Ambient-condition acetylene hydrogenation to ethylene over WS2-confined atomic Pd sites

Ambient-condition acetylene hydrogenation to ethylene (AC-AHE), integrated with acetylene production from coal and natural gas, offers a low-energy, non-oil route to ethylene compared with high-temperature naphtha steam cracking (>750 °C). However, conventional Pd-, Cu-, and Ni-based catalysts typically require elevated temperatures to achieve full acetylene conversion, leading to high energy consumption and undesirable by-products (ethane, green oil, coke) that shorten catalyst lifetime. Strategies to improve ethylene selectivity—such as atomically dispersed Pd sites, bimetallic alloys (Pd-Ag, Pd-Cu, Pd-In, Pd-Au, Pd-Ga), site poisoning/coverage (Lindlar), or p-block dopants (S, Sb)—often improve selectivity at the expense of activity. Even catalysts that break the activity–selectivity trade-off (e.g., Pd1Au1@MOF, Ni-S/C) face compromises in temperature requirement, reaction rate, or stability. Hence, the research aims to develop a catalyst that concurrently achieves high activity, high selectivity, and superior stability for AC-AHE at room temperature, overcoming the seesaw relationships among these metrics.

The study surveys prior approaches to acetylene semi-hydrogenation. Pd-based catalysts are efficient but struggle with room-temperature conversion and over-hydrogenation at higher temperatures. Alternative metals (Cu, Ni) and numerous tactics have been explored: single-atom Pd sites, Pd-based bimetallic alloys (Pd-Ag, Pd-Cu, Pd-In, Pd-Au, Pd-Ga), Lindlar-type poisoning, and tuning π-adsorption via S and Sb. While these can enhance ethylene selectivity, they frequently reduce activity. Recent advances (Pd1Au1@MOF dimers, Ni–S/C) have alleviated trade-offs but introduce new challenges, such as higher operating temperatures, lower rates, or insufficient stability. The literature underscores the persistent trade-offs between activity and selectivity and between activity and stability in AC-AHE, motivating the present work.

Catalyst synthesis and preparation: WS2 was synthesized by a direct chemical route using Na2WO4·2H2O and SiO2 colloids, followed by CS2 treatment in an autoclave at 400 °C for 4 h, HF etching (5% HF, 5 h), washing, and drying. WO3 was prepared by oxidizing WS2 at 350 °C in 2% O2/Ar for 3 h. Pd/WS2 catalysts with different Pd loadings (0.2, 0.8, 1.1, 2.2 wt%) were prepared by wet impregnation using H2PdCl4 solution, followed by drying. A 0.7 wt% Pd/WO3 and a Pd1Ag3/Al2O3 reference catalyst (impregnation of Pd(NO3)2 and AgNO3, calcination at 400 °C, 2 h in air) were also prepared. Catalyst pretreatment involved H2 reduction at 300 °C for 3 h. Catalytic testing: Selective hydrogenation of acetylene was performed at 0.1 MPa and 25 °C in a fixed-bed flow reactor using 10 mg catalyst without diluent. Feed: 1 vol% C2H2, 20 vol% H2, He balance. Typical GHSV 360,000 mL g_cat−1 h−1; stability tests at 480,000 mL g_cat−1 h−1. Products were analyzed by online GC (FID, KB-Al2O3/Na2SO4 capillary column). Conversions and selectivities were calculated via standard formulas; carbon balance was ~100.5% (0.8 Pd/WS2) and >95% for others. Characterization: XRD (Rigaku Ultima IV, Cu Kα), TEM (FEI Tecnai20, 200 kV), HAADF-STEM (JEOL ARM-200F, 200 kV) with image simulations (QSTEM), atomic-resolution EELS (Nion HERMES-100, 60 kV), in-situ XAFS (EXAFS/XANES) at SSRF BL14W1 after H2 reduction, XPS (Omicron, Al Kα; in-situ reduction at 300 °C), ICP-OES for Pd content (PerkinElmer 7300DV). H–D exchange experiments were conducted at 25 °C using pulsed D2 in 20% H2 with MS detection of m/z 2, 3, 4 to assess H2 dissociation activity. Computational details: DFT calculations were performed in VASP using PBE-GGA with PAW, 400 eV plane-wave cutoff, 1×1×1 k-point for a 6×6 WS2 monolayer supercell (vacuum >15 Å). Geometry optimizations used energy/force criteria of 1e−5 eV and −0.02 eV Å−1. D3 (zero-damping) dispersion corrections were included. Transition states were located via Fixed Bond Length method in ASE, verified by a single imaginary frequency. Pd(111) was modeled as a p(4×4) slab with three layers (top two relaxed). Free energies included ZPE and vibrational corrections at 25 °C; barriers computed as ΔEact + ΔZPE + ΔUvib. Bader charge analysis and charge density difference maps quantified charge transfer upon Pd binding to WS2. Models considered Pd at W-top, S-top, and hollow sites.

• Performance: 0.8 wt% Pd/WS2 achieved >99% C2H2 conversion with 70% C2H4 selectivity at 25 °C and GHSV up to 360,000 mL g_cat−1 h−1. Space-time yield (STY) of ethylene reached 1123 mol_C2H4 mol_Pd−1 h−1 at 25 °C, nearly 4× that of Pd1Ag3/Al2O3. Ethylene selectivity remained ≥70% across conversion levels and increased to 74% at lower H2/C2H2 ratio (5% C2H2:10% H2). Performance was insensitive to excess ethylene in feed. Superior stability was demonstrated over >500 h at GHSV 480,000 mL g_cat−1 h−1 without notable changes in activity, morphology, surface carbon, or electronic state.
• Loading dependence: 0.8 wt% Pd loading was optimal; lower or higher loadings reduced conversion or selectivity, respectively. 2.2 wt% Pd/WS2 formed Pd nanoparticles after H2 reduction, correlating with increased over-hydrogenation to C2H6 and lower selectivity.
• Structure and dispersion: XRD showed no Pd crystallites for 0.8 Pd/WS2, 2.2 Pd/WS2, 0.8 Pd/WO3 post-reduction. TEM/EDX indicated highly dispersed Pd in 0.8 Pd/WS2 before and after reduction; 2.2 Pd/WS2 exhibited Pd particles (~4.7 nm, Pd(111) 0.23 nm spacing) after reduction. HAADF-STEM resolved atomically dispersed Pd on two WS2 sites: 77.2% at W-top sites (S-coordinated) and 22.8% at S-vacancy/hollow sites. EELS confirmed Pd M4,5 edges at bright atom positions.
• Electronic structure: In-situ EXAFS showed dominant Pd–S coordination and absence of Pd–Pd for 0.2 and 0.8 Pd/WS2; 2.2 Pd/WS2 and 0.7 Pd/WO3 displayed Pd–Pd and Pd–S/O, indicating particle formation. In-situ XPS (Pd 3d) after H2 reduction revealed Pd+ (~336.3 eV), Pd2+ (~337.3 eV), and Pd0 (~335.5 eV). 0.2 and 0.8 Pd/WS2 contained >50% Pd+, whereas 2.2 Pd/WS2 and 0.7 Pd/WO3 had >76% Pd0 and <6% Pd+. In-situ Pd K-edge XANES placed 0.8 Pd/WS2 closer to PdO than Pd foil, indicating positively charged Pd in low-loading samples.
• H2 activation: H–D exchange at 25 °C showed higher HD/D2 ratios for Pd/WS2 than for WS2 or Pd/WO3 at equal Pd amounts, indicating superior H2 dissociation activity on Pd/WS2, consistent with higher acetylene conversion.
• DFT mechanism and selectivity: For single-atom Pd on WS2, adsorption free energies for C2H2 were stronger on Pd/W-site (−0.78 eV) and Pd/Hollow-site (−1.08 eV) than on Pd/S-site (−0.42 eV). The rate-limiting barrier (C2H2 → C2H3) was 0.80 eV (Pd/W-site) and 0.76 eV (Pd/Hollow-site), versus 1.29 eV (Pd/S-site), indicating higher activity at W and hollow sites. On Pd/W-site, C2H4 desorption barrier (0.49 eV) was lower than further hydrogenation to C2H5 (0.83 eV), favoring ethylene selectivity. In contrast, on Pd(111) nanoparticles, C2H4 desorption was difficult (1.14 eV) while hydrogenation to C2H5 was easier (0.68 eV), promoting over-hydrogenation. PDOS and charge analysis showed Pd–S coordination induces electron-deficient Pdδ+ with hybridization of Pd 4d with neighboring S 3p and W 5d states, enabling co-adsorption/activation of H2 and C2H2 while weakening C2H4 binding.

The study addresses the central challenge of achieving high activity, selectivity, and stability for acetylene semi-hydrogenation at room temperature. Sulfur-confined, atomically dispersed Pdδ+ sites on WS2 provide concurrent activation of H2 and C2H2 at ambient conditions while weakening ethylene adsorption. This site-specific electronic tuning—substantiated by EXAFS/XANES/XPS, HAADF-STEM/EELS, H–D exchange, and DFT—shifts the reaction pathway toward facile C2H4 desorption and suppresses over-hydrogenation to C2H6. The predominance of Pd at W-top and hollow sites, stabilized by Pd–S coordination, explains the high ethylene productivity and long-term stability without aggregation or coking. Compared to conventional Pd catalysts and bimetallic/alloy strategies, Pd/WS2 uniquely balances the activity–selectivity–stability triad at room temperature with record STY, offering a practical, low-energy route for ethylene production.

A robust AC-AHE process was demonstrated over 0.8 wt% Pd/WS2 featuring S-confined, positively charged Pdδ+ single sites. The catalyst delivers >99% conversion, 70% ethylene selectivity, and a record ethylene STY of 1123 mol_C2H4 mol_Pd−1 h−1 at 25 °C with >500 h stability. Operando/structural characterizations and DFT reveal that Pd–S coordination stabilizes electron-deficient Pd single atoms, enabling efficient co-activation of H2 and acetylene and favoring ethylene desorption over over-hydrogenation. The work establishes an efficient, non-oil, low-energy pathway for ethylene production with simultaneous high productivity and stability.