In situ surface reconstruction of silver leads to competent activity for the electrocatalytic hydrogenation of 5-hydroxymethylfurfural

Abstract

Electrochemical reconstruction is a well-documented phenomenon in electrocatalysis, yet it has rarely been purposefully harnessed to enhance catalyst performance in a controllable manner. In this study, we introduce a rational electrochemical reconstructing strategy using Ag as a model electrocatalyst and the electrocatalytic hydrogenation of 5-hydroxymethylfurfural (HMF) to 2,5-bis(hydroxymethyl)furan (BHMF) as a benchmark reaction. Through a simple yet effective electrochemical reconstruction approach, we generated preferentially oriented Ag nanoparticles (PO-Ag NPs) with more exposure of the Ag(110) facets, which exhibited significantly enhanced activity and selectivity compared to pristine Ag. The faradaic efficiency of PO-Ag NPs for HMF hydrogenation reached above 95% with a selectivity exceeding 98%. Comprehensive electrochemical, spectroscopic, and computational investigations identify the preferential adsorption of HMF on the Ag(110) facet as the primary factor contributing to the enhanced performance. Furthermore, PO-Ag NPs demonstrated excellent long-term stability, retaining high activity and selectivity over extended electrolysis cycles. These findings highlight electrochemical reconstruction as a powerful and controllable strategy for designing highly efficient electrocatalysts, with broad implications for a wide range of electrochemical transformations beyond hydrogenation.

---

Nan Jiang

Nan Jiang earned her doctorate (2018) in Chemistry from Utah State University, United States. She then joined the University of Nevada, Las Vegas (UNLV) and Advanced Light Source (ALS) at Lawrence Berkeley National Laboratory (LBL) as a postdoctoral fellow. Currently, she is an Assistant Professor at Guizhou University. Her current research focuses on designing and developing electrode materials for various energy storage and conversion devices. She has published over 40 peer-reviewed papers in international journals. She was awarded by High-Level Talents Returning to China Program in 2024 and was included on the “World's Top 2% Scientists” list for 2023 by Mendeley Data.

Introduction

To reduce reliance on fossil resources, biomass upgrading using renewable energy has emerged as an appealing alternative for the sustainable production of fuels and green chemicals.^1,2 5-Hydroxymethylfurfural (HMF), a versatile platform molecule derived from biomass, can be efficiently obtained through the hydrolysis of cellulose with high yields.^3,4 Given the presence of aldehyde and alcohol functional groups anchored to the furan ring, HMF can be further transformed into a variety of commodity chemicals through processes, such as oxidation,⁵ hydrogenation,^6–8 hydrogenolysis,⁹ reductive amination,^10,11 open-ring reaction,¹² cleavage,¹³ and esterification.¹⁴ Among these upgraded products, 2,5-bis(hydroxymethyl)furan (BHMF), produced by hydrogenating the aldehyde group of HMF, plays a vital role in manufacturing polyurethane foams, polyesters, resins, artificial fibers, and pharmaceuticals.^7,15 To date, the industrial production of BHMF still depends on conventional thermocatalytic methods, which demand noble metal catalysts, polluting organic solvents, and harsh operating conditions including high H₂ pressures (28–350 bar) and temperatures (403–423 K),¹⁵ raising concerns regarding safety, energy consumption, and environmental issues.

Due to its environmentally friendly and carbon-neutral nature, the electrocatalytic hydrogenation of HMF, conducted under ambient conditions with H₂O as the proton source, has gained significant attention.^9,15–20 The electrocatalytic hydrogenation of HMF towards BHMF can proceed through either a direct electroreduction pathway or an adsorbed hydrogen (H*) route (Scheme 1).^21,22 In the direct electroreduction pathway (blue line), the carbonyl group (C=O) in HMF acquires one H⁺/e⁻ pair via proton-coupled electron transfer (PCET), leading to the formation of a radical intermediate (C*–OH). The generated C*–OH can then either react with another H⁺/e⁻ pair to produce BHMF or undergo dimerization with another C*–OH radical through C–C coupling (red line in Scheme 1), resulting in the production of bis(hydroxymethyl)hydrofuroin (BHH). In particular, dimerization tends to occur at high HMF concentrations and very negative potential, which favor the over-accumulation of C*–OH and limit the PCET process.¹⁷ In contrast, the H* route generates in situ H* on the surface of the electrode via the electrochemical reduction of H⁺ or H₂O (Volmer step). The subsequent hydrogenation of the adjacent adsorbed HMF* with H* follows the Langmuir–Hinshelwood (L–H) mechanism, yielding BHMF (green line in Scheme 1). However, the H₂ evolution reaction (HER), which consumes H* via Tafel or Heyrovsky steps, has been considered as the major competing reaction, thereby lowering the faradaic efficiency (FE). This issue is particularly pronounced for platinum-group metals, which possess low hydrogen activation barriers. As a result, the two routes may compete with each other. The preference highly depends on the relative potentials for the formation of H* and C*–OH.

Scheme 1 Schematic illustration of the electrocatalytic hydrogenation of HMF towards BHMF through two pathways.

Owing to their exceptional catalytic ability in the electrocatalytic hydrogenation of HMF towards BHMF and relatively low HER activity, Ag-based electrocatalysts have garnered considerable research attention.^23–25 Previous studies have demonstrated the outstanding electrocatalytic hydrogenation performance of Ag foil over other metal foil compositions in slightly alkaline electrolytes.^26,27 Combined experimental and computational investigations elucidate that the moderate adsorption of HMF on Ag foil leads to the selective hydrogenation of HMF to BHMF.²⁸ Moreover, HMF hydrogenation on Ag-based electrodes is more favorable than the HER, offering a wide potential window and a high FE. To further boost BHMF productivity and selectivity, Ag-based bimetallic catalysts and Ag nanocomposites with synergistic effects have been developed. For instance, AgCu bimetallic electrocatalysts demonstrate superior catalytic performance for the hydrogenation of HMF to BHMF, primarily due to the stronger adsorption of HMF on Cu and a synergistic effect between Ag and Cu.²⁹ Additionally, Li et al. designed a composite catalyst by immobilizing Ag nanoparticles (Ag NPs) on SnO₂ sheets, promoting the electrocatalytic HMF hydrogenation to BHMF with a high FE.¹⁷ The oxygen vacancies on SnO₂ function as electrophilic sites to selectively adsorb the carbonyl group in HMF, which then reacts with the adsorbed H* species generated by Ag NPs to produce BHMF. However, it is highly desired but remains a grand challenge to achieve efficient BHMF formation with high HMF concentrations due to the tendency for dimerization and the severe competition from the HER at various negative potentials.

