Journal of Catalysis 385 (2020) 176–182

Contents lists available at ScienceDirect

Journal of Catalysis

journal homepage: www.elsevier .com/locate / jcat

Oxidative dehydrogenation of propane using layered borosilicate zeoliteas the active and selective catalyst

https://doi.org/10.1016/j.jcat.2020.03.0210021-9517/� 2020 Elsevier Inc. All rights reserved.

⇑ Corresponding author.E-mail address: anhuilu@dlut.edu.cn (A.-H. Lu).

Bin Qiu, Fan Jiang, Wen-Duo Lu, Bing Yan, Wen-Cui Li, Zhen-Chao Zhao, An-Hui Lu ⇑State Key Laboratory of Fine Chemicals, School of Chemical Engineering, Dalian University of Technology, Dalian 116024, PR China

a r t i c l e i n f o

Article history:Received 11 January 2020Revised 12 March 2020Accepted 18 March 2020

Keywords:Borosilicate zeolitePropaneOxidative dehydrogenationBoron species

a b s t r a c t

Oxidative dehydrogenation (ODH) of propane to propylene offers a promising large-scale alternative todirect dehydrogenation. The reported metal oxides catalysts generally cause over-oxidation of propyleneto COx, thus hindering the commercialization of ODH processes. In this study, a layered borosilicate zeo-lite was demonstrated as highly active and selective catalyst for the oxidative dehydrogenation of pro-pane, showing a propylene selectivity of 83.6% and the total light olefins (propylene and ethylene)selectivity of 91.2% at a propane conversion of 3.9%. Further, at a propane conversion of 15.6%, the catalystexhibited a propylene selectivity of 80.4% and the total light olefins selectivity of 91.6% at 530 �C.Structural characterization revealed that the abundance of defective trigonal boron species (B[3]a andB[3]b) was responsible for the activity of propane ODH.

� 2020 Elsevier Inc. All rights reserved.

1. Introduction

Oxidative dehydrogenation (ODH) of propane to propyleneoffers a promising large-scale alternative to direct dehydrogena-tion. The major advantages of ODH are the absence of equilibriumlimitations on propane conversion and the suppression of catalystcoking [1–3]. Supported metal oxides [4–7], especially those basedon VOx as the active sites, have been extensively studied aspropane ODH catalysts in the past decades [8,9]. However, thesecatalysts generally cause over-oxidation of propylene to COx

[10–12], which hinders the commercialization of ODH processes.Recently, several catalysts based on defective hexagonal boron

nitride for the ODH of light alkanes have been reported. This newfamily of catalysts shows superior selectivity to olefins and lowCO2 formation [13–19]. Our previous work revealed that oxygen-containing boron species at the surface of boron nitride are criticalto their superior catalytic performance [15,19]. Hermans and co-workers proposed that amorphous oxidized boron layer wereresponsible for high activity in boron nitride [20] and the sameconclusion was obtained in the supported B/SiO2 catalysts [21].Inspired by the previous studies, besides boron nitride, thermallystable silicon boride has also been proved as the active catalystfor ODH [22]. Nevertheless, these catalysts are usually havinglow surface area. For heterogeneous catalysis, high surface areacatalyst usually provides more active sites. Hence, the design of

new catalysts with well dispersed boron using high surface areasupports is desirable to further improve the catalytic performance.

Zeolites are one of the most common catalyst supports, andwidely used in industry due to its large internal surface areas,micropores of molecular dimension, and high thermal stabilities[23,24]. In this study, we demonstrated that layered borosilicatezeolite (ERB-1, an MWW borosilicate zeolite) [25] was an activeand selective catalyst for ODH of propane, showing a propyleneselectivity of 80.4% at a propane conversion of 15.6%. Noticeably,the total selectivity to light olefins can reach 91.6% at 530 �C. Thecatalytic active sites were attributed to the abundant of defectivetrigonal boron structures.

