Field-induced reagent concentration and sulfur adsorption enable efficient electrocatalytic semihydrogenation of alkynes.

Efficient electrocatalytic alkyne semihydrogenation with potential/time-independent selectivity and Faradaic efficiency (FE) is vital for industrial alkene productions. Here, sulfur-tuned effects and field-induced reagent concentration are proposed to promote electrocatalytic alkyne semihydrogenation. Density functional theory calculations reveal that bulk sulfur anions intrinsically weaken alkene adsorption, and surface thiolates lower the activation energy of water and the Gibbs free energy for H* formation. The finite element method shows high-curvature structured catalyst concentrates K+ by enhancing electric field at the tips, accelerating more H* formation from water electrolysis via sulfur anion-hydrated cation networks, and promoting alkyne transformations. So, self-supported Pd nanotips with sulfur modifiers are developed for electrochemical alkyne semihydrogenation with up to 97% conversion yield, 96% selectivity, 75% FE, and a reaction rate of 465.6 mmol m-2 hour-1. Wide potential window and time irrelevance for high alkene selectivity, good universality, and easy access to deuterated alkenes highlight the promising potential.

INTRODUCTION

Selective semihydrogenation of alkynes is an extremely important process in both industry and academia because the produced alkenes display wide applications in pharmaceuticals and materials science (, ). Transfer semihydrogenation of alkynes, which mainly adopts alcohols, amines, silanes, or metal hydrides as the hydrogen donors, has emerged as a promising alternative for alkenes syntheses because of the mild reaction conditions and the avoidance of flammable hydrogen gas (H2) (, ). In contrast, there are few examples of using water as a hydrogenating agent for transfer semihydrogenation of alkynes despite the advantages on low cost and high safety (, ), which may be due to the ineffective activation of water. Recently, electrochemistry has aroused increasing attention from chemists (, ), and electrochemical water splitting provides an efficient route to hydrogen production (). Therefore, electrochemical transfer hydrogenation via water electrolysis can serve as an environmentally friendly and sustainable method for producing hydrogenated products, as witnessed by electrochemical CO2 and NO3− reduction reactions (, ). Although electrochemical transfer semihydrogenation has been reported (, ), it usually suffers from the need for corrosive sulfuric acid or toxic amines as the hydrogen sources, limited substrate scope, overreduction to alkanes, and unclear reaction mechanisms. Thus, further exploring electrochemical transfer semihydrogenation of alkynes with green and more expedient water as the hydrogen donor to overcome the above drawbacks and unveil the reaction mechanism is urgently needed.

Recently, we have realized an electrochemical transfer semihydrogenation of alkynes with H2O as the hydrogen source under mild conditions over a Pd-P cathode (). Its Faraday efficiencies (FEs) of alkenes are highly dependent on the precise controls on cathodic potentials and reaction time. To address these issues, a low-coordinated copper cathode with surface sulfur (S) was then found that allowed the highly selective synthesis of alkenes across a wide potential range (). However, its reaction rate was relatively low because of the weak hydrogenation ability of Cu itself, thus a higher potential is required to drive the semihydrogenation with only 3.6 to 10% FEs due to competitive hydrogen evolution reaction (HER). Therefore, to stimulate the overall practicability of transfer semihydrogenation of alkynes with water even to industrial productions, an efficient electrocatalyst that speeds up the reaction rate and maintains high selectivity of alkenes without being time and potential dependent is highly desirable.

Pd-based catalysts hold wide dominance in alkyne hydrogenation owing to the affinity to alkynes and high efficiency in H2 activation and H retainment (, ). Pd-based cathodes also demonstrate excellent performance for electrochemical H2 production from water splitting (, ). Usually, regulating the electronic structure of the Pd center can greatly boost the alkene selectivity or HER activity (, ). In particular, sulfur modification has been proved to be an efficient route to improve the intrinsic activity and stability of Pd toward electrochemical HER and thermocatalytic organic hydrogenation reactions (–). In addition, high-curvature structures exhibit a high local electric field that increases the concentration of reactants or electrolytes around the tips, hence enhancing electrochemical CO2RR and HER performances (–). For example, Liu et al. () have demonstrated that the nanostructured gold nanoneedles could efficiently concentrate K+, which in turn led to a high local concentration of CO2 surrounding the catalyst’s surface, thus promoting the electrocatalytic CO2RR (). In these regards, S-modified high-curvature Pd nanostructures are supposed to induce reagent concentration and accelerate the formation of active hydrogen (H*) from water electrolysis, thus enabling rapid electrocatalytic alkyne semihydrogenation. However, the synthesis and exploration of a S-modified high-curvature Pd electrode remain to be a challenging task.

