English

• With the increasing expand of world population and rapid consumption of global fossil fuels, energy crisis and environmental issues are serious [1]. Hydrogen has aroused researchers' much attention because of its high energy density and no-pollution. Electrochemical water splitting is an efficient technology to produce hydrogen [2]. However, the sluggishly anodic oxygen evolution reaction (OER) has seriously hinders its energy efficiency and brings about extra energy loss. Moreover, the production, O₂, is less value and dangerous when mixed with hydrogen [3]. Recently, replacing OER with organics oxidation reaction, which usually processes low overpotential and high value-added products, has catch widespread research interest [4]. 5-Hydroxymethyl-furfural (HMF), which is an important platform chemical and with rich resources, can be converted into various valuable chemicals [5]. Electro-oxidation of HMF can obtain 2,5-furandicarboxylic acid (FDCA), which is the promising candidate to terephthalic acid that commonly used in the manufacture of polymers such as polyester and also an intermediate in the synthesis of other important polymers, fine chemicals, phamaceuticals and pesticides [6-9]. Thus, it is of great importance to develop efficient electrocatalysts for HMF oxidation reaction (HMFOR).

  Co₃O₄, which is a typical spinel oxide, has been widely explored as an attractive electrocatalyst for HMFOR owing to its high stability and tunable valance state [10-12]. However, poor electrical conductivity and limited intrinsic ability make its electrochemical ability unsatisfactory. The electrical conductivity can be improved by coupled with conductive substrate such as carbon nanomaterials and Ni foam [13-15]. The strong electron interaction between Co₃O₄ and conductive substrate could facilitate electron transfer leading to enhanced electrochemical capacity [14]. Meanwhile, the intrinsic ability of Co₃O₄ could be enhanced by modulating its electronic structure, which has a significant effect on the affinity of active sites to reaction species [16]. Heteroatom-doping is an effective strategy to regulate the electronic structure of electrocatalyst [17]. The introduction of foreign metal or non-metal element could not only provide extra active sites but also make charge redistribution resulting in optimized binding energy to reaction intermediates [18-20]. For example, Zhang et al. have synthesized N-doped Co₃O₄, which exhibited superior electrocatalytic ability for HMFOR. The introduction of N benefited the formation of defects and active sites [21]. Wang et al. have decorated Co₃O₄ with single-atom Ir. The introduction of single-atom Ir has enhanced the adsorption with C=C group of HMF resulting in dramatically improved electrocatalytic HMFOR ability [22]. Various 3d transition metal ions such as Fe³⁺, Ni²⁺ and Cu²⁺ have been doped into Co₃O₄ to tailor its electrochemical ability [23-25]. Moreover, doping with high-valence non-3d transition metal ions such as Zr⁴⁺, W⁶⁺, Mn⁴⁺ and Mo⁶⁺ also is an effective method to enhance the electrochemical ability of Co₃O₄ [20,26-28]. Mo⁶⁺, which is a typical high-valence 4d transition metal ion, processed similar ions radius with Co³⁺ and can be used as heteroatoms to replace Co³⁺ in Co₃O₄ [29]. It has been reported that Mo-doping could modulate the electronic property of Co₃O₄ and accelerate charge transfer accounting for superior OER ability [30]. However, the HMFOR electrocatalyst based on Mo-doped Co₃O₄ has not been reported yet. It's of great research value to explore the effects of Mo⁶⁺-doping on the electrochemical HMFOR ability of Co₃O₄.

  Herein, in this work we have successfully synthesized Mo-doped Co₃O₄ (Mo-Co₃O₄) nanowires to investigate the influence of Mo⁶⁺-heteroatoms on electronic structure and electrochemical ability of HMFOR. With the introduction of Mo⁶⁺, the Co³⁺ content was decreased and metal-oxygen bond has been strengthened. Electrochemical results demonstrated that Mo-Co₃O₄ exhibited more excellent electrochemical ability with higher current density and lower overpotential than Co₃O₄ for HMFOR. Density functional theory (DFT) revealed that with the introduction of Mo⁶⁺, the electron was assembled at surrounding cobalt sites accounting for enhanced adsorption ability with HMF, which results in rich HMF concentration at local active sites (Scheme 1). Post characterization revealed that Mo-Co₃O₄ still kept pristine crystalline structure after four successive electrolysis cycles and displayed good durability. CoOOH species were formed after electrolysis and worked as active sites. This work provides a valuable reference for developing efficient heteroatom-doped electrocatalysts.

  Scheme 1

  

  Scheme 1.  An illustration about the core relationship between Mo⁶⁺ doped Co₃O₄, modulated electronic structure, enhanced adsorption ability and its superior electrocatalytic ability towards HMFOR. The blue, pink, red, brown and white balls are represented as Co, Mo, O, C and H, respectively.

  DownLoad: Full-Size Img PowerPoint

  Synthesis of Co₃O₄ doped with different Mo content: Ni foam was washed with 2 mol/L HCl, ultra-pure water and ethanol, subsequently, before using. 232.5 mg Co(NO₃)₂·6H₂O, calculated amount of Na₂MoO₄·2H₂O (molar ratio of Co/Mo of 10:0.1, 10:0.5 and 10:1) and 242.4 mg urea were dissolved in 30 mL ultra-pure water with continuously stirring for 30 min. Then, a piece of washed Ni foam (3 × 4 cm) and the mixture were placed into a 50 mL autoclave and suffered from 393 K for 10 h. After cooling to room temperature, the sample was taken out and washed with ultra-pure water and ethanol followed by drying at 333 K for 2 h. Finally, Mo-doped Co₃O₄ could be obtained by annealing at 623 K for 2 h at a heating rate of 278 K/min. The samples with feeding Co/Mo ratio of 10:0.1, 10:0.5 and 10:1 were named as Mo-Co₃O₄ (10:0.1), Mo-Co₃O₄, and Mo-Co₃O₄ (10:1). Mo-Co₃O₄ was explored in detail.

  All the electrochemical measurements including cyclic voltammetry (CV), linear sweep voltammetry (LSV) and constant potential were tested in CHI 760E CH Instrument electrochemical workstation. The samples that grown on Ni foam were cut into 1 × 1 cm and used as working electrode. The counter electrode and reference electrode were carbon rode and saturated calomel electrode (SCE), respectively. LSV curves were recorded at 5 mV/s. All potentials were referenced to a reversible hydrogen electrode (RHE) according to: E_RHE = E_SCE + 0.242 + 0.059 × pH. The nyquist plots were measured at the frequency range from 0.01 Hz to 100 kHz.

  The morphology of samples were investigated by scanning electron microscopy (SEM, Zeiss Signma 500). Transmission electron microscopy (TEM) images were obtained in JEM-2100. The bulk crystal structure of the obtained samples is characterized using powder X-ray diffraction (XRD, D/MAX-3D diffractometer). X-ray photoelectron spectroscopy (XPS) were tested using ESCALAB 250 system (Thermo Fisher Scientific, England) and were used to explore the electronic structure. BET analysis was performed on Belsorp Max. The Raman spectra were collected on labRAM HR Evolution-HORIBA Raman system. The temperature-programmed desorption in ammonia (NH₃-TPD) was tested in a Micromeritics Autochem Ⅱ 2920 chemisorption analyzer. The samples were heated up to 423 K at 10 K/min and kept for 30 min in He flow to remove adsorbed impurities. Afterwards, the samples were heated up to 313 K for the adsorption of NH₃. After flushing with He for 1 h, the physically adsorbed NH₃ could be removed. TPD data was recorded from 313 K to 873 K with 10 K/min. The quantify of NH₃ adsorption was calculated by integration of peak area and the adsorption of exhaust NH₃.

  High-performance liquid chromatography (HPLC, Shimadzu Prominence LC-2030C system) with an ultraviolet-visible detector was applied to detect and analyze HMF oxidation products. 20 µL of electrolyte was sampling during potentiostatic electrolysis and diluted to 2 mL with ultrapure water and analyzing it by HPLC. The wavelength of the UV detector was set to 265 nm, mobile phase A and phase B was methanol and 5 mmol/L ammonium formate aqueous solution, respectively. The ratio of A: B was 3:7 and flow rate was 0.6 mL/min. A 4.6 mm × 250 mm Ultimate 5 µm AQ-C18 column was used and each separation lasts 10 min.

  The HMF conversion, FDCA yield and faradaic efficiency were calculated using the following equations, respectively.

  (1)

  (2)

  (3)

  where F is the Faraday constant (96,485 C/mol) and n is the mol of reactant calculated from the concentration measured by HPLC.

  All DFT calculations were performed using Vienna ab initio simulation package (VASP). The interaction between ion and electron was described with the projector-augmented-wave (PAW) method. A 400 eV cut off energy for the plane wave expansion was adopted in all the calculations. Hubbard U correction (DFT+U) was employed using the generalized gradient approximation (GGA) of Perdew−Burke−Ernzerhof (PBE) methods for accounting for the correlation energy of the strongly localized 3d orbital, and setting U-J = 3.5 eV and 2.0 eV for Co and Mo atoms in this work. A vacuum layer of 20 Å was added into two adjacent two consecutive slabs for eliminating their interaction. The total energy and force convergence criteria were set as 1 × 10⁻⁴ eV and 0.02 eV/Å, respectively. In addition, the dipole correction was applied in all slab calculations.

  SEM was performed to explore the morphology of Mo-doped Co₃O₄. As shown in Fig. S1a (Supporting information), pure Co₃O₄ was consistent with nanowires. With the introduction of Mo, the nanowires microscopy was still kept and a few nanosheets were appeared when the feeding molar ratio of Co/Mo was 10:1 (Fig. 1a and Fig. S1 in Supporting information). XRD was used to detect the crystalline structure of samples. As shown in Fig. S2 (Supporting information), the doping of Mo-heteroatoms did not change the spinel Co₃O₄ structure. The diameter distribution of Mo-Co₃O₄ nanowires was shown in Fig. S3 (Supporting information). High-resolution TEM (HRTEM) was applied to explore the crystalline structure of Mo-Co₃O₄ at high spatial resolution. In Fig. 1b, the measured lattice fringe of 2.86 Å was attributed to (220) crystal face of Co₃O₄. The inset image was its fast Fourier transform (FFT) image, which can be well indexed to (040) and (220) lattice plane of Co₃O₄. The selected area electron diffraction (SAED) image of Mo-Co₃O₄ was shown in Fig. S4 (Supporting information) and indicated to spinel Co₃O₄ phase. Element mapping images of Mo-Co₃O₄ (Fig. 1c) revealed the uniform distribution of Co, Mo and O, respectively. The atom ratio of Co/Mo in Mo-Co₃O₄ was detected by Energy dispersive spectra (EDS) to be about 37:1 (Table S1 in Supporting information). Inductively coupled plasma-optical emission spectroscopy (ICP-OES) was also used to detect the Co/Mo ratio in Mo-Co₃O₄. As shown in Table S2 (Supporting information), the Co/Mo was to be 31:1, which was close to EDS data. Raman spectra was carried out to investigate the bond vibration mode in Co₃O₄ and Mo-Co₃O₄. It can be clearly seen from Fig. 1d that Co₃O₄ displayed a typical spinel vibration mode. The Raman band at 188.0 and 673.9 cm⁻¹ were corresponded to F_2g¹ symmetry of tetrahedral sites and A_1g mode of octahedral sites, respectively. The band at 470.5 and 512.6 cm⁻¹ were assigned to F_2g² and E_g mode owing to the vibration of tetrahedral and octahedral sites, respectively. The weak band at 608.4 cm⁻¹ was associated with F_2g mode. The Raman band of A_1g mode in Mo-Co₃O₄ showed a shift to higher wavenumber compared with pristine Co₃O₄, indicating the screwy octahedron sites owing to the doping of Mo [31]. XPS was used to explore the electronic structure of Co₃O₄ and Mo-Co₃O₄. It can be seen from Fig. S5 (Supporting information) that Mo has been successfully doped into Co₃O₄. For Co 2p XPS spectra (Fig. 1e), it can be divided into eight contributions. The peaks at 779.8, 781.7, 784. 5 and 788.8 eV formed Co 2p_3/2. While Co 2p_1/2 were constituted with four peaks at 794.8, 796.7, 799.5 and 803.8 eV, respectively. The peak at 779.8 and 781.7 eV were assigned to Co³⁺ and Co²⁺, respectively. The ratio of Co³⁺/Co²⁺ in Co₃O₄ and Mo-Co₃O₄ was calculated to be about 1.59 and 1.40, respectively. The decreased Co³⁺ content may be owing to the substitution of Co³⁺ at octahedron sites with Mo⁶⁺ [30]. The ratio of Co³⁺/Co²⁺ was 1.48 and 1.31 in Mo-Co₃O₄ (10:0.1) and Mo-Co₃O₄ (10:1), respectively (Fig. S6a in Supporting information). The Mo 3d XPS spectra can be deconvoluted into two peaks at 231.9 and 235.1 eV, respectively, which corresponded to Mo⁶⁺ species (Fig. 1f and Fig. S6b in Supporting information) [30]. For O 1s XPS spectra, it can be divided into three peaks. The peaks at 529.6 and 531.1 eV were attributed lattice oxygen and metal oxyhydroxides, respectively. While, the peak at 533.2 eV was corresponding to adsorbed water species. The O 1s XPS spectra in Mo-Co₃O₄ exhibited a shift towards higher binding energy compared with Co₃O₄, indicating a stronger metal-oxygen interaction (Fig. S7 in Supporting information) [32]. Based on above, the electronic structure of Co₃O₄ has been well modulated with the introduction of Mo⁶⁺. Wang et al. have reported that the Co²⁺ in Co₃O₄ was responsible for the adsorption of HMF [33,34]. Thus, it can be expected that Mo-Co₃O₄ would exhibited an enhanced adsorption ability and preferable electrochemical ability for HMF oxidation.

  Figure 1

  

  Figure 1.  (a) SEM image, (b) HRTEM image and (c) corresponding element mapping image of Mo-Co₃O₄. (d) Raman spectra of Co₃O₄ and Mo-Co₃O₄. (e) Co 2p XPS spectra of Co₃O₄ and Mo-Co₃O₄. (f) Mo 3d XPS spectra of Mo-Co₃O₄.

  DownLoad: Full-Size Img PowerPoint

  The electrochemical abilities of samples were evaluated in a three-electrode system using 1.0 mol/L KOH as electrolyte. The LSV curves with 50 mmol/L HMF and 0 mmol/L HMF were recorded to compare the electrochemical ability of HMFOR with OER. It is clearly seen from Fig. 2a that with the addition of HMF, the current density of both Co₃O₄ and Mo-Co₃O₄ were obviously increased suggesting the preferential oxidation of HMF than OH⁻. What is more, the current density of Mo-Co₃O₄ was significantly higher than that of Co₃O₄ at the same overpