A Mesoporous Cobalt Aluminate Spinel Catalyst for

DOI: 10.1002/cctc.201700647

Full Papers

A Mesoporous Cobalt Aluminate Spinel Catalyst for Nonoxidative Propane Dehydrogenation Bo Hu,[a] Wun-Gwi Kim,[a] Taylor P. Sulmonetti,[a] Michele L. Sarazen,[a] Shuai Tan,[a] Jungseob So,[a] Yujun Liu,[b] Ravindra S. Dixit,[b] Sankar Nair,*[a] and Christopher W. Jones*[a] A mesoporous CoAl2O4 spinel (Co-Al) is synthesized by a onestep evaporation-induced self-assembly (EISA) method. N2 physisorption and TEM are used to demonstrate the presence of mesopores within the Co-Al material. The spinel crystal structure of Co-Al, in which Co occupies tetrahedral (Td) sites, is confirmed by using XRD and UV/Vis spectroscopy. In nonoxidative propane dehydrogenation at 550 8C, a propane conversion of approximately 8 % is observed for Co-Al with a > 80 % propylene selectivity, which corresponds to a turnover frequency of 5.1 h@1 based on an estimation of the number of active

Co sites by using NH3 temperature-programmed desorption. A much higher propane conversion rate and a circa 80 % propylene selectivity is observed upon reaction at 600 8C. Continuous deactivation of the catalyst is observed for Co-Al at this elevated temperature. In situ X-ray absorption spectroscopy results suggest that Co remains as a Td Co2 + species under the reaction conditions. The Td Co2 + sites within the Co-Al material are thus proposed to act as Lewis acidic active sites; this acidity is verified using IR spectroscopy with pyridine as a probe molecule.

Introduction Light olefins, for example, ethylene and propylene, are the most important building blocks in the chemical industry for the production of a variety of chemical products.[1] As the production of shale gas, which contains a large amount of ethane and propane, continues to increase,[2] interest in its use as a cheap alternative hydrocarbon source for the production of ethylene and propylene by catalytic dehydrogenation is strong. To date, many of the best alkane dehydrogenation catalysts are based on Cr[3] and Pt,[4] which are also utilized in industrial alkane dehydrogenation processes. However, these catalysts often undergo severe deactivation throughout the catalytic process to result in relatively complicated industrial processes to allow for effective catalyst regeneration. To this end, much work has been directed to optimize the catalytic performance of commercially relevant Cr- and Pt-based catalytic materials. In parallel, a significant amount of work has been directed towards alternative catalytic materials that might exhibit superior activity, selectivity, or stability for alkane dehy[a] Dr. B. Hu, Dr. W.-G. Kim, T. P. Sulmonetti, Dr. M. L. Sarazen, Dr. S. Tan, J. So, Prof. S. Nair, Prof. C. W. Jones School of Chemical & Biomolecular Engineering Georgia Institute of Technology 311 Ferst Dr., Atlanta, GA 30332 (USA) Fax: (+ 1) 404-894-2866 E-mail: [email protected] [email protected] [b] Dr. Y. Liu, Dr. R. S. Dixit Engineering & Process Sciences The Dow Chemical Company Freeport, TX 77541 (USA) Supporting information and the ORCID identification number(s) for the author(s) of this article can be found under: https://doi.org/10.1002/ cctc.201700647.

ChemCatChem 2017, 9, 3330 – 3337

drogenation. With a focus on propane dehydrogenation, noble metal alloys,[5] noble-metal-promoted oxides,[6] binary/ternary Group IIIA metal oxides,[7] acidic zeolites,[8] isolated transition metal ions on oxide supports,[9] and coordinatively unsaturated Zr on La-modified ZrO2[10] are examples of promising catalytic materials reported recently. Among these catalytic materials, catalysts based on 3d transition metals have been explored widely in alkane dehydrogenation because of their versatile chemical reactivity, earth abundance, and relatively low toxicity. Recently, framework Fe2 + in ZSM-5 was found by Yun and Lobo to provide a 78–85 % propylene selectivity at a modest propane conversion (ca. 3.6 %).[11] Furthermore, a family of active Fe catalysts was described recently by us for which a good activity and high propylene selectivity was obtained by modifying the Al2O3 support with phosphate species.[12] Similarly, the sulfation of Al2O3 was found by Li et al. to be a viable strategy to improve the activity and selectivity of supported transition metals, for example, Fe and Co, for propane dehydrogenation.[13] Silica-supported isolated transition metal ions (that is, Co, Fe) were found by Hock et al. to be catalytically active and selective for propane dehydrogenation.[9b, e] A similar catalytic dehydrogenation activity was also reported by Cop8ret et al. for a highly dispersed Cr/SiO2 catalyst.[3b] Herein, we report the synthesis and characterization of a mesoporous CoAl2O4 spinel (Co-Al) as an active and selective catalyst for propane dehydrogenation between 550 and 600 8C. The Co-Al catalyst was characterized by using XRD, N2 physisorption, and TEM to demonstrate its mesoporous spinel structure. The local coordination environment of Co was found to be tetrahedral (Td) by using UV/Vis and X-ray absorption spectroscopy (XAS). We used in situ XAS to indicate that Co-Al catalyzes propane dehydrogenation through a non-redox pro-

3330

T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers cess, presumably in which Co acts as a Lewis acid. The Lewis acidic character of Td Co2 + was verified by using IR spectroscopy with pyridine as a probe, and the number of surface acid sites was measured by using NH3 temperature-programmed desorption (TPD) for the calculation of the turnover frequency (TOF).