Here, the S-modified high-curvature Pd is theoretically designed as a cathode candidate for electrocatalytic selective hydrogenation of alkynes. Then, carbon-supported sulfur anions and thiolate-modified Pd nanotips (ArS-Pd4S NTs) are synthesized and experimentally confirmed to be an excellent electrocatalyst to convert alkynes with water to alkenes with up to 97% conversion yield, 96% selectivity, 75% FE, and the reaction rate of 465.6 mmol m−2 hour−1 at −1.1 V versus Hg/HgO, outperforming Pd NTs, ArS-Pd NTs, Pd4S NTs, and ArS-Pd4S nanoparticles (NPs). The high reaction rate and FE at low potentials, robust selectivity of alkenes in a wide range of potential, and facile access to deuterated alkenes by using D2O endow our method with great potential for a green and economic industrial alkene synthesis.

RESULTS

Screening of Pd-based electrocatalysts

To achieve relatively high selectivity, FE, and reaction rate of alkenes through electrochemical semihydrogenation of alkynes in an aqueous electrolyte, three factors need to be taken into account for an electrode choice: (i) ease of alkene desorption, (ii) low water activation energy and small Gibbs free energy (∆GH*) for H* formation, and (iii) high concentration of reactants around the surface of the electrode. Density functional theory (DFT) calculations are first conducted to provide theoretical guidance. We use pure Pd as a model. Pd4S, surface thiolate–modified Pd (denoted as ArS-Pd), and their combination (denoted as ArS-Pd4S) are chosen for comparisons because Pd4S is the most stable among various palladium sulfides with high hydrogenation activity (fig. S1) (, ), while the organic sulfur modification is a common and efficient route to improve the thermocatalytic performance of Pd catalysts ().

To simplify the simulation process, we use acetylene (C2H2) as the model substrate. Figure 1A displays the energy profiles for the hydrogenation of acetylene on Pd and ArS-Pd (100), as well as Pd4S and ArS-Pd4S (110). The exothermal adsorptions of acetylene imply the favorable acetylene hydrogenation of all these four catalysts. Although the desorption of adsorbed ethylene (C2H4*) from the surface of these four catalysts to gaseous C2H4 are all exothermic, much less energy is required for both Pd4S and ArS-Pd4S, reflecting much easier desorption of C2H4 from Pd4S-based catalysts (Fig. 1B). However, the demanded energy for ethylene desorption from Pd4S and ArS-Pd4S is quite similar, showing that Pd4S is mainly responsible for weakening ethylene adsorption. In addition, further hydrogenating C2H4* to C2H5* on both Pd4S and ArS-Pd4S is energetically unfavorable, hindering the overhydrogenation of C2H4 to ethane (C2H6). Furthermore, the energy barrier for cleaving H─OH bond over ArS-Pd4S is lower than that of Pd4S (1.36 eV versus 1.49 eV) (Fig. 1C), demonstrating an easier water dissociation process due to the ArS modification. The formations of *H and *OH over both ArS-Pd4S and Pd4S are spontaneous, while the more negative ΔG(H− −OH → *H + *OH) of ArS-Pd4S (−0.87 eV) than Pd4S (−0.82 eV) suggests a more thermodynamically favorable process for forming H*. Notably, ArS-Pd4S presents a weaker affinity of H* (−0.53 eV) compared with Pd4S (−0.55 eV), conducive to the subsequent semihydrogenation of the alkynes. These obtained results reveal that water electrolysis and H* release more readily occur over ArS-Pd4S, promoting a high reaction rate of acetylene hydrogenation.

Fig. 1.

DFT and finite element method calculations.

(A) Free energy diagram for alkyne semihydrogenation reactions and (B) calculated ΔG(*C2H4 → C2H4(g)) over different samples. (C) Free energy diagram for water-splitting process over ArS-Pd4S and Pd4S. (D) Electric field on the surface of different Pd-based catalysts. The tip radii of the structures are 1.5 nm (top) and 4.5 nm (bottom), respectively. (E) Surface K+ concentration and (F) OH− concentration distributions adjacent to the surface of Pd NTs.

Moreover, high-curvature structured electrocatalysts have been proved to concentrate local electric fields and affect ion distributions (). The finite element method simulation is used to explore whether a high-curvature structure could affect the electrocatalytic performance of the Pd catalyst. Cones with different radii are separately created to represent Pd electrodes immersed in 1.0 M KOH electrolytes. The electric fields around the tips obviously increase as the cones sharpen from 4.5 to 1.5 nm (Fig. 1D), which is caused by the migration of free electrons to the tips of the charged Pd electrode due to the electrostatic repulsion (). The impact of the enhanced local electric field on the distribution of ions around the surface is then estimated by the Gouy-Chapman-Stern model, which provides the mapping of the surface K+ and OH− concentrations in the Helmholtz layer of the electrical double layer close to the surface of the Pd electrode. As shown in Fig. 1E, there is a leap in K+ concentration adjacent to the tips due to the enhanced local electric field. Consequently, more water molecules will gather on the electrode surface via forming hydrated cations (). In addition, the surface-adsorbed OH− concentration sharply drops (Fig. 1F) because of the mutual repulsion effect. It conduces to the leave of adsorbed OH−, thus accelerating the formation of H* (H2O + e− + * → H* + OH−), the hydrogen source of the alkyne semihydrogenation. Thus, the high-curvature structured Pd catalyst will increase the K+ concentration and speed up water electrolysis, therefore enhancing the electrochemical hydrogenation of alkynes. On the basis of these theoretical predictions, we speculate that the high-curvature structured ArS-Pd4S NTs can serve as an excellent electrocatalyst for the water-participated transfer semihydrogenation of alkynes with a high reaction rate, FE, and selectivity.

