Introduction

The precise large-area controlled growth of atoms provides a possible basis for the unrealized scientific dreams, such as atomic manipulation, single-atom devices, highly ordered atomic arrangement, atomic assembly, selective processing, and single-atom layer devices. Atomically precise growth provides a transformative approach to sufficiently activate the loaded metal atoms to achieve high-performance catalysis, since the formation of catalytical active sites is thermodynamically unfavorable compared to the formation of inert basal plane sites1,2,3,4,5,6. Unfortunately, in traditional nanoparticle catalysts developed so far, only a few edge sites exposed at the surface are active during catalysis, resulting in very low-density loading of active sites7,8,9,10,11,12,13,14,15. The solution to this issue is to synthesize a single-atom-metal layer with a large area, which should theoretically have 100 % exposed active atoms, a maximum number of active sites, and also provide an ideal platform to understand the structure-activity relationship at the atomic level16,17,18,19,20,21,22,23,24,25,26. However, traditional methods are largely limited by complex synthesis conditions, such as high temperatures, special gas pressures, and the use of plasma-assisted deposition. Therefore, it is very different to prepare catalysts based on two or more metals27,28,29,30,31,32,33,34,35. To develop the precise preparation of large-area, metal single-atom layers containing different metals, some traditional methods have been studied, (such as precipitation, impregnation, and high-temperature annealing), in which the supports used are difficult to control the anchoring of metal atoms, undetermined valence states with metals, and severe aggregation, which makes the optimization of catalytic activity and elucidation of the catalytic process and evolution almost impossible36,37,38,39,40,41,42,43,44,45,46,47,48.

A typical example is graphdiyne (GDY)49,50,51,52, containing sp- and sp2−cohybridized carbon atoms, which is a rising-star of carbon allotrope with large π conjugated structures, rich in acetylene bonds, and with infinite natural pores, inhomogeneous charge distribution, and excellent stability53,54,55,56,57,58. The abundance of sp-hybridized alkyne-rich pores, space confined and incomplete charge transfer effects with metals, and superior electrochemical properties of GDY are conducive to the design and synthesis of structurally accurate and high-performance catalytic system59,60,61,62,63,64,65. Increasing evidence shows that GDY materials can be extensively used as supports for anchoring metal atoms to obtain high-performance catalysts63,66,67,68,69,70,71.

In this study, we realized the atomically precise growth of platinum-manganese bimetallic nanoplates on graphdiyne (PtMn/GDY) through the simple GDY-induced adsorption/anchoring method (Fig. 1). The well-defined structure of PtMn/GDY offers a single-atom-thick planar morphology with well-arranged active sites and tunable metal atom electronic states, which are expected to fully activate the loaded metal atoms and achieve maximum activated sites. The incomplete charge transfer effect between GDY and metal atoms, and the synergistic effect between the two atoms could generate specific adsorption sites for olefins/intermediates and improve the electrocatalytic efficiency. As a result, zero-valent PtMn/GDY (Pt0Mn0/GDY) could selectively and efficiently electrooxidize styrene (ST) into 1-phenyl-1,2-ethanediol (PED) by extracting oxygen/hydrogen species from water, achieving a high conversion efficiency of ~100% and high PED selectivity of ~100% at ambient temperature and pressure. Benefiting from this GDY catalytic platform, we can rationally regulate the structure of bimetal metal atom array (MAA) catalysts and explore variations in the active sites during the reaction process. We believe that GDY can provide a distinctive platform for green sustainable energy catalytic systems.

Fig. 1: Schematic illustration of the synthesis routes of Pt0Mn0/GDY.
figure 1

a Synthesis of GDY. b The controlled anchoring of Pt and Mn atoms on GDY.

Full size image

Results

Synthesis and morphological characterizations

Controlled growth of highly ordered Pt and Mn bimetallic single atomic layers was achieved by using a natural atomic growth bimetallic nanoplate method (Fig. 1). In brief, a film of GDY nanosheets was first grown on the surface of a carbon cloth via the coupling of hexacetylenebenzene (HEB), forming the 3D GDY foam. The as-synthesized GDY foam was subsequently used as the substrate for the controlled anchoring of Pt and Mn atoms through a four-step strategy (Fig. 1b) from the initial anchoring of individual metal atoms (step 1) to several metal atoms (step 2), small nanoplates (step 3) and finally, a highly ordered Pt and Mn bimetallic single atomic layer on GDY (step 4).

SEM images show that the GDY foam is composed of a film of vertically aligned and interconnected GDY nanosheets (Fig. 2a, b and Supplementary Fig. 1) with a lattice spacing distance of 0.46 nm (Fig. 2c, d and Supplementary Fig. 2). The GDY foam has a porous morphology which endows it with a large surface area and provides many sites for the anchoring of metal atoms. After the growth of Pt0Mn0, the surface of GDY became rough (Supplementary Fig. 3). High-angle annular dark-field imaging scanning transmission electron microscopy (HAADF-STEM) images show that the Pt/Mn metal atoms can be controllably anchored on the surface of GDY from single Pt-Mn atoms (30 s; Fig. 2e), Pt-Mn clusters of small size (1 min; Fig. 2f), Pt-Mn clusters of larger size (2 min; Fig. 2g), to highly ordered Pt0Mn0 single-atom layer (3 min; Fig. 2h) by simply tuning the reaction time from 30 s to 3 min. Atomic force microscopy (AFM) measurements show that the pristine GDY nanosheets have a thickness of ~1.1 nm (Fig. 2i). Combined with high-resolution transmission electron microscopy (HRTEM) results, the GDY nanosheet was calculated to be three layers thick), and the Pt0Mn0/GDY has a thickness of ~1.6 nm (Fig. 2j, k). The thickness of the grown Pt0Mn0 is ~ 0.42 nm (Fig. 2l and Supplementary Fig. 4), consistent with the thickness of a Pt0Mn0 single-atom layer. The mass loading of Pt and Mn in Pt0Mn0/GDY was 13.14 wt% and 2.36 wt%, respectively, determined by inductively coupled plasma spectrometry (ICP).

Fig. 2: Morphologic characterization of GDY foam and the growth of Pt0Mn0 on GDY.
figure 2

a Low- and b high-magnification SEM, c HRTEM and d SAED images of GDY. HAADF-STEM images of the sample obtained at the reaction time of e 30 s, f 1 min, g 2 min, and h 3 min, respectively. AFM images of i GDY and j, k Pt0Mn0/GDY samples. l Schematic illustration on the measurement of the thickness Pt0Mn0 single-atom layer on the surface of GDY.

Full size image

HRTEM images further demonstrate the uniform dispersion of the Pt0Mn0 single-atom-layer on the surface of GDY (Fig. 3a and Supplementary Fig. 5). The diameter of the Pt0Mn0 single-atom-layer was around 2.4 ± 0.9 nm (Supplementary Fig. 6). Figure 3b shows the copresence of two types of lattices spacings of 0.46 and 0.23 nm for GDY and Pt0Mn0 layers in Pt0Mn0/GDY, respectively, indicating the excellent structural stability of GDY during the anchoring process, which is also confirmed by Raman spectra results (Supplementary Fig. 7). Elemental mapping analysis revealed that only C, Pt, and Mn were present and uniformly distributed in the samples (Supplementary Figs. 8 and 9). HAADF-STEM characterization shows the highly ordered anchoring of metal atoms on GDY (Fig. 3c) and the formation of high-density atomic sites at the edges of the Pt0Mn0 bimetallic layers (Fig. 3d). Using the 3D transformation method, the Pt and Mn atoms can be clearly distinguished (Fig. 3e, f and Supplementary Fig. 10), with Pt atoms densely arranged around Mn atoms (Fig. 3g, h).

Fig. 3: Structural characterizations of the samples.
figure 3

a TEM and b HRTEM images of Pt0Mn0/GDY. c HAADF-STEM image of Pt0Mn0 on GDY. d The enlarged images of the marked regions in c showing the presence of atomic steps at the edge of Pt0Mn0 layers. Distinction of Pt and Mn atoms (red circle) in e HAADF-STEM image and f–h 3D transformation images (g, h are the enlarged images of the dashed dark and red square in f respectively). i Schematic representation of the XPS depth profiling analysis. j Pt 4 f, k Mn 2p and l C 1 s XPS spectra of Pt0Mn0/GDY recorded at different etching times (ET). m Illustration of the charge transfer between metal atoms and GDY.