2. Experimental methods

2.1. Catalyst preparation

The MWW borosilicate lamellar precursor was synthesizedhydrothermally according to earlier report, with slight modifica-tions [25]. First, 8.67 g hexamethyleneimine (HMI) was dissolvedin 15.3 g deionized water with strong stirring at room temperature.Then, boric acid was added and the mixture was stirred for 1 h.Finally, silica sol (9.2 g) was added and then the mixture was stir-red for another hour to obtain a gel with a molar composition of1SiO2:xB2O3:1.4HMI:19H2O (x = 0.3, 0.5 and 1.0). The gel wassealed into a 50 mL Teflon-lined stainless steel autoclave and sub-jected to hydrothermal treatment at 175 �C under tumbling condi-tion for 7 days. After cooling down to room temperature, the

B. Qiu et al. / Journal of Catalysis 385 (2020) 176–182 177

product was filtered, washed, dried and finally calcined at 550 �C inair for 6 h to remove the templating agent. The materials aredenoted BZEO-1, BZEO-2 and BZEO-3, accordingly with boron con-tent 1.54 wt%, 2.48 wt% and 4.88 wt%. The used catalyst BZEO-2(20 h-spent, 2.40 wt% B) was treated through washing and filteringwith 600 mL deionized water, to obtain the sample denoted BZEO-2-W of which the boron content is determined as 0.85 wt%.

2.2. Catalyst characterization

The actual boron content of the zeolites before and after varioustreatments was measured by inductively coupled plasma opticalemission spectroscopy (ICP-OES) on Optima2000DV. Before themeasurements, the sample (10 mg) was dissolved by hydrochloricacid (3 mL) and hydrofluoric acid (3 mL) mixed solution in a teflon-lined autoclave at 150 �C for 1.5 h. After evaporating up the solu-tion that contained hydrochloric acid and hydrofluoric acid, werepeatedly added deionized water into the teflon container andthen collected the solution into a polypropylene volumetric flask,finally fixed the solution volume as 25 mL.

Powder X-ray diffraction (XRD) was recorded on PANalyticalX’Pert3 Powder diffractometer using Cu Ka radiation (k = 0.15406 nm). The zeolite powder was placed inside a quartz-glass sampleholder for testing. The tube voltage was 40 kV, and the current was40 mA.

N2 adsorption-desorption isotherms were measured with anASAP 2020 sorption analyzer (Micromeritics). Prior to the mea-surement, the sample was degassed by evacuation at 200 �C for4 h. The Brunauer-Emmett-Teller (BET) method was used to calcu-late the specific surface area (SBET). Total pore volume (Vtotal) wascalculated from the amount of gas adsorbed at a relative pressureP/P0 of 0.99. Micropore volume (Vmicro) was calculated using thet-plot method.

Transmission electron microscopy (TEM) images were recordedon a Hitachi HT7700, operating at an accelerating voltage of120 kV. The scanning electron microscopy investigation were car-ried out with Hitachi FESEM SU8220 instrument.

IR spectra were recorded on a Nicolet 6700 FT-IR spectrometerequipped with mercury cadmium telluride (MCT) detector.

11B MAS NMR spectra were recorded on Agilent DD2-500 MHzspectrometer with an 11.7T magnet and Bruker AVANCE III600 MHz spectrometer with a 14.1T magnet, using a 4-mm MASNMR probe with a spinning rate of 10 kHz. Quantitative 1D MAS11B one pulse NMR spectra acquired at 14.1T were obtained witha small flip angle of p/12 at corresponding 11B rf field of about36 kHz with a recycle delay of 2 s [26]. Two-dimensional (2D)11B multiple-quantum (MQ) MAS NMR experiments were per-formed on Agilent DD2-500 MHz spectrometer using a three-pulse sequence incorporating a z-filter at a spinning speed of10 kHz. The spectra were acquired with 1024 scans per increment,a recycle delay of 1 s. Chemical shifts were referenced to a 1 MH3BO3 aqueous solution at 19.6 ppm. The DMFIT programwas usedto deconvolute the spectra and fit the peaks [27]. The peak areasrepresented the amount of nuclear spins, i.e., the amount of corre-sponding B species; therefore, the relative amount of B species ineach spectrum could be derived. We assumed that all B specieswere observed in the 11B MAS NMR, and the amount of each spe-cies could be obtained for each sample with a given B contentderived from ICP-OES. Considering the fact that under the ODHconditions, water molecules can be formed in situ, indicating thatthe environment of the catalysts located were not free of water.Therefore, when we measured the NMR spectra, we took our sam-ples (that were stored after reactions under inert atmosphere) andpacked them into NMR rotors, and then conducted the measure-ments at room temperature.