Syntheses and characterizations of Pd-based catalysts

The abovementioned four Pd-based NTs are synthesized for comparisons. Pure Pd NTs are facilely prepared on the carbon fiber paper (CP) through a simple electrodeposition process. Then, thiolate treatment with/without heat treatment is applied to synthesize Pd@ArS-Pd4S NTs and ArS-Pd NTs, respectively, and Na2S treatment is applied to obtain Pd@Pd4S NTs. The electrodeposited Pd NTs present a well-defined crystalline structure [Joint Committee on Powder Diffraction Standards (JCPDS) 44-1043] with the NT morphology (fig. S2, A to C). In addition, the NT morphology remains well for the Pd@ArS-Pd4S NT sample (Fig. 2A and fig. S3A). For the formation of Pd@ArS-Pd4S NTs, partial C─S bonds cleave during the modification of Pd NTs by thiolate upon heating, and S anions release to form Pd4S thin layer with adsorbing ArS on the surface (). The interface between the ArS-Pd4S thin layer (~10 nm) and the Pd core can be clearly seen (inset image of Fig. 2A). The energy-dispersive x-ray (EDX) mapping (fig. S3B) and EDX line scanning result (fig. S3C) reveal the homogeneous distribution of abundant Pd and a small amount of S throughout the Pd@ArS-Pd4S. The x-ray diffraction (XRD) results (fig. S4) demonstrate that the diffraction peaks correspond to the standard Pd (JCPDS 44-1043) and Pd4S (JCPDS 10-0335). The Pd 3d x-ray photoelectron spectroscopy (XPS) spectra can be deconvoluted to 335.0 and 340.2 eV for Pd0 3d5/2 and Pd0 3d3/2 and 336.1 and 341.4 eV for Pd2+ 3d5/2 and Pd2+ 3d3/2, respectively (Fig. 2B). The S 2p XPS spectra confirm the copresence of S2− and thiolate on the Pd@ArS-Pd4S NTs (Fig. 2C) (). Notably, the peaks of Pd2+, S2−, and thiolate decrease (135 s) and then disappear (270 s) with the increment of sputtering time, indicating that only a thin layer of Pd4S formed at the surface of Pd NTs. To further understand their electronic configuration and local environment, we perform x-ray absorption near-edge structure spectra (XANES) of Pd@Pd4S NTs and Pd@ArS-Pd4S NTs (Fig. 2D). Peak A of the two samples can be assigned to the S─Pd resonance of Pd4S, indicating the incorporation of S2− into the inner lattice of Pd after being treated by either Na2S or sodium thiophenolate. Peak B at ~2475 eV is attributed to the S─C resonance of the adsorbed thiolate of Pd@ArS-Pd4S NTs (). The existence of the S─C bond in Pd@Pd4S is attributed to the interaction between Na2S and the CP substrate. In addition, the peaks at ~2483 eV may be ascribed to SO42− because of the oxidation of S2− and thiolate in air. Besides, Pd@Pd4S NTs and ArS-Pd NTs are also synthesized and characterized as shown in figs. S5 and S6.

Fig. 2.

Characterizations of Pd@ArS-Pd4S NTs.

(A) Transmission electron microscopy (TEM) and high-resolution TEM (inset) image of Pd@ArS-Pd4S NTs. (B) High-resolution Pd 3d spectra and (C) S 2p spectra at different sputtering times: 0, 135, and 270 s. (D) XANES spectra of Pd@ArS-Pd4S NTs and Pd@Pd4S NTs. a.u., arbitrary units.

