Homogeneous-like Alkyne Selective Hydrogenation Catalyzed by Cationic Nickel Confined in Zeolite

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE1 Mar 2022Homogeneous-like Alkyne Selective Hydrogenation Catalyzed by Cationic Nickel Confined in Zeolite Xin Deng, Ruihao Bai, Yuchao Chai, Zhenpeng Hu, Naijia Guan and Landong Li Xin Deng School of Materials Science and Engineering & National Institute for Advanced Materials, Nankai University, Tianjin 300350 Google Scholar More articles by this author , Ruihao Bai School of Materials Science and Engineering & National Institute for Advanced Materials, Nankai University, Tianjin 300350 Google Scholar More articles by this author , Yuchao Chai School of Materials Science and Engineering & National Institute for Advanced Materials, Nankai University, Tianjin 300350 Google Scholar More articles by this author , Zhenpeng Hu *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] School of Physics, Nankai University, Tianjin 300071 Google Scholar More articles by this author , Naijia Guan School of Materials Science and Engineering & National Institute for Advanced Materials, Nankai University, Tianjin 300350 Google Scholar More articles by this author and Landong Li *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] School of Materials Science and Engineering & National Institute for Advanced Materials, Nankai University, Tianjin 300350 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.021.202100820 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail The selective hydrogenation of alkynes to their corresponding alkenes is an important type of organic transformation, which is currently accomplished by modified palladium catalysts. Herein, we report that coordinatively unsaturated Ni(II) sites confined in faujasite zeolite, that is, [email protected], can efficiently catalyze the selective hydrogenation of various alkynes in both gas- and liquid phase. Spectroscopic and kinetic analyses explicitly reveal that alkyne hydrogenation over [email protected] follows the homogeneous associative mechanism instead of the classic heterogeneous Horiuti–Polanyi mechanism. Density functional theory calculations confirm that the preferential adsorption of alkynes on coordinatively unsaturated Ni(II) sites is crucial to initiate dihydrogen activation for the subsequent hydrogenation. [email protected] represents a true heterogeneous catalyst acting as a homogeneous catalyst and shares the advantages of both heterogeneous and homogeneous catalysis. These findings may shed light on the rational design of robust catalysts and new catalytic routes by linking heterogeneous and homogeneous catalysis. Download figure Download PowerPoint Introduction The selective hydrogenation of alkynes to alkenes is of great significance in the industry of polymers, fine chemicals, and pharmaceuticals.1,2 For instance, the selective hydrogenation of acetylene and propyne into the corresponding olefins is widely used in the purification of pyrolysis gas.1,3,4 As the main precursors for the production of vitamin E, methyl-3-buten-2-ol and linalool are generally obtained via the selective hydrogenation of the relevant alkynols.2,5 Various catalysts have been explored for the selective hydrogenation of alkynes and alkynols, both in homogeneous and heterogeneous systems. Despite the high selectivity and explicit mechanism of homogeneous catalysts, the infeasibility of recovery suppresses their extensive applications. Benefitting from advantages in separation and recycle, heterogeneous catalysts gradually evolve as promising candidates for the selective hydrogenation of alkynes, especially for industrial applications. Among heterogeneous hydrogenation catalysts, palladium appears to be a primary choice due to its intrinsic ability to dissociate dihydrogen.6,7 However, the undesired reactions of overhydrogenation and oligomerization may be induced by the presence of bulk hydride and carbide in the subsurface region of Pd nanoparticles, which are detrimental to product selectivity and lead to catalytic deactivation.8 Consequently, S,9,10 Pb,11 quinoline, or even CO gas12 are widely employed as modifiers in the selective hydrogenation process, aiming to enhance the catalytic performance by covering the corner or edge active sites of Pd nanoparticles. That is, the so-called “site isolation” strategy.13 In fact, the Lindlar catalyst, typically Pd/CaCO3 modified by both Pb and quinoline, has been recognized as a benchmark catalyst in the semihydrogenation industry.14 Intermetallic compounds like PdxGay15 and Al13Fe416 have been explored to exhibit superior