2.3. Catalytic tests of propane and ethane oxidative dehydrogenation

Selective oxidation of propane and ethane was studied in a fixedbed reactor (I. D. = 8 mm, length = 42 cm) packed with 200 mg cat-alyst (without diluent, 40–60 mesh) and heated to 500–590 �Cunder atmospheric pressure. Quartz wool was put in the middleof the reactor tube to support the catalysts. A thermocouple wasinserted into the center of the catalyst bed to control the reactiontemperature. The feed gas contains C3H8/O2/N2 with a volume ratioof 1:1.5:3.5 and C2H6/O2/N2 with a volume ratio of 1:1:4 at a fixedtotal flow rate of 48 mL/min. Reactants and products were ana-lyzed using an online gas chromatograph (Techcomp, GC 7900)equipped with a GDX-102 and molecular sieve 5A column. A TCDwas used to detect O2, N2, C3H8, C3H6, C2H6, C2H4, CH4, CO, andCO2. Conversion was defined as the number of moles of carbonconverted divided by the number of moles of carbon present inthe feed. Selectivity was defined as the number of moles of carbonin the product divided by the number of moles of carbon reacted.The turnover frequency (TOF) was calculated based on the hypoth-esis that B[3]a and B[3]b were the active sites. The quantity of B[3]a

and B[3]b species was calculated from the results of 11B MAS NMRspectra and ICP-OES. The carbon balance was checked by compar-ing the number of moles of carbon in the outlet stream to the num-ber of moles of carbon in the feed. Under our typical evaluatingconditions, the carbon balance was generally higher than 95%.The blank experiment was conducted using 200 mg quartz sandunder the same reaction conditions.

3. Results and discussions

3.1. Structure and property of MWW borosilicate

Layered borosilicate zeolites were synthesized. Fig. 1a–cshowed the SEM images of BZEO-1, BZEO-2 and BZEO-3 catalysts,and the BZEO zeolites were composed of intergrown, randomly-oriented, flaky crystals with particles of about 1–2 lm in lengthand around 100 nm in thickness. TEM image further revealed thatthe lamellar zeolite consisted of rectilinear sheets with a clear crys-tallized structure (Fig. 1d).

As seen in Fig. 2a, the X-ray diffraction (XRD) patterns showedfour main diffraction peaks at 2h = 7.21, 8.02, 10.10 and 26.33�, cor-responding to the known (1 0 0), (1 0 1), (1 0 2) and (3 1 0) reflec-tions of ERB-1, which confirmed the MWW structure [28]. The N2

adsorption-desorption isotherms of different boron-containingMWW-typed zeolites exhibited nearly type-I shape (Fig. 2b), fea-turing predominantly micropores with a small proportion of inter-stitial mesopores. Detailed information was listed in Table S1. Thepresence of boron in the framework was confirmed by the FT-IRspectra (Fig. 2c, Fig. S1). The bands at 1405 and 926 cm�1 wereassigned to trigonally and tetrahedrally coordinated boron respec-tively [29,30], making it clear that the heteroatom boron insertedinto the framework of zeolites.

3.2. Catalytic performance of MWW borosilicate

The catalytic performance of the layered borosilicate zeolites inpropane ODH was evaluated using a reaction gas consisting ofC3H8, O2 and N2 in ratios of 1:1.5:3.5, at a total WHSV of 4.7gC3H8 gcat�1h�1. The propane conversions of 5.6%, 9.3% and 30.6% atabout 520 �C were corresponded to BZEO-1, BZEO-2 and BZEO-3respectively, together with the boron content increased from1.54 wt% to 4.88 wt% (Fig. 3a). The results turned out that higherconversion was obtained at higher boron content under identicalreaction condition. While BZEO-3 with more boron content exhib-ited higher propane conversion than BZEO-2 and BZEO-1 at same

Fig. 1. SEM images of (a) BZEO-1, (b) BZEO-2 and (c) BZEO-3 and (d) TEM image of BZEO-2.

Fig. 2. (a) XRD patterns, (b) N2 adsorption-desorption isotherms and (c) FT-IR spectra of boron contained MWW-typed zeolites. N2 sorption isotherm of BZEO-1 was verticallyshifted an additional 30 cm3 g�1 (STP) for easy view.

178 B. Qiu et al. / Journal of Catalysis 385 (2020) 176–182

reaction temperature, its activity degraded after long term running(Fig. 3b). The XRD pattern of the used BZEO-3 catalyst showed thatthe crystallinity was declined to some extent (Fig. S2). Taking intoaccount the fact that the boron content of BZEO-3 before and afterreaction remained almost identical, the deactivation process ofBZEO-3 was probably caused by the declined zeolite structure.The apparent activation energies of propane over BZEO-1, BZEO-2and BZEO-3 catalysts were of 212, 204 and 205 kJ/mol, respectively(Fig. 3c). These values were quite similar indicating that identicalreaction pathway involving same active sites occurred on thesecatalysts. Therefore, BZEO-2 was used for further study.

Fig. 3d shows the temperature dependence of propane conver-sion over BZEO-2 sample. As expected, higher temperatures affordhigher propane conversions. For example, as the temperature wasincreased from 500 to 530 �C, the propane conversion increased

from 3.9% to 15.6%. The catalyst also showed good stability withboth propane conversion and propylene selectivity remaining con-stant (~30% and ~70%, respectively) for over 20 h at 540 �C (Fig. 2b).The XRD patterns of the used catalysts still retained the typicalMWW structure under the ODH conditions for 20 h (Fig. S2) andthe morphology of the used catalyst was almost unchanged afterexposure to the reaction conditions for 20 h (Fig. S3). The majorproducts of propane ODH consist of C3H6, C2H4, and CO, with verysmall amount of CO2 (<1%). The selectivity towards the desiredproduct, propylene, decreases slightly with elevated temperatures(Fig. 3d). However, the total selectivity for light olefins (ethylene +propylene, C2-3

= ) remained almost unchanged up to 530 �C. Forexample, the total selectivity to light olefins is 91.2% (propylene:83.6%, ethylene: 7.6%) at 500 �C, and 91.6% (propylene: 80.4%, ethy-lene: 11.2%) at 530 �C. The selectivity of ethylene was measured to

Fig. 3. (a) Dependence of propane conversion on the reaction temperature over BZEO catalysts. (b) Time on stream profiles of propane conversion and propylene selectivityover BZEO-2 (540 �C) and BZEO-3 (515 �C) catalysts. (c) Arrhenius plots for the reaction rate (activation energy, Ea) of C3H8. (d) Effect of reaction temperature on C3H8

conversion and corresponding product distribution of BZEO-2 catalyst. Reaction rates of propane was measured in the range of 470–520 �C with a feed gas of 16.7 vol% C3H8/25.2 vol% O2/N2 at the space velocity of 4.7 gC3H8 gcat�1h�1.

B. Qiu et al. / Journal of Catalysis 385 (2020) 176–182 179

constantly increase at elevated temperature (Fig. 2d). In the case ofethane ODH, the conversion of ethane from 3.5% increased to 37.6%when the temperature from 560 �C reached to 590 �C with theyield of ethylene from 5.1% up to 30.6% (Fig. S4). The layeredborosilicate zeolite catalyze ethane ODH at a slightly higher tem-perature range of 560–590 �C, which is due to more energeticallydemanded activation of primary CAH bonds in ethane.

Moreover, the dependence of reactant partial pressure on thereaction rate was measured to investigate the kinetic behaviors(Fig. 4a, b). The kinetic analysis exhibited that the reaction orderis approximately 0.5 for oxygen and a near second-order depen-dence of propane concentration. Similar kinetic behaviors were

Fig. 4. Kinetic behaviors of propane ODH over the BZEO-2 catalyst.(a) and (b) Dependenthe BZEO-2 catalyst. Reaction conditions: mcat = 50 mg, T = 530 �C, feed: 0–25.2 vol% O2

observed in BNOH which suggested that the BZEO-2 catalyst hasan analogous reaction mechanism [15].

3.3. Identification of the possible active boron species

Benefited from the sensitivity of solid-state 11B MAS NMR spec-troscopy to the local coordination environment of boron, 11B MASand MQ MAS NMR spectra of as-prepared BZEO-2 and after differ-ent reaction intervals were measured and shown in Fig. 5 andFig. S5. All of the one-dimensional 11B MAS NMR spectra showedbroad and overlapped signals due to quadrupolar interaction.These overlapped signals can be effectively discriminated along

ce of propane conversion on O2 and C3H8 partial pressure during propane ODH over, 16.7 vol% C3H8, N2 balance, WHSV = 18.8 gC3H8 gcat�1h�1.

Fig. 5. 11B MAS NMR spectra (14.1 T) of BZEO-2: (a) two-dimensional 11B MQ MAS NMR spectrum of BZEO-2 after 1 h spent. And recorded after: (b) 1 h on-stream, (b) 20 hon-stream, (c) washing the recovered catalyst with water. The top lines represent the experimental spectra, the gray dashed lines represent total analytical simulations of thespectra, and the overlapping individual peaks below represent the features of four types of boron species.