Results and Discussion The as-prepared CoAl2O4 catalyst was found to have a Co loading of 11.1 wt % by using inductively coupled plasma optical emission spectroscopy (ICP-OES) to give a formula of 0.3 CoAl2O4·Al2O3 in which 0.3 is the molar ratio of the CoAl2O4 phase to the Al2O3 phase. We used XRD to probe the structure of Co-Al. The wide-angle (2 q = 10–708) XRD pattern of mesoporous Co-Al is shown in Figure 1. The peaks at 2 q = 31.2, 36.7, 44.8, 59.3, and 65.28 (marked with black dots) matched the pattern of a reference spinel CoAl2O4 sample from Alfa Aesar, which indicates the existence of a spinel cobalt aluminate structure in the Co-Al catalyst. No peaks related to other crystalline phases were observed. However, no XRD peaks were found in the small-angle XRD region (2 q = 0.6–38), which suggests that there was no ordered mesoporous structure within this material.

Figure 2. a) N2 isotherm and b) pore size distribution of the Co-Al catalyst.

Figure 1. XRD patterns of the mesoporous Co-Al catalyst and commercial CoAl2O4.

The porosity of the Co-Al catalyst was characterized by using N2 physisorption. The Co-Al catalyst had a type IV adsorption/desorption isotherm (Figure 2 a), which suggests that it possesses mesopores. The average pore size was approximately 8 nm, and the mesopore volume was 0.3 cm3 g@1 (Figure 2 b; Supporting Information, Table S1), consistent with the hypothesized mesoporous structure of the Co-Al catalyst. The porous Co-Al catalyst had a surface area of 138 m2 g@1 based on BET calculations. ChemCatChem 2017, 9, 3330 – 3337

www.chemcatchem.org

The porosity of the Co-Al catalyst was further characterized by using TEM. The mesopores of Co-Al could be visualized clearly in the TEM images (Figure 3), and the pore size was approximately 5–7 nm. This is consistent with the pore size distribution measured by using N2 physisorption (Figure 2 b). However, some TEM images of different sample areas showed nonporous Co-Al phases as well (Supporting Information, Figure S1), which explains the moderate BET surface area obtained by using N2 physisorption. The crystalline planes were analyzed in the highlighted square region shown in Figure 3, and electron diffraction patterns of the (111), (220), and (311) planes were recorded (Supporting Information, Figure S2). The observation of the (220) and (311) crystalline planes is consistent with the peaks in the XRD pattern at 2 q = 31.2 and 36.78. Scanning tunneling electron microscopy with energy-dispersive Xray spectroscopy (STEM-EDX) was performed to analyze the mesoporous parts of the samples, and the Co loading was found to be 12.5 wt % (Supporting Information, Figure S3), which suggests that CoAl2O4 also exists in the porous domains.

3331


Full Papers

Figure 4. UV/Vis spectra of the Co-Al catalyst and a reference commercial CoAl2O4 sample.

Figure 3. (Top and bottom) TEM images at different magnifications that demonstrate the presence of mesopores in the Co-Al material.

As it is a late transition metal, Co, which has different oxidation states and coordination geometries, will exhibit characteristic UV/Vis signals. Thus, UV/Vis spectroscopy was applied to characterize the Co-Al catalyst to give insight into the chemical structure of the Co species. The mesoporous Co-Al catalyst exhibited similar UV/Vis features as the reference CoAl2O4 spinel in the range of l = 200–800 nm (Figure 4). Three characteristic peaks were observed at l = 551, 586, and 623 nm and they were assigned to the 4A2(F)!4T1(P) d–d transition of Td Co2 + , which corresponds to the “cobalt blue” color of the sample.[14] The peak at l = 482 nm with a weak intensity was assigned to ChemCatChem 2017, 9, 3330 – 3337

www.chemcatchem.org

the spin-forbidden d–d transition of Td Co2 + .[14e, 15] No other peaks related to other Co species were found in the spectrum, which indicated that Co-Al possesses a normal spinel structure in which Co2 + occupied Td sites and Al3 + was in the octahedral (Oh) sites. The catalytic activity and selectivity of the Co-Al catalyst were assessed using 5 % propane/N2 with a total flow rate of 20 mL min@1 at 550 and 600 8C. At such high temperatures, propane can also be converted into methane and ethylene by thermal cracking or to carbonaceous coke, which are undesired side reactions. These byproducts increase the cost of downstream separation and catalyst regeneration. Thus, a high propylene selectivity is desired for a well-performing propane dehydrogenation catalyst. The Co-Al catalyst provides a constant propane conversion of 8–10 % with a selectivity above 80 % at 550 8C (except for the first data point; Figure 5). No loss of activity and selectivity was found under the specific reaction conditions for approximately 5 h on stream. The carbon balance was above 98 % over the course of the test, which indicates a low level of coke formation over the Co-Al catalyst under such conditions. At the higher temperature of 600 8C, a much higher propane conversion was observed. The maximum propane conversion was 39 % during the initial stage of the reaction (Figure 6 a). The conversion decreased gradually to 15 % after approximately 5 h under the reaction conditions. Other than a high catalytic activity, Co-Al also exhibited a reasonably high propylene selectivity at 600 8C. The selectivity of propylene remained at around 80 % at 600 8C with a greater than 95 % carbon balance (Figure 6 b). At 600 8C, thermal cracking is not negligible and it may lead to a slightly lower propylene selectivity compared to that at 550 8C. Furthermore, at the high propane conversion observed at 600 8C, the high concentration of propylene may result in more coke formation, which is a possible reason for the catalyst deactivation observed at 600 8C. The overall observation of the Co-Al catalytic reactivity indicates that the spinel

