conversion of 48.9 % and a high DMM selectivity of 89.9 % at 393 K. The catalyst with the ball-milling method has also relatively high methanol conversion (43 %) and DMM selectivity (90.7 %), which is surpassed only by the V2O5/TiO2–Al2O3-sol sample. The excellent catalytic performance of the V2O5/TiO2– Al2O3 samples for menthol oxidation to DMM is attributed to the effectively modified interaction between the active component and the support coupled with the appropriate acidic sites provided by the mixed TiO2–Al2O3 support. Proper preparation methods can further promote the interaction between active V component and the mixed support as well as provide weaker acidic sites. The interaction between active V component and support changes the chemical states of supported active V components, producing more V4?
1. Zhao Y, Qin Z, Wang G et al (2013) Catalytic performance of V2O5/ZrO2–Al2O3 for methanol oxidation. Fuel 104:22–27 2. Thavornprasert KA, Capron M, Jalowiecki-Duhamel L et al (2014) Highly productive iron molybdate mixed oxides and their relevant catalytic properties for direct synthesis of 1,1-dimethoxymethane from methanol. Appl Catal B 145:126–135 3. Nikonova OA, Capron M, Fang G et al (2011) Novel approach to rhenium oxide catalysts for selective oxidation of methanol to DMM. J Catal 279:310–318 4. Zhao H, Bennici S, Shen J et al (2010) Nature of surface sites of catalysts and reactivity in selective oxidation of methanol to dimethoxymethane. J Catal 272:176–189 5. Yuan Y, Iwasawa Y (2002) Performance and characterization of supported rhenium oxide catalysts for selective oxidation of methanol to methylal. J Phys Chem B 106:4441–4449 6. Zhao H, Bennici S, Cai J et al (2010) Effect of vanadia loading on the acidic, redox and catalytic properties of V2O5–TiO2 and V2O5–TiO2/SO42- catalysts for partial oxidation of methanol. Catal Today 152:70–77 7. Kaichev VV, Popova GY, Chesalov YA et al (2014) Selective oxidation of methanol to form dimethoxymethane and methyl formate over a monolayer V2O5/TiO2 catalyst. J Catal 311:59–70 8. Liu H, Iglesia E (2003) Selective one-step synthesis of dimethoxymethane via methanol or dimethyl ether oxidation on H3?nVnMo12-nPO40 Keggin structures. J Phys Chem B 107:10840–10847 9. Chen S, Wang S, Ma X et al (2011) Selective oxidation of methanol to dimethoxymethane over bifunctional VOx/TS-1 catalysts. Chem Commun 47:9345–9347 10. Chen S, Meng Y, Zhao Y et al (2013) Selective oxidation of methanol to dimethoxymethane over mesoporous Al-P-V-O catalysts. AlChE J 59:2587–2593
123
Sci. Bull. 11. Royer S, Se´cordel X, Brandhorst M et al (2008) Amorphous oxide as a novel efficient catalyst for direct selective oxidation of methanol to dimethoxymethane. Chem Commun 7:865–867 12. Gornay J, Se´cordel X, Tesquet G et al (2010) Direct conversion of methanol into 1,1-dimethoxymethane: remarkably high productivity over an FeMo catalyst placed under unusual conditions. Green Chem 12:1722–1725 13. Matralis HK, Ciardelli M, Ruwet M et al (1995) Vanadia catalysts supported on mixed TiO2–Al2O3 supports: effect of composition on the structure and acidity. J Catal 157:368–379 14. Pophal C, Kameda F, Hoshino K et al (1997) Hydrodesulfurization of dibenzothiophene derivatives over TiO2–Al2O3 supported sulfided molybdenum catalyst. Catal Today 39:21– 32 15. Reddy BM, Chowdhury B, Reddy EP et al (2001) X-ray photoelectron spectroscopy study of V2O5 dispersion on a nanosized Al2O3-TiO2 mixed oxide. Langmuir 17:1132–1137 16. Martı´nez S, Morales R, Ca´rdenas-Galindo MG et al (2005) Reducibility, heats of re-oxidation, and structure of vanadia supported on TiO2 and TiO2–Al2O3 supports used as vanadium traps in FCC. Thermochim Acta 434:74–80 17. Goodrow A, Bell AT (2007) A theoretical investigation of the selective oxidation of methanol to formaldehyde on isolated vanadate species supported on silica. J Phys Chem C 111:14753–14761 18. Benjaram M, Biswajit C, Panagiotis G (2001) An XPS study of the dispersion of MoO3 on TiO2–ZrO2, TiO2–SiO2, TiO2–Al2O3, SiO2–ZrO2, and SiO2–TiO2–ZrO2 mixed oxides. Appl Catal A 211:19–30 19. Kim T, Wachs IE (2008) CH3OH oxidation over well-defined supported V2O5/Al2O3 catalysts: influence of vanadium oxide loading and surface vanadium-oxygen functionalities. J Catal 255:197–205
123
20. Linacero R, Rojas-Cervantes M, Lopez-Gonzalez JDD (2000) Preparation of xTiO2(1-x)Al2O3 catalytic supports by the sol– gel method: physical and structural characterisation. J Mater Sci 35:3279–3287 21. Lapina OB, Shubin AA, Nosov AV et al (1999) Characterization of V2O5-TiO2 catalysts prepared by milling by ESR and solid state 1H and 51V NMR. J Phys Chem B 103:7599–7606 22. Galan CA, Ortiz AL, Guiberteau F et al (2009) Crystallite size refinement of ZrB2 by high-energy ball milling. J Am Ceram Soc 92:3114–3117 23. Habel D, Goerke O, Tovar M et al (2008) Phase relations in the system TiO2–V2Ox under oxidizing and reducing conditions. J Phase Equilib Diff 29:482–487 24. Ghosh J, Bysakh S, Mazumder S (2014) Multiphase transformation and hybrid nanostructure under non-equilibrium and equilibrium condition during high-energy ball milling of BaTiO3 powders. Phase Transit 87:325–335 25. Gao X, Bare SR, Fierro JLG et al (1999) Structural characteristics and reactivity/reducibility properties of dispersed and bilayered V2Ox/TiO2/SiO2 Catalysts. J Phys Chem B 103:618–629 26. Chenakin S, Prada Silvy R, Kruse N (2005) Effect of X-rays on the surface chemical state of Al2O3, V2O5, and aluminovanadate oxide. J Phys Chem B 109:14611–14618 27. Zhu HQ, Qin ZF, Shan WJ et al (2004) Pd/CeO2-TiO2 catalyst for CO oxidation at low temperature: a TPR study with H2 and CO as reducing agents. J Catal 225:267–277 28. Coste S, Bertrand G, Coddet C et al (2007) High-energy ball milling of Al2O3–TiO2 powders. J Alloys Compd 434–435:489–492 29. Trudeau M, Schulz R, Dussault D et al (1990) Structural changes during high-energy ball milling of iron-based amorphous alloys: is high-energy ball milling equivalent to a thermal process? Phys Rev Lett 64:99
www.scibull.com www.springer.com/scp
Article
Chemistry
Selective oxidation of methanol to dimethoxymethane over V2O5/TiO2–Al2O3 catalysts Tuo Wang • Yali Meng • Liang Zeng Jinlong Gong
•
