Home Journals Articles Updates About
CN
image
收藏切换
Comparative study of CO2 hydrogenation to methanol on cubic bixbyite-type and rhombohedral corundum-type indium oxide
收藏切换
PDF
Bin Yanga, Longtai Lia, Ziye Jiaa, Xiping Liub, Chunjie Zhangb, Limin Guo*, a
Chinese Chemical Letters | 2020, 31(10) : 2627 - 2633
Less
收藏切换
Chinese Chemical Letters | 2020, 31(10): 2627-2633
• Communication •
Comparative study of CO2 hydrogenation to methanol on cubic bixbyite-type and rhombohedral corundum-type indium oxide
Full
Bin Yanga, Longtai Lia, Ziye Jiaa, Xiping Liub, Chunjie Zhangb, Limin Guo*, a
Affiliations
  • a School of Environmental Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, China
  • b Research Institute of Air Purification Equipment, Shanxi Xinhua Chemical Co., LTD, Taiyuan 03008, China
Published: 2020-10-15 doi: 10.1016/j.cclet.2020.05.031
Outline
收藏切换

Hydrogenation of CO2 to value-added chemicals has attracted much attention all through the world. In2O3 with cubic bixbyite-type (denoted as c-In2O3) is well known for its high CO2 hydrogenation activity and CH3OH selectivity at high temperature. However, the other structure of In2O3 with rhombohedral corundum-type (denoted as rh-In2O3) rarely been investigated as catalyst. Herein, c-In2O3 and rh-In2O3 were prepared and comparatively studied for CO2 hydrogenation. The results indicated that c-In2O3 showed higher CO2 conversion activity than rh-In2O3 due to the impressive reducibility and reactivity. Whereas rh-In2O3 had higher CH3OH selectivity due to weaker CH3OH and stronger CO adsorption on rh-In2O3. Although c-In2O3 and rh-In2O3 catalysts showed different CO2 hydrogenation performance, in-situ diffuse reflectance infrared Fourier transform spectroscopy showed CO2 can be reduced to CO through redox cycling and hydrogenation to CH3OH through formate path.

CO2 hydrogenation  /  In2O3  /  Methanol  /  Structure effect
Bin Yang, Longtai Li, Ziye Jia, Xiping Liu, Chunjie Zhang, Limin Guo. Comparative study of CO2 hydrogenation to methanol on cubic bixbyite-type and rhombohedral corundum-type indium oxide[J]. Chinese Chemical Letters, 2020 , 31 (10) : 2627 -2633 . DOI: 10.1016/j.cclet.2020.05.031
Full Text
Less
CO2 emission caused by human activities results in global warming and ecological issues [1, 2]. The conversion of CO2 into value-added hydrocarbon is a promising way to eliminate CO2 [3]. Hydrogenation of CO2 to methanol (CH3OH) is one of the most promising approaches for its transformation [1, 4]. Moreover, CH3OH can be directly used to synthesize olefin and high value-added hydrocarbons through zeolite catalysis [5-8]. By coupling of methanol-synthesis and methanol-to-hydrocarbons reaction with a bifunctional catalyst can realize the direct conversion of CO2 to hydrocarbon products, e.g., gasoline (C5-C11), lower olefin (C2-C4) and aromatic hydrocarbon [5, 9]. It contains two consecutive processes: methanol synthesis from CO2 through metal-oxide catalysts and methanol to hydrocarbons through zeolites catalysts [10]. It is an essential prerequisite that the reaction temperature of methanol to hydrocarbons should be over 340 ℃ [9, 10], which is thermodynamically restrained for methanol synthesis from CO2 and H2. The CH3OH selectivity of conventionally Cu-based catalysts is lower than 5% over 320 ℃ [5, 7]. Obviously, significant and effective catalysts are required for this strategy. As Oliver Martin reported, the methanol selectivity of In2O3 with cubic bixbyite-type can be tuned up to 100% from 200 ℃ to 300 ℃ and over 25% at 340 ℃, which was closed to thermodynamic limit [11, 12]. Further researches showed In2O3 had emerged as an outstanding catalytic system for CO2 hydrogenation to methanol with highly selectivity under industrially relevant conditions [12].