3332


Full Papers

Figure 6. a) Propane conversion over Co-Al at 600 8C and b) propylene selectivity over Co-Al at 600 8C.

Figure 5. a) Propane conversion over Co-Al at 550 8C and b) propylene selectivity over Co-Al at 550 8C.

CoAl2O4 material could serve as a promising catalyst for alkane dehydrogenation. To obtain an in-depth understanding of the active sites in the Co-Al catalyst during propane dehydrogenation, XAS was performed to characterize the catalyst under both ex situ and in situ conditions. Specifically, the oxidation state and coordination geometry of Co were characterized by using X-ray absorption near edge structure (XANES) and extended X-ray adsorption fine structure (EXAFS), respectively. As the X-ray energy approaches the binding energy of the Co 1s electrons, the 1s electrons can be excited to the unoccupied 3d orbitals to give a pre-edge feature in the XANES spectrum of Co-Al (Figure 7 a). As the 3d orbitals are the valence shell of Co, the pre-edge energy of such a 1s!3d transition is indicative of the oxidation state of Co. By calibrating the adsorption energy with Co foil, the pre-edge energy of Co-Al was found to be 7.7092 keV (Table 1). To determine the oxidation state of Co in the Co-Al catalyst, a series of Co standard samples with differChemCatChem 2017, 9, 3330 – 3337

www.chemcatchem.org

ent oxidation states was also measured, and their pre-edge energies are listed in Table 2. By comparison, the pre-edge energy of Co in the Co-Al material is similar to that of Co2 + standards and is different from that of the Co3 + references, which suggests that Co is in the + 2 oxidation state in the CoAl catalyst. The EXAFS spectrum of Co in the Co-Al material (Figure 7 b) was utilized for fitting to obtain the coordination number and bond distances of the first-shell Co@O. The peak at around 1.5 a (phase uncorrected) is assigned to the first-shell Co@O, and the fitting results of the first-shell peak are listed in Table 3. In air at room temperature, Co was found to have an average coordination number of 4.2 : 0.4, consistent with a Td geometry of Co, as observed by using UV/Vis spectroscopy. The average Co@O bond length was 1.96 a, similar to that found previously.[16] To determine the catalytically active Co species for propane dehydrogenation, the Co-Al catalyst was also characterized by using in situ XANES and EXAFS measurements. Under either a H2 or propane atmosphere at 550 8C for 30 min, the pre-edge

3333


Full Papers Table 3. Fitting results for Co in Co-Al under different conditions (k range = 3.0–12.85; R range = 1–4).[a]

Treatment conditions

CN[b]

RCo@O [a]

s2 V 10@3[c]

DE [eV]

Air, RT H2, 550 8C Propane, 550 8C

4.2 : 0.4 4.2 : 0.4 3.9 : 0.1

1.96 : 0.01 1.96 : 0.01 1.96 : 0.01

7:1 7:1 5:1

1.5 : 1.1 1.7 : 1.2 2.0 : 0.9

[a] The R factors of three fittings are 0.0076, 0.0096, and 0.0058, respectively. [b] CN is the coordination number of the first Co@O shell. [c] s2 is the Debye–Waller factor.

Figure 7. a) Co k-edge XANES spectrum of Co-Al in air at room temperature and b) Co k-edge k2-weighted Fourier-transformed EXAFS spectrum of Co-Al in air at room temperature.

Table 1. Pre-edge energies of Co-Al under different treatment conditions. Treatment conditions

Pre-edge energy [keV]

Oxidation state

Air, RT H2, 550 8C Propane, 550 8C

7.7092 7.7092 7.7091

+2 +2 +2

Table 2. Pre-edge energies of Co standards with different oxidation states. Co standard

Pre-edge energy [keV]

Oxidation state

Co(Ac)2 Co(OH)2 Co(NO3)2·6 H2O CoO Co(acac)3 Co(NH3)6Cl3

7.7095 7.7096 7.7096 7.7096 7.7101 7.7102

+2 +2 +2 +2 +3 +3

energies of Co in the Co-Al catalyst were the same as those of the sample at room temperature in air. No change of oxidation state was observed under in situ conditions. Similar to the oxiChemCatChem 2017, 9, 3330 – 3337