Received: 7 February 2015 / Accepted: 6 March 2015 Ó Science China Press and Springer-Verlag Berlin Heidelberg 2015
Abstract This paper describes the effect of the preparation method of binary oxide supports (TiO2–Al2O3) on catalytic performance of V2O5/TiO2–Al2O3 catalysts for methanol selective oxidation to dimethoxymethane (DMM). The TiO2–Al2O3 supports are synthesized by a number of methods including mechanical mixing, ball milling, precipitation, co-precipitation, and sol–gel method, which is followed by incipient wetness impregnation to produce V2O5/TiO2–Al2O3 catalysts. Among these samples, the V2O5/TiO2–Al2O3 catalyst prepared by the sol–gel method has the best catalytic performance with a maximum methanol conversion of 48.9 % and a high DMM selectivity of 89.9 % at 393 K, showing superior performance than V2O5/TiO2 and V2O5/Al2O3. The excellent catalytic performance of V2O5/TiO2–Al2O3 is attributed to the effective interaction between the active component and the mixed support. Such interaction changes the chemical states of supported active V components, produces an increased amount of V4? species, and facilitates the electron transfer between support and active component. Additionally, the incorporation of titanium cation into the alumina structure could also help produce an appropriate amount of acidic sites, which increases the DMM selectivity. The coordinated environment of the dispersed vanadia on TiO2–Al2O3 mixed support improves the catalytic efficiency on methanol oxidation to DMM. T. Wang Y. Meng L. Zeng J. Gong Key Laboratory for Green Chemical Technology of Ministry of Education, School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China T. Wang Y. Meng L. Zeng J. Gong (&) Collaborative Innovation Center of Chemical Science and Engineering, Tianjin 300072, China e-mail: [email protected]
Keywords Methanol selective oxidation Dimethoxymethane V2O5/TiO2–Al2O3 Binary oxide support
1 Introduction Dimethoxymethane (DMM) has successfully attracted increasing attention as a downstream chemical product of methanol in industrial practice. DMM is a useful intermediate in organic synthesis and an excellent solvent in perfume and pharmaceutical industries. It can also act as an environmentally friendly fuel with high efficiency and low toxicity [1]. Moreover, it has been used as a starting monomer in the synthesis of polyoxymethylene dimethylether (POMM) [2]. The traditional method for DMM synthesis includes two consecutive steps: gas-phase methanol partial oxidation to formaldehyde (FA) and the subsequent acetalization of methanol and formaldehyde over acid catalysts [3]. However, this reaction scheme suffers from high reaction temperatures, complicated procedures, and severe equipment corrosion [2]. In addition to the target product DMM in methanol oxidation, several byproducts could form, including methyl formate (MF), dimethyl ether (DME), FA, and oxycarbide. An alternative synthetic pathway by one-step selective oxidation of methanol to DMM under mild reaction conditions becomes more desirable with both environmental and economic advantages. To obtain a high DMM yield in one step, the design of the catalyst is followed by the principle of appropriately balanced redox ability and acidity [4]. Excessive redox sites of the catalyst could lead to the abundant amount of FA, MF, and oxycarbide, while superfluous acidic sites result in a large amount of DME.
123
Sci. Bull.
Several catalytic systems have been reported to be active for the methanol oxidation to DMM by carefully coupling the redox sites with suitable supports. For example, Re oxide has been considered as a potential active site for methanol oxidation, and the supported Re oxide like Re/c-Fe2O3 catalyst has a DMM selectivity of 91 % and a methanol conversion of 48.4 % at 513 K [5]. But a bulk ReO3 catalyst without support has a rather low methanol conversion, only *12.4 % [5]. The V2O5/TiO2based catalysts by acidic modification have shown superior performance. For instance, V2O5/TiO2/SO42- achieved a 47 % methanol conversion at 393 K, which is much better than the conventional V2O5/TiO2 catalyst with 17 % methanol conversion [4, 6, 7]. The bulk H5PV2Mo10O40 only showed a 41.1 % methanol conversion with a 40.2 % DMM selectivity at 453–513 K, while the supported heteropoly acid catalysts exhibited 60.3 %–68.4 % methanol conversion with 67 %–51 % DMM selectivity. This is benefitted from the Keggin structures of H3?nPVnMo12-nO40 providing both redox and Bronsted acid sites required for producing DMM [8]. Additionally, several other catalytic systems have been reported to be active for the methanol oxidation to DMM, including modified V/TS-1 catalysts [9], amorphous multi-metal oxides [10, 11], Fe2(MoO4)3–MoO3 catalysts [2, 12]. Among these catalysts, the supported V2O5 catalysts have shown excellent catalysis performance in the selective oxidation of methanol to DMM, owing to the outstanding redox properties of V2O5. The catalytic activity and selectivity of the catalyst would be improved by the use of an appropriate support. Due to its high specific surface area, weak acidity, and good thermal stability, c-Al2O3 has been widely used as a support for a number of catalysts. Recently, some interesting results have been obtained for TiO2 as a promoter on high surface area supports. The chemical states of supported metal or metal oxides, adsorption properties of the catalyst and their catalytic activity are influenced by the incorporation of TiO2 with the support. Particularly, TiO2– Al2O3 support has drawn special attention owing to its versatile catalytic properties for various commercially important reactions such as isomerization of alkenes, dehydration of alcohols, and hydrodesulphurization [13–15]. Particularly, the TiO2–Al2O3 supported catalysts show better performance than single Al2O3 supported catalyst in the hydrodesulfurization (HDS) reaction [14]. It is also well established that the supported vanadium oxides are effective catalysts for partial oxidation of hydrocarbons and alcohols [13, 16, 17]. And the structure of the dispersed vanadium species is closely related to the nature of the specific oxide support, the loading amount, and the preparation procedure [18].