Electrocatalytic performance of Pd-based catalysts

The performances of electrochemical transfer alkyne semihydrogenation over these four Pd-based electrocatalysts are evaluated in a divided three-electrode system. All the potentials are referred to as Hg/HgO unless otherwise stated. 4-Ethynylaniline (1a), KOH aqueous solution (1.0 M), and 1,4-dioxane (Diox) work as the model substrate, electrolyte, and cosolvent, respectively. After addition of 0.2 mmol of 1a, the linear sweep voltammetry (LSV) curve presents a lower onset potential and a larger current density before reaching −1.24 V (Fig. 3A), demonstrating that the reduction of 1a is more dominant than HER, while HER takes priority at more negative potentials than −1.24 V in the absence of 1a. Thus, the potential-dependent measurements are carried out in the potential range of −0.9 to −1.3 V. Pd@ArS-Pd4S NTs achieve the optimal conversion of 1a (97%), selectivity of alkene product 2a (96%), and the FE (75%) at −1.1 V when 48 C electrons are passed (Fig. 3B), surpassing the other Pd-based electrodes. Deuterium labeling experiments confirm H2O to be the sole hydrogen source for this reaction (fig. S16). Notably, Pd@ArS-Pd4S NTs exhibit high conversions of 1a (more than 92%), selectivity of 2a (more than 90%), and FE (more than 67%) in a wide potential window from −1.0 to −1.3 V, indicating that the surface-adsorbed thiolate and Pd4S synergistically contribute to the efficient semihydrogenation of 1a. Investigations on the contact angles of these samples (figs. S7 and S8 and table S1) reveal that Pd@Pd4S NTs are more hydrophilic than other samples, indicative of the favorable affinity of H2O, while thiolate-treated samples become superwetting after replacing H2O with a mixed 1a solution of H2O and Diox, conducive to the contact with 1a. Therefore, Pd@ArS-Pd4S NTs with modified hydrophilicity and lipophilicity allow its superior performance.

Fig. 3.

Electrocatalytic alkyne semihydrogenation performances of Pd@ArS-Pd4S NTs.

(A) LSV curves of Pd@ArS-Pd4S NTs in 1.0 M KOH solution [Diox/H2O, 2:5 (v/v)] at a scan rate of 10 mV s−1 with and without 0.2 mmol of 1a. (B) Potential-dependent conversions (Conv.) of 1a, selectivity (Sel.) of 2a, and 2a FEs over Pd@ArS-Pd4S NTs, Pd@Pd4S NTs, ArS-Pd NTs, and Pd NTs. (C) Time-dependent conversions of 2a over Pd@ArS-Pd4S NTs, Pd@Pd4S NTs, ArS-Pd NTs, and Pd NTs. Error bars correspond to the SD of three independent measurements.

Time-dependent results show that the selectivity of 2a did not vary apparently with the reaction time during the conversion of 1a, reaching up to 95% (fig. S9A). Noticeably, Pd@ArS-Pd4S NTs keep much higher selectivity with increasing reaction time than ArS-Pd NTs, while it achieves a higher conversion of alkynes than Pd@Pd4S NTs (fig. S9, B to D). Accordingly, the thiolate adsorption and Pd4S may contribute to improved 1a conversion and enhanced 2a selectivity corporately, matching well with the DFT results. Moreover, Pd@ArS-Pd4S NTs show the lowest conversion of 2a to the overhydrogenated alkane product 3a among the samples according to the time-dependent results (Fig. 3C), confirming its intrinsic inhibition on the undesired 3a. Furthermore, Pd@ArS-Pd4S NTs exhibit high alkyne conversion yield and excellent alkene selectivity for 6 cycles (fig. S10). Although the conversion yield gradually decreases after the seventh cycle (still exceeding 70%), there is no obvious decay in alkene selectivity for up to 9 cycles, which may hold a potential for long-term and repetitive electrosynthesis of alkenes. We attribute the decrease of conversion yield mainly to the stripping off of the electrocatalyst from the seventh run.