www.chemcatchem.org

dation state, the exposure of the Co-Al catalyst to H2 and propane at 550 8C each yielded a very similar spectrum to that obtained in air at room temperature. The EXAFS fitting results indicate that Co remained coordinated tetrahedrally with an average Co@O bond length of approximately 1.96 a under all conditions. The results obtained by using in situ XAS support the hypothesis that Td Co2 + is the catalytically active species and that such Co species catalyze the reaction through a nonredox process, presumably by acting as a Lewis acid. A similar observation was reported by Iglesia et al. previously for Co/ ZSM-5 as a propane dehydrogenation/dehydrocyclization catalyst in which the exchanged Co2 + ions remained in the + 2 oxidation state up to 500 8C in H2, as indicated by using XANES, and no reduction was found for Co/ZSM5 even up to 1000 8C by using temperature-programmed reduction (TPR).[17] Furthermore, a silica-supported Co catalyst was reported recently by Hock et al. to be catalytically active for propane dehydrogenation by a non-redox mechanism.[9e] Their DFT calculations suggest that the heterolytic cleavage of the C@H bond is the critical step for propane activation by a non-redox dehydrogenation mechanism over isolated Co2 + /SiO2. Recently, a detailed experimental study was reported by Cop8ret et al. to probe the reaction mechanism of propane dehydrogenation over silica-supported isolated Co2 + , and a similar heterolytic cleavage pathway was proposed for the C@H bond activation of propane.[18] Based on previous findings on Co activity towards alkane C@H activation at high temperatures, a similar nonredox mechanism is proposed for our Co-Al catalyst in which Td Co2 + is the catalytically active species that acts as a Lewis acid. Lewis acids, such as Zn2 + or Ga3 + ,[6b, 9c, d, 19] are catalytically active for propane dehydrogenation. Based on the hypothesis that Td Co2 + catalyzed propane dehydrogenation through a non-redox process by acting as a Lewis acid, IR spectroscopy with pyridine as a probe was utilized to probe the acidity of the Co-Al catalyst. The observation of IR peaks at n˜ = 1450, 1492, and 1615 cm@1 is consistent with the existence of Lewis acid sites within the Co-Al catalyst (Figure 8). The weak peak at n˜ = 1575 cm@1 is assigned to adsorbed pyridine species that interact with surface hydroxyl groups. A similar IR spectrum with pyridine as the probe was observed for the commercial CoAl2O4 reference sample (Supporting Information, Figure S7). The number of acid sites was measured by using NH3 temperature-programmed desorption. The Co-Al catalyst exhibited a broad NH3 desorption peak from 100 to 500 8C, which gave

3334


Full Papers and some Al species from the Al2O3 phase in the catalyst, which may overestimate the number of exposed Td Co2 + , leading to an even higher TOF. Nonetheless, the TOF obtained here at least gives a lower bound estimate of the activity and it is similar to the TOF values of some Ga- and Cr-based catalysts published previously.[1, 3b, 9d, 21] Thus, the discovery of the Co-Al catalyst in this study emphasizes the potential of supported first-row transition metals as effective alkane dehydrogenation catalysts and it may contribute to a new strategy for the design of better alkane dehydrogenation catalysts based on earth-abundant metals.

Conclusions

Figure 8. FTIR spectrum of pyridine adsorbed on Co-Al.

the total number of desorbed NH3 species as 228 mmol g@1. Based on the assumptions that NH3 only titrated Co2 + and that the stoichiometry between NH3 and Td Co2 + is 1:1, the turnover frequency (TOF) of propane dehydrogenation over the Co-Al catalyst was calculated to be an average of 5.1 h@1 at 550 8C. This TOF is higher than that of a Co/SiO2 catalyst prepared by the strong electrostatic adsorption of Co(NH3)63 + onto SiO2,[9e] which may be because of support effects (CoAl2O4 for Co-Al vs. SiO2). Recently, the catalyst support has been shown to affect the activity and selectivity of Cr-based dehydrogenation catalysts.[20] Although the TOF of a Co/SiO2 catalyst prepared by grafting organometallic precursors[18] was higher than that of Co-Al at 550 8C (from an initial value of 12.5 to 5 h@1 after 10 h), the formation of metallic Co was observed by XANES at 550 8C for Co/SiO2 prepared by grafting molecular precursors, which could be the reason for the deactivation.[18] For the Co-Al catalyst, the Co species remained as Td Co2 + at 550 8C because of the high thermal stability of CoAl2O4, which led to a stable activity and a high propylene selectivity. At 600 8C, a much higher initial TOF was observed and a maximum TOF of 23.0 h@1 was obtained. Under those conditions, the deactivation of Co-Al at 600 8C led to a final TOF of 8.7 h@1 after approximately 5 h on stream. All the Co species should be Td with a + 2 oxidation state in CoAl2O4 if the material adopts a normal spinel structure. XRD was used to determine that the crystal structure of the Co domain in Co-Al is spinel. UV/Vis and XAS spectra further indicated that Co is Td with a + 2 oxidation state, consistent with the normal spinel structure. These data are consistent with Td Co2 + as the only Co species that exists in this catalytic material. Based on the porosity and TEM results that suggest that some dense domains without mesopores are present in this material, it is believed that the majority of Lewis acidic Td Co2 + sites may be buried in the bulk of the material rather than on the surface, which leads to a relatively small fraction of Co sites titrated by using NH3 TPD (12 % of the total Co). However, we cannot exclude the possibly of interactions between NH3 ChemCatChem 2017, 9, 3330 – 3337

www.chemcatchem.org

We describe a mesoporous CoAl2O4 (Co-Al) material as an active and selective catalyst for nonoxidative propane dehydrogenation at 550 and 600 8C. The Co-Al catalyst provided a steady-state propane conversion of approximately 8 % at 550 8C with a > 80 % propylene selectivity, whereas a maximum propane conversion of approximately 39 % was observed at 600 8C at an identical weight hourly space velocity with a slightly lower propylene selectivity of approximately 80 %. The deactivation of the Co-Al catalyst was observed during the reaction at 600 8C. The coordination geometry of Co was revealed to be tetrahedral with a + 2 oxidation state by using Xray absorption spectroscopy and UV/Vis spectroscopy. Tetrahedral Co2 + was proposed to be the active site, which acts as a Lewis acid, as no significant change of oxidation state and coordination environment was observed for Co under the reaction conditions by using in situ X-ray absorption spectroscopy. The significant Lewis acidity and minimal Brønsted acidity of the material were further demonstrated by using IR spectroscopy with pyridine as the probe. We utilized NH3 temperatureprogrammed desorption to measure the number of exposed, acidic Co sites, and the turnover frequency was calculated to be 5.1 h@1 at 550 8C and up to 23.0 h@1 at 600 8C. It is suggested that the catalytic performance can be further improved by increasing the amount of exposed Co species by introducing more mesopores into the material by optimizing the catalyst synthesis.

Experimental Section General information: Cobalt nitrate hexahydrate (Co(NO3)2·6 H2O, ACS reagent, + 98 %), cobalt(II) acetate (Co(Ac)2), cobalt(II) oxide (CoO), cobalt(II) hydroxide (Co(OH)2), cobalt(III) acetylacetonate (Co(acac)3), hexaaminecobalt(III) chloride (Co(NH3)6Cl3), aluminum isopropoxide (Al(OiPr)3, + 98 %), Pluronic P-123 (average Mn & 5800), and nitric acid (HNO3, ACS reagent, 70 %) were purchased from Sigma Aldrich. Bulk CoAl2O4 as the reference sample was purchased from Alfa Aesar. Ethanol (KOPTEC, 200 Proof) was purchased from VWR. All chemicals were used as received without further purification. Synthesis of Co-Al: The Co-based catalyst was synthesized by a one-step EISA method.[14f, 22] Briefly, P123 (2.0 g) was dissolved in ethanol (40 mL) and concentrated nitric acid (3.2 mL). Then, a predetermined amount of aluminum isopropoxide (4.09 g) and cobalt nitrate (0.44 g) were added, and the solution was stirred vigorously

3335


Full Papers for 6 h at RT. The reaction mixture was then heated to 60 8C to remove the solvent without stirring. The resulting solid material was calcined at 700 8C with a ramp rate of 1 8C min@1 in air for 5 h to obtain CoAl2O4, denoted as Co-Al. Elemental analysis demonstrated the material to be 11.1 wt % Co. Catalytic measurements: The propane dehydrogenation reaction was performed by using a U-shape quartz reactor of approximately 1/8-inch diameter and 80 cm length. Typically, 0.2 g catalyst was loaded onto a quartz wool bed in air at RT, and the quartz reactor with the catalyst was loaded into a Techne FB-08 Al2O3 fluidized bath at 200 8C. The catalyst was then purged with N2 (Airgas, UHP) for 10 min at 200 8C with a flow rate of 20 mL min@1. Then catalyst was purged with H2 (Airgas, UHP) for another 10 min at 200 8C with a flow rate of 20 mL min@1. The reactor was then heated to the desired temperature (550 or 600 8C) under 20 mL min@1 H2 flow. Once the reactor reached the desired temperature, the H2 flow was maintained for 10 min at a flow rate of 20 mL min@1. Next, the sample was purged with 20 mL min@1 N2 for at least 5 min. Then the feed gas was switched to a flow of 20 mL min@1 of 5 vol % C3H8 balanced with N2. The hydrocarbon products and H2 were analyzed by using online GC (Shimadzu GC-2014) equipped with a RESTEK column (Rt-Alumina BOND/Na2SO4, 30 m V 0.25 mm V 4 mm), a flame ionization detector, and a thermal conductivity detector. The first data point was collected after the feed gas had flowed for 10 min. The carbon mass balance was calculated for every time point collected and had an average deviation from closure of 5 % or less. After the reaction, the inlet and the outlet of the quartz reactor were sealed quickly, and then the reactor was removed from the fluidized bath to cool to RT. Once the catalyst had cooled to RT, it was removed from the reactor and stored in a glass vial for additional characterization. The conversion and atomic selectivity were calculated as follows [Eqs. (1)–(3)]: C3 H8 conversion ð%Þ ¼

C3 H8,in @C3 H8,out > 100 % C3 H8,in

Atomic selectivity ð%Þ ¼

Yi X i,out > > 100 % 3 C3 H8,in @C3 H8,out

no: of carbonout > 100 % no: of carbonin P Y i ? X i,out ¼ > 100 % 3 ? C3 H8,in

ð1Þ ð2Þ

Carbon balance ð%Þ¼

ð3Þ

in which Xi,out refers to the concentration of product i (that is, CH4, C2H6, C2H4, C3H6) and Yi is the number of carbon atoms in the product i molecule. Catalyst characterization: XRD patterns were measured by using an X’pert Pro X-ray diffractometer from PANalytical with Ni-filtered CuKa radiation (l = 1.5418 a) with generator settings of 45 kV and 40 mA. A scanning region of 10–708 with a scanning step size of 0.028 was used. Textural properties were determined from the N2 adsorption–desorption isotherms measured by using a Micromeritics Tristar II 3020. Approximately 0.1 g of sample was degassed under vacuum and preheated at 120 8C for 12 h by using a Schlenk line before the measurement. Elemental analysis for Co within the catalysts was performed by using ICP-OES provided by ALS Environmental. NH3 chemisorption was performed by using a Micromeritics Autochem II Chemisorption Analyzer to titrate the number of acid sites. Approximately 50 mg of sample was preheated to 550 8C with a ramp of 5 8C min@1 under a 20 mL H2/Ar flow and held at 550 8C for 1 h. Then the sample was cooled to 100 8C in a 20 mL He flow with a ramp rate of 20 8C min@1. After exposure to ChemCatChem 2017, 9, 3330 – 3337

www.chemcatchem.org

NH3 at 100 8C for 60 min and purging with He at 100 8C for another 60 min, the NH3-adsorbed sample was heated to 800 8C with a ramp of 10 8C min@1 while recording the signal of desorbed NH3. High-resolution TEM was performed by using a FEI Tecnai G2 F30 Transmission Electron Microscope operated at 300 kV. To prepare samples for imaging, a well-dispersed catalyst suspension in H2O was dropped onto a Lacey carbon-coated copper grid and dried in ambient air for sample preparation. Co K-edge (7709.0 eV) XAS spectra were acquired by using the bending magnet station 12-BM-B at the Advanced Photon Source (Argonne National Laboratory) with a Rh focusing mirror and Si(111) monochromator.[23] A Co foil and a third X-ray detector were placed in the beam path beyond the transmission detector to allow the acquisition of a reference spectrum concurrent with each sample measurement. The use of a custom-built reactor cell with a six-port sample holder allowed multiple samples to be pretreated at up to 550 8C under flowing air, 4 % H2 in He, or 3 % propane in He (Airgas) and maintained under a static pretreatment atmosphere as they were cooled to RT and examined by using XAS without exposure to air. To facilitate data analysis, XANES reference spectra were collected of Co(Ac)2, CoO, Co(OH)2, Co(NO3)2·6 H2O, Co(acac)3, and Co(NH3)6Cl3. The XAS data were processed using Athena software for background removal, postedge normalization, and XANES analysis. EXAFS was analyzed using Artemis software. Pyridine adsorption followed by IR spectroscopy was performed by using a Thermo-Nicolet 8700 FTIR spectrometer with a MCT/A detector. Every spectrum took 64 scans at a resolution of 4 cm@1. The KBr was first pressurized into a wafer, and the catalyst was placed on the top and bottom on the wafer. The prepared KBr wafer with the catalyst was repressurized to fix the catalyst on the surface of the wafer, which was mounted in a custom-built vacuum chamber with a ZnSe window. The activation of the catalyst wafer was conducted at 450 8C under less than 10@6 mbar for 1 h, and a background spectrum was collected after the temperature was reduced to 250 8C. Then pyridine was introduced into the chamber at a pressure of 0.1 mbar of pyridine for 30 min. A spectrum was collected after 30 min of the evacuation of the chamber.

Acknowledgements This work was financially supported by the Dow Chemical Company. XAS studies were conducted using resources of the Advanced Photon Source, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC0206CH11357. The authors especially thank Dr. Guanghui Zhang (Purdue University), Dr. Carlo U. Segre (Illinois Institute of Technology), and Dr. Sungsik Lee (Argonne National Laboratory) for fruitful discussions.

Conflict of interest The authors declare no conflict of interest. Keywords: cobalt · dehydrogenation · mesoporous materials · self-assembly · spinel phases

3336

[1] J. J. H. B. Sattler, J. Ruiz-Martinez, E. Santillan-Jimenez, B. M. Weckhuysen, Chem. Rev. 2014, 114, 10613 – 10653.