catalytic performance compared with their monometallic counterparts. Alternatively, by regulating the synergistic relationship between the active metal site and the support, isolated active sites can be achieved, which are denoted as “supported single-atom catalysts (SACs).” Plenty of SACs with tailored electronic and geometric properties have been successfully designed, for example, Cu-Pd/SiO2,17 Cu/Al2O3,18 Pd1-ND/G,19 Pd/C3N4,20 [email protected],21 and Pd/Fe3O4.22 These catalysts possess a superior ability to dissociate dihydrogen and, accordingly, exhibit good hydrogenation performance. Notably, Pd1/Cu(111),23 Pt1/Cu(111),24 and Pd1/Cu(100)25 are disclosed to be efficient catalysts for selective hydrogenations by taking advantage of hydrogen spillover. Heterogeneous hydrogenation catalysts generally comply with the classic Horiuti–Polanyi mechanism (or so-called dissociative mechanism) (Scheme 1), that is, dihydrogen initially dissociates on metal active sites followed by the successive addition of hydrogen atoms to the adsorbed hydrocarbon species.26 In the Horiuti–Polanyi mechanism, the dissociation of dihydrogen is nearly barrierless on many metals like Pd and Ni,12 which is practically facile for overhydrogenations. In contrast, the associative mechanism containing the cleavage of H–H bonds under the joint effort of adsorbed alkyne and metal site is proposed (Scheme 1),27 which applies to metals with limited reactivity toward dihydrogen dissociation and is widely observed in organometallic chemistry and homogeneous catalysis.28–31 Under the guidance of the associative mechanism, many metal complexes like (IMes)Ag*Rp,32 Ni*Ln,33 and Cp*Ru34 show considerable selectivity to trans-alkenes in the hydrogenation of internal alkynes, which is rarely achieved via the heterogeneous Horiuti–Polanyi mechanism. The associative hydrogenation mechanism is recently disclosed on Ag/Au nanoparticles.27,35 Moreover, it is proved to be feasible on Pd/polymer under the premise of dynamic metal–polymer interaction. In this regard, dihydrogen cannot be activated alone on the Pd surface but can be activated in the presence of coadsorbed acetylene.36 Specifically, the associative mechanism is discovered to be suitable for the heterogeneous ammonia synthesis by both theoretical and experimental evidence.37 According to these results, it is possible to design heterogeneous catalysts for the selective hydrogenation of alkynes via the associative mechanism, utilizing the advantages of both heterogeneous and homogeneous catalysis. Scheme 1 | Selective hydrogenation of alkynes by dihydrogen: heterogeneous Horiuti–Polanyi mechanism versus homogeneous associative mechanism. Download figure Download PowerPoint In our previous work, a series of coordinatively unsaturated cation sites have been successfully confined in faujasite zeolite, that is, [email protected]38 With the cation site as a central ion and zeolite framework as the inorganic ligand, the structure of [email protected] is analogous to organometallic complexes. Surprisingly, [email protected] exhibits prominent performance in the selective hydrogenation of alkynes, following the typical homogeneous associative mechanism. [email protected] is a true heterogeneous catalyst working in the reasonable manner of homogeneous organometallic complexes, bridging heterogeneous and homogeneous catalysis. Experimental Methods Nickel species encapsulated in faujasite zeolite ([email protected]) were synthesized through a direct hydrothermal process.38 Various characterization methods were applied to confirm the structure of [email protected], including powder X-ray diffraction, temperature-programmed reduction, UV–vis spectra, X-ray photoelectron spectroscopy (XPS), X-ray absorption spectroscopy (XAS), and the neutron powder diffraction (NPD) experiments shown in the Supporting Information. The catalytic performance was performed on a fixed-bed. Detailed kinetic measurements are displayed in the Supporting Information. Subsequently, all the structures and simulation processes were performed based on the spin-polarized density functional theory (DFT) as implemented in the Vienna Ab initio Simulation Package (VASP). (All the parameters are shown in the Supporting Information.) Results and Discussion Construction and characterization of [email protected] Following the optimized hydrothermal route in our previous work,38 [email protected] with faujasite topologic structure was synthesized ( Supporting Information Figures S1 and S2). The calcined sample showed a standard type I isotherm with a total surface area of 530 cm3 g−1 ( Supporting