123
Considering versatile catalytic properties of TiO2–Al2O3 support and good redox properties of V2O5, we have developed V2O5/TiO2–Al2O3 catalysts for methanol oxidation to DMM. The aim of this study is to explore the influence of preparation method of mixed TiO2–Al2O3 support on the catalytic activity of V2O5. The V2O5/TiO2– Al2O3 catalysts prepared by five different methods have been employed in selective methanol oxidation to DMM, including mechanical mixing, ball milling, precipitation, co-precipitation, and sol–gel methods. The materials are characterized with the N2 adsorption/desorption, X-ray diffraction (XRD), temperature-programmed reduction (H2-TPR), Raman spectroscopy, temperature-programmed desorption of ammonia (NH3-TPD), and X-ray photoelectron spectroscopy (XPS) techniques. A set of activity tests has been conducted in a fixed-bed micro-reactor for collecting and comparing their kinetic data. The relationship between catalytic activities and the behaviors of both redox and acidic sites of V2O5/TiO2–Al2O3 catalysts is analyzed and discussed.
2 Experimental 2.1 Catalyst preparation A series of 15V2O5/TiO2–Al2O3 catalysts were prepared by a wetness impregnation method with the TiO2–Al2O3 complex support. The V2O5 content is all fixed at 15 % for well dispersion and effective comparison reason, as the monolayer coverage of V2O5/TiO2 and V2O5/Al2O3 catalysts is about 12.5 % and 20 %, respectively [7, 19]. The TiO2–Al2O3 complex support were prepared by five different methods—mechanical mixing, ball milling, precipitation, sol–gel, and co-precipitation method—with the uniform constitute of Al2O3 and TiO2 (Al2O3:TiO2 = 2:1 by weight). A c-Al2O3 powder support and an industrial TiO2 (P25) (80 % anatase and 20 % rutile) powder were calcined in air at 573 K for 2 h prior to the catalyst preparation. For the complex support with the mechanical mixing method (denoted as TiO2–Al2O3-mix), the required amount of Al2O3 and P25 were mechanically mixed with each other until smooth powder was formed. For the support with the ball-milling method (denoted as TiO2–Al2O3mil), the required amount of Al2O3 and P25 were spread in the high energy ball-milling pot and milled for 12 h. The complex support with the precipitation method (denoted as TiO2–Al2O3-imp) was prepared by the method of gelsupported precipitation [13]. In the case with the sol–gel method (denoted as TiO2–Al2O3-sol), it was prepared by the procedure adapted from the literature [20]. For the complex support with the co-precipitation method (denoted
Sci. Bull.
as TiO2–Al2O3-cop), the mixed oxide support was prepared by a homogeneous co-precipitation method using urea as hydrolyzing agent [18]. All the oxide supports were calcined in air at 773 K for 4 h before the impregnation of ammonium metavanadate. The V2O5/TiO2–Al2O3 catalysts were obtained by the impregnation of calcined TiO2–Al2O3 support with the mixed ammonium metavanadate and oxalic acid aqueous liquid solution. Subsequently, samples were obtained upon the evaporation of the solvent at 353 K and then dried at 393 K for 12 h and calcined at 773 K for 6 h in furnace.
2.2 Catalyst characterization N2 adsorption/desorption analysis was performed at 77 K with a Micromeritics Tristar 300 surface area and porosity analyzer. The specific surface area was calculated by the Bruauer–Emmett–Teller (BET) method. The pore parameters (pore volume and pore diameter) were evaluated from the desorption branch of isotherms based on the Barrett–Joyner–Halenda (BJH) model. XRD patterns were collected on a Rigaku D/Max-2500 diffractometer using C-filtered Cu Ka radiation ˚ ), operated at 40 kV and 200 mA, at a scan (k = 1.54056 A rate of 0.02°/s from 10° to 90°. Raman spectra were collected at ambient temperature using a Thermo Scientific DXRTM Raman Microscope with a green semiconductor laser (532 nm) as the excitation source. Samples were pretreated in a vacuum drier at 393 K for 12 h before any measurements. NH3-TPD spectra were recorded on an Autochem 2920 chemisorption analyzer. The sample was pretreated at 573 K under Ar flow (30 mL min-1) for 1 h and was then cooled down to 333 K. NH3 was subsequently introduced into the flow system for 30 min. The TPD spectra were recorded at a ramp rate of 10 K min-1 from 333 to 973 K under a He flow. H2-TPR measurements were carried out on an Autochem 2920 chemisorption instrument with a thermal conductivity detector (TCD). The sample (100 mg) was pretreated at 573 K under Ar flow (30 mL min-1) for 1 h and cooled to 333 K. The sample was then contacted with a H2–Ar mixture (10 vol% H2) at a flow rate of 30 mL min-1 and heated at a rate of 10 K min-1 from 333 to 950 K. XPS measurements were performed on a Physical Electronics PHI-1600 XPS system equipped with a hemispherical electron analyzer and an Mg Ka X-ray source (1,253.6 eV). The samples were vacuum-dried at 393 K for 4 h prior to any measurements. The energy region of the photoelectrons was scanned at a pass energy of 29.35 eV. The binding energies were referenced to the C 1s signal at 284.6 eV. The data were processed using the PHI Multipack Program (Gaussian peak fitting).
2.3 Catalyst activity measurements Catalyst activity tests were carried out in a fixed-bed micro-reactor at ambient pressure. The reactor tube was made of quartz with an inner diameter of 7 mm. Methanol was introduced into the reaction zone by bubbling a gas mixture of O2 and N2 (99.99 %) through a glass saturator filled with methanol (99.9 %) at 288 K. In each test, 0.3 g of tableted catalysts was loaded. The catalyst was pretreated with a 25 vol% O2 and 75 vol% N2 mixture gas (40 mL min-1) at 673 K for 2 h before reaction test. Upon cooling to the reaction temperature, a mixture of methanol, N2, and O2 was introduced to the catalyst bed with a flow rate of 48 mL min-1. The gas hourly space velocity (GHSV) was kept at 8,000 h-1 with a feed gas of CH3OH:O2:N2 = 1:2.5:7.5 in volume. The reaction temperature ranged from 373 to 453 K. The products were analyzed by an online SP-2100 gas chromatography (GC) using a Propack T column connected with a flame ionization detector (FID) and a TCD detector. The original products like methanol, MF, FA, DME, and DMM were analyzed on the FID line, while the carbon monoxide (CO) and carbon dioxide (CO2) were quantified on the TCD line. The gas lines were kept at 373 K to prevent condensation of the reactants and products. The carbon balance was 99.0 % ± 1 % for all GC measurements.
3 Results 3.1 Structural properties Physical properties of the catalysts with different preparation methods are presented in Table 1. The 15V2O5/TiO2– Al2O3 catalyst with the impregnation method has the largest BET surface area, which is almost the same as pure Al2O3 support (221 m2 g-1 BET surface area, 0.34 m3 g-1 pore volume, and 6.0 nm average pore diameter). The BET surface areas of the catalysts with mechanical mixing and ball-milling methods are similar, about 110 m2 g-1, whereas the average pore radius of the ball-milling catalyst is larger than the mechanical mixing catalyst, about 9.3 nm, indicating the ball milling enhances the interaction between TiO2 and Al2O3 [21]. The BET surface areas of the catalysts with sol–gel and co-precipitation methods are lower than samples with other methods. The pore volumes of sol–gel and co-precipitation catalysts decrease substantially compared with other catalysts, while the average pore diameters are much greater than other catalysts. XRD patterns of the 15V2O5/TiO2–Al2O3 catalysts with different preparation methods are shown in Fig. 1. It is apparent that vanadium oxide species are well dispersed on the mixed TiO2–Al2O3 support, except for the sample of
123
Sci. Bull. Table 1 Porous properties of V2O5/TiO2–Al2O3 catalysts with different preparation methods Sample
SBET (m2 g-1)
Pore volume (m3 g-1)
15V2O5/TiO2–Al2O3-mix
110.7
0.24
15V2O5/TiO2–Al2O3-imp
214.0
0.37
6.8
32.7
15V2O5/TiO2–Al2O3-cop
45.3
0.10
23.4
64.3
15V2O5/TiO2–Al2O3-mil
109.1
0.31
9.3
15.2
15V2O5/TiO2–Al2O3-sol
61.3
0.13
14.4
18.1
a
Average pore diameter (nm) 7.4
Crystallite size (nm)a 26.4
Calculated by the TiO2 peak of XRD data
Fig. 1 (Color online) XRD patterns of the catalysts with different preparation methods
Fig. 2 (Color online) Raman spectra of catalysts with different preparation methods
15V2O5/TiO2–Al2O3-mix. Compared with 15V2O5/TiO2– Al2O3-mix, the Al2O3 crystal phase (characteristic peaks appear at 45.8°, 67°, and 37.6° (JCPDS No. 10-425)) is not presented on the 15V2O5/TiO2–Al2O3-mil, which only possesses TiO2 phases (sharp peaks at 22°, 23°, 28°, 31°, 34°) (JCPDS No. 21-1272). And there are no sharp V2O5 phase peaks at 20° and 31° (JCPDS No. 45-1074) on the 15V2O5/TiO2–Al2O3-mil. It indicates that the ball-milling process might promote the dispersion of active species and decrease the crystallinity of the Al2O3 in the mixed support [22]. For the sol–gel and co-precipitation catalysts, there are a large amount of TiO2 crystal phases and traces of V2Ti7O17 on the surface [23], and the intensity of the lines due to TiO2 phase on the co-precipitation catalyst is sharper than that of the sol–gel method. The crystallite size of the catalysts is also calculated from the XRD data, and the results are listed in Table 1; it is apparent that the ballmilling and sol–gel catalysts have smaller crystallite sizes than other samples. Raman spectra of V2O5/TiO2–Al2O3 catalysts with different preparation methods are presented in Fig. 2 for more detailed structural information. The pure Al2O3 support exhibits almost no absorption peak, and the band at 1,003 cm-1 is assigned to V=O stretches in monovanadates [19]. The sharp intense Raman bands at 639, 513, 292, and
152 cm-1 are attributed to crystallized anatase TiO2, and the bands at 705, 413, and 223 cm-1 are corresponded to rutile TiO2 [16]. For the 15V2O5/TiO2–Al2O3-imp sample, the degree of TiO2 crystallization is the lowest, which indicates that most of the Ti species are highly dispersed on the Al2O3 surface. Compared with 15V2O5/TiO2–Al2O3mix, it is apparent that the intensity of TiO2 phase for 15V2O5/TiO2–Al2O3-mil sample is decreased significantly and the sharpest bands at 639 cm-1 gradually disappeared, implying that the ball milling promotes the non-equilibrium phase transition of TiO2 phase and formation of lattice defects [24]. For the 15V2O5/TiO2–Al2O3-sol sample, there is no apparent peak at 1,003 cm-1, illustrating that vanadium oxide species are well dispersed on the TiO2–Al2O3 support. Moreover, the bands at 152 cm-1 shift to higher wave number for about 3 cm-1, declaring that the interaction of Ti and Al species is enhanced. For the 15V2O5/ TiO2–Al2O3-cop sample, in addition to the sharper intensity of crystallized TiO2 phase, the shift of band at 152 cm-1 is more distinct; the interaction of Ti and Al species on the mixed TiO2–Al2O3-cop support is much stronger than that of samples prepared by other methods. The Raman bands at 248 cm-1 of 15V2O5/TiO2–Al2O3cop is assigned to the stretching and bending modes of V– O–Ti bridging bonds as the result of the interaction
123
Sci. Bull.