In2O3 is known as a conductive transparent layer and thin-film transistors [13, 14]. There are two representative crystal structures, namely cubic bixbyite-type (denoted as c-In2O3) and rhombohedral corundum-type (denoted as rh-In2O3) [14]. As the most studied structure, c-In2O3 was easily lost oxygen atoms to form defective surface with oxygen vacancy [11, 15, 16]. These surface sites assist CO2 activation and hydrogenation by stabilizing the key reaction intermediates, such as surface bound formate (HCOO), dioxymethylene (H2COO), and H2CO species; the hydrogenation of the latter gives surface methoxy species, supposedly via the rate-determining step [15, 17]. H2 can re-generate the vacancy which is replenished during methanol formation. This catalytic cycle may play an important role on the formation of methanol from CO2[12, 18, 19]. Due to the excellent CH3OH selectivity of In2O3, CO2 had been successfully hydrogenated to lower olefin through In2O3/SAPO-34 catalyst [10] and aromatic hydrocarbon through In2O3/HZSM-5 catalyst [14]. Except for CO2 hydrogenation, c-In2O3 also can be used for methanol steam reforming, water-gas shift reaction, formaldehyde reforming and so on [20, 21].
However, there is little knowledge about the intrinsic adsorption and catalytic property of rh-In2O3 except the electronic and structural characterization by theoretical calculations [13, 22]. The prevalent scientific interest is the transition between these two phases as a case of ccp (cubic close-packed structure) to hcp (hexagonal closed-packed) transition at high-pressure and high-temperature condition [14, 23, 24]. As a meta-stability structure, rh-In2O3 was considered to be unstable at high temperature and pressure. However, Simon Penner et al. found rh-In2O3 remained stable over 430 ℃ in He or CO2 atmospheres [25]. Thus stability of rh-In2O3 is satisfied for most of catalysis reactions. Rh-In2O3 have been proved that this nano-rhombodedra are terminated by (012) facets due to the geometric arrangement of face-sharing octahedral [26]. There is little information about the properties of surface chemistry, e.g., CO2 hydrogenation activity, adsorption-desorption properties or reaction mechanisms.
Herein, the c-In2O3 and rh-In2O3 were synthesized for comparative study of CO2 hydrogenation to elucidate the structure-activity relationship. The crystal phase of catalysts was characterized by X-ray diffraction (XRD) and shown in Fig. 1a. The diffraction peaks of black line were well matched with JCPDS No. 06-0416, which can be indexed to cubic bixbyite-type structure In2O3 (c-In2O3) [11]. Another diffraction pattern could be assigned to rh-In2O3, which was well matched with JCPDS No. 22-0336 [27]. There were no impurity peaks, indicating the single phase of the product. The average crystal size of c-In2O3 and rh-In2O3 calculated by Scherer equation in Fig. 1a were 19.8 and 24.5 nm, respectively.
During synthetic process, the difference between the catalysts was the solvent, i.e., methanol-based route for rh-In2O3 and water-based route for c-In2O3. As shown in Fig. 1b, In(OH)3 was formed in the water-based route and then decomposed to c-In2O3 after calcination. In contrast, InOOH was formed in the methanol-based route and then rh-In2O3 was obtained afterthermal decomposition of InOOH. During the calcination, the decomposition of In(OH)3 was accompanied by water vapor and toward the stable phase. But for methanol-based route, InOOH with the residual methanol favored a transition coordination sphere of In3+ ions by avoiding fast hydrolysis and condensation, which evolved to the rhombohedral corundum-type [28, 29]. The formation of precursors was critical for the successful synthesis of c-In2O3 and rh-In2O3. The rh-In2O3 was terminatedby (012) plane and thepresence of ammonia/methanol had strongly influence on the surface energy of the exposed facet [26]. However, the thermodynamically stable facet of c-In2O3 (110) face with surface energy of 0.969 J/m2 was formed in the presence of water [11].
As Table S1 (Supporting information) and diagrammatic sketch (Figs. 1c and d) showed, c-In2O3 was Ia space group with lattice constant a = b = c = 10.126 Å and described as an oxygen-vacancy fluorite structure with O2- anions missing in an ordered way [28]. There were two types of In3+ ions in c-In2O3 located at the end of a face-diagonal and a body-diagonal, which occupied octahedral and trigonal prismatic interstices of the O2- anions [28, 30]. The lattice parameters of rh-In2O3 with RC space group were a = b = 5.491 Å and c = 14.526 Å. In comparison with c-In2O3, rh-In2O3 showed different atomic arrangement with a rather regular octahedron, including trigonal bi-prism coordinated In3+ with O2- anions [26]. This structural difference was mainly due to the O2- anions lattice change and In3+ ions shifted from octahedral and trigonal prismatic to trigonal bi-prism sites.