Information Figure S3) and an average pore size of ∼7 Å ( Supporting Information Figure S4), implying a typical microporous structure. The Si/Al ratio of [email protected] zeolite was determined to be 3.5 by inductively coupled plasma (ICP) analysis and the 29Si NMR spectrum ( Supporting Information Figure S5). The fine structure of [email protected] was first determined by the NPD patterns and corresponding Rietveld refinement ( Supporting Information Figure S6), which revealed the formation of three-coordinated Ni species sitting in the six-membered rings.38 The hydrogen temperature-programmed reduction profile of [email protected] exhibited a characteristic hydrogen consumption peak centered at high temperature of ∼1010 K ( Supporting Information Figure S7), suggesting the presence of cationic Ni species stable against reduction. For further information on the structure of [email protected], XAS analyses were performed. The X-ray absorption near-edge structure (XANES) spectrum of [email protected] showed a broad white-line peak at 8350 eV (Figure 1a), which was close to that of NiO and suggested the +2 valence state of Ni species in [email protected] This is in line with the Ni 2p3/2 binding energy of 856.4 eV from the XPS ( Supporting Information Figure S8). In the Fourier transforms of the extended X-ray absorption fine structure (EXAFS) spectrum, [email protected] showed a dominant peak at 2.04 Å, corresponding to the first shell of Ni–O (Figure 1b and Supporting Information Table S1). This is also in line with the previous work of Guesmi et al.,39 where the Ni–O bond range lies between 1.97 and 2.05 Å and the coordination number lies in the range of 3∼4 over Ni2+/Y system. Meanwhile, the suppressed second coordination shell of Ni–Si/Al could be observed at 2.90 Å. No scattering peak of Ni–Ni coordination was observed, ruling out the presence of metallic Ni species. The coordination numbers of the center Ni atoms with surrounding O atoms and Si/Al atoms were determined to be ∼3.8 and ∼2.1, respectively ( Supporting Information Table S1). Noteworthily, the in situ UV–vis spectra ( Supporting Information Figure S9) indicated the presence of five-coordinated Ni species in [email protected] when exposed to ambient environment, due to the adsorption of water molecules on the coordinatively unsaturated Ni species. Thus, the [email protected] was treated in dry helium with ascending temperatures and the d–d transition bands shifted from 23,750 to 22,885 cm−1 during the dehydration process. Specifically, the five-coordinated Ni species in square-pyramidal (C4v) symmetry transformed through four-coordinated Ni species in distorted square-planar (D4h) symmetry and finally to three-coordinated Ni species in trigonal (D3h) symmetry during dehydration. The local atomic structure of [email protected] was also optimized by DFT calculations,38 which suggested that Ni(II) species preferred to sit in the six-membered rings from sodalite (SOD) cage with Al pairs and the average Ni–O bond length was ∼2.03 Å (Figure 1c), consistent with the EXAFS results ( Supporting Information Table S1). Figure 1 | (a) Normalized XANES spectra at Ni K-edge of NiO, Ni foil, and [email protected] samples. (b) The Fourier transform of k3-weighted EXAFS spectra for [email protected] (c) Schematic illustration of Ni(II) confined in Y zeolite by DFT calculations. O atoms in red, Si atoms in yellow, Al atoms in magenta, and Ni atoms in dusty blue. The Ni–O bond length shown in angstroms (Å). Download figure Download PowerPoint Selective hydrogenation of alkynes over [email protected] To clarify broad applicability of the [email protected] catalyst, the selective hydrogenation of various alkynes in both the gas- and liquid phase was conducted. The catalytic performance of [email protected] in acetylene selective hydrogenation is shown in Figure 2a. Full acetylene conversion and high ethylene selectivity of 92% could be obtained simultaneously under optimized reaction conditions, that is, at 468 K and with a residence time of 0.48 s, in great contrast to the Ni-containing zeolites with similar Ni loadings of ∼4.5% prepared by ion exchange (Ni-Y) or wet impregnation (NiO/Y) ( Supporting Information Figure S10). Notably, acetylene conversion showed a significant fluctuation while the selectivity to ethylene remained almost unchanged with varied residence time, indicating intrinsic selectivity in acetylene hydrogenation over [email protected] Besides, the [email protected] showed a limited “comfort zone,” where the full acetylene conversion rapidly decreased with increasing temperature. The unique behaviors