between the active component and the support [25]. The Raman analysis is consistent with the XRD results. 3.2 Acidic properties NH3-TPD profiles of the V2O5/TiO2–Al2O3 catalysts with different preparation methods are shown in Fig. 3. In the NH3-TPD profiles, there are three acidic sites, which are weak acidic sites (350–600 K), medium strong acidic sites (600–750 K), and strong acidic sites (750–900 K), respectively. The profiles show that the strength and distribution of acidic sites vary significantly with different preparation methods. The strong acidic sites could be observed distinctly on the 15V2O5/TiO2–Al2O3-mix sample, which is similar with the c-Al2O3 support profiles, indicating the invariance of the Al2O3 structure. For the catalysts prepared by the ball milling, impregnation, and sol– gel methods, there are mainly weak acidic sites on these catalysts. The induction of weaker acid sites is related to the lowered intensity of Al–O–Al bonds, which is a consequence of the incorporation of TiO2 into Al2O3 with these methods, changing the original Al2O3 structure [20]. The middle strong acidic sites of the 15V2O5/TiO2–Al2O3cop catalysts tend to increase compared with the other samples, which might be related to the stronger interaction of Ti and Al species by the formation of intensely bridged Ti–O–Al bonds as well as the higher electron acceptability by the close interaction between vanadium oxide on the TiO2–Al2O3 supports [20]. 3.3 Redox properties The chemical states of Ti 2p, Al 2p, and O 1s on the catalyst support surface significantly influence the oxidation activity of methanol to DMM, which also could reflect the degree of the mutual function between the active
Fig. 3 (Color online) NH3-TPD profiles of the catalysts with different preparation methods
component and the support [15]. From the Ti 2p, Al 2p, and O 1s XPS analysis shown in Fig. 4, it is evident that the binding energies of Ti 2p and Al 2p shift to lower values, especially on the V2O5/TiO2–Al2O3-sol catalyst, indicating there is lower positive charge density on the alumina and titanium centers, due to the mutual interaction between the active component and the support [15, 18]. Such enhanced interaction could induce more oxygen vacancies on the V2O5/TiO2–Al2O3 catalyst [24]. From binding energies results of O 1s shown in Fig. 4, the binding energy at around 530.0–530.4 eV is the characteristic of lattice oxygen of Ti–O–Ti in TiO2, while the binding energy at around 531.1–531.4 eV is the lattice oxygen of Al–O–Al in Al2O3 and 529.6–529.9 eV is the lattice oxygen of V–O or V=O in V2O5. The O 1s binding energies of V2O5/TiO2– Al2O3-sol sample shift to lower values, implying that the amount of active lattice oxygen species of the catalyst is increased, which could account for better performance in methanol oxidation [18, 26]. XPS analysis also reveals the chemical states of active V species on the support surface, which has significant influence on the redox capability of the catalyst. Binding energies and peak fitting results of V 2p3/2 obtained from XPS spectra of V2O5/TiO2–Al2O3 catalysts are shown in Fig. 5 and Table 2. From the V 2p3/2 deconvolution results, it is clearly seen that a greater amount of V4? species is presented on the V2O5/TiO2–Al2O3-mil and V2O5/TiO2– Al2O3-sol surface than other samples, indicating that the degree of reduction of V2O5 is enhanced and the formation of more oxygen vacancies with the ball-milling and sol–gel methods. The reducibility of the V2O5/TiO2–Al2O3 catalysts with different preparation methods was also characterized by H2-TPR (Fig. 6). The H2-TPR profiles for the V2O5/TiO2– Al2O3 samples are presented as a peak in the 650–850 K temperature range, which is assigned to reduction of V5?– V3? in various polyvanadates species. It is apparent that the temperature of the first peak (Tmax) for V2O5/TiO2– Al2O3-sol is the lowest, which demonstrates that its redox capability is the strongest among all the samples. Additionally, 15V2O5/TiO2–Al2O3-sol sample has an additional reduction peak in the range from 850 to 920 K, which is attributed to reduction of Ti4? to Ti3? [27]. The V2O5/ TiO2–Al2O3-mil sample also has strong redox ability that is preceded only by the V2O5/TiO2–Al2O3-sol sample, while the redox capability of V2O5/TiO2–Al2O3-mix is the poorest one. Most interestingly, for the 15V2O5/TiO2– Al2O3-cop sample, there is large peak presented at about 869 K, which is the overlapped reduction peaks of V5? to V3? and Ti4? to Ti3? in V–O–Ti structures, as a result of the strong interaction between active species and interior supports.
123
Sci. Bull.
Fig. 4 (Color online) XPS spectra of Al 2p3/2 (a), Ti 2p3/2 (b), and O 1s (c) signals of catalysts
Fig. 5 (Color online) XPS spectra and peak fitting results of V 2p3/2 signals of catalysts
3.4 Catalytic activities Catalytic activities of V2O5/TiO2–Al2O3 catalysts with different preparation methods are shown in Table 3. It is apparent that most V2O5/TiO2–Al2O3 catalysts have higher DMM yield than the 15V2O5/TiO2 and 15V2O5/Al2O3 catalysts, implying the positive effects of the mixed TiO2– Al2O3 support. Meanwhile, the activity of V2O5/TiO2– Al2O3-sol catalyst is significantly higher than other V2O5/ TiO2–Al2O3 samples, which reaches a maximum conversion of 48.9 % with a DMM selectivity of 89.9 % at 393 K. The 15V2O5/TiO2–Al2O3-mil and 15V2O5/TiO2– Al2O3-cop samples also have relatively higher methanol conversions, about 43.0 % and 42.7 %, respectively. However, the DMM selectivity of the 15V2O5/TiO2–
123