Raman was also conducted to understand the structural difference and the result was shown in Fig. 1e. The representative Raman spectra of the indium oxides (black line) was corresponded to c-In2O3. The peaks at 303, 495 and 629cm-1 were associated with InO6 octahedral, 133 and 363cm-1 were associated with the bending vibrations of In4O4 and stretching vibrations of In-O-In [31]. Rh-In2O3 also exhibited the typical Raman-active phonons. The peaks at 161, 218 and 267cm-1 were assigned to the symmetric stretching vibration of O-In-O, 377cm-1 was the symmetric bending vibration of O-In-O [31]. The other frequencies were caused by the symmetric stretching and symmetric bending vibration of O-In-O [32].
Transmission electron microscopy (TEM) showed the average diameter of c-In2O3 and rh-In2O3 was around 17.9 nm and 24.9 nm, respectively. The summary of particle size distribution was showed in Fig. S1 (Supporting information). High resolution transmission electron microscopy (HRTEM) of c-In2O3 (Fig. S2 in Supporting information) exhibited the characteristic spacing of 0.505, 0.417 and 0.264 nm, corresponding to the (010), (211) and (104) faces of c-In2O3, respectively [33]. The spacing of 0.475 and 0.221 nm of rh-In2O3 (Fig. S2) well matched with the representative lattice distance of (211) and (222) faces of rh-In2O3 [32]. The selected area electron diffraction patterns were presented in Fig. S3 (Supporting information). In both catalysts, the discrete spots indicated well crystallinity and the distinct diffraction rings can be ascribed to many tiny crystallites.
X-ray photoelectron spectroscopy (XPS) was used to understand the surface chemical state and the results were enclosed in Fig. S4 (Supporting information). The binding energy was calibrated using C 1s energy of 285 eV in Fig. S5 (Supporting information). As Fig. S4 showed, the broad and symmetric peak locked at 444.6 eV and 452.2 eV were belonged to In 3d5/2 and 3d3/2 of In2O3 [27, 34]. Negligible binding energy value difference for both catalysts suggested similar surface chemical electronic property. O 1s core level of catalysts was composed of two components and the intensity was determined by curve fitting procedure. 530.1 eV can be assigned to lattice oxygen (O lattice) and 531.7 eV can be assigned to oxygen defect (O defect) [2]. This oxygen defects may be the sub-coordinated In atom that was activated by thermal calcination. The oxygen vacancy concentration was the fraction of surface oxygen atoms adjacent to a defect calculated from the de-convoluted O 1s XPS signal. It revealed that the oxygen vacancy amount of c-In2O3 (24.5%) was more than the amount of rh-In2O3 (21.2%).
The CO2 hydrogenation performance was shown in Fig. 2. Both catalysts showed CO2 catalytic activity and highly methanol selectivity at all tested temperatures. In addition to CH3OH formation, these catalysts also generated CO as a primary byproduct (The detailed products information was enclosed in Fig. S6 in Supporting information). However, CO2 conversion over catalysts under 300 ℃ was lower than 5% and gradually increased with reaction temperature increasing (Fig. 2a). Meanwhile, the selectivity of CH3OH decreased and CO gradually became the dominate product, which was thermodynamically favorable with the increasing reaction temperature. In Fig. S6, the presence of CH4 for c-In2O3 and absence for rh-In2O3 at high reaction temperature indicated the different hydrogenation ability. The hydrocarbon products of rh-In2O3 was merely methanol and c-In2O3 can further hydrogenate CO2 to CH4 at high temperature, suggesting higher hydrogenation ability of c-In2O3.
C-In2O3 showed higher CO2 conversion than rh-In2O3. At 360 ℃(Fig. 2a), CO2 conversion of c-In2O3 was 17.02%, which was about three times higher than rh-In2O3. It was worthwhile mentioned that the specific surface area obtained from N2-sorption (Fig. S7 in Supporting information) was 24.8 m2/g for c-In2O3 and 22.8 m2/g for rh-In2O3. The similar surface area for the two catalysts indicated that the catalytic difference was not the result of surface area. In addition, the rh-In2O3 showed higher methanol selectivity than c-In2O3. At 260 ℃, the methanol selectivity of both catalysts were all over 80% and decreased gradually with the temperature increasing. Especially at 340 ℃, the methanol selectivity of rh-In2O3 maintained approximate 30%. But the value of c-In2O3 was only 20%. As Table S2 showed, the two catalytic processes of methanol synthesis and methanol to hydrocarbon can combined into one integrated process and directly hydrocarbon molecule from CO2 and H2, which need >30% methanol selectivity at the optimized temperature over 340 ℃ [21]. Thus, both c-In2O3 and rh-In2O3 showed wide temperature range for high methanol selectivity and promising catalysts for hydrocarbon production coupling with zeolite catalysts. The apparent activation energy for CO2 conversion was 55.33 and 55.39 kJ/mol over c-In2O3 and rh-In2O3, respectively in Fig. 2b. The quite close activation energy may indicate the same CO2 hydrogenation mechanism, which was further confirmed by the following results of in-situ DRIFTS.