More intriguingly, morphological effects of Pd@ArS-Pd4S NTs have been found to show a notable enhancement in electrocatalytic performance. Pd NPs have been synthesized through chemical reduction by NaBH4 (fig. S11) and then treated by sodium thiophenolate through the same method to obtain Pd@ArS-Pd4S NPs (fig. S12). As shown in Fig. 4A, both Pd@ArS-Pd4S NPs and Pd@ArS-Pd4S NTs have high selectivity of more than 96% at −1.1 V. However, Pd@ArS-Pd4S NTs present a higher conversion compared to Pd@ArS-Pd4S NPs, indicating the enhanced activity aroused by the high-curvature structure. To confirm the performance promotion, we first compare the intrinsic activity of Pd@ArS-Pd4S NPs and Pd@ArS-Pd4S NTs by eliminating the influence of different electrochemical surface areas (ECSAs; fig. S13). Pd@ArS-Pd4S NTs present an enhanced activity compared with Pd@ArS-Pd4S NPs (Fig. 4B). In addition, the electrochemical impedance spectra (EIS) show a much smaller charge transfer resistance (Rct) of Pd@ArS-Pd4S NTs than that of Pd@ArS-Pd4S NPs (Fig. 4C), revealing the accelerated charge-transfer kinetics offered by the high-curvature NTs. It can be deduced that the intensive negative local electric field could effectively repel the negatively charged OH−, facilitating the release of active sites occupied by *OH and the supply of H* (*H + *OH → *H + OH−), thus promoting the alkyne semihydrogenation. Moreover, inductively coupled plasma atomic emission spectroscopy (ICP-AES) is applied to measure the K+ ions adsorbed away from the electrolytes (Fig. 4D). The Pd NTs adsorb much more K+ ions (~19.3%) than that of Pd NPs, highlighting the important role of the high-curvature structure for concentrating K+ and validating the initial theoretical prediction. Notably, there is still a 7.8% increase of K+ ions adsorbed by Pd@ArS-Pd4S NTs compared with Pd@ArS-Pd4S NPs after the thiolate modification. The concentrated K+, which formed more anion-hydrated cation networks [S2−-K+(H2O)] with surface S2− anions through noncovalent Coulomb interactions, enhances the dissociation of H2O for forming H* (). Control experiments verify the promoting effect of K+ ions in this semihydrogenation reaction. When tetramethylammonium hydroxide (TMAOH) and NaOH are used to replace KOH electrolytes, Pd@ArS-Pd4S NTs exhibit lower current density and a sluggish alkyne hydrogenation process under other identical conditions (fig. S14), due to the weaker interaction between S2− and TMA+ and the larger ionic hydration number and radius of the hydrated cation of Na+(H2O) (n = 7 for K+ versus 13 for Na+) (). Consequently, the preliminary results suggest that both the NT structure and the concentrated K+ are essential in promoting the electrochemical transfer semihydrogenation of alkynes with water, rationalizing our speculation and theoretical prediction.

Fig. 4.

Performance comparison of Pd@ArS-Pd4S NTs and Pd@ArS-Pd4S NPs.

(A) Conversions of 1a and selectivity of 2a at −1.1 V versus Hg/HgO. (B) ECSA-normalized LSV curves. (C) EIS Nyquist plots at −1.0 V versus Hg/HgO of Pd@ArS-Pd4S NPs and Pd@ArS-Pd4S NTs. (D) Field-induced K+ concentration loss in the electrolytes caused by Pd NPs, Pd NTs, Pd@ArS-Pd4S NPs, and Pd@ArS-Pd4S NTs. Error bars are the SD of three independent measurements.

Mechanism and universality of electrocatalytic semihydrogenation

The origin of enhanced selectivity and stability toward electrocatalytic semihydrogenation of alkynes over Pd@ArS-Pd4S NTs is further explored. First, we conduct the LSV tests in anhydrous N,N-dimethylformamide (DMF) over Pd@ArS-Pd4S NTs (fig. S15A). When 0.1 mmol of 1a is added into the cathodic cell, no obvious changes are observed in the LSV curve. After further addition of 1.0 ml of H2O, the significantly increased current density indicates the promoting effect of water for electrochemical hydrogenation of 1a. Second, the kinetic isotope effect experiment with the kH/kD value of 2.0 manifests that H2O dissociation is the rate-determining step for 1a semihydrogenation to 2a (fig. S16). This further emphasizes the vital roles of thiolate modification and NT structure of Pd-based catalysts in lowing water activation energy and facilitating water electrolysis. Third, the surface-adsorbed atomic hydrogen (H*ads) is responsible for aqueous electrochemical hydrogenation reactions (). The cyclic voltammogram (CV) curve of Pd@ArS-Pd4S NTs in 0.5 M Na2SO4 solution presents a distinct peak of surface-adsorbed H*ads (Fig. 5A). After adding 1a, the H*ads peak decreases obviously, demonstrating the H*ads-participated electrocatalytic semihydrogenation of 1a. The reduction of subsurface-absorbed atomic hydrogen (H*abs) may be ascribed to its outmigration to the surface and then is used for the semihydrogenation of alkynes or HER. Furthermore, when we add tertiary butanol (t-BuOH) to the reaction system, the conversion of 1a is inhibited, again supporting an H*ads-involved transformation of 1a (fig. S15B). Therefore, a possible mechanism is proposed in Fig. 5B. The reaction begins with the adsorption of 1a and water on the surface of Pd@ArS-Pd4S NTs. After electrolysis begins, H*ads is generated from water electroreduction. Then, H*ads adds to the C≡C bond of a nearby 1a to form the carbon radical intermediate, which abstracts another H*ads to produce the alkene product 2a. The carbon and hydrogen radicals (also referred to H*ads) are detected by the electron paramagnetic resonance (EPR) measurements using 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) as the trapping agent (Fig. 5C) (, ). Last, desorption of 2a regenerates the active sites for the next reaction cycle.

Fig. 5.

Electrocatalytic alkyne semihydrogenation mechanism of Pd@ArS-Pd4S NTs.