occurring on [email protected] catalyst may imply an unusual hydrogenation mechanism. The stability of [email protected] catalyst was evaluated and ∼98% acetylene conversion with ∼89% ethylene selectivity could be maintained for over 1200 min ( Supporting Information Figure S11). Figure 2 | (a) Temperature-dependent behaviors of [email protected] catalyst in the selective hydrogenation of acetylene. Reaction conditions: catalyst = 0.3 g, temperature = 353–523 K, 1% C2H2, 20% H2, balanced with He, residence time = 0.48, 0.36, 0.29, and 0.24 s; (b) Temperature-dependent behaviors of [email protected] catalyst in the selective hydrogenation of phenylacetylene. Reaction conditions: catalyst = 50 mg, phenylacetylene = 1 mmol, cyclohexane = 3 mL, tetradecane = 0.2 mmol, H2 pressure = 0.6 MPa, time = 6 h, and stirring speed = 1000 rpm. (c) Time-on-stream behaviors of [email protected] catalyst in the selective hydrogenation of phenylacetylene. Reaction conditions: catalyst = 50 mg, phenylacetylene = 1 mmol, cyclohexane = 3 mL, tetradecane = 0.2 mmol, H2 pressure = 0.6 MPa, temperature = 433 K, and stirring speed = 1000 rpm. (d) Recycling of [email protected] in the selective hydrogenation of phenylacetylene. Reaction conditions: catalyst = 50 mg, phenylacetylene = 1 mmol, cyclohexane = 3 mL, tetradecane = 0.2 mmol, H2 pressure = 0.6 MPa, time = 6 h, temperature = 433 K, and stirring speed = 1000 rpm. Download figure Download PowerPoint The selective hydrogenation of liquid alkynes catalyzed by [email protected] was then investigated. The temperature-dependent behavior of phenylacetylene hydrogenation over [email protected] is shown Figure 2b. Typically, a high selectivity to styrene of 92% was obtained at a phenylacetylene conversion of 80% within 6 h at 433 K. The full conversion of phenylacetylene could be achieved with the selectivity kept at <90% after 12 h of reaction (Figure 2c). The catalytic performance of [email protected] remained nearly unchanged after six cycles (Figure 2d), indicating a true heterogeneous catalytic process and good catalytic stability in liquid-phase hydrogenation. Expectedly, [email protected] exhibited remarkable catalytic performance in the selective hydrogenation of liquid-phase alkynes, like 3-chlorophenylacetylene and 1,8-nonadiyne (Table 1). The performance was comparable with the commercial Lindlar catalyst ( Supporting Information Table S2) although the reaction temperature is higher. Specifically, much higher selectivity toward enol was achieved on [email protected] (83.6% at a conversion of 69.5%) than that on Lindlar catalyst (50.1% at a conversion of 100%) in 2-methyl-2-butynol hydrogenation, demonstrating the advantage of [email protected] in chemoselective alkynol hydrogenation. It should be noticed that a certain amount of trans-methyl styrene (∼59%) and trans-stilbene (∼39%) can be obtained in the hydrogenation of 1-phenylpropyne and diphenylacetylene, respectively, which is rarely seen in heterogeneous hydrogenation. Apparently, the characteristics of [email protected] differ from traditional heterogeneous catalysts and appear to be more like organometallic complexes in homogeneous hydrogenations. Table 1 | [email protected] Selective Hydrogenation of Alkynes and Alkynols in the Liquid Phase Spectroscopic and kinetic analyses on alkyne selective hydrogenation Spectroscopic analyses and kinetic studies were performed for insight into the selective hydrogenation of alkynes over [email protected] catalyst. As shown in the temperature-programmed-desorption (TPD) profiles in Figure 3a, [email protected] exhibited a stronger affinity toward acetylene than ethylene, with the desorption temperatures of 363 and 328 K, respectively. Meanwhile, no dihydrogen desorption signals could be detected in the whole temperature range up to 600 K, indicating the very weak adsorption of dihydrogen on [email protected] Noteworthily, the TPD profiles of [email protected] are totally different from [email protected] (cationic Ni confined in chabazite zeolite, Si/Al = 3.3), in which stronger adsorption of dihydrogen over acetylene is observed (Figure 3b). It indicates that hydrogen can easily dissociate on [email protected] without acetylene, and some hydrogen may migrate around the chabazite after the dissociation. Therefore, in contrast to the direct heterolytic activation of dihydrogen occurring on [email protected],21 a new pattern of dihydrogen activation on [email protected] can be expected. The competition between acetylene and dihydrogen adsorption on [email protected] was further investigated by Fourier transform infrared (FTIR) spectroscopy (Figures 3c and 3d). The stretching bands of C–H (νs at 2925 cm−1, νas at ∼3010 cm−1) due to the chemisorbed acetylene strongly bonded to Ni species40,41 are clearly observed without respect to pretreatment via He or H2 flow. Thereupon, strong bonding between acetylene molecules and Ni(II) sites in [email protected] under the reaction conditions is confirmed, in great contrast to the case of [email protected]21 Figure 3 | TPD profiles of H2, C2H2, and C2H4 on (a) [email protected] and (b) [email protected] FTIR spectra of C2H2 adsorption over [email protected] recorded after (c) He or (d) H2 purging. Download figure Download PowerPoint In situ FTIR spectra of acetylene hydrogenation at various temperatures were conducted to monitor the hydrogenation process on [email protected] ( Supporting Information Figure S12). The IR bands of the gas phase or physisorbed acetylene (FTIR bands at 3310, 3260, and 730 cm−1) were clearly observed at various temperatures. Likewise, the spectral features of gas-phase ethylene (FTIR bands range between 3020–2900 cm−1 and 1000–900 cm−1) gradually emerged with increasing temperature to 453 K. These observations are consistent with the catalytic performance of acetylene selective hydrogenation, that is, ethylene is produced at elevated temperatures accompanied by the consumption of acetylene. Besides, the IR bands at 2928 and 2875 cm−1 gradually appeared with increasing temperatures, which could be ascribed to the symmetric stretching vibrations of CH2 and CH3, respectively.42,43 The stretching vibrations of CH2 and CH3 were speculatively derived from the adsorbed C2H3* and C2H5* species, respectively.21 The intensity difference between the IR bands of CH2 and CH3 is in accordance with the catalytic results, which means that ethylene is the dominant product accompanied by the formation of trace ethane. The in situ FTIR spectra suggest that acetylene hydrogenation occurs in a stepwise fashion, namely the dissociated hydrogen atom reacts with acetylene and ethylene sequentially to produce C2H3* and C2H5* intermediates, respectively. Considering the strong adsorption of acetylene on Ni(II) under reaction conditions, it is assumed that the cleavage of dihydrogen proceeds via a delayed pathway. Speaking more explicitly, acetylene first bonds to Ni species followed by dihydrogen dissociation to produce C2H3* intermediates. Detailed kinetic analyses were performed to disclose the mechanism of acetylene hydrogenation over [email protected] Both the internal and external diffusion limitations could be eliminated, as shown in Supporting Information Figures S13 and S14, respectively. The mass transfer within the crystalline of [email protected] was evaluated by a frequency response technique ( Supporting Information Figure S15), and the Weisz–Prater and Mears criterion was finally adopted to assess intracrystalline mass transport limitations in kinetic studies ( Supporting Information Table S3 and corresponding discussions). The kinetic isotope effect (KIE) was measured to demonstrate the rate-determining step (RDS) in acetylene hydrogenation. A KIE value of 3.74 was obtained when performing acetylene hydrogenation over [email protected] in H2 and D2, as shown in Supporting Information Figure S16. This result confirms that H2 dissociation intervenes in the RDS of the reaction.3,44,45 Our kinetic studies reveal the values of apparent activation energy (Ea) (see detailed steps in Supporting Information Table S4) for acetylene and phenylacetylene hydrogenation over [email protected] are 66 ( Supporting Information Figures S17 and S18) and 67 kJ mol−1 ( Supporting Information Figures S19 and S20), respectively. These values are considerably higher than the average activation energy value of the heterogeneous catalysts (44 kJ mol−1) and very close to that of homogeneous catalysis systems (64 kJ mol−1) ( Supporting Information Figure S21). In this context, a homogeneous-like mechanism for alkyne hydrogenation over [email protected] can be rationally proposed. The reaction orders of dihydrogen 1.15 ± 0.14 ( Supporting Information Figure S22) and acetylene −1.52 ± 0.47 ( Supporting Information Figure S23) for acetylene hydrogenation were measured in the kinetic region. The positive orders of dihydrogen and the negative orders of acetylene imply a typical Langmuir–Hinshelwood pathway,46 in accordance with the results from competitive adsorption (Figure 3). According to the kinetic and isotopic results, an associative mechanism is proposed containing several elementary steps (Table 2). First, acetylene molecules strongly adsorb on the Ni(II) active sites (step 1). Then the dihydrogen molecules undergo dissociation under the joint efforts of Ni(II) species and adsorbed acetylene (step 2). The dissociated H atoms are added onto the adsorbed acetylene molecules, forming the C2H3* and C2H4* species, in sequence (steps 3 and 4). Finally, the C2H4* desorb from the Ni(II) sites to give the ethylene product selectively (step 5). Table 2 | The Associative Mechanism with Simplified Reaction Steps Step Reaction Pathway 1 C 2 H 2 + * ⇄ C 2 H 2 * 2 H 2 + 2 * → 2 H* 3 C 2 H 2 * + H * ⇄ C 2 H 3 * + * 4 C 2 H 3 * + H * ⇄ C 2 H 4 * + * 5 C 2 H 4 * ⇄ C 2 H 4 + * Step 2 is the RDS according to the KIE experiments. Then the reaction rate can be expressed as (detailed evolution steps are shown in the Supporting Information): r = k 2 [ H 2 ] ( 1 + ( K − 4 K − 5 [ C 2 H 4 ] K 3 K 1 [ C 2 H 2 ] ) 1 2 + K 1 [ C 2 H 2 ] + K − 5 [ C 2 H 4 ] + ( K 3 K 1 K − 4 K − 5 [ C 2 H 2 ] [ C 2 H 4 ] ) 1 2 ) 2 (1)After evolution and substitution, the following reaction rate can be obtained: r = k 2 · [ H 2 ] ( 1 + ( K − 4 K − 5 r · 0.00148 K 3 K 1 [ C 2 H 2 ] ) 1 2 + K 1 [ C 2 H 2 ] + K − 5 r · 0.00148 + ( K 3 K 1 K − 4 K − 5 [ C 2 H 2 ] r · 0.00148 ) 1 2 ) 2 (2) According to the data measured for acetylene hydrogenation by H2 (Figure 4a) and D2 (Figure 4b) at 413 K, respectively, the parameters in eq. 1 were calculated by nonlinear regression analysis. As shown in Supporting Information Table S5, the rate constants (k2) for the RDSs in H2 and D2 flow were 6.92 and 1.34, respectively, to give a KIE value of 5.16, which is higher than the experimental result but still lies in the range of the primary KIE of 2–7. The RDS of acetylene hydrogenation is further confirmed. Obviously, the predicted turnover frequency (TOFs) are in good agreement with the measured TOFs in acetylene hydrogenation by both H2 (Figure 4c) and D2 (Figure 4d), demonstrating that the proposed elementary steps fit well with the associative mechanism. Figure 4 | Dependence of acetylene hydrogenation rates on acetylene concentrations for [email protected] catalyst under (a) H2 and (b) D2 at 413 K. Parity plots for the measured and predicted rates in acetylene hydrogenation (eq 2) on [email protected] catalyst under (c) H2 and (d) D2 with regression-fitted parameters shown in Supporting Information Table S5. Download figure Download PowerPoint Considering the low conversion of acetylene within the kinetic region, the reaction rate can be simplified as: r ≈ k 2 [ H 2 ] ( 1 + K 1 [ C 2 H 2 ] ) 2 (3)The reaction orders of dihydrogen and acetylene are simplified to be 1 ( Supporting Information Equation S23) and −2∼0 ( Supporting Information Equation S24), respectively, which matches well with the measured reaction orders ( Supporting Information Figures S22 and S23). As a whole, kinetic analyses reveal that acetylene hydrogenation over [email protected] follows the homogeneous associative mechanism consisting of elementary steps of initial acetylene adsorption and the subsequent dihydrogen activation as the RDS. The hydrogen–deuterium (H–D) exchange experiments in the absence and presence acetylene were performed over [email protected] and [email protected], and the results are shown in Figure 5. For [email protected], clear H–D signal (m/z = 3) was observed in H2–D2 stream at 423 K, indicating the effective activation of dihydrogen and the H–D exchange thereof. Upon the introduction of acetylene pulses, the H–D signal declined due to the reaction between H–D and acetylene, accompanied by the formation of ethylene. While for [email protected], no H–D signal was observed in H2–D2 stream at 423 K, revealing that dihydrogen could not be effectively activated. The introduction of acetylene pulses greatly promoted the dihydrogen activation over [email protected] and the H–D signal appeared. Overall, the H–D exchange experiments provide clear and direct evidence on the different patterns of dihydrogen activation and reaction pathways of acetylene hydrogenation over [email protected] and [email protected] That is, the direct dihydrogen activation occurs on [email protected] while the acetylene-promoted dihydrogen activation occurs on [email protected] Figure 5 | Pulse-response experiments of acetylene feeding to blank quartz sand, [email protected], and [email protected] in H2–D2 stream at 423 K. Reaction conditions: 0.2 g catalyst, 10 mL/min H2, 10 mL/min D2, 50 mL/min of N2, with two acetylene pulses introduced intermittently. Download figure Download PowerPoint Theoretical interpretations of alkyne selective hydrogenations DFT calculations were performed on alkyne-selective hydrogenations with [ema