It is widely recognized that the crystal facet effect strongly modulates the chemisorption of active reactants, hence controlling the reaction pathways and product selectivity/distribution. In 2021, Choi and co-workers investigated the orientation dependence of the electrocatalytic HMF reduction on Ag single-crystal surfaces using density functional theory (DFT) calculations.²⁸ Their findings indicate that adsorbed HMF molecules exhibit the strongest binding affinity to the Ag(110) facet, with decreasing binding strength on the Ag(100) and Ag(111) facets. However, to date, it has been challenging to experimentally modify the arrangement of surface Ag atoms to generate preferentially oriented Ag nanoparticles (PO-Ag NPs) as highly active electrocatalysts for HMF hydrogenation. Herein, we employ a facile in situ electrochemical oxidation–reduction method to rearrange the surface Ag atoms of Ag foil, exposing crystal orientations dominated by the Ag(110) facet and significantly enhancing the performance of electrocatalytic hydrogenation of HMF to BHMF. Multiple characterization techniques and electrochemical measurements have confirmed the successful reconstruction of the Ag foil surface to generate a higher proportion of the Ag(110) facet. The resulting PO-Ag NPs show a high FE and a wide potential window for BHMF production, even at high HMF concentrations. Operando characterization experiments together with DFT calculations reveal that the stronger adsorption of HMF on Ag(110) is the key factor for the enhanced electrocatalytic performance of PO-Ag NPs.

Experimental section

Chemicals and materials

Ag foil (thickness: 0.25 mm, 99.99%) and single-crystal surfaces of Ag(111), Ag(110), and Ag(100) were purchased from Hefei Kejing Materials Technology Co., Ltd. 5-Hydroxymethylfurfural (HMF, 99.5%), boric acid (H₃BO₃, 99.9%), 2,5-bishydroxymethylfuran (BHMF, >99.5%), potassium hydroxide (KOH, 99.99%), and potassium bicarbonate (KHCO₃, 99.5%) were obtained from Shanghai Aladdin Biochemical Technology Co., Ltd and used without any further purification.

Synthesis of PO-Ag NPs

Ag foil (0.5 cm × 0.5 cm) was employed as the working electrode, with a Ag/AgCl (saturated KCl) electrode as the reference electrode and a carbon rod as the counter electrode. Linear sweep voltammetry (LSV) was performed on Ag foil within a potential range from −0.06 to −1.26 V vs. RHE at a scan rate of 5 mV s⁻¹ for 5 cycles in a 0.5 M KHCO₃ solution. Subsequently, the potential was set at 8.74 V vs. RHE for 5 seconds to produce a uniform layer of Ag₂CO₃ on the surface of Ag foil. Finally, PO-Ag NPs were synthesized by conducting LSV within the same potential range (–0.06 to −1.26 V vs. RHE, 5 mV s⁻¹) for 5 cycles, followed by applying a negative potential of −1.76 V vs. RHE for 120 seconds in a fresh 0.5 M KHCO₃ solution.

Material characterization studies

Scanning electron microscopy (SEM) and elemental mapping tests were performed using a Thermo Scientific Quattro S instrument equipped with an EDAX ELECT PIUS energy-dispersive X-ray spectroscopy (EDS) system. PO-Ag NPs were detached from Ag foil and further characterized by transmission electron microscopy (TEM) using a FEI-Talos F200S microscope operating at 200 kV. X-ray diffraction (XRD) patterns were recorded on a Bruker D8 Advance diffractometer with Cu Kα radiation. X-ray photoelectron spectroscopy (XPS) measurements were conducted using a Thermo Fisher Scientific K-Alpha spectrometer. The C 1s peak was calibrated to 284.8 eV to ensure accurate alignment of all XPS data.

Operando Raman measurements

Operando Raman tests were conducted using a LabRAM HR Raman microscope (Horiba Jobin Yvon, HR550) equipped with a 532 nm laser for excitation, a Synapse CCD detector, a monochromator with a 1800 grooves per mm grating, and a water immersion objective (Olympus LUMFL, 60×, numerical aperture = 1.10). Each spectrum was acquired over a period of 60 seconds. The electrochemical Raman study was carried out in a custom-made cell, using a Ag/AgCl electrode as the reference electrode and a Pt wire as the counter electrode. The potential was precisely controlled with an Ivium-n-Stat electrochemical workstation.

Quasi in situ electron paramagnetic resonance spectroscopy

Electron paramagnetic resonance (EPR) spectra were acquired using a Bruker EMXplus (298 K, 9.843483 GHz) with a sweep width of 100.0 G and a sweep time of 60 seconds. The EPR measurements were performed in an H-type cell, with a Ag/AgCl electrode functioning as the reference electrode and a carbon rod serving as the counter electrode. After electrolysis at –0.51 V vs. RHE for 3 min, the sample was collected and subjected to EPR tests with 5,5-dimethyl-1-pyrroline-N-oxide (DMPO, 50 mM) as the trapping agent.

Electrochemical measurements

Electrochemical measurements were conducted using a BioLogic SP-150e potentiostat. Both HER and HMF hydrogenation experiments were carried out in an H-type cell (with 15 mL capacity per chamber), separated by a Nafion 211 membrane. In these experiments, Ag foil (0.5 cm × 0.5 cm) and PO-Ag NPs (0.5 cm × 0.5 cm) were used as working electrodes, while a carbon rod worked as the counter electrode and a Ag/AgCl electrode acted as the reference electrode. The electrolyte consisted of a N₂-saturated 1 M borate buffer solution (pH = 9.2, 10 mL). All HER and HMF hydrogenation tests were carried out under magnetic stirring and continuous N₂ bubbling. Unless otherwise specified, LSV measurements and electrolysis reported in this work were corrected with 85% iR-correction. Operando electrochemical impedance spectroscopy (EIS) tests were conducted at –0.41 V vs. RHE in a frequency range from 0.1 to 100 000 Hz. Potential values were converted to the reversible hydrogen electrode (RHE) scale using the equation:

E(V vs. RHE) = E(V vs. Ag/AgCl) + 0.059 × pH + 0.197 (1)

The overpotentials of the HER and HMF hydrogenation were calculated using the following equations:²⁵

Overpotential of HER = 0 − E_HER(V vs. RHE) (2)

Overpotential of HMF hydrogenation = 0.12 − E_HMF(V vs. RHE) (3)

where E_HER and E_HMF represent the applied potentials for the HER and HMF hydrogenation, respectively.

For rotating disk electrode (RDE) measurements, PO-Ag NPs or Ag foil was fixed onto a glassy carbon electrode with an exposed area of 0.5 cm × 0.5 cm, which served as the working electrode. LSV tests were performed at rotating rates ranging from 1000 rpm to 2000 rpm in the presence of 50 mM HMF. The electron transfer number (n) for HMF hydrogenation was determined from the slope of the Koutecký–Levich plot, defined by the equation:

Slope = (0.62nFD^2/3ν^−1/6C)⁻¹ (4)

where F is the Faraday constant (96 500 C mol⁻¹), D is the diffusion coefficient of HMF in water (9.169 × 10⁻⁶ cm² s⁻¹), ν is the kinematic viscosity of water (0.01 cm² s⁻¹), and C is the concentration of HMF (50 mM).²⁴

Product analysis

The liquid samples were initially diluted 25 times or 50 times and then quantitatively analyzed by high-performance liquid chromatography (HPLC, Essentia LC-16) equipped with an ultraviolet photometric detector at 225 nm. HMF and its reductive products were determined using a Shim-pack GIST C18 column (150 mm and 5 μm) maintained at 45 °C. The analysis utilized a gradient elution method: starting with a 15 : 85 ratio of CH₃CN to H₂O for 4 minutes, followed by a 2-minute gradient elution that increased the CH₃CN ratio to 40% and concluding with a 4-minute gradient elution that reduced the CH₃CN ratio back to 15%. In addition, BHH was identified using liquid chromatography mass spectrometry (LC-MS, Agilent 6545B), which was equipped with a 1290 Infinity liquid chromatography system. The flow rate was set at 0.5 mL min⁻¹, and the detection wavelength was 225 nm. The FE, selectivity, and productivity of BHMF were calculated with the following equations:

The FE of BHH was calculated by the following equation, as it is not commercially available.³⁰

Computational method

DFT calculations were performed using the generalized gradient approximation (GGA) with the Perdew–Burke–Ernzerhof (PBE) functional to calculate the adsorption energy of substrate on the catalyst.^31,32 The ionic cores were described using projected augmented wave (PAW) potentials, while the valence electrons were treated with a plane wave basis set, applying a kinetic energy cutoff of 450 eV.³³ The convergence criteria for electronic energy and force were set to 10⁻⁶ eV and 0.03 eV Å⁻¹, respectively. A supercell was constructed for the face-centered cubic Ag(100), Ag(110), and Ag(111) facets as the support structure. To minimize the interactions between repeating images, a vacuum layer of at least 20 Å was introduced. During geometry optimization, the bottom half layers of the Ag substrate were fixed, while the upper half was allowed to relax. The Brillouin zone was sampled with a 2 × 2 × 1 Monkhorst–Pack k-point mesh. Dispersion interactions were incorporated into the calculations by using the DFT-D3 correlation correction. Adsorption energies (E_ads) were calculated using the following equation:

E_ads = E_total − E_Ag-sub − E_HMF (9)

where E_total, E_Ag-sub, and E_HMF represent the total energies of the optimized HMF/Ag-substrate system, the Ag-substrate, and the HMF molecule, respectively.

Results and discussion

Catalyst synthesis and characterization

PO-Ag NPs were synthesized using an in situ electrochemical oxidation–reduction approach, as illustrated in Fig. 1a. Initially, Ag foil was immersed in a 0.5 M KHCO₃ solution and subjected to a potential of 8.74 V vs. RHE for 5 seconds. This rapid and intense anodization oxidizes the surface Ag atoms of Ag foil (AD-Ag), forming a layer of monoclinic Ag₂CO₃. During this process, the crystallographic structure is disrupted, and Ag atoms are selectively removed from the Ag(111) plane due to anisotropic etching and crystallographic strain. In the subsequent reduction step, monoclinic Ag₂CO₃ transformed back into face-centered cubic (fcc) metallic Ag, accompanied by the accumulation of crystal defects, which results in the preferential exposure of the Ag(110) facet. SEM and XRD were used to characterize Ag foil, AD-Ag, and PO-Ag NPs. Ag foil shows a flat surface (Fig. 1b), and its XRD pattern (Fig. 1e) reveals five peaks corresponding to fcc metallic Ag. Following the electrochemical oxidation of the surface Ag to Ag₂CO₃, a stacked cube structure emerged (Fig. 1c), along with the formation of interparticle pores, which is confirmed as the monoclinic phase of Ag₂CO₃ by the XRD pattern (Fig. 1f). Elemental mapping images of AD-Ag (Fig. S1†) demonstrate a homogeneous distribution of C, O, and Ag, indicating the formation of a uniform layer of Ag₂CO₃. As revealed by XRD (Fig. 1g), monoclinic Ag₂CO₃ is converted back to metallic Ag during the reduction process. Notably, the Ag(110) and Ag(100) facets are not visible in the XRD pattern due to forbidden reflections. However, their existence can be reasonably inferred proportionally to the Ag(220) and Ag(200) peaks, respectively, since these planes are aligned parallel to one another in a fcc structure. Thus, the increased intensity of the Ag(220) peak in PO-Ag NPs implies more exposure of the Ag(110) facet. The resulting PO-Ag NPs exhibit a coral-like nanocolumn morphology (Fig. 1d) with a thickness of approximately 2.5 μm (inset in Fig. 1d), leading to a higher electrochemically active surface area (ECSA) than that of the pristine Ag foil (Fig. S2†).

Fig. 1 (a) Schematic illustration of the synthesis process for PO-Ag NPs. SEM images of (b) Ag foil, (c) AD-Ag, and (d) PO-Ag NPs (inset: cross-section of PO-Ag NPs). XRD patterns of (e) Ag foil, (f) AD-Ag, and (g) PO-Ag NPs.

The detailed surface structure of PO-Ag NPs was further investigated by high-resolution transmission electron microscopy (HRTEM), XPS, and HCO₃⁻ adsorption/desorption experiments. According to the HRTEM images (Fig. 2a), PO-Ag NPs consist of interconnected nanoparticles with an average diameter of ∼60 nm. These interconnected nanoparticles create numerous grain boundaries (Fig. S3†), which contain uncoordinated Ag atoms that can facilitate the adsorption of both H₂O and HMF molecules, thereby promoting HMF hydrogenation.³⁴ Additionally, clear lattice fringes (Fig. 2a) with interplanar spacings of 0.24, 0.20, and 0.15 nm are observed, corresponding to the (111), (200), and (220) planes of Ag, respectively. Elemental mapping images of PO-Ag NPs (Fig. S4†) display a decrease in O content due to the reduction of Ag₂CO₃ to metallic Ag (Tables S1 and S2†). Moreover, the presence of C and O confirms the residual CO₃²⁻ in PO-Ag NPs.

Fig. 2 (a) HRTEM images of PO-Ag NPs. (b) CV curve of PO-Ag NPs in 0.5 M KHCO₃ saturated with N₂ at a scan rate of 50 mV s⁻¹ (inset: the corresponding facet ratio obtained from HCO₃⁻ adsorption/desorption experiments). XPS spectra of (c) Ag 3d and (d) C 1s.

To precisely quantify the Ag(110) facet on the surface of PO-Ag NPs, HCO₃⁻ adsorption/desorption tests were conducted.³⁵ As shown in the cyclic voltammetry (CV) curve (Fig. 2b), the peaks observed around −0.65 to −0.75 V vs. SHE, −0.45 to −0.55 V vs. SHE, and −0.25 to −0.35 V vs. SHE can be attributed to the electro-adsorption/desorption of HCO₃⁻ on Ag(110), Ag(100), and Ag(111) facets, respectively. The ratio of Ag(110), Ag(100), and Ag(111) facets calculated from the integral peak area is 3.7 : 1.4 : 1, indicating a predominance of Ag(110) over Ag(100) and Ag(111). In contrast, Ag foil exhibits only one peak around −0.25 to −0.35 V vs. SHE (Fig. S5†), confirming the dominance of the Ag(111) facet.

XPS analysis was performed to examine the electronic structures of Ag foil, AD-Ag, and PO-Ag NPs (Fig. S6†). The high-resolution XPS spectra of Ag 3d (Fig. 2c) reveal that both Ag foil and PO-Ag NPs are primarily in the metallic state of Ag⁰, indicating the reduction of AD-Ag to metallic Ag during electroreduction. Notably, the binding energy of PO-Ag NPs shifts to a lower value compared to that of Ag foil, indicating an upshift of the d-band center, which could facilitate the selective adsorption and activation of the C=O group in HMF, thereby enhancing the catalytic performance.³⁶ Additionally, the lower binding energy of PO-Ag NPs implies a higher electron density on the surface, which may be ascribed to the charge transfer from the residual CO₃²⁻ to Ag (Fig. 2d). This increased electron density could lead to stronger adsorption of the C=O group in HMF through efficient back donation, thus activating the C=O group and improving the catalytic activity.³⁷

Electrocatalytic tests

To evaluate the performance of the electrocatalytic hydrogenation of HMF on PO-Ag NPs, LSV was carried out in a 1 M borate buffer solution (pH = 9.2) containing 50 mM HMF. As shown in Fig. 3a, both Ag foil and PO-Ag NPs demonstrate significantly enhanced electrocatalytic activity for HMF hydrogenation compared to the HER, implying more favorable thermodynamics for HMF hydrogenation. In particular, PO-Ag NPs exhibit the most pronounced shifts in both the onset potential and current density for HMF hydrogenation compared to Ag foil. PO-Ag NPs reach a current density of −32.5 mA cm⁻² at −0.45 V vs. RHE, which is 6.5 fold higher than that of Ag foil (Fig. S7†). More importantly, the electrocatalytic activities of Ag foil and PO-Ag NPs for the HER are comparable (Fig. 3a and b). The slightly higher current density of PO-Ag NPs right after the onset potential might stem from its larger surface area (Fig. S2†). Therefore, while the crystal facet effect has a very limited promoting effect on the HER, it significantly boosts HMF hydrogenation, establishing a wide potential window (>300 mV) between the HER and HMF hydrogenation.

Fig. 3 (a) LSV curves for Ag foil and PO-Ag NPs in 1 M borate buffer solution (pH = 9.2) with and without 50 mM HMF saturated with N₂ at a scan rate of 5 mV s⁻¹. (b) Tafel slopes for the HER and HMF hydrogenation over Ag foil and PO-Ag NPs. (c) FE and selectivity of BHMF for 50 mM and 100 mM HMF hydrogenation at −0.51 V vs. RHE using Ag foil and PO-Ag NPs as working electrodes. (d) FE and selectivity of BHMF for 50 mM HMF hydrogenation at −0.51 V vs. RHE using single-crystal Ag(111), Ag(110), and Ag(100) as working electrodes.

Subsequently, Tafel plots were utilized to assess the kinetics of the HER and the electrocatalytic HMF hydrogenation (Fig. 3b). The Tafel slopes of Ag foil and PO-Ag NPs for the HER are 116.4 and 112.9 mV dec⁻¹, respectively, which illustrate that the rate-determining step (RDS) is the Volmer step for both electrodes. When 50 mM HMF was introduced into the electrolyte, the Tafel slope of PO-Ag NPs decreased slightly to 103.0 mV dec⁻¹, suggesting more favorable kinetics for the electrocatalytic HMF hydrogenation compared to the HER. Similar Tafel slopes for the HER and HMF hydrogenation on PO-Ag NPs indicate that the introduction of HMF holds a minimal impact on the reaction kinetics; thus, the RDS likely remains as the Volmer step during HMF hydrogenation. In contrast, Ag foil presented a considerably larger Tafel slope (175.5 mV dec⁻¹) upon the addition of HMF, demonstrating sluggish kinetics and a change in the RDS. The distinct kinetic behaviors observed for HMF hydrogenation between Ag foil and PO-Ag NPs presumably imply different reaction mechanisms. Additionally, operando EIS was conducted at –0.41 V vs. RHE in the presence of 50 mM HMF. Fig. S8† presents the Nyquist plots for the HMF hydrogenation over Ag foil and PO-Ag NPs. The semicircular portion of the plots corresponds to the charge transfer resistance (R_ct). Compared to Ag foil, PO-Ag NPs exhibit a significantly lower R_ct value for HMF hydrogenation, indicating faster charge transport kinetics on PO-Ag NPs.

To quantitatively identify the reductive products, chronoamperometry experiments were performed over Ag foil and PO-Ag NPs at −0.51 V vs. RHE in an H-type cell with a fixed passing charge equivalent to converting half of the HMF to BHMF with a 100% FE. The liquid products and the remaining HMF were analyzed by HPLC (Fig. S9 and S10†). Additionally, BHH was identified through LC-MS (Fig. S11†). Fig. 3c illustrates that PO-Ag NPs exhibit significantly higher FEs for the hydrogenation of 50 mM and 100 mM HMF (94.0 and 91.2%, respectively) compared to Ag foil (76.8 and 68.4%, respectively). Notably, the selectivity of BHMF on PO-Ag NPs is nearly 1.5 times and 2 times larger than that on Ag foil for 50 mM and 100 mM HMF hydrogenation, respectively. To further demonstrate the facet effect, single-crystal Ag(111), Ag(110), and Ag(100) were employed directly as working electrodes for HMF hydrogenation. As depicted in Fig. S12a,† single-crystal Ag(111), Ag(110), and Ag(100) demonstrated comparable overpotentials for the HER. The Tafel slopes of single-crystal Ag(111), Ag(110), and Ag(100) for the HER are 204.7, 210.5, and 222.1 mV dec⁻¹, respectively (Fig. S12b†), which illustrate that the RDS is the Volmer step for three single-crystal electrodes, and Ag(111) has the fastest kinetics for the HER. Regarding HMF hydrogenation (Fig S12c and d†), Ag(110) displays a lower overpotential and Tafel slope than those of Ag(100) and Ag(111), indicating more favorable thermodynamics and kinetics for HMF hydrogenation on Ag(110). Furthermore, controlled potential electrolysis (Fig. 3d) reveals that the FEs and selectivity of the three single-crystal electrodes follow the order: Ag(110) > Ag(100) > Ag(111). This result confirms that the facet effect plays a crucial role in promoting HMF hydrogenation. Interestingly, even the single-crystal Ag(110) shows lower FE and selectivity for BHMF compared to PO-Ag NPs, indicating that additional factors such as morphology and the presence of residual CO₃²⁻ may also contribute to the observed performance.

To investigate the potential dependence of electrocatalytic HMF hydrogenation, chronoamperometry experiments were carried out at various applied potentials ranging from −0.41 V to −0.71 V vs. RHE, with a fixed passing charge equivalent to converting half of the HMF to BHMF with a 100% FE. For 50 mM HMF hydrogenation catalyzed by PO-Ag NPs, only two reduction products (BHMF and BHH) were detected under the investigated conditions, whereas multiple species were observed for Ag foil (Fig. S13†). As shown in Fig. 4a and S14,† PO-Ag NPs exhibit higher FE and selectivity for BHMF compared to Ag foil across the investigated potentials. Importantly, PO-Ag NPs displayed enhanced performance, achieving a near-100% FE for BHMF at −0.46 V vs. RHE. Upon further increasing the applied potentials, the production of BHH decreased and eventually disappeared beyond −0.56 V vs. RHE, leading to a near-unity selectivity for BHMF. Specifically, PO-Ag NPs achieved a productivity of 0.872 mmol cm⁻² h⁻¹ at −0.51 V vs. RHE, which was 23 times higher than that of Ag foil (0.037 mmol cm⁻² h⁻¹). Although the selectivity of BHMF increased as the applied potential became more negative, a continuous decline in the FE of BHMF was observed due to the competition of the HER at higher potentials. These results demonstrate that PO-Ag NPs not only realize outstanding performance in the electrocatalytic hydrogenation of HMF but also effectively suppress the HER at relatively low potentials. We also performed chronoamperometry tests over PO-Ag NPs with a higher HMF concentration of 100 mM at different potentials (Fig. 4b). At relatively low potentials, an increase in HMF concentration led to a decrease in the selectivity for BHMF owing to the adequate supply of the C*–OH intermediate for dimerization. At high potentials, PO-Ag NPs achieved remarkable selectivity for BHMF compared to those at lower potentials. This can be attributed to the sufficient H* produced on the catalyst surface at higher potentials. As a consequence, the competing HER becomes more severe. Compared to the reported studies (Table S3†), PO-Ag NPs achieve superior catalytic activity for electrocatalytic HMF hydrogenation. Moreover, to demonstrate the scalability of this approach, HMF hydrogenation was conducted using a larger electrode (1 cm × 1 cm) with a proportionally increased electrolyte volume (40 mL) at various potentials. Within the operating potential range, both the FE and the selectivity for BHMF slightly increase compared to the smaller electrode (0.5 cm × 0.5 cm) (Fig. S14c†). After 10 reaction cycles at −0.51 V vs. RHE, the FE, selectivity, and current density remain essentially unchanged (Fig. S14d and e†). These results prove that PO-Ag NPs exhibit both scalability and outstanding durability.