(A) CV curves of Pd@ArS-Pd4S NTs in 0.5 M Na2SO4 solution with and without 1a. (B) The proposed reaction mechanism for alkyne semihydrogenation reaction over Pd@ArS-Pd4S NTs cathode. (C) EPR trapping for hydrogen (*) and carbon (#) radicals.

The applicability of this electrochemical transfer semihydrogenation of alkynes using water as the hydrogen source over Pd@ArS-Pd4S NTs is investigated (Table 1). A series of aryl alkynes bearing either electron-withdrawing or electron-donating substituents on the aryl ring and the heterocyclic aryl alkyne (2a-k) work well under standard reaction conditions, delivering the corresponding alkenes with excellent selectivity (from 94:6 to 98:2). In addition, the aliphatic alkynes are also amenable to our strategy (2l-n), demonstrating the good universality of this semihydrogenation method. Impressively, our method can be further developed to the facile synthesis of functionalized deuterated alkenes, ones of potential building blocks for synthesizing deuterated drugs, with high yields, selectivity, and deuterated efficiency (99%) using cheap and safe D2O as the deuterated source (fig. S29 and note S1), avoiding the usage of D2 and other toxic deuterated reagents. Notably, the semihydrogenation method over Pd@ArS-Pd4S NTs is also performed by applying a flow cell to replace the batch H-type electrolyzer (fig. S17). We successfully achieve a conversion yield of 92% and an alkene selectivity of 98%. The universality and scalability of Pd@ArS-Pd4S NTs reveal the promising potential in the application of electrocatalytic semihydrogenation of alkynes.

Table 1.

Substrate scope for electrocatalytic transfer semihydrogenation of alkynes with H2O or D2O over Pd@ArS-Pd4S NT cathode.

Reaction conditions: alkyne substrates (0.1 mmol), Pd@ArS-Pd4S NTs (working area: 1.0 cm2), 1.0 M KOH [Diox/H2O, 2:5 (v/v), 7 ml], 25° ± 0.5°C, −1.1 V versus Hg/HgO. Conversion yields were reported, and the data in parentheses were the alkene selectivity.

*K2CO3 (0.5 M) was used as the electrolyte using Diox/D2O [2:5 (v/v), 7 ml] as cosolvents.

DISCUSSION

This study systematically demonstrates the advantages of bulk/surface sulfur modification and high-curvature structure for enhancing electrocatalytic semihydrogenation performance. DFT calculation and finite element method were initially used to rationally design the electrocatalyst for alkyne semihydrogenation, and then Pd@ArS-Pd4S NTs are successfully synthesized. For Pd@ArS-Pd4S NTs, the bulk sulfur modification (Pd4S) contributes to the weakened adsorption of alkenes, suppressing undesirable overhydrogenation to alkanes. Simultaneously, surface thiolates (ArS) facilitate the water dissociation and the fading affinity of H*, efficiently producing H* for subsequent semihydrogenation. Thus, Pd@ArS-Pd4S NTs exhibit a more robust time-dependent alkene selectivity than ArS-Pd NTs (fig. S9) and a faster alkyne conversion than Pd@Pd4S NTs (Fig. 3B). Moreover, the sharp tips can create an enhanced local electric field to concentrate the opposite-charged K+ ions, which promotes the generation of H* via forming more Sδ−-K+(H2O) networks, leading to a higher alkyne conversion yield of Pd@Pd4S NTs compared with that of Pd@Pd4S NPs (Fig. 4A).

In addition, flow electrolysis holds great feasibility and suitability in industrial production, because it can realize continuous production and is scalable in size and number (used in parallel or series). Notably, Pd@ArS-Pd4S NTs have been successfully applied in a flow cell and exhibit a high conversion yield of 92% and an alkene selectivity of 98%, revealing its possibility for large-scale alkene production. However, the optimizing of reaction parameters (e.g., the concentration of the substrate, applied current, and the flow rate) are demanded to further improve space-time yield to meet the industrial requirements. Moreover, universality is an important factor to evaluate a reaction system. The semihydrogenation method over Pd@ArS-Pd4S NTs can be developed to semihydrogenate various kinds of alkynes including conjugated/unconjugated aryl alkyne, aliphatic alkyne, and heterocyclic alkyne, highlighting its universality and scalability. However, internal alkynes cannot be effectively semihydrogenated to internal alkenes under our reaction conditions even prolonging the reaction time to 10 hours, which may be caused by the steric effect of such substrates. Consequently, we deduce that the electrocatalytic semihydrogenation of internal alkynes may be realized by changing a smaller ligand than ─ArS to modify the catalyst’s surface, which could be studied in our future work.