The space-time yield (STY) of methanol over the catalysts was also shown in Fig. 2c. C-In2O3 generated about 3.0 mmol g-1 h-1 CH3OH at 340 ℃, which was higher than 1.8 mmol g-1 h-1 of rh-In2O3. The stability of both catalysts showed slight deactivation after 12 h reaction on stream. The XRD patterns of used catalysts was shown in Fig. S8 (Supporting information). In compared with Fig. 1, the intensities of peaks in both catalysts were decreased. However, the diffraction peaks of both samples were still sharp and intense, indicating their highly crystalline nature. No impurity peaks were observed, confirming the high purity of the used catalysts. Moreover, the transformation between c-In2O3 and rh-In2O3 was not observed, which showed the structural stability under reaction condition. Moreover, the influence of experimental conditions on CH3OH selectivity was also evaluated (Figs. 2d and e), which showed higher gas hourly space velocity (GHSV) and pressure were beneficial for CH3OH selectivity. And rh-In2O3 had higher CH3OH selectivity during all experimental conditions.
Temperature-programmed reduction (TPR) in hydrogen (H2-TPR) was carried out to understand the reducibility of catalysts and the results were shown in Fig. 3a. Prior to the bulk reduction into metallic indium over 580 ℃ (c-In2O3) or 626 ℃ (rh-In2O3), there was one primary reduction peak [11, 17]. The diverse reduction peak between c-In2O3 (212 ℃) and rh-In2O3 (307 ℃) was determined by the reducible oxygen atoms and exposing facets [35]. The reduction peak of rh-In2O3 mainly was associated with lattice oxygen atoms from (012) planes [27]. The reduction peak of c-In2O3 at 212 ℃ was associated with oxygen reduction from (110) planes [22, 23]. Oxygen vacancy in In2O3-based catalysts was considered to be the active sites for CO2 hydrogenation [11]. TPR profiles suggested that the oxygen of c-In2O3 was more active and easy to release than rh-In2O3. Thus, oxygen vacancy was easily generated in c-In2O3, which was linked with the CO2 catalytic activity.
Temperature-programmed desorption (TPD) of CO2 (CO2-TPD) in Fig. 3b was used to understand the interaction between CO2 and In2O3. CO2 adsorbed as bicarbonate bridging two In atoms around oxygen vacancy sites and carbonate formed over oxygen atom [11, 15, 16, 19]. Aside from the peak of physically adsorbed CO2 appeared at approximately 100 ℃, two additional peaks were observed in both catalysts. The desorption peaks at approximate 379 and 454 ℃ in c-In2O3 can be assigned to bicarbonate and carbonate, respectively. In-situ DRIFTS later found the transformation from bicarbonate to carbonate during heating-up. CO2 desorption peaks of rh-In2O3 were appeared at 275 and 325 ℃. Lower desorption temperature of rh-In2O3 suggested weaker basicity and interaction between CO2 and rh-In2O3 than c-In2O3. Moreover, c-In2O3 with a large CO2 desorption area featured higher density of adsorption sites. This strong and high density of basicity sites of c-In2O3 was conducive to enhance the CO2 conversion.
The reason influenced the products selectivity can be explored by TPD experiment by using the main byproducts (CH3OH and CO) as adsorbates 33. Before the TPD measurements, the samples were treated at 300 ℃ and then cooling to room temperature. CH3OH was introduced through Ar bubbling at 60 ℃ and the adsorption procedure was last 2 h. As Fig. 3c showed, CH3OH desorbed from rh-In2O3 was more difficult than c-In2O3 (peak at 384 ℃ versus 250 ℃), suggesting stronger CH3OH adsorption ability of rh-In2O3. In addition, the methanol adsorption area over rh-In2O3 was far larger than that of c-In2O3, which showed the more adsorptive sites for CH3OH over rh-In2O3.