Al2O3-mil sample is 90.7 %, which is much higher than the DMM selectivity of 15V2O5/TiO2–Al2O3-cop catalyst (77.6 %). The 15V2O5/TiO2–Al2O3-mix sample has much lower methanol conversion (25.1 %) and DMM selectivity (79.7 %) than 15V2O5/TiO2–Al2O3-mil sample, which illustrates the important role of the ball-milling process. The selectivity of DME and FA is low since the reaction temperature is as low as 373 K and is unfavorable for the formation of DME and FA [7]. It is acknowledged that the support prepared via the ball-milling method and the sol– gel method could promote the dispersion of active species and decrease the average particle size of catalyst, which is beneficial for increasing the catalytic activity [28, 29]. Additionally, both methods strengthen the mutual function between TiO2 and Al2O3, increasing weaker acid sites by incorporating TiO2 into Al2O3 and improving the redox capability of vanadates by producing more oxygen vacancies [24]. As the temperature has great influence on the activation of reactants and the performance of catalysts, the reactivity of the 15V2O5/TiO2–Al2O3-sol sample is evaluated as a function of temperature, and the results are listed in Table 4. The conversion increases gradually with the rise of reaction temperature from 373 to 453 K. However, accompanied with the increased conversion, the DMM selectivity presents a tendency of rapid decline. When the temperature reaches 393 K, methanol conversion increases to 48.9 % with a high DMM selectivity of 89.9 %. However, the selectivity of MF and DME increases rapidly with rising temperature, especially for MF, which indicates that high reaction temperatures facilitate the formation of MF and DME instead of DMM. Since methanol oxidation to DMM is an exothermic reaction, low temperature is favorable for the DMM production. The formation of FA and COx is also thermodynamically unfavorable at low temperatures. From our results and previous studies [3, 4, 7], it is not difficult to find that the reaction temperature has
Sci. Bull. Table 2 Peak-fitting results of V 2p3/2 XPS spectra for V2O5/TiO2/Al2O3 catalysts V 2p3/2 BE (eV)
Sample
V(IV)/(V(IV)?V(V))
V(IV) 2p3/2
V(V) 2p3/2
(Area %)
15V2O5/TiO2/Al2O3-mix
515.8
517.3
15.9
15V2O5/TiO2/Al2O3-imp
515.8
517.1
17.2
15V2O5/TiO2–Al2O3-cop
515.6
517.1
24.6
15V2O5/TiO2/Al2O3-mil
515.5
517.0
31.2
15V2O5/TiO2–Al2O3-sol
516.6
517.0
26.3
significant influence on the product distribution. Between 343 and 413 K, DMM is the major product with a selectivity of 68 %–95 %. Between 423 and 453 K, MF becomes the main product with a selectivity of 47 %–86 %. The formation of FA, DME, CO, and CO2 is not observed until the temperature is higher than 473 K.
4 Discussion The catalysts supported on the TiO2–Al2O3 support result in better catalysis performance in methanol oxidation to DMM reaction systems with respect to catalysts supported
over single oxides. As shown in the previous section, the 15V2O5/TiO2–Al2O3-sol sample has a 48.9 % methanol conversion with a DMM selectivity of 89.9 % at 393 K, but the 15V2O5/TiO2 only exhibits a 23.2 % methanol conversion with a DMM selectivity of 86 %, and the 15V2O5/Al2O3 has a 34.6 % methanol conversion with a DMM selectivity of 67.9 %. The comparison of the catalytic performance implies the significant role of mixed TiO2–Al2O3 oxide support for the enhanced activity in the methanol oxidation. Due to the coexistence of TiO2 and Al2O3, the chemical properties of V species on the catalyst are effectively influenced by modifying the interaction between active component and mixed support. Meanwhile, the acid properties of the catalyst have been changed by the interaction of TiO2 with Al2O3, and the catalysts of sol–gel and ball-milling methods have generated more weaker acidic sites that have notable positive effect on DMM selectivity, which is presented in Table 3 and Fig. 3. The mixed TiO2–Al2O3 oxide support makes an important role in the activity of methanol oxidation to DMM through its effect on the acidity of the catalyst and the chemical properties of V species as follows:
4.1 The effect of mixed support on the acidity of the catalysts Fig. 6 (Color online) H2-TPR profiles of catalysts with different preparation methods
The NH3-TPD profiles show that the acidity of mixed oxides is quite different from that of individual component
Table 3 Catalytic activities of V2O5/TiO2–Al2O3 catalysts with different preparation methods Sample
Methanol conversion (%)
Selectivity (%) DMM
MF
DME
FA 0.0
15V2O5/TiO2–Al2O3-mix
25.1
79.7
11.6
8.7
15V2O5/TiO2–Al2O3-imp
32.0
84.9
9.1
6.0
0.0
15V2O5/TiO2–Al2O3-cop
42.7
77.6
13.9
8.2
0.3
15V2O5/TiO2–Al2O3-mil
43.0
90.7
3.4
5.6
0.3
15V2O5/TiO2–Al2O3-sol
48.9
89.9
4.3
5.7
0.1
15V2O5/TiO2
23.2
86.0
10.3
3.7
0.0
15V2O5/Al2O3
34.6
67.9
16.5
14.6
1.0
-1
T = 393 K; feed gas CH3OH/O2/N2 = 1/2.5/7.5 (v:v); gas hourly space velocity (GHSV) = 8,000 h ; reaction duration = 8 h
123
Sci. Bull. Table 4 Catalytic performance at different temperatures for 15V2O5/TiO2–Al2O3-sol Temperature (K)
Methanol conversion (%)
Selectivity (%) DMM
DME
MF
FA
COx
373
39.3
90.4
2.8
6.8
0.0
0.0
393
48.9
89.9
4.3
5.7
0.1
0.0
413
55.5
68.3
6.8
23.8
1.1
0.0
433
59.9
39.4
8.9
46.9
4.8
0.0
453
64.3
18.0
14.3
63.3
3.7
0.7
metal oxide, and such difference could be caused by the excess charges from the formation of Ti–O–Al bonds [20]. The extent of the formation of the bridged hetero metal– oxygen bonds depends on the mixing degree between Ti and Al, which varies with the preparation methods. If proper preparation methods are used, the bonds of Al–O– Al become weaken with the incorporation of Ti into Al2O3 structure, and weaker acidic sites can thus be produced. For example, the catalysts with sol–gel, ball milling, and precipitation methods possess more weaker acidic sites. But the acidity might tend to increase if the interaction between Ti and Al species is so intense that highly bridged Ti–O–Al bonds are formed [13]. As a result, the catalyst with the co-precipitation method has a large amount of middle strong acidic sites. The resulting acidic site strength and distribution significantly affect the product selectivity. The catalytic test results show that the catalysts with more weaker acidic sites, including 15V2O5/ TiO2–Al2O3-sol, 15V2O5/TiO2–Al2O3-mil, and 15V2O5/ TiO2–Al2O3-imp, favor the formation of DMM with a selectivity higher than 85 %. The low DMM selectivity of 15V2O5/TiO2–Al2O3-cop might be due to the large amount of middle strong acid sites, which favors the formation of MF and DME [9]. Although the 15V2O5/ TiO2–Al2O3-imp sample has a comparatively lower methanol conversion (32.0 %) than 15V2O5/TiO2–Al2O3cop sample (42.7 %), the DMM selectivity of 15V2O5/ TiO2–Al2O3-imp (84.9 %) is higher than 15V2O5/TiO2– Al2O3-cop (77.6 %), attributing to the weaker acid sites on the catalyst surface [10]. 4.2 The effect of mixed support on the chemical properties of V species In this work, it is apparent that the V2O5 supported on mixed TiO2–Al2O3 support with different preparation methods show huge differences on the catalytic conversion of methanol to DMM. To some extent, such differences reflect the chemical properties of V species affected by the TiO2–Al2O3 mixed support. In many catalytic systems, the support not only allows for a high dispersion of the active