Full size image

Structural analysis

High-resolution C 1 s X-ray photoelectron spectroscopy (XPS) spectra (Supplementary Fig. 11) show that GDY has four peaks at 284.2 (sp2–C), 285.2 (sp–C), 286.4 (C–O), and 288.5 (C = O). Pt0Mn0/GDY presents a new peak at ~290.1 eV corresponding to a π-π* transition which indicating the presence of the strong interactions between Pt0Mn0 and GDY. The peak area ratio of sp-C/sp2−C for Pt0Mn0/GDY is 2, evidencing the high structural stability of GDY during the anchoring process. Compared with GDY, the negative shift in binding energy for the sp-C peak of the Pt0Mn0/GDY reveals charge transfer from Pt0Mn0 to GDY. In addition, the binding energies of the Pt 4 f (Supplementary Fig. 12) and Mn 2p (Supplementary Fig. 13) are slightly higher than those of pure Pt and Mn, confirming the incomplete charge transfer between Pt0Mn0 and the acetylenic bond. Pt 4 f (Supplementary Fig. 12) and Mn 2p (Supplementary Fig. 13) XPS profiles confirm the zero valence of the Pt and Mn species. Depth-profiling XPS analysis (Fig. 3i) of the obtained Pt 4 f (Fig. 3j), Mn 2p (Fig. 3k), and C 1 s (Fig. 3l) spectra show that there are no changes in the chemical states or composition from the surface to the inner part of the samples. The slight shift in the binding energy in Fig. 3k is mainly due to surface charging effects during the depth profiling of ion sputtering. Efficient charge transfer between GDY and the metal species in Pt0Mn0/GDY (Fig. 3m) may be beneficial for improving the intrinsic activity. Pt0Mn0/GDY possesses a hydrophilic surface (Supplementary Fig. 14), which promotes direct contact of reactants and electrolytes with the surface of the catalysts to maximize the activation of the active sites. In addition, EIS measurements demonstrate that Pt0Mn0/GDY has better conductivity than GDY (Supplementary Figs. 15 and 16). These advantages can improve the catalytic activity of the Pt0Mn0/GDY samples.

The content and the coordination environment of the metal atoms can be easily modulated by simply changing the molar ratio of Pt and Mn precursors (mPt:mMn). As revealed by HAADF-STEM, samples with mPt:mMn of 10:0 (Pt/GDY, Fig. 4a) and 8:2 (Pt8Mn2/GDY, Fig. 4b) exhibit the same lattice spacing of 0.23 nm. As the Pt content decreased to 5:5 (Pt5Mn5/GDY, Fig. 4c) and 2:8 (Pt2Mn8/GDY, Fig. 4d), a bimetallic single-atom layer with larger lattice spacing was obtained. Remarkably, the single-atom layer morphology was well maintained during the variation in mPt:mMn. The extended X-ray absorption fine structure (EXAFS) and X-ray absorption near edge structure (XANES) spectra confirmed the presence of zero-valent Pt nanoplates (Fig. 4e, 4f and Supplementary Fig. 17) and positively charged single Mn atoms (Fig. 4g, h and Supplementary Fig. 18). Wavelet transforms (WT) results offer more powerful resolution in both R and k spaces, and the intensity maximum ascribed to the coordination of Pt can be observed at approximately 10.6 Å−1 in all the samples of Pt8Mn2/GDY (Pt0Mn0/GDY), Pt8Mn2/GDY and Pt8Mn2/GDY (Fig. 4i–l), while the coordination of Mn disappears (Fig. 4m–p). With an increase in the Mn content, the coordination environment of Mn changes significantly. In Pt8Mn2/GDY, Mn-C coordination (Fig. 4n) guarantees a partial positive charge on the Mn metal, whereas in Pt5Mn5/GDY and Pt2Mn8/GDY, the emergence and enhancement of Mn-O coordination (Fig. 4o, p) indicate that Mn metal tends to interact with oxygen.

Fig. 4: Structural modulation of PtMn/GDY.
figure 4

HAADF images of a Pt/GDY, b Pt8Mn2/GDY (Pt0Mn0/GDY), c Pt5Mn5/GDY, and d Pt2Mn8/GDY. e Pt L-edge XANES and f EXAFS spectra and g Mn K-edge XANES and h EXAFS spectra of the samples. Wavelet transform images from Pt L-edge EXAFS spectra of i Pt foil, j Pt8Mn2/GDY (Pt0Mn0/GDY), and k Pt5Mn5/GDY and l Pt2Mn8/GDY, respectively. Wavelet transform images from Mn K-edge EXAFS spectra of m Mn foil, n Pt8Mn2/GDY (Pt0Mn0/GDY), o Pt5Mn5/GDY, and p Pt2Mn8/GDY, respectively.