CO-TPD result was shown in Fig. 3d. c-In2O3 showed weak interaction with CO due to the absence of chemical desorption peak at high temperature except the physical desorption peak at about 100 ℃. For rh-In2O3, there was a width desorption peak at about 432 ℃ aside from the physical adsorption peak. CO2 hydrogenation experiments indicated higher CH3OH selectivity of rh-In2O3 than c-In2O3, which could be partially explained by the results of by-products TPD experiment. For c-In2O3, CO was more easily desorbed from the surface than CH3OH due to the weaker interaction between CO and c-In2O3. The rh-In2O3 showed CH3OH desorbed at a lower temperature (384 ℃ and showed in Fig. 3c) than CO (432 ℃ and showed in Fig. 3d), which may be beneficial for CH3OH selectivity improvement over rh-In2O3.
To further investigate the interaction between CO2 and catalysts, in-situ DRIFTS experiment of CO2 adsorption were carried out. It represented the characteristic peaks of adsorbed CO2 on the surface at 40 ℃ and reaction intermediates along with the temperature increasing. There were three strong peaks located at 1356, 1420, 1649 cm-1 and a weak peak at 1227 cm-1 (Fig. 4a and Table S3 in Supporting information). The intensity of these peaks gradually strengthened after 30 min exposure.In detail, the peak at 1649, 1420 and 1227 cm-1 could be identified as asymmetric stretching vibration υas(O—C—O), bending vibration δ(OH) and symmetric stretching vibration υs(O—C—O) of bicarbonate species (HCO3-) (as summarized in Table S1), respectively [36]. The other one at 1356 cm-1 can be assigned to υs(O—C—O) of carbonate (CO32-) species and υas(O—C—O) of carbonate may hide or overlap in the peak shoulder of bicarbonate species (1400-1500 cm-1). The surface of In2O3 contained lots of hydroxyl (-OH group) and easily reacted with CO2 to form bicarbonate species as Scheme S1 (Supporting information) briefly illustrated [15, 37].
Once reached 150 ℃, it can be clearly seen that the vibration of bicarbonate weakened and the vibration of carbonate strengthened due to the υas(O—C—O) at 1501 cm-1 of carbonate appeared. This transformation meant bicarbonate turned to carbonate and was significantly enhanced until bicarbonate disappeared as the temperature up to 300 ℃. Meanwhile, a new peak at 2077 cm-1 that assigned to adsorbed CO species appeared [25]. This CO species suggested CO2 can be directly reduced to CO by In2O3. This result was consistent with the density functional theory (DFT) simulation of previous reports, which indicated CO2 reacted with -OH on the In2O3 surface to form a bicarbonate specie [15, 16]. CO2 gradually turned to CO and leaved O atom quenching the oxygen vacancy, which could be regenerated by hydrogen or thermal treatment, as the temperature increased. The similar adsorption and transformation phenomena could be also found in rh-In2O3(Fig. 4b and Table S3). It was worthwhile mentioned that the intensity of carbonate for rh-In2O3 was much lower and even disappeared at 300 ℃ compared to c-In2O3, which was consistent with the weak interaction between CO2 and rh-In2O3 resulted from CO2-TPD experiments.
There was an urgent need but it was a significant challenge to directly study the reaction between H2 or CO2/H2 and In2O3 by in-situ DRIFTS. Because the interaction of In2O3 with H2 strongly decreased the IR diffuse reflection and disturbed the surface stability (Fig. S9 in Supporting information), which prevented the reliable study by in situ DRIFTS [11, 38, 39]. Thus, the indirect strategy of HCOOH reaction was usually adopted instead. The mechanism of methanol synthesis from CO2 hydrogenation on the In2O3 surface was formate-path as the previous reports, which meant CO2 firstly hydrogenation to formate (HCOO-) and subsequently to methoxy and CH3OH [11, 15, 16, 33]. The interaction of formate and catalysts can be used to indirectly testify CO2 hydrogenation mechanism.
After treatment with formic acid and Ar sweeping at 40 ℃, c-In2O3 showed typical stretching vibration of formate species in Fig. 4c. The characteristic absorption bands at 1619, 1563, 1384 and 1359 cm-1 can be assigned to formate [40]. The band at 1359 cm-1 was υs(O—C—O) vibration and 1384 cm-1 was δ(CH) characteristic vibration of formate species [40]. 1619 and 1563 cm-1 were υas(O—C—O) vibration which indicated different adsorption sections of formate species, i.e., 1619 cm-1 was characteristic of the vibration of unsymmetrical formate and 1563 cm-1 was the symmetrical bidentate formate [40], as illustrated in Scheme S2 (Supporting information).