180 B. Qiu et al. / Journal of Catalysis 385 (2020) 176–182

the F1 dimension using two-dimensional 11B MQ MAS NMR. Asshown in Fig. 5a, at least five types of major B species (B[3]a, B[3]b, B[3]c, B[4]a and B[4]b) present in the catalysts. Making furtherefforts to obtain best fit, the one-dimensional 11B MAS NMR spec-tra of these catalysts acquired at magnetic fields of 11.7T and 14.1Twere deconvoluted by the same parameters (diso, CQ and gQ) guidedby the sliced spectra from 11B MQ MAS NMR, and the parameterswere listed in Table S2.

One-dimensional 11B MAS NMR were performed to distinguishthe B species and identify the active sites. Each spectrum can bedeconvoluted into five peaks with chemical shifts at ca.18, 15,11, 0 and �2.8 ppm. The peaks at ~18, ~15 and ~11 ppm wereascribed to tri-coordinated boron which denoted as B[3]a, B[3]b

and B[3]c respectively. Normally, the 11B isotropic chemical shiftof trigonal boron correlates with the number of hydroxyl on Batom, and the peak at ~18 ppm was often identified as B(OH)3[31]. Since the BZEO-2 catalyst was calcined at 550 �C, at this tem-perature B(OH)3 is unstable due to the dehydration reaction. Thus,the B[3]a is probably corresponding to the aggregated BAO unitsoriginated from the condensation of BAOH. Therefore, site B[3]a

was tentatively attributed to the terminal boron species AOBAOH[32]. Sites B[3]b and B[3]c were assigned to frameworkB(OSi)xOH3-x(x=1,2) and B(OSi)3 structure [33,34]. The sharp peaksat around 0 ppm and �2.8 ppm were accordingly assigned to B(OH)2(OSi)2 and B(OSi)4 tetrahedral framework, donated as B[4]a

and B[4]b [35–37]. It can be seen that the relative amounts of boronsites in fresh and spent catalyst at 1 h TOS were similar (Fig. S5).After 20 h reaction, there was no significant change of B[3]c spe-cies, while the presence of B[3]a was increased at the cost of B

[3]b species (Fig. 5c). Considering the fact that propane conversionincreased from 19.9% to 29.8% during reaction period of 1 to 20 h, itcan be inferred that the amount of B[3]a and B[3]b species (defec-tive trigonal boron species) may be correlated with the catalyticactivity of the catalyst in propane ODH reaction.

To further validated the assumption, the spent BZEO-2 after20 h reaction was washed thoroughly with deionized water toremove any soluble species [38,39]. As a consequence, the boroncontent declined from 2.48 wt% to 0.85 wt%. In the 11B MAS NMR(Fig. 5d) spectra, the content of B[3]a and B[3]b signals decreaseddramatically (by ca.72%), whereas the amount of B[4]a and B[4]b

species changed marginally (Table 1). Actually, both the extraframework and incorporated boron species were removable by awashing step [31]. Therefore, it is quite difficult to distinguishwhich boron species belong to extra or incorporated frameworkonly based on the results fromwashing experiment. The significantloss of activity by about 87% after washing implicated that therewas a correlation between the amount of defective trigonal boronspecies and the propane conversion, i.e., in the order of blank(1.4%) < washed BZEO-2 (3.8%) < BZEO-2 after 20 h on-stream(29.8%). To elucidate the origin of active sites, we calculated theTOF of BZEO catalysts at 500 �C based on the fraction of B[3]a

and B[3]b (Fig. S6, Table S3). The TOF value of BZEO-1, BZEO-2and BZEO-3 were 4.3, 4.3 and 3.8 h�1, respectively, indicated theintrinsic catalytic activities for the BZEO catalysts are identicalunder same reaction conditions. The tetrahedral boron can betransformed to trigonal structure with carefully dehydration treat-ment. However, the reaction conditions of ODH process was oper-ated at ambient pressure with the presence of alkane and the

Table 111B MAS NMR signals of BZEO-2 after exposed to various conditions and corresponding activity of ODH of propane.

Catalysts Magnetic field (T) Content of boron speciesa/(mgboron species/gcat) Propane conversionb (%)

B[3]a B[3]b B[3]c B[4]a+b

1 h-spent 14.1 3.01 6.83 14.76 0.20 19.920 h-spent 14.1 7.31 3.18 11.64 1.88 29.8washed 14.1 0.30 2.63 4.60 0.97 3.8blank experiment – – – – – 1.4

a Content of boron species: boron content (by ICP-OES) � ratio of boron species (in Table S2).b Reaction conditions: 200 mg catalyst, 540 �C, gas feed: 16.7 vol% C3H8, 25 vol% O2, N2 balance; total flow rate, 48 mL/min.

Scheme 1. Illustration on the active sites of borosilicate zeolite for propane ODH.

B. Qiu et al. / Journal of Catalysis 385 (2020) 176–182 181

in situ formed water molecules. Besides, the relative amounts of B[4]a+b in the fresh catalysts were much lower (0.2 mgboron species/gcat) compared with that of B[3]a and B[3]b, which significantlyincreased to 1.88 mgboron species/gcat after 20 h running on stream.Being aware of that B[4]a+b did not contribute to the reactivity,while B[3]a+b did, we can then deduce that the transformation fromB[4]a+b to B[3]a+b can be ignored during the constant reaction pro-cess. Thus, we postulate that B[3]a and B[3]b are responsible for theODH activity. The results agreed well that the defective trigonalstructures were the active sites, being consistent with the previousreports that BO3 rather than BO4 species were required for propaneODH to propylene (Scheme 1) [31,40].

4. Conclusion

In summary, an MWW-type zeolite containing boron specieswas demonstrated as a promising catalyst for the ODH of propane.Structural characterization revealed that the abundance of defec-tive trigonal boron species (B[3]a and B[3]b) were correlated withthe activity for propane ODH. This well-defined catalyst systemprovided a platform for further investigation of the relationshipbetween boron coordination environment and ODH performance.

Declaration of Competing Interest

The authors declare that they have no known competing finan-cial interests or personal relationships that could have appearedto influence the work reported in this paper.

Acknowledgements

This study was supported by state key program of National Nat-ural Science Foundation of China (21733002), Joint Sino-GermanResearch Project (21761132011), Cheung Kong Scholars Pro-gramme of China (T2015036).

Appendix A. Supplementary material

Supplementary data to this article can be found online athttps://doi.org/10.1016/j.jcat.2020.03.021.

References

[1] S. Vajda, M.J. Pellin, J.P. Greeley, C.L. Marshall, L.A. Curtiss, G.A. Ballentine, J.W.Elam, S.C. Mucherie, P.C. Redfern, F. Mehmood, P. Zapol, Subnanometreplatinum clusters as highly active and selective catalysts for the oxidativedehydrogenation of propane, Nat. Mater. 8 (2009) 213–216.

[2] Y. Liu, L. Luo, Y. Gao, W. Huang, CeO2 morphology-dependent NbOx-CeO2

interaction, structure and catalytic performance of NbOx/CeO2 catalysts inoxidative dehydrogenation of propane, Appl. Catal. B 197 (2016) 214–221.

[3] T.D. Gould, A.M. Lubers, A.R. Corpuz, A.W. Weimer, J.L. Falconer, J.W. Medlin,Controlling nanoscale properties of supported platinum catalysts throughatomic layer deposition, ACS Catal. 5 (2015) 1344–1352.

[4] B.Y. Jibril, S. Ahmed, Oxidative dehydrogenation of propane over Co, Ni and Momixed oxides/MCM-41 catalysts: effects of intra-and extra-frameworklocations of metals on product distributions, Catal. Commun. 7 (2006) 990–996.

[5] M. Cherian, M.S. Rao, A.M. Hirt, I.E. Wachs, G. Deo, Oxidative dehydrogenationof propane over supported chromia catalysts: influence of oxide supports andchromia loading, J. Catal. 211 (2002) 482–495.

[6] S. Wang, S. Du, W. Tang, S. Hoang, X. Lu, W. Xiao, B. Zhang, J. Weng, E. Schneer,Y. Guo, J. Ding, Mesoporous perovskite nanotube-array enhanced metallic-state platinum dispersion for low temperature propane oxidation,ChemCatChem 10 (2018) 2184–2189.

[7] L. Sun, Y. Chai, W. Dai, G. Wu, N. Guan, L. Li, Oxidative dehydrogenation ofpropane over Pt-Sn/Si-beta catalysts: key role of Pt-Sn interaction, Catal. Sci.Technol. 8 (2018) 3044–3051.

[8] S. Zhang, H. Liu, Insights into the structural requirements for oxidativedehydrogenation of propane on crystalline Mg-VO catalysts, Appl. Catal. A 568(2018) 1–10.

[9] X. Rozanska, R. Fortrie, J. Sauer, Size-dependent catalytic activity of supportedvanadium oxide species: oxidative dehydrogenation of propane, J. Am. Chem.Soc. 136 (2014) 7751–7761.

[10] M.D. Argyle, K. Chen, C. Resini, C. Krebs, A.T. Bell, E. Iglesia, In situ UV-visibleassessment of extent of reduction during oxidation reactions on oxidecatalysts, Chem. Commun. 16 (2003) 2082–2083.

[11] R.P. Singh, M.A. Bañares, G. Deo, Effect of phosphorous modifier on V2O5/TiO2

catalyst: ODH of propane, J. Catal. 233 (2005) 388–398.[12] H. Zhang, S. Cao, Y. Zou, Y.M. Wang, X. Zhou, Y. Shen, X. Zheng, Highly efficient

V-Sb-O/SiO2 catalyst with Sb atom-isolated VOx species for oxidativedehydrogenation of propane to propene, Catal. Commun 45 (2014) 158–161.

[13] R. Huang, B. Zhang, J. Wang, K.H. Wu, W. Shi, Y. Zhang, Y. Liu, A. Zheng, R.Schlogl, D. Su, Direct insight into ethane oxidative dehydrogenation overboron nitrides, ChemCatChem 9 (2017) 3293–3297.

[14] J.T. Grant, C.A. Carrero, F. Goeltl, J. Venegas, P. Mueller, S.P. Burt, S.E. Specht, W.P. McDermott, A. Chieregato, I. Hermans, Selective oxidative dehydrogenationof propane to propene using boron nitride catalysts, Science 354 (2016) 1570–1573.

[15] L. Shi, D. Wang, W. Song, D. Shao, W.-P. Zhang, A.-H. Lu, Edge-hydroxylatedboron nitride for oxidative dehydrogenation of propane to propylene,ChemCatChem 9 (2017) 1788–1793.

[16] Y. Zhou, J. Lin, L. Li, X. Pan, X. Sun, X. Wang, Enhanced performance of boronnitride catalysts with induction period for the oxidative dehydrogenation ofethane to ethylene, J. Catal. 365 (2018) 14–23.

[17] P. Chaturbedy, M. Ahamed, M. Eswaramoorthy, Oxidative dehydrogenation ofpropane over a high surface area boron nitride catalyst: exceptional selectivityfor olefins at high conversion, ACS Omega 3 (2018) 369–374.

[18] J. Tian, J. Lin, M. Xu, S. Wan, J. Lin, Y. Wang, Hexagonal boron nitride catalyst ina fixed-bed reactor for exothermic propane oxidation dehydrogenation, Chem.Eng. Sci. 186 (2018) 142–151.

[19] L. Shi, B. Yan, D. Shao, F. Jiang, D. Wang, A.-H. Lu, Selective oxidativedehydrogenation of ethane to ethylene over a hydroxylated boron nitridecatalyst, Chin. J. Catal. 38 (2017) 389–395.

182 B. Qiu et al. / Journal of Catalysis 385 (2020) 176–182

[20] A.M. Love, B. Thomas, S.E. Specht, M.P. Hanrahan, J.M. Venegas, S.P. Burt, J.T.Grant, M.C. Cendejas, W.P. McDermott, A.J. Rossini, I. Hermans, Probing theTransformation of boron nitride catalysts under oxidative dehydrogenationconditions, J. Am. Chem. Soc. 141 (2019) 182–190.

[21] A.M. Love, M.C. Cendejas, B. Thomas, W.P. McDermott, P. Uchupalanun, C.Kruszynski, S.P. Burt, T. Agbi, A.J. Rossini, I. Hermans, Synthesis andcharacterization of silica-supported boron oxide catalysts for the oxidativedehydrogenation of propane, J. Phys. Chem. C 123 (2019) 27000–27011.

[22] B. Yan, W.-C. Li, A.-H. Lu, Metal-free silicon boride catalyst for oxidativedehydrogenation of light alkanes to olefins with high selectivity and stability,J. Catal. 369 (2019) 296–301.

[23] S. Grundner, M.A. Markovits, G. Li, M. Tromp, E.A. Pidko, E.J. Hensen, A. Jentys,M.S. Sanchez, J.A. Lercher, Single-site trinuclear copper oxygen clusters inmordenite for selective conversion of methane to methanol, Nat. Commun. 6(2015) 7546.

[24] P. Serna, D. Yardimci, J.D. Kistler, B.C. Gates, Formation of supported rhodiumclusters from mononuclear rhodium complexes controlled by the support andligands on rhodium, Phys. Chem. Chem. Phys. 16 (2014) 1262–1270.

[25] R. Millini, G. Perego, W.O. Parker Jr, G. Bellussi, L. Carluccio, Layered structureof ERB-1 microporous borosilicate precursor and its intercalation propertiestowards polar molecules, Micropor. Mater. 4 (1995) 221–230.

[26] D. Massiot, C. Bessada, J.P. Coutures, F. Taulelle, A quantitative study of 27A1MAS NMR in crystalline YAG, J. Magn. Reson. 90 (1990) 231–242.

[27] D. Massiot, F. Fayon, M. Capron, I. King, S. Le Calvé, B. Alonso, J.-O. Durand, B.Bujoli, Z. Gan, G. Hoatson, Modelling one- and two-dimensional solid-stateNMR spectra, Magn. Reson. Chem. 40 (2002) 70–76.

[28] M.K. Rubin, P. Chu, US Patent 4 954 325 (1990).[29] R. Millini, G. Perego, G. Bellussi, Synthesis and characterization of boron-

containing molecular sieves, Top. Catal. 9 (1999) 13–14.[30] M. Sayed, A. Auroux, J.C. Védrine, The effect of boron on ZSM-5 zeolite shape

selectivity and activity II. coincorporation of aluminium and boron in thezeolite lattice, J. Catal. 116 (1989) 1–10.

[31] H. Koller, C. Fild, R.F. Lobo, Variable anchoring of boron in zeolite beta,Micropor. Mesopor. Mat. 79 (2005) 215–224.

[32] W.D. Lu, D. Wang, Z. Zhao, W. Song, W.-C. Li, A.-H. Lu, Supported boron oxidecatalysts for selective and low temperature oxidative dehydrogenation ofpropane, ACS Catal. 9 (2019) 8263–8270.

[33] P.V. Wiper, J. Amelse, L. Mafra, Multinuclear solid-state NMR characterizationof the Brønsted/Lewis acid properties in the BP HAMS-1B (H-[B]-ZSM-5)borosilicate molecular sieve using adsorbed TMPO and TBPO probe molecules,J. Catal. 316 (2014) 240–250.

[34] H.T.T. Tong, H. Koller, Control of Al for B framework substitution in zeolite Betaby counterions, Micropor. Mesopor. Mat. 148 (2012) 80–87.

[35] X. Ouyang, S.J. Hwang, R.C. Runnebaum, D. Xie, Y.J. Wanglee, T. Rea, S.I. Zones,A. Katz, Single-step delamination of a MWW borosilicate layered zeoliteprecursor under mild conditions without surfactant and sonication, J. Am.Chem. Soc. 136 (2014) 1449–1461.

[36] M.A. Sanhoob, O. Muraza, E.N. Shafei, T. Yokoi, K.H. Choi, Steam catalyticcracking of heavy naphtha (C12) to high octane naphtha over B-MFI zeolite,Appl. Catal. B 210 (2017) 432–443.

[37] L. Chen, M. Zhang, Y. Yue, C. Ye, F. Deng, NMR and theoretical studies of boron-modified mordenite, Micropor. Mesopor. Mat. 76 (2004) 151–156.

[38] A. Cichocki, W. Łasocha, M. Michalik, Z. Sawłowicz, M. Bus, Release of boronduring conversion of MFI boralite to ammonium and hydrogen forms, Zeolites10 (1990) 583–587.

[39] T.I. Korányi, J.B. Nagy, Distribution of aluminum and boron in the periodicalbuilding units of boron-containing b zeolites, J. Phys. Chem. B. 110 (2006)14728–14735.

[40] O.V. Buyevskaya, D. Müller, I. Pitsch, M. Baerns, Selective oxidative conversionof propane to olefins and oxygenates on boria-containing catalysts, Stud. Surf.Sci. Catal. 119 (1998) 671–676.