Full size image

Alkene-to-diol conversion performance

PtMn/GDY was then used for the electro-catalyzed oxidation of ST to PED by a three-electrode system (Fig. 5a) in an aqueous electrolyte containing 125 μmol of ST at room temperature and ambient pressure. The obtained electrolyte was directly analysed using proton nuclear magnetic resonance (1H NMR: Supplementary Figs. 19–21). Figure 5b shows that Pt0Mn0/GDY had the best catalytic performance among all the PtMn/GDY catalysts, commercial 20wt.% Pt/C (selectivity: 86.9%), GDY (selectivity: 43.7%), and carbon cloth (selectivity: 34.0%), with 100% ST conversion (CPED) and selectivity for PED (SPED) (Supplementary Fig. 22 and Supplementary Table 1). To evaluate the more detailed catalytic performance of Pt0Mn0/GDY, the chronoamperometry tests at 10 mA cm−2 were conducted. As shown in Fig. 5c and Supplementary Table 2, most of the ST molecules were transformed into SO (61.46%) and PED (33.32%) in the first 30 min, followed by an increase in the PED content with a further increase in reaction time. The oxidation of ST to PED reached 100% and the total reaction selectivity reached nearly 100% when the reaction time was increased to 3 h (Supplementary Fig. 22). The PED yield (YPED) and Faradaic efficiency (FE) were calculated to be 2.4 mol gPtMn−1 h−1 and 36.3%, respectively (Fig. 5d and Supplementary Table 3), and decreased with increased reaction time (Supplementary Table 3) and increased working current density applied to the catalytic process (Supplementary Table 5 and Supplementary Figs. 23–29). Remarkably, the PED yield was significantly higher than that of previously reported benchmark catalysts (Supplementary Table 4). It is noticeable that the catalytic performance of Pt0Mn0/GDY was well maintained for 17 cycles and there was little decrease in the FE and yield, with 60% retention even after 100 cycles, surpassing all reported catalysts for alkene-to-diol conversion. (Fig. 5e).

Fig. 5: Electrochemical alkene-to-diol conversion performances of Pt0Mn0/GDY.
figure 5

a Schematic illustration of the electro-oxidation process of Pt0Mn0/GDY. b Catalytic selectivity and conversion of the PtxMny/GDY with different raw ratio. c The proportion of each component during the catalytic process. d Time course of product yields and FE by Pt0Mn0/GDY. e Stability test of the catalyst.

Full size image

The Pt 4 f (Fig. 6a) and Mn 2p (Fig. 6b) XPS spectra of Pt0Mn0/GDY recorded at different stages of electrocatalysis show that the valence states of Pt (Fig. 6a) and Mn atoms (Fig. 6b) in Pt0Mn0/GDY increased to higher valence states (Pt0 to Pt2+, Mn0 to Mn4+) during the reaction process, which might be the main reason for the decrease in FE during the reaction process. To reveal further variations on the surface of the catalyst, electrochemical in situ Raman spectra and in situ attenuated total reflectance infrared (ATR-IR) spectra were obtained to gain more insight into the electrooxidation process72. In-situ Raman spectra (Fig. 6c) show that the two characteristic peaks for D and G bands of GDY were well maintained during the catalytic process. The variation of the peak for the acetylenic bond was observed upon connection of the circuit, and the original single broad peak split into two sharp peaks; two new peaks at approximately 550 and 1000 cm−1 correspond to the ST. These indicate an electrically responsive configuration change of the acetylenic bond73,74,75,76,77,78, which benefits the strong adsorption of ST molecules on Pt0Mn0/GDY, facilitates the activation of the ST molecule, and improves the catalytic performance. The in-situ ATR-IR spectra (Fig. 6d) show the presence of the absorption bands at 1044 cm−1 for hydroxy CH-OH and CH2-OH bonds during the electrochemical process (Fig. 6e) and the two peaks at 1060 cm−1 and 1040 cm−1 for styrene at the beginning of the process. Whereas the peaks located at 3010 cm−1 and 2923 cm−1 corresponding to styrene vibration (Fig. 6f) disappeared gradually with the reaction proceeds. These results demonstrated the completely conversion of styrene into PED. After electrolysis, a Pt-Mn bimetallic single atomic layer remained (Supplementary Fig. 30), indicating high stability of the Pt0Mn0/GDY. The as-prepared Pt0Mn0/GDY was stable to air exposure for several days. Based on the above experimental results, we speculate the alkene-to-diol conversion proceeds a four-step process on Pt0Mn0/GDY catalysts (Fig. 6g), including (i) the selectively adsorption of ST molecules on Pt0Mn0 in Pt0Mn0/GDY followed by being activated to *ST; (ii) the hydroxyl species extracted from water can then easily attack the terminal carbon atom in *ST immediately to form hydroxy-styrene *(OH-ST); (iii) the attack of the *(OH-ST) by negatively charged bromide ions, forming the intermediate SO product; and (iv) ring-opening of SO by the attack of H2O molecules results in the formation of PED32,79.