Full Papers [2] a) E. McFarland, Science 2012, 338, 340 – 342; b) T. Ren, M. Patel, K. Blok, Energy 2006, 31, 425 – 451. [3] a) M. Alc#ntara-Rodr&guez, E. Rodriguez-Castellon, A. Jimenez-Lopez, Langmuir 1999, 15, 1115 – 1120; b) M. P. Conley, M. F. Delley, F. NﬄÇezZarur, A. Comas-Vives, C. Cop8ret, Inorg. Chem. 2015, 54, 5065 – 5078; c) K. H. Kang, T. H. Kim, W. C. Choi, Y.-K. Park, U. G. Hong, D. S. Park, C.-J. Kim, I. K. Song, Catal. Commun. 2015, 72, 68 – 72; d) P. P. Li, W. Z. Lang, K. Xia, L. Luan, X. Yan, Y. J. Guo, Appl. Catal. A 2016, 522, 172 – 179. [4] a) X. Fan, J. Li, Z. Zhao, Y. Wei, J. Liu, A. Duan, G. Jiang, Catal. Sci. Technol. 2015, 5, 339 – 350; b) L. L. Long, W. Z. Lang, X. Yan, K. Xia, Y. J. Guo, Fuel Process. Technol. 2016, 146, 48 – 55; c) E. A. Redekop, V. V. Galvita, H. Poelman, V. Bliznuk, C. Detavernier, G. B. Marin, ACS Catal. 2014, 4, 1812 – 1824; d) K. Xia, W. Z. Lang, P. P. Li, L. L. Long, X. Yan, Y. J. Guo, Chem. Eng. J. 2016, 284, 1068 – 1079; e) Y. Zhang, Y. Zhou, S. Zhang, S. Zhou, X. Sheng, Q. Wang, C. Zhang, J. Mater. Sci. 2015, 50, 6457 – 6468; f) H. Zhu, D. H. Anjum, Q. Wang, E. Abou-Hamad, L. Emsley, H. Dong, P. Laveille, L. Li, A. K. Samal, J.-M. Basset, J. Catal. 2014, 320, 52 – 62. [5] a) D. J. Childers, N. M. Schweitzer, S. M. K. Shahari, R. M. Rioux, J. T. Miller, R. J. Meyer, J. Catal. 2014, 318, 75 – 84; b) J. R. Gallagher, D. J. Childers, H. Y. Zhao, R. E. Winans, R. J. Meyer, J. T. Miller, Phys. Chem. Chem. Phys. 2015, 17, 28144 – 28153; c) Z. Wu, E. C. Wegener, H.-T. Tseng, J. R. Gallagher, J. W. Harris, R. E. Diaz, Y. Ren, F. H. Ribeiro, J. T. Miller, Catal. Sci. Technol. 2016, 6, 6965 – 6976. [6] a) G. Liu, L. Zeng, Z. J. Zhao, H. Tian, T. F. Wu, J. L. Gong, ACS Catal. 2016, 6, 2158 – 2162; b) J. J. H. B. Sattler, I. D. Gonzalez-Jimenez, L. Luo, B. A. Stears, A. Malek, D. G. Barton, B. A. Kilos, M. P. Kaminsky, T. W. G. M. Verhoeven, E. J. Koers, M. Baldus, B. M. Weckhuysen, Angew. Chem. Int. Ed. 2014, 53, 9251 – 9256; Angew. Chem. 2014, 126, 9405 – 9410. [7] a) S. Tan, L. B. Gil, N. Subramanian, D. S. Sholl, S. Nair, C. W. Jones, J. S. Moore, Y. J. Liu, R. S. Dixit, J. G. Pendergast, Appl. Catal. A 2015, 498, 167 – 175; b) S. Tan, S. J. Kim, J. S. Moore, Y. J. Liu, R. S. Dixit, J. G. Pendergast, D. S. Sholl, S. Nair, C. W. Jones, ChemCatChem 2016, 8, 214 – 221. [8] a) R. Gounder, E. Iglesia, J. Catal. 2011, 277, 36 – 45; b) T. Bucˇko, L. Benco, O. Dubay, C. Dellago, J. Hafner, J. Chem. Phys. 2009, 131, 214508; c) R. Gounder, E. Iglesia, J. Am. Chem. Soc. 2009, 131, 1958 – 1971; d) S.-W. Choi, W.-G. Kim, J.-S. So, J. S. Moore, Y. Liu, R. S. Dixit, J. G. Pendergast, C. Sievers, D. S. Sholl, S. Nair, C. W. Jones, J. Catal. 2017, 345, 113 – 123. [9] a) U. Das, G. H. Zhang, B. Hu, A. S. Hock, P. C. Redfern, J. T. Miller, L. A. Curtiss, ACS Catal. 2015, 5, 7177 – 7185; b) B. Hu, N. M. Schweitzer, G. H. Zhang, S. J. Kraft, D. J. Childers, M. P. Lanci, J. T. Miller, A. S. Hock, ACS Catal. 2015, 5, 3494 – 3503; c) N. M. Schweitzer, B. Hu, U. Das, H. Kim, J. Greeley, L. A. Curtiss, P. C. Stair, J. T. Miller, A. S. Hock, ACS Catal. 2014, 4, 1091 – 1098; d) A. Getsoian, U. Das, J. Camacho-Bunquin, G. H. Zhang, J. R. Gallagher, B. Hu, S. Cheah, J. A. Schaidle, D. A. Ruddy, J. E. Hensley, T. R. Krause, L. A. Curtiss, J. T. Miller, A. S. Hock, Catal. Sci. Technol. 2016, 6, 6339 – 6353; e) B. Hu, A. Getsoian, N. M. Schweitzer, U. Das, H. Kim, J. Niklas, O. Poluektov, L. A. Curtiss, P. C. Stair, J. T. Miller, A. S. Hock, J. Catal. 2015, 322, 24 – 37. [10] T. Otroshchenko, S. Sokolov, M. Stoyanova, V. A. Kondratenko, U. Rodemerck, D. Linke, E. V. Kondratenko, Angew. Chem. Int. Ed. 2015, 54, 15880 – 15883; Angew. Chem. 2015, 127, 16107 – 16111. [11] J. H. Yun, R. F. Lobo, J. Catal. 2014, 312, 263 – 270.

ChemCatChem 2017, 9, 3330 – 3337

www.chemcatchem.org

[12] S. Tan, B. Hu, W. G. Kim, S. H. Pang, J. S. Moore, Y. J. Liu, R. S. Dixit, J. G. Pendergast, D. S. Sholl, S. Nair, C. W. Jones, ACS Catal. 2016, 6, 5673 – 5683. [13] a) Y. A. Sun, Y. M. Wu, H. H. Shan, G. W. Wang, C. Y. Li, Catal. Sci. Technol. 2015, 5, 1290 – 1298; b) Y. N. Sun, Y. N. Gao, Y. M. Wu, H. H. Shan, G. W. Wang, C. Y. Li, Catal. Commun. 2015, 60, 42 – 45. [14] a) N. El Habra, L. Crociani, C. Sada, P. Zanella, M. Casarin, G. Rossetto, G. Carta, G. Paolucci, Chem. Mater. 2007, 19, 3381 – 3386; b) M. Gaudon, A. Apheceixborde, M. Menetrier, A. Le Nestour, A. Demourgues, Inorg. Chem. 2009, 48, 9085 – 9091; c) F. Meyer, R. Hempelmann, S. Mathur, M. Veith, J. Mater. Chem. 1999, 9, 1755 – 1763; d) U. Lavrencˇicˇ Sˇtangar, B. Orel, M. Krajnc, J. Sol-Gel Sci. Technol. 2003, 26, 771 – 775; e) M. Zayat, D. Levy, Chem. Mater. 2000, 12, 2763 – 2769; f) L. L. Xu, J. Zhang, F. G. Wang, K. D. Yuan, L. J. Wang, K. Wu, G. Q. Xu, W. Chen, RSC Adv. 2015, 5, 48256 – 48268. [15] M. Taguchi, T. Nakane, K. Hashi, S. Ohki, T. Shimizu, Y. Sakka, A. Matsushita, H. Abe, T. Funazukuri, T. Naka, Dalton Trans. 2013, 42, 7167 – 7176. [16] a) F. Bosi, U. Halenius, V. D’Ippolito, G. B. Andreozzi, Am. Mineral. 2012, 97, 1834 – 1840; b) C. Maurizio, N. El Habra, G. Rossetto, M. Merlini, E. Cattaruzza, L. Pandolfo, M. Casarin, Chem. Mater. 2010, 22, 1933 – 1942. [17] W. Li, S. Y. Yu, G. D. Meitzner, E. Iglesia, J. Phys. Chem. B 2001, 105, 1176 – 1184. [18] D. P. Estes, G. Siddiqi, F. Allouche, K. V. Kovtunov, O. V. Safonova, A. L. Trigub, I. V. Koptyug, C. Cop8ret, J. Am. Chem. Soc. 2016, 138, 14987 – 14997. [19] a) Z. X. Chen, A. Derking, W. Koot, M. P. van Dijk, J. Catal. 1996, 161, 730 – 741; b) N. S. Nesterenko, O. A. Ponomoreva, V. V. Yuschenko, I. I. Ivanova, F. Testa, F. Di Renzo, F. Fajula, Appl. Catal. A 2003, 254, 261 – 272; c) M. Saito, S. Watanabe, I. Takahara, M. Inaba, K. Murata, Catal. Lett. 2003, 89, 213 – 217; d) B. J. Xu, B. Zheng, W. M. Hua, Y. H. Yue, Z. Gao, J. Catal. 2006, 239, 470 – 477; e) M. Chen, J. Xu, F. Z. Su, Y. M. Liu, Y. Cao, H. Y. He, K. N. Fan, J. Catal. 2008, 256, 293 – 300; f) Y. Liu, Z. H. Li, J. Lu, K. N. Fan, J. Phys. Chem. C 2008, 112, 20382 – 20392; g) C. Coperet, Chem. Rev. 2010, 110, 656 – 680. [20] M. F. Delley, M.-C. Silaghi, F. NuÇez-Zarur, K. V. Kovtunov, O. G. Salnikov, D. P. Estes, I. V. Koptyug, A. Comas-Vives, C. Cop8ret, Organometallics 2017, 36, 234 – 244. [21] V. J. Cybulskis, S. U. Pradhan, J. J. Lovjn-Quintana, A. S. Hock, B. Hu, G. Zhang, W. N. Delgass, F. H. Ribeiro, J. T. Miller, Catal. Lett. 2017, 147, 1252 – 1262. [22] a) S. M. Morris, P. F. Fulvio, M. Jaroniec, J. Am. Chem. Soc. 2008, 130, 15210 – 15216; b) Q. Yuan, A. X. Yin, C. Luo, L. D. Sun, Y. W. Zhang, W. T. Duan, H. C. Liu, C. H. Yan, J. Am. Chem. Soc. 2008, 130, 3465 – 3472. [23] M. A. Beno, M. Engbretson, G. Jennings, G. S. Knapp, J. Linton, C. Kurtz, U. Rutt, P. A. Montano, Nucl. Instrum. Methods Phys. Res. Sect. A 2001, 467, 699 – 702.

Manuscript received: April 16, 2017 Revised manuscript received: May 17, 2017 Accepted manuscript online: May 18, 2017 Version of record online: August 7, 2017

3337