123
phase, but also plays a role in determining the catalytic activity by interacting with the active phase. From the literature [13, 16, 19], both the V5? and V4? species are the active sites in the oxidation process. However, as V4? could provide more oxygen vacancies for improving the electron-transfer capability, more V4? species lead to better catalytic activity. The redox process of V species between high oxidation state and low oxidation state allows the reactant to be oxidized with the release of lattice oxygen in V species. From the XPS results in Table 2, the sol–gel, ball milling, and co-precipitation methods generate more V4? species for methanol oxidation. The generation of lower valence state of V species is related to the interaction between the active component and the mixed support, which is confirmed by the appearance of traces of V2Ti7O17 phases on these catalysts in Fig. 1 and the shift of Al 2p and Ti 2p binding energy in Fig. 4. The definite degree of interaction of active component and the support is in favor of producing more oxygen vacancies to facilitate the electron transfer between support and active components [23], which leads to stronger redox capability. The catalysts with high redox capability, including 15V2O5/ TiO2–Al2O3-sol, 15V2O5/TiO2–Al2O3-mil, and 15V2O5/ TiO2–Al2O3-cop, favor the methanol oxidation with conversions higher than 43 %. 4.3 The nature of the V2O5/TiO2–Al2O3 catalyst with different preparation methods The metal oxide–support interaction cannot behave well without specific support structure. Different support structure is generated from the variation in preparation methods of the mixed TiO2–Al2O3 support. The substance of the V2O5/TiO2–Al2O3 catalyst with different preparation method could reflect the metal oxide–support interaction to some extend. The theoretical monolayer capacity of supported TiO2 species on the Al2O3 is about 0.66 mmol/ 100 m-2 Al2O3 (about 0.168 g/g c-Al2O3). Thus, with the given high TiO2 mass content of 0.33 g/g c-Al2O3 (TiO2:Al2O3 = 1:2 wt%), TiO2 might have been crystallized into phases on the Al2O3 support surface or interacted
Sci. Bull.
with Al2O3 [20]. For example, the sol–gel method allows a portion of Ti species to incorporate into Al2O3 lattice to change the electronic properties, as shown in Fig. 5, which further enhances the interaction between the active component and the support. The ball-milling method facilitates the non-equilibrium phase transition of Al2O3 crystal and promotes the interface interaction of TiO2 with Al2O3 companied with the decrease in crystallite size [24]. Additionally, the interaction between the support and active components of sol–gel and ball-milling samples is stronger than other samples, which is helpful to promote the electron-transfer properties, which could improve the redox properties for methanol oxidation. However, the precipitation and mechanical mixing method have little influence on the improvement of the interaction between the support and active component, not to mention the enhancement of redox ability, as confirmed by the XPS profiles in Fig. 4 and the structure of catalysts in Fig. 1. In brief, a proper preparation method would allow one species to incorporate into the lattice of another species, enhancing the interaction between support and active component, changing the electronic properties, and facilitating the electron transfer between support and active component, thereby improving the redox and acidic properties of the V2O5/TiO2–Al2O3 catalyst for methanol oxidation.
species. The V4? could offer more oxygen vacancies to facilitate the electron transfer between the support and active component, thereby improving the redox properties of the V2O5/TiO2–Al2O3 catalyst for increased methanol conversion. Providing weaker acidic sites is enabled by the incorporation of titanium cation into the alumina structure, weakening the Al–O–Al bonds. The resulted weaker acidic sites are favorable for improving the DMM selectivity. In summary, the study emphasizes the role of the mixed support on the chemical properties of active vanadium oxide species and the acidic properties. The coordinated environment of the V chemical state and the mixed TiO2–Al2O3 support is significant for improving the catalytic efficiency on selective methanol oxidation to DMM. Acknowledgments This work was supported by the National Natural Science Foundation of China (21006068, 21222604), the Program for New Century Excellent Talents in University (NCET-100611), Specialized Research Fund for the Doctoral Program of Higher Education (20120032110024), the Scientific Research Foundation for the Returned Overseas Chinese Scholars (MOE), and the Program of Introducing Talents of Discipline to Universities (B06006). Conflict of interest of interest.
The authors declare that they have no conflict
References 5 Conclusions We have studied the catalytic properties of V2O5/TiO2– Al2O3 catalysts for the selective methanol oxidation to DMM. In order to investigate the mechanism of V2O5/TiO2– Al2O3 catalysts in methanol oxidation to DMM, we have explored five different preparation methods and obtained diversified catalytic performance on the V2O5/TiO2–Al2O3 samples. The V2O5/TiO2–Al2O3 samples have superior catalytic properties than the V2O5/TiO2 and V2O5/Al2O3 samples. The catalyst prepared by the sol–gel method has the best catalytic performance with a maximum conversion of 48.9 % and a high DMM selectivity of 89.9 % at 393 K. The catalyst with the ball-milling method has also relatively high methanol conversion (43 %) and DMM selectivity (90.7 %), which is surpassed only by the V2O5/TiO2–Al2O3-sol sample. The excellent catalytic performance of the V2O5/TiO2– Al2O3 samples for menthol oxidation to DMM is attributed to the effectively modified interaction between the active component and the support coupled with the appropriate acidic sites provided by the mixed TiO2–Al2O3 support. Proper preparation methods can further promote the interaction between active V component and the mixed support as well as provide weaker acidic sites. The interaction between active V component and support changes the chemical states of supported active V components, producing more V4?