Fig. 4 (a) FE and selectivity of BHMF for 50 mM HMF hydrogenation on PO-Ag NPs at various potentials. (b) FE and selectivity of BHMF for 100 mM HMF hydrogenation on PO-Ag NPs at various potentials. (c) Recycling test of PO-Ag NPs for 50 mM HMF hydrogenation at −0.51 V vs. RHE. (d) Conversion of HMF and yield of BHMF during the chronoamperometry experiment conducted at −0.51 V vs. RHE in 1 M borate buffer containing 50 mM HMF.

To investigate the robustness of PO-Ag NPs, 10 consecutive electrolysis cycles were performed at −0.51 V vs. RHE. Even after 10 reaction cycles, the FE, selectivity, and catalytic current density remained essentially unchanged (Fig. 4c and S14f†), indicating the excellent durability of PO-Ag NPs. Post-electrolysis characterization experiments further confirmed the robustness of PO-Ag NPs. SEM images and elemental mappings (Fig. S15†) indicate no significant changes in the nanostructure of PO-Ag NPs after electrolysis. The XRD pattern (Fig. S16a†) proves that the post-electrolysis PO-Ag NPs still maintain a fcc Ag structure. The XPS spectrum of the Ag 3d region for the post-electrolysis PO-Ag NPs shows a similar binding energy to that of the fresh sample (Fig. S16b†), suggesting that the electronic structure of PO-Ag NPs remains unchanged after electrolysis. Fig. S17† presents the CV curve and the corresponding facet ratio obtained from HCO₃⁻ adsorption/desorption experiments using post-electrolysis PO-Ag NPs as the working electrode. The calculated ratios of Ag(110), Ag(100), and Ag(111) were 55.75%, 31.58%, and 12.67%, respectively. Thus, Ag(110) still remains dominant after the reaction. To further explore the charge-dependence of BHMF production, we monitored the concentrations of BHMF and HMF at −0.51 V vs. RHE. It was calculated that ∼96.5 C of charge would be required to completely convert 50 mM HMF, assuming a 100% FE for BHMF production. As shown in Fig. 4d, after passing 96.5 C of charge, the yield of BHMF reaches approximately 43.8 mM, with 5.8 mM of HMF remaining, resulting in a selectivity of 99.1% for BHMF.

Mechanism studies

To elucidate the mechanisms of HMF hydrogenation, the kinetic isotopic effect (KIE) was utilized to gain insights into whether the dissociation of H₂O to generate H* is the RDS. LSV measurements for the HER and HMF hydrogenation were conducted on both Ag foil and PO-Ag NPs in H/D isotopically substituted environments. In the case of the HER, a substantial negative shift is observed in the LSV curve for both Ag foil and PO-Ag NPs (Fig. 5a and b), implying more sluggish kinetics for the D₂ evolution reaction (DER) compared to the HER. This suggests that the dissociation of H₂O to produce H* is indeed the rate-limiting step. For HMF hydrogenation, PO-Ag NPs display a similar KIE to that of the HER, confirming the crucial role of the formation of H* in the reaction mechanism. In sharp contrast, Ag foil shows only a 51 mV cathodic shift at 10 mA cm⁻², which is extremely smaller than a typical KIE. This reduced KIE indicates a fundamentally different mechanism for HMF hydrogenation over Ag foil, where the generation of H* is no longer the RDS. Furthermore, Koutecký–Levich analysis (Fig. S18†) demonstrates that the electron transfer number (n) of HMF hydrogenation catalyzed by PO-Ag NPs is 1.84, which is close to 2. This result supports a two-electron L–H mechanism, leading to nearly 100% selectivity for BHMF production. For Ag foil, the n value is 1.39, implying a dominant one-electron reaction pathway and accounting for the increased production of BHH.

Fig. 5 LSV curves of (a) Ag foil and (b) PO-Ag NPs in pH = 9.2 or pD = 9.2 electrolyte with and without 50 mM HMF. (c) EPR spectra of Ag foil and PO-Ag NPs in pure 1 M borate buffer (top) and in the presence of 50 mM HMF (bottom) at −0.51 V vs. RHE using DMPO as the spin-trapping agent. (d) EPR spectra of Ag foil and PO-Ag NPs in 1.0 M borate buffer containing 50 mM HMF at −0.51 V vs. RHE using DMPO as the spin-trapping agent.

To gain a better understanding of the reaction process, quasi in situ EPR measurements were conducted to detect reaction intermediates using DMPO as the trapping agent, that is, DMPO–H* and DMPO–C*–OH (Fig. S19†).^16,17,38 As displayed in Fig. 5c (top part), H* intermediates are detected in both Ag foil and PO-Ag NPs systems with similar signal intensities in the pure electrolyte, suggesting that the facet effect has a minimal influence on the generation of H* (Volmer step). This observation aligns with the LSV measurements, which present comparable HER catalytic performance between Ag foil and PO-Ag NPs. Upon the addition of HMF, the intensity of the H* signal decreased significantly for both Ag foil and PO-Ag NPs (Fig. 5c bottom part), likely due to the competition from adsorbed HMF molecules or the consumption of H* for HMF hydrogenation. Notably, the H* signal for PO-Ag NPs is slightly weaker than that for Ag foil, implying stronger adsorption of HMF on PO-Ag NPs or more consumption of H* for HMF hydrogenation, consistent with the superior catalytic performance of PO-Ag NPs. In addition, C*–OH intermediates were observed for both Ag foil and PO-Ag NPs (Fig. 5d). It is worth mentioning that C*–OH can either dimerize via C–C coupling to produce BHH or hydrogenate with H* or a pair of H⁺/e⁻ to yield BHMF. Previous studies have demonstrated that a low concentration of C*–OH on the electrode surface is less favorable for dimerization. Instead, they are more likely to react with H* or another H⁺/e⁻ pair to produce BHMF.³⁹ As shown in Fig. 5d, PO-Ag NPs present a lower concentration of C*–OH compared to that of Ag foil, which contributes to the higher selectivity for BHMF and the lower yield of BHH.