Further investigation explored the temperature dependence of the surface coverage with unsymmetrical and symmetrical bidentate formate species. At 100 ℃, the adsorption bands of unsymmetrical formate species at 1619 cm-1 decreased. On the contrary, the symmetrical bidentate formate species at 1563 cm-1 enhanced. This opposite change for these formates on the surface could be indicative of the transformation of unsymmetrical to symmetrical bidentate formate species. The unsymmetrical formate was metastable state and easily changed to symmetrical bidentate formate under thermal treatment [41].
When temperature increased to 200 ℃, the unsymmetrical bidentate formate became obscured and symmetrical bidentate formate remained stable. The band at 1171 cm-1 can be assigned to υ(C—O) of methoxy (CH3O) species, 1730 cm-1 assigned to υ(C—O) vibration of aldehyde and the other two bands between 2300–2400 assigned to CO2 gas phase [42, 43]. These new bands became stronger and the HCOO species vanished as the temperature increased to 250 ℃. These new species suggested the formate transformation on c-In2O3, which contained decomposition and hydrogenation reaction. Aldehyde and carbon dioxide were decomposed from formate species. The hydrogen was generated from the reaction between In-H and proton of formic acid gas (Scheme S2) [44]. Methoxyl, which was apparently as the precursor of methanol, may come from the hydrogenation of HCOO species.
As for rh-In2O3, it was easily found the bands were similar to c-In2O3 in spite of some peaks shift in Fig. 4d. The peak position of all bands were summarized in Table S4 (Supporting information). At 250 ℃, the formate species on rh-In2O3 was completely disappeared and even formed the negative spikes, suggesting that formic acid was easily decomposed over rh-In2O3. Overall, the oxygen vacancy on the In2O3 surface assists CO2 activation and hydrogenation and also stabilized the key intermediates involved in methanol formation. In addition, methanol formation replenishes the oxygen vacancy sites whereas H2 helps to generate the vacancies. The cycle between the perfect and defective states of the surface catalyzes the formation of methanol from CO2 hydrogenation [15, 16].
In2O3 showed significant advantage in CH3OH synthesis at high temperature. The active sites of In2O3 catalyst was ascribed to oxygen vacancy sites under reaction conditions. These surface sites assisted CO2 activation and hydrogenation by stabilizing the key reaction intermediates, such as surface bound formate (HCOO), and further hydrogenation to CH3OH. This mechanism suggested that oxygen vacancy sites could be replenished during CO2 hydrogenation. As a reducible oxide, In2O3 was different from conventional binary or ternary metal-metal oxide catalysts. CO2 activation, H2 activation and CH3OH synthesis can be completed by In2O3, not the combination or interface of metal-metal oxide catalysts. This feature successfully restrained the CO formation and enhanced the CH3OH selectivity. Both c-In2O3 and rh-In2O3 catalysts showed catalytic activity and thermal stability at high reaction temperature. And the activation energy and reaction mechanism of CO2 hydrogenation was identical. The results of in-situ DRIFTS showed CO can be directly generated from CO2 on the surface of In2O3. This redox reaction was not sustainable because the oxygen vacancy was gradually vanished by CO2. So the cyclic creation and annihilation of oxygen vacancy was necessary for the continuous methanol formation from CO2 hydrogenation. The intermediate formate was used to indirectly explore the reaction mechanism, which showed formate can decompose to CO2 and hydrogenate to CH3OH. Meanwhile, the absence of CO signal may be ascribed to its very low concentration, indicating the generation of CO from formate can be well restrained on In2O3. The oxygen vacancy on the In2O3 were beneficial for CO2 activation and stabilization of the key intermediates. CH3OH generation replenished the oxygen vacancy and H2 regenerated the vacancy to achieve the catalytic cycle. Thus, the generation of CO and CH3OH on the surface of In2O3 can separate to two pathways that redoxpath reaction from CO2 to CO and formate-path reaction from CO2 to CH3OH. CO2 and H2 favored to form formate and then hydrogenation to CH3OH on the surface of In2O3, suggesting the high CH3OH selectivity.
However, due to the thermodynamic restrictions, CO still was the main product at high temperature. With the higher reducibility and basicity, c-In2O3 showed higher CO2 conversion and CH3OH productivity than rh-In2O3. And the adsorbed CH3OH molecules desorbed at lower temperature than that of the adsorbed CO molecules on rh-In2O3. In contrast, CO was more easily desorbed from the surface than CH3OH on c-In2O3. So the rh-In2O3 showed higher CH3OH selectivity during CO2 hydrogenation compared with the c-In2O3.
Above all, the comparative study of c-In2O3 and rh-In2O3 for CO2 hydrogenation was carried out. C-In2O3 showed higher CO2 conversion activity than rh-In2O3, which was ascribed to the higher reducibility and basicity within c-In2O3 and associated with the generation of the oxygen vacancy. rh-In2O3 showed higher methanol selectivity than c-In2O3. Although different structures showed different CO2 hydrogenation performance, the hydrogenation mechanism was identical for the two catalysts. CO2 can be directly reduced to CO through redox mechanism and CO2 hydrogenation to CH3OH was through formate-path. This work demonstrated the fundamental understanding of the structureactivity relationship for rh-In2O3 and c-In2O3 catalysts and offered some hint for superior catalytic system for CO2 hydrogenation with high CH3OH selectivity over 340 ℃.
Declaration of competing interest
Less
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
Less
The work was financially supported by the National Natural Science Foundation of China (No. 21878116) and Natural Science Foundation of Hubei Province (No. 2019CFA070). The authors thank the Analysis and Testing Center of Huazhong University of Science and Technology for analytical support.
Appendix A. Supplementary data
Less
Supplementary material related to this article can be found, inthe online version, at doi:https://doi.org/10.1016/j.cclet.2020.05.031.
References
Less
[1]
S. Li, Y. Wang, B. Yang, L. Guo, Appl. Catal. A: Gen. 571(2019) 51-60.
[2]
B. Yang, L. Wang, Z. Hua, L. Guo, Ind. Eng. Chem. Res. 58(2019) 9838-9843.
[3]
B. Yang, W. Deng, L. Guo, T. Ishihara, Chin. J. Catal. 41(2020) 1348-1359.
[4]
R.W. Dorner, D.R. Hardy, F.W. Williams, H.D. Willauer, Energy Environ. Sci. 3(2010) 884.
[5]
Z. Li, J. Wang, Y. Qu, et al., ACS Catal. 12(2017) 8544-8548.
[6]
J. Wei, Q. Ge, R. Yao, et al., Nat. Commun. 8(2017) 15174.
[7]
L. Qi, Y. Wei, L. Xu, Z. Liu, ACS Catal. 5(2015) 3973-3982.
[8]
X. Zhu, J.P. Hofmann, B. Mezari, et al., ACS Catal. 6(2016) 2163-2177.
[9]
K. Cheng, B. Gu, X. Liu, et al., Angew. Chem., Int. Ed. 55(2016) 4725-4728.
[10]
P. Gao, S. Dang, S. Li, et al., ACS Catal. 8(2017) 571-578.
[11]
O. Martin, A.J. Martin, C. Mondelli, et al., Angew. Chem., Int. Ed. 55(2016) 6261-6265.
[12]
M.S. Frei, C. Mondelli, R. García-Muelas, et al., Nat. Commun. 10(2019) 3377.
[13]
J.M. Yang, Z.P. Qi, Y.S. Kang, L. Qing, W. Sun, Chin. Chem. Lett. 27(2016) 492-496.
[14]
P. Gao, S. Li, X. Bu, et al., Nat. Chem. 9(2017) 1019-1024.
[15]
A. Tsoukalou, P. Abdala, D. Stoian, et al., J. Am. Chem. Soc. 141(2019) 13497-13505.
[16]
M.F. Bekheet, M.R. Schwarz, S. Lauterbach, et al., Angew. Chem., Int. Ed. 52(2013) 6531-6535.
[17]
A. Gurlo, Angew. Chem. Int. Ed. 49(2010) 5610-5612.
[18]
T. Bielz, H. Lorenz, P. Amann, B. Klötzer, S. Penner, J. Phys. Chem. C 115(2011) 6622-6628.
[19]
J. Wang, H. Wang, P. Hu, Sci. China Chem. 61(2017) 336-343.
[20]
S. Shu, D. Yu, Y. Wang, et al., J. Cryst. Growth 312(2010) 3111-3116.
[21]
E.M. Kock, M. Kogler, C. Zhuo, et al., Phys. Chem. Chem. Phys.19(2017) 19407-19419.