In summary, the high-curvature Pd@ArS-Pd4S NTs have been proved to be an excellent cathode for electrocatalytic alkyne semihydrogenation by using water as the hydrogen source with a high reaction rate and FE at low potentials and robust selectivity of alkenes in a wide range of potential. The performance enhancement aroused by bulk/surface modification and high-curvature structure can be expected to design efficient catalysts for boosting the reaction efficiency and selectivity in various electrocatalytic reactions (e.g., organic oxidation, nitrate reduction, and other hydrogenation reactions).

MATERIALS AND METHODS

DFT calculations

All the computations were performed on the basis of the DFT methods, as implemented in the plane wave set Vienna ab initio Simulation Package code (, ). The pseudopotential was generated from the projector-augmented wave method. The exchange-correlation functional in the Perdew-Burke-Ernzerhof form within a generalized gradient approximation was used (). The climbing image nudged elastic band method was used to calculate the transition states of water splitting (). A kinetic energy cutoff of 450 eV was set. The convergence threshold for the iteration in the self-consistent field was set as 10−4 eV. The geometry optimization within the conjugate gradient method was performed with forces on each atom of less than 0.05 eV/Å. A p (4 by 4) unit cell of Pd (111) surface with three atomic layers and a p (2 by 2) unit cell of Pd4S (110) surface with three Pd-S layers and two Pd layers were modeled to ensure the lateral lattice larger than 1 nm. To prevent periodic image interactions, a vacuum layer of 15 Å was inserted in the c direction. The bottom Pd atomic layer of Pd (111) surface and bottom Pd-S layer and the sub-bottom Pd layer of Pd4S (110) surface were fixed, while other layers and the adsorbates were fully relaxed during structural optimizations. The Brillouin zone was sampled by a k-point mesh of 4 by 4 by 1.

The reaction Gibbs free energy for each elementary step was calculated according to the following expression: ∆G = ∆EDFT + ∆EZPE − T∆S, where ∆EDFT is the total energy difference between the products and the reactants of each reaction step and ∆EZPE and ∆S are the differences of zero-point energy and entropy, respectively. The zero-point energy of free molecules and adsorbates were obtained from the vibrational frequency calculations. The reaction free energy of each step that involves an electrochemical proton-electron transfer was described by the computational hydrogen electrode model proposed by Nørskov et al. ().

Finite element method simulations

Nernst-Planck-Poisson calculations of the Gouy-Chapman model were used to elucidate the underlying mechanisms of charge transfer and storage, as well as ion diffusion, governed by the Poisson equation (–)and the Nernst-Plank equationwhere ε0 is the permittivity of vacuum, εr is the relative permittivity of the medium, φ is the electrostatic potential, D is the diffusivity of chemical species i, C is the density of the species, z is the valency of the species, e is the elementary charge, kB is Boltzmann’s constant, and T is the temperature. We used the “Electrostatics” and “Transport of Diluted Species” physics of COMSOL5.5 to obtain the electrochemical behavior of all species. The tip radius of the structure in the panel is 5 nm. The thickness of the Helmholtz layer was taken as the radius of a hydrated potassium ion (0.33 nm) (), the absolute temperature T was taken as 297.3 K, and the diffusion coefficients D of the potassium ion and the hydroxide ion were taken to be 2.14 × 10−9 and 5.30 × 10−9 m2 s−1 in water (). This indicates a 600,000-fold increased surface-adsorbed K+ ion concentration at the Pd needle tip due to a locally enhanced electrostatic field.

Synthesis of CP-supported Pd NTs

Pd NTs were synthesized via electrodeposition. The conventional three-electrode system was used: CP (1 cm by 1 cm), saturated calomel electrode, and carbon rod were used as the working, reference, and counter electrodes, respectively. The electrodeposition was conducted in the electrolyte containing 25 mM PdCl2 and 200 mM HCl at the potential of 100 mV for 600 s. The resulting product was washed by distilled water and absolute ethanol several times and dried in air.

Synthesis of CP-supported ArS-Pd NTs

ArS-Pd NTs were prepared by immersing the freshly prepared Pd NTs in an ethanolic solution of 9 mM sodium thiophenolate at room temperature for 6 hours. The resulting catalyst was rinsed several times by absolute ethanol and dried naturally.

Synthesis of CP-supported ArS-Pd4S@Pd NTs

ArS-Pd NTs were immersed into 30 ml of DMF and kept at 60°C for 12 hours. After cooling down to room temperature, the resulting product was washed by absolute ethanol several times and dried in air.

Synthesis of CP-supported Pd4S@Pd NTs

The freshly prepared Pd NTs were immersed into the aqueous solution containing 9 mM Na2S and kept at 80°C for 12 hours. After cooling down to room temperature, the resulting product was washed by distilled water and absolute ethanol several times and then dried in air.

Synthesis of Pd NPs

Poly(vinyl pyrrolidone) (molecular weight = 40,000, 0.4 g) was added into 30 ml of absolute ethanol and heated to 60°C under stirring to form a homogeneous solution. Then, Pd(acac)2 (50 mg) was dissolved, and the freshly prepared solution of NaBH4 (200 mg in 8 ml of ethanol) was then added dropwise. After 30 min, the resulting product was isolated by centrifugation and washed with ethanol.

Synthesis of ArS-Pd4S@Pd NPs

The freshly prepared Pd NPs were immersed in an ethanolic solution of 9 mM sodium thiophenolate at room temperature for 12 hours. Then, the collected products were immersed into 30 ml of DMF and kept at 60°C for 12 hours. After cooling down to room temperature, the resulting product was isolated by centrifugation. The loading mass of ArS-Pd4S@Pd NPs for each electrode is 2.0 mg.

Characterizations

The morphology of the catalysts was observed by a FEI Apreo S LoVac scanning electron microscope with an accelerating voltage of 10.0 kV. The nanostructure and element distribution of the catalysts were characterized using a FEI Tecnai G2 F20 transmission electron microscope equipped with EDX spectroscopy and operating at 200.0 kV. The XRD patterns were analyzed in the range of 10° to 90° at a scan rate of 20° min−1 using a Rigaku Smartlab 9KW diffraction system with a Cu Kα source (λ = 1.54056 Å). The XPS measurements were performed on a Thermo Fisher Scientific ESCALAB-250Xi spectrometer using a monochromatic Al Kα x-ray beam (1486.60 eV). All the peaks were calibrated by the binding energy of 284.8 eV of the C 1s spectrum. The x-ray absorption spectroscopy (XAS) of S K-edge was undertaken under an ultrahigh vacuum at the 1W1B beamline of the Beijing Synchrotron Radiation Facility. The XAS spectra were analyzed with the ATHENA software package as reported (). The nuclear magnetic resonance (NMR) spectra were recorded on Varian Mercury Plus 400 instruments at 400 MHz (1H NMR) and 101 MHz (13C NMR) with chloroform-d as the solvent. Chemical shifts were reported in parts per million downfield from internal tetramethylsilane. Multiplicity was indicated as follows: s (singlet), d (doublet), t (triplet), m (multiplet), br (broad). Coupling constants were reported in hertz. The quantitative analysis of the liquid products was conducted with a gas chromatograph (Agilent, 7890A) with thermal conductivity, flame ionization detector, and HP-5ms capillary column (0.25 mm in diameter, 30 m in length). Identification of the reactants and products was performed using a gas chromatography (GC)–mass spectrometry (TRACE DSQ) with HP-5ms capillary column (0.25 mm in diameter, 30 m in length). The injection temperature was set at 300°C. Nitrogen was used as the carrier gas at 1.5 ml min−1. The quantitative analysis of the products for every 20 min during the time-dependent test was performed by the high-performance liquid chromatography (Agilent 1220).

Electrochemical measurements

Electrochemical measurements were carried out in a divided three-compartment electrochemical cell consisting of a working electrode, a carbon rod counter electrode, and a Hg/HgO reference electrode. The cathode cell (10 ml) and anode cell (10 ml) containing 1.0 M KOH aqueous solution (5.0 ml) were separated by the membrane. Alkynes (0.2 mmol) dissolved in 2.0 ml of Diox were rapidly added into the cathode cell and stirred to form a homogeneous solution. Then, chronoamperometric measurement was carried out at −1.1 V versus Hg/HgO under stirring until the starting substrates were depleted. After that, the products at the cathode were extracted with dichloromethane (DCM). The DCM phase was removed, and the residuals were subjected to be separated either by flash column chromatography or using a thin-layer chromatography plate to give the isolated yields or were analyzed by GC to provide the GC yields. The yields were calculated by dividing the amount of the obtained desired product by the theoretical yield. The GC yields were calculated according to standard calibration curves. The procedure and reaction setup of electrochemical semideuteration of alkynes was similar to the semihydrogenation process except that the H2O and 1.0 M KOH were replaced by D2O and 0.5 M K2CO3, respectively.

Scavenge of high active *H with t-BuOH

To ensure the reliability of the comparison experiment, a piece of ArS-Pd4S@Pd NT electrode was cut into two halves for electrocatalytic semihydrogenation of alkynes with/without t-BuOH. A total of 0.1 mmol of 1a was added into the electrolytic cell for the following semihydrogenation. Chronoamperometry was carried out at a given constant potential −1.1 V versus Hg/HgO for 160 min. The content of 2a was detected every 20 min.

Surface-adsorbed K+

ArS-Pd4S@Pd NTs and NPs were run in 20 mM KOH at −1.1 V versus Hg/HgO. After 10 min, the K+ ion concentrations in the electrolytes were detected by ICP-AES. The adsorbed K+ was then estimated on the basis of the loss of K+ concentrations in the electrolytes.