1. Zhao Y, Qin Z, Wang G et al (2013) Catalytic performance of V2O5/ZrO2–Al2O3 for methanol oxidation. Fuel 104:22–27 2. Thavornprasert KA, Capron M, Jalowiecki-Duhamel L et al (2014) Highly productive iron molybdate mixed oxides and their relevant catalytic properties for direct synthesis of 1,1-dimethoxymethane from methanol. Appl Catal B 145:126–135 3. Nikonova OA, Capron M, Fang G et al (2011) Novel approach to rhenium oxide catalysts for selective oxidation of methanol to DMM. J Catal 279:310–318 4. Zhao H, Bennici S, Shen J et al (2010) Nature of surface sites of catalysts and reactivity in selective oxidation of methanol to dimethoxymethane. J Catal 272:176–189 5. Yuan Y, Iwasawa Y (2002) Performance and characterization of supported rhenium oxide catalysts for selective oxidation of methanol to methylal. J Phys Chem B 106:4441–4449 6. Zhao H, Bennici S, Cai J et al (2010) Effect of vanadia loading on the acidic, redox and catalytic properties of V2O5–TiO2 and V2O5–TiO2/SO42- catalysts for partial oxidation of methanol. Catal Today 152:70–77 7. Kaichev VV, Popova GY, Chesalov YA et al (2014) Selective oxidation of methanol to form dimethoxymethane and methyl formate over a monolayer V2O5/TiO2 catalyst. J Catal 311:59–70 8. Liu H, Iglesia E (2003) Selective one-step synthesis of dimethoxymethane via methanol or dimethyl ether oxidation on H3?nVnMo12-nPO40 Keggin structures. J Phys Chem B 107:10840–10847 9. Chen S, Wang S, Ma X et al (2011) Selective oxidation of methanol to dimethoxymethane over bifunctional VOx/TS-1 catalysts. Chem Commun 47:9345–9347 10. Chen S, Meng Y, Zhao Y et al (2013) Selective oxidation of methanol to dimethoxymethane over mesoporous Al-P-V-O catalysts. AlChE J 59:2587–2593
123
Sci. Bull. 11. Royer S, Se´cordel X, Brandhorst M et al (2008) Amorphous oxide as a novel efficient catalyst for direct selective oxidation of methanol to dimethoxymethane. Chem Commun 7:865–867 12. Gornay J, Se´cordel X, Tesquet G et al (2010) Direct conversion of methanol into 1,1-dimethoxymethane: remarkably high productivity over an FeMo catalyst placed under unusual conditions. Green Chem 12:1722–1725 13. Matralis HK, Ciardelli M, Ruwet M et al (1995) Vanadia catalysts supported on mixed TiO2–Al2O3 supports: effect of composition on the structure and acidity. J Catal 157:368–379 14. Pophal C, Kameda F, Hoshino K et al (1997) Hydrodesulfurization of dibenzothiophene derivatives over TiO2–Al2O3 supported sulfided molybdenum catalyst. Catal Today 39:21– 32 15. Reddy BM, Chowdhury B, Reddy EP et al (2001) X-ray photoelectron spectroscopy study of V2O5 dispersion on a nanosized Al2O3-TiO2 mixed oxide. Langmuir 17:1132–1137 16. Martı´nez S, Morales R, Ca´rdenas-Galindo MG et al (2005) Reducibility, heats of re-oxidation, and structure of vanadia supported on TiO2 and TiO2–Al2O3 supports used as vanadium traps in FCC. Thermochim Acta 434:74–80 17. Goodrow A, Bell AT (2007) A theoretical investigation of the selective oxidation of methanol to formaldehyde on isolated vanadate species supported on silica. J Phys Chem C 111:14753–14761 18. Benjaram M, Biswajit C, Panagiotis G (2001) An XPS study of the dispersion of MoO3 on TiO2–ZrO2, TiO2–SiO2, TiO2–Al2O3, SiO2–ZrO2, and SiO2–TiO2–ZrO2 mixed oxides. Appl Catal A 211:19–30 19. Kim T, Wachs IE (2008) CH3OH oxidation over well-defined supported V2O5/Al2O3 catalysts: influence of vanadium oxide loading and surface vanadium-oxygen functionalities. J Catal 255:197–205
123
20. Linacero R, Rojas-Cervantes M, Lopez-Gonzalez JDD (2000) Preparation of xTiO2(1-x)Al2O3 catalytic supports by the sol– gel method: physical and structural characterisation. J Mater Sci 35:3279–3287 21. Lapina OB, Shubin AA, Nosov AV et al (1999) Characterization of V2O5-TiO2 catalysts prepared by milling by ESR and solid state 1H and 51V NMR. J Phys Chem B 103:7599–7606 22. Galan CA, Ortiz AL, Guiberteau F et al (2009) Crystallite size refinement of ZrB2 by high-energy ball milling. J Am Ceram Soc 92:3114–3117 23. Habel D, Goerke O, Tovar M et al (2008) Phase relations in the system TiO2–V2Ox under oxidizing and reducing conditions. J Phase Equilib Diff 29:482–487 24. Ghosh J, Bysakh S, Mazumder S (2014) Multiphase transformation and hybrid nanostructure under non-equilibrium and equilibrium condition during high-energy ball milling of BaTiO3 powders. Phase Transit 87:325–335 25. Gao X, Bare SR, Fierro JLG et al (1999) Structural characteristics and reactivity/reducibility properties of dispersed and bilayered V2Ox/TiO2/SiO2 Catalysts. J Phys Chem B 103:618–629 26. Chenakin S, Prada Silvy R, Kruse N (2005) Effect of X-rays on the surface chemical state of Al2O3, V2O5, and aluminovanadate oxide. J Phys Chem B 109:14611–14618 27. Zhu HQ, Qin ZF, Shan WJ et al (2004) Pd/CeO2-TiO2 catalyst for CO oxidation at low temperature: a TPR study with H2 and CO as reducing agents. J Catal 225:267–277 28. Coste S, Bertrand G, Coddet C et al (2007) High-energy ball milling of Al2O3–TiO2 powders. J Alloys Compd 434–435:489–492 29. Trudeau M, Schulz R, Dussault D et al (1990) Structural changes during high-energy ball milling of iron-based amorphous alloys: is high-energy ball milling equivalent to a thermal process? Phys Rev Lett 64:99