[22]
J. Ye, C. Liu, D. Mei, Q. Ge, ACS Catal. 3(2013) 1296-1306.
[23]
J. Ye, C. Liu, Q. Ge, J. Phys. Chem. C 116(2012) 7817-7825.
[24]
N. Rui, Z. Wang, K. Sun, et al., Appl. Catal. B 218(2017) 488-497.
[25]
M. Neumann, D. Teschner, A. Knop-Gericke, W. Reschetilowski, M. Armbrüster, J. Catal. 340(2016) 49-59.
[26]
J. Ye, Q. Ge, C. Liu, Chem. Eng. Sci. 135(2015) 193-201.
[27]
A. Gurlo, S. Lauterbach, G. Miehe, H. Kleebe, R. Riedel, J. Phys. Chem. C 112(2008) 9209-9213.
[28]
K. Reyes-Gil, E. Reyes-García, D. Raftery, J. Phys. Chem. C 111(2007) 14579-14588.
[29]
M. Epifani, P. Siciliano, A. Gurlo, N. Barsan, U. Weimar, J. Am. Chem. Soc. 126(2004) 4078-4079.
[30]
A. Gurlo, G. Miehe, R. Riedel, Chem. Commun. (Camb.) 19(2009) 2747-2749.
[31]
A. Gurlo, P. Kroll, R. Riedel, Chem. Eur. J. 14(2008) 3306-3310.
[32]
J.A. Sans, R. Vilaplana, D. Errandonea, et al., Nanotechnology 28(2017)205701.
[33]
D. Yu, S. Yu, S. Zhang, et al., Adv. Funct. Mater. 13(2003) 497-501.
[34]
M.S. Frei, M. Capdevila-Cortada, R. García-Muelas, et al., J. Catal. 361(2018) 313-321.
[35]
W. Yin, D. Esposito, S. Yang, et al., J. Phys. Chem. C 114(2010) 13234-13240.
[36]
F. Liao, Y. Huang, J. Ge, et al., Angew. Chem., Int. Ed. 50(2011) 2162-2165.
[37]
J.C. Matsubu, S. Zhang, L. DeRita, et al., Nat. Chem. 9(2017) 120-127.
[38]
T. Bielz, H. Lorenz, W. Jochum, et al., J. Phys. Chem. C 114(2010) 9022-9029.
[39]
E.M. Köck, M. Kogler, M. Grünbacher, et al., J. Phys. Chem. C 120(2016) 15272-15281.
[40]
G. Popova, T. Andrushkevich, Y. Chesalov, E. Stoyanov, React. Kinet. Catal. Lett. 41(2000) 805-811.
[41]
S. Kattel, B. Yan, Y. Yang, J.G. Chen, P. Liu, J. Am. Chem. Soc. 138(2016) 12440-12450.
[42]
S.V. Ryazantsev, V.I. Feldman, L. Khriachtchev, J. Am. Chem. Soc. 139(2017) 9551-9557.
[43]
K. Larmier, W.C. Liao, S. Tada, et al., Angew. Chem., Int. Ed. 56(2017) 2318-2323.
[44]
S.T. Teklemichael, M.D. McCluskey, J. Phys. Chem. C 116(2012) 17248-17251.
Appendix
Less
Year 2020 volume 31 Issue 10
PDF
37
20
Cite this Article
BibTeX
Article Info
doi: 10.1016/j.cclet.2020.05.031
  • Receive Date：2020-04-15
  • Online Date：2026-01-31
  • Published：2020-10-15
Article Data
Affiliations
History
  • Received：2020-04-15
  • Revised：2020-05-09
  • Accepted：2020-05-21
Affiliations
    a School of Environmental Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, China
    b Research Institute of Air Purification Equipment, Shanxi Xinhua Chemical Co., LTD, Taiyuan 03008, China
References
Share
https://castjournals.cast.org.cn/joweb/ccl/EN/10.1016/j.cclet.2020.05.031
Share to
QR

Scan QR to access full text

Cite this article
BibTeX
Citations
表12种不同金属材料的力学参数
科
Family		 属数
Number of
genus		 种数
Number of
species		 占总种数比例
Percentage of
total species (%)		 属
Genus		 种数
Number of
species		 占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏
  • BibTeX
  • EndNote
  • RefWorks
  • TxT
Links
Articles
Latest Articles
Most Read
Collections
Updates
Events
News
Multimedia
About
About Us
Contact
No. 86 Xueyuan South Road, Haidian District, Beijing
100081
010-62199257
qkjq@cast.org.cn

Copyright © 2025 China Association for Science and Technology. All rights reserved. For all open access content, the relevant licensing terms apply.
Sponsored by the Office of the Leading Group for Cybersecurity and Informatization of CAST, and supported by Science and Technology Review Publishing House