Fig. 6: In-situ structural characterizations of Pt0Mn0/GDY.
figure 6

a Pt 4 f and b Mn 2p XPS spectra of Pt0Mn0/GDY recorded during the reaction processes. c In-situ Raman and FTIR d In-situ measurements on Pt0Mn0/GDY during the reaction processes and the corresponding enlarged spectra of the marked area. The enlargement of e yellow and f red dashed box section in d. g Proposed mechanism of the catalytic process.

Full size image

Discussion

In summary, we realized the controllable growth of single-atom layered Pt-Mn bimetallic nanoplate through a graphdiyne platform, the space-confined alkyne-rich pores of GDY facilitate the ordered arrangement of Pt and Mn atoms. By carefully controlling the content ratio of these two metal atoms, the tunable composition and valent states of the catalysts can be easily obtained. With both Pt and Mn are zero-valent states (Pt0Mn0/GDY), the catalyst has the highest catalytic selectivity and performances for ST-to-PED conversion at ambient temperatures and pressures. The specific incomplete charge transfer between GDY and metal atoms, the synergistic effect between the two atoms, and the formation of sp–C ~ Pt0Mn0 bonds in Pt0Mn0/GDY guarantee the excellent catalytic activity and stability. These results demonstrate the excellent intrinsic performance of GDY as an ideal platform for green sustainable energy catalytic system.

Methods

Synthesis of 3D GDY

The cleaned carbon cloth (3 cm × 4 cm) and two pieces of Cu foils (3 cm × 5 cm × 0.1 mm) were immersed into 30 mL pyridine solution of hexacetylenebenzene (0.3 mg mL−1) in 50 mL Teflon-lined stainless-steel autoclave. Cu foils were used as the catalyst for the coupling reaction. The autoclave was kept at 110 °C under the protection of Ar for 12 h. After the reaction completing, the obtained 3D GDY were thoroughly washed by 3 M HCl at least three times to remove the possible copper residues, following washed by hot DMF and acetone for three time to remove the oligomer. Finally, the 3D GDY electrode was obtained.

Synthesis of PtxMny/GDY

The PtxMny/GDY were fabricated by a facile microwave irradiated deposition method. Totally 10 mg K2PtCl4 and MnCl2 mixture dissolved in 4 mL ethylene glycol and 3D GDY (1 cm × 2 cm) was added into the 2–5 mL quartz tubes and saturated with Ar. The reduction reaction was carried out at 160 °C for 3 min, The resulting PtxMny/GDY were cleaned by ethanol and deionized water for three times, respectively. Finally, dried under the vacuum at an ambient temperature overnight. (10 mg K2PtCl4 and 0 mg MnCl2 for Pt10Mn0/GDY (Pt0/GDY); 8 mg K2PtCl4 and 2 mg MnCl2 for Pt8Mn2/GDY (Pt0Mn0/GDY); 5 mg K2PtCl4 and 5 mg MnCl2 for Pt5Mn5/GDY; 2 mg K2PtCl4; and 8 mg MnCl2 for Pt2Mn8/GDY).

Characterizations

Scanning electron microscope (SEM; Apreo, Thermo Scientific), high-resolution transmission electron microscope (HRTEM; Talos F200X G2 TEM, Thermo Scientific), and High-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM, JEM-ARM200F, JEOL) were used to characterize the morphologies of the samples. Energy-dispersive X-ray spectroscopy (EDX) was collected with an energy-dispersive X-ray detector in Apreo SEM and Talos F200X G2 TEM. inductively coupled plasma optical emission spectrometry (ICP-OES, Agilent ICPOES730) was used to determine the metal content. In-situ attenuated total reflection infrared absorption spectroscopy (ATR-IR) was conducted in the study. The electrochemical workstation was attached to the sample cell, the Pt0Mn0/GDY, Pt wire and Ag/AgCl electrode were used as the working electrode, counter electrode and reference electrode, respectively. A mixed solvent of dimethyl sulfoxide and deioned water (1:1, ~2 mL for wetting the Pt0Mn0/GDY electrode) which contained 0.06 M styrene and 0.06 M sodium bromide was used as electrolyte. In-situ FTIR spectra were recorded during the electrolysis at 10 mA cm−2. (In-situ) Raman spectra was taken through the LabRAM HR Evolution spectrometer with the excitation laser source of 473 nm, all the samples were scanned with a static raster scan type and a 600 lines per millimeter (L/mm) grating at 50 times magnification. For in situ spectral acquistion, the exposure time and accumulative cycles are setted as 6 s and two, the electrochemical workstation was attached to the sample cell, Pt0Mn0/GDY, Pt wire and Ag/AgCl electrode were used as the working electrode, counter electrode and reference electrode, respectively. A mixed solvent of deuterium dimethyl sulfoxide and deuteroxide (1:1, ~0.5 mL) which containing 0.06 M styrene and 0.06 M sodium bromide was used as electrolyte. The electrolysis was conducted at 10 mA cm−2, the in situ spectrum was collected at 30 s intervals during the electrocatalysis. X-ray photoelectron spectroscopy (XPS) measurements were carried through a Thermo Scientific Nexsa instrument with monochromatic Al Kα X-ray radiation, the XPS spectra of the catalyst during the electrocatalysis were collected by characterzing the sample with different electrocatalysis time. Depth-profiling XPS coupled with ion sputtering was conducted, the etch rate is setted as 0.28 nm s−1. Nuclear magnetic resonance spectroscopy (1H NMR, AVANCE NEO) was used to determine the yield and purity of final products.

Electrochemical measurements

The electrochemical measurements were conducted on CHI 660E electrochemical workstation (Chenhua, Shanghai), an undivided electrolytic cell was used. The Pt0Mn0/GDY (Pt-Mn mass loading: 15.5%), carbon rod, and Ag/AgCl electrode were used as the working electrode, counter electrode, and reference electrode, respectively. A mixed solvent of 1 mL deuterium dimethyl sulfoxide and 1 mL deuteroxide, which contains 0.06 M styrene and 0.06 M sodium bromide, was used as an electrolyte. All potentials were recorded against the Ag/AgCl. The iR-compensation was not performed. The galvanostat method was used for electrolysis in corresponding time (0-3 h) with a stirring rate of 200 rpm and conducted at a constant current of 10 mA cm−2 for continuous ST conversion. Pt/GDY (Pt mass loading: 17.1%), Pt6Mn4/GDY (Pt-Mn mass loading: 16.4%), Pt5Mn5/GDY (Pt-Mn mass loading: 16.1%), Pt4Mn6/GDY (Pt-Mn mass loading: 15.8%), Pt2Mn8/GDY (Pt-Mn mass loading: 15.2%) and Pt/C (20% wt) were used as control.

Calculation of the conversion, selectivity, and Faradaic efficiency

For the oxidation product of 1-Phenyl-1,2-ethanediol, the conversion and selectivity were calculated from the 1H NMR results based on Eq. 1:

$${Conv}.=\frac{{S}_{{SPEDt}}}{{S}_{{St}}+{S}_{{SO}}+{S}_{{PED}}};{Select}.=\frac{{S}_{{SO}}}{{S}_{{SO}}+{S}_{{PED}}}$$
(1)

Where SSt, SSO, and SPED are the characteristic peak areas of the styrene, styrene oxide, and 1-phenyl-1, 2-ethanediol which integrated from the 1H NMR results.

The Faradaic efficiency of 1-Phenyl-1,2-ethanediol was calculated according to Eq. 2:

$${{FE}}_{{PED}}=\frac{1\,\times \,F\,\times \,{n}_{{PED}}}{Q\,}$$
(2)

Where nPED is the molar number of 1-Phenyl-1,2-ethanediol, F is the Faradaic constant (96485 C mol-1), Q is the total charge passing the electrode.

XAFS measurements and analysis

The X-ray absorption fine structure spectra were collected at 1W1B station in Beijing Synchrotron Radiation Facility (BSRF). The storage rings of BSRF was operated at 2.5 GeV with an average current of 250 mA. Using Si(111) double-crystal monochromator, the data collection were carried out in transmission/fluorescence mode using ionization chamber. All spectra were collected in ambient conditions. The acquired EXAFS data were processed by using the ATHENA module implemented in the IFEFFIT software packages according to the standard procedures. To obtain the quantitative structural parameters around central atoms, least-squares curve parameter fitting was performed using the ARTEMIS module of IFEFFIT software packages.