The HMF hydrogenation process was further evaluated using operando Raman spectroscopy in the presence of 50 mM HMF. Fig. 6a and b display operando Raman signals for Ag foil and PO-Ag NPs in a custom-made cell, respectively. At the open-circuit potential (OCP), both Ag foil and PO-Ag NPs exhibited distinct vibration peaks at 1522 cm⁻¹ and 1665 cm⁻¹, corresponding to the C=C symmetric stretching vibration in the furan ring and the C=O vibration of the adsorbed HMF molecules, respectively. Additionally, peaks at 1397 cm⁻¹ and 1367 cm⁻¹ were assigned to the C–H scissoring vibration of the furan ring.^30,40 Upon applying a slight negative potential of −0.06 V vs. RHE, the ν(C=O) on the PO-Ag NPs displayed a noticeable red shift to a lower wavenumber, specifically shifting from 1665 cm⁻¹ to 1640 cm⁻¹. Similar red shifts of approximately 20 cm⁻¹ were observed for the ν(C=C) and ν(C–H) peaks on the PO-Ag NPs compared to those on Ag foil. These shifts are attributed to the enhanced adsorption of HMF on the PO-Ag NPs under operating conditions. Notably, the intensity of the HMF vibration peaks on the PO-Ag NPs is stronger than that on Ag foil. Moreover, a new peak at 1560 cm⁻¹ appeared, which was assigned to the asymmetric stretching vibration of the C=C bond in the furan ring. These findings suggest stronger HMF adsorption on the PO-Ag NPs. Regarding Ag foil, the intensity of the HMF vibration peaks gradually decreased as the potential increases, while a more pronounced C=C vibration peak (1559 cm⁻¹) attributed to BHMF emerged at −0.46 V vs. RHE. In stark contrast, a significantly enhanced HMF signal on the PO-Ag NPs was observed, and the BHMF signal appeared at a relatively low potential of −0.26 V vs. RHE. These observations indicate that PO-Ag NPs promote HMF adsorption under the given operating conditions, which has a positive effect on HMF hydrogenation.

Fig. 6 Operando Raman spectra of the electrocatalytic hydrogenation of HMF on (a) Ag foil and (b) PO-Ag NPs in 1.0 M borate buffer containing 50 mM HMF at different potentials. (c) The most stable configurations of HMF adsorbed on the Ag(111), Ag(110), and Ag(100) surfaces.

To elucidate the remarkable HMF hydrogenation performance of PO-Ag NPs, DFT calculations using the PBE functional were conducted to investigate the adsorption of HMF on three Ag facets: (111), (110), and (100). Fig. 6c illustrates the most stable adsorption configuration of HMF on these surfaces. The adsorption energy of HMF via the aldehyde oxygen atom on the three Ag facets follows this order: Ag(110) > Ag(100) > Ag(111), which aligns with the findings from operando Raman analysis, suggesting stronger interactions between HMF and PO-Ag NPs. Additionally, the C=O bond length of HMF adsorbed on Ag(110) is 1.251 Å (Table S4†), which is longer than those on Ag(100) and Ag(111). This elongation of the C=O bond on Ag(110) indicates a weakened C=O bond, potentially facilitating the hydrogenation of HMF to BHMF.

Conclusions

In summary, we successfully synthesized preferentially oriented Ag nanoparticles, predominantly featuring the Ag(110) facet, using a simple electrochemical oxidation–reduction method. These PO-Ag NPs enable the selective hydrogenation of HMF to BHMF at high HMF concentrations and large current densities, achieving superior FE and selectivity across a wide potential range. By combining operando Raman spectroscopy and DFT calculations, we found that the stronger interaction between the C=O group in HMF and PO-Ag NPs promotes the electrocatalytic hydrogenation of HMF to BHMF. Furthermore, electrochemical measurements and EPR analysis indicate that the hydrogenation of HMF on PO-Ag NPs primarily follows the L–H mechanism. This study elucidates the facet effect of Ag in the electrochemical conversion of HMF and provides fundamental insights for designing efficient Ag-based electrocatalysts.

Data availability

The data supporting the results of this study have been included as part of the ESI,† and additional data are available from the corresponding authors upon reasonable request.

Author contributions

D. Q. synthesized the PO-Ag NPs catalyst, conducted the electrochemical measurements, and drafted the original manuscript under the guidance of N. J. and Y. S. N. J. conceptualized the project and designed the experiments. N. J. and Y. S. reviewed and revised the final manuscript. All authors discussed the content and reviewed the manuscript.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 52362028) and Guizhou Provincial Basic Research Program (Natural Science) (No. [2023]042). NJ acknowledges the support from the Natural Science Research Project of the Education Department of Guizhou Province (No. QJJ[2022]001).

References

1. C. Li, X. Zhao, A. Wang, G. W. Huber and T. Zhang, Chem. Rev., 2015, 115, 11559–11624.
2. Y. Jing, Y. Guo, Q. Xia, X. Liu and Y. Wang, Chem, 2019, 5, 2520–2546.
3. J. J. Bozell and G. R. Petersen, Green Chem., 2010, 12, 539–554.
4. C. O. Tuck, Science, 2012, 338, 604.
5. N. Jiang, B. You, R. Boonstra, I. M. T. Rodriguez and Y. Sun, ACS Energy Lett., 2016, 1, 386–390.
6. S. Li, X. Sun, Z. Yao, X. Zhong, Y. Coo, Y. Liang, Z. Wei, S. Deng, G. Zhuang, X. Li and J. Wang, Adv. Funct. Mater., 2019, 29, 1904780.
7. X.-L. Liu, W. Zhong, Y.-F. Jin, T.-J. Wang, X. Xiao, P. Chen, Y. Chen and X. Ai, Chin. J. Catal., 2025, 69, 241–248.
8. Z. Zhao, R. Yu, S. Wang, L. Guo, L. Zhang, M. Chen, J. Chen and H. Zheng, ChemSusChem, 2025, 18, e202401278.
9. P. Hauke, T. Merzdorf, M. Klingenhof and P. Strasser, Nat. Commun., 2023, 14, 4708.
10. H. Liu, W. Tang, D. Xue, J. Xiao and C. Wang, Catal. Sci. Technol., 2024, 14, 5764–5769.
11. M. Zhang, S. Xu, M. Boubeche, D. Decarolis, Y. Huang, B. Liu, E. K. Gibson, X. Li, Y. Wang, H. Luo, C. R. A. Catlow and K. Yan, Green Chem., 2022, 24, 9570–9578.
12. J. J. Roylance and K.-S. Choi, Green Chem., 2016, 18, 2956–2960.
13. Y.-R. Zhang, B.-X. Wang, L. Qin, Q. Li and Y.-M. Fan, Green Chem., 2019, 21, 1108–1113.
14. M. Balakrishnan, E. R. Sacia and A. T. Bell, Green Chem., 2012, 14, 1626–1634.
15. M. G. Sendeku, K. Harrath, F. T. Dajan, B. Wu, S. Hussain, N. Gao, X. Zhan, Y. Yang, Z. Wang, C. Chen, W. Liu, F. Wang, H. Duan and X. Sun, Nat. Commun., 2024, 15, 5174.
16. K. Ji, M. Xu, S.-M. Xu, Y. Wang, R. Ge, X. Hu, X. Sun and H. Duan, Angew. Chem., Int. Ed., 2022, 61, e202209849.
17. X. Guo, H. Fu, J. Yang, L. Luo, H. Zhou, M. Xu, X. Kong, M. Shao, H. Duan and Z. Li, ACS Catal., 2023, 13, 13528–13539.
18. B. Chen, Y. Feng, S. Ma, W. Xie, G. Yan, Z. Li, J. Sperry, S. Yang, X. Tang, Y. Sun, L. Lin and X. Zeng, J. Energy Chem., 2023, 76, 421–428.
19. C. Tian, J. Yu, D. Zhou, H. Ze, H. Liu, Y. Chen, R. Xia, P. Ou, W. Ni, K. Xie and E. H. Sargent, Adv. Mater., 2024, 36, 2312778.
20. X. Yuan, K. Lee, J. R. Schmidt, K.-S. Choi and J. Am, Chem. Soc., 2023, 145, 20473–20484.
21. J. J. Roylance, T. W. Kim and K.-S. Choi, ACS Catal., 2016, 6, 1840–1847.
22. X.-Q. Pan, X.-Y. Zhang, G.-X. Huang, S.-C. Mei, J.-W. Huang, J.-J. Chen, W.-J. Liu and H.-Q. Yu, Appl. Catal., B, 2023, 323, 122191.
23. G. S. de Luna, P. H. Ho, A. Sacco, S. Hernandez, J.-J. Velasco-Velez, F. Ospitali, A. Paglianti, S. Albonetti, G. Fornasari and P. Benito, ACS Appl. Mater. Interfaces, 2021, 13, 23675–23688.
24. X. H. Chadderdon, D. J. Chadderdon, T. Pfennig, B. H. Shanks and W. Li, Green Chem., 2019, 21, 6210–6219.
25. H. Liu, T.-H. Lee, Y. Chen, E. W. Cochran and W. Li, Green Chem., 2021, 23, 5056–5063.
26. Y. Kwon, Y. Y. Birdja, S. Raoufmoghaddam and M. T. M. Koper, ChemSusChem, 2015, 8, 1745–1751.
27. Y. Kwon, E. de Jong, S. Raoufmoghaddam and M. T. M. Koper, ChemSusChem, 2013, 6, 1659–1667.
28. D. K. Lee, S. R. Kubota, A. N. Janes, M. T. Bender, J. Woo, J. R. Schmidt and K.-S. Choi, ChemSusChem, 2021, 14, 4563–4572.
29. G. S. de Luna, P. H. Ho, A. Lolli, F. Ospitali, S. Albonetti, G. Fornasari and P. Benito, ChemElectroChem, 2020, 7, 1238–1247.
30. X. Cao, Y. Ding, D. Chen, W. Ye, W. Yang and L. Sun, J. Am. Chem. Soc., 2024, 146, 25125–25136.
31. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1997, 77, 3865–3868.
32. K. Mathew, V. S. C. Kolluru, S. Mula, S. N. Steinmann and R. G. Hennig, J. Chem. Phys., 2019, 151, 234101.
33. P. E. Blöchl, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 50, 17953–17979.
34. Y.-C. Chen, Y.-Z. Chen, C.-M. Wang and Y.-K. Hsu, Int. J. Hydrogen Energy, 2023, 48, 3316–3327.
35. N. Hoshi, M. Kato and Y. Hori, J. Electroanal. Chem., 1997, 440, 283–286.
36. K. Mariandry, R. Kokate, S. V. Somerville, L. Gloag, S. Cheong, L. R. Carroll, P. V. Kumar, J. J. Gooding and R. D. Tilley, Chem. Mater., 2023, 35, 8636–8644.
37. Y. Ye, H. Yang, J. Qian, H. Su, K.-J. Lee, T. Cheng, H. Xiao, J. Yano, W. A. Goddard and E. J. Crumlin, Nat. Commun., 2019, 10, 1875.
38. S. Han, Y. Shi, C. Wang, C. Liu and B. Zhang, Cell Rep. Phys. Sci., 2021, 2, 100337.
39. X. H. Chadderdon, D. J. Chadderdon, J. E. Matthiesen, Y. Qiu, J. M. Carraher, J.-P. Tessonnier and W. Li, J. Am. Chem. Soc., 2017, 139, 14120–14128.
40. M. Li, T. Zheng, D. Lu, S. Dai, X. Chen, X. Pan, D. Dong, R. Weng, G. Xu and F. Wang, J. Energy Chem., 2023, 84, 101–111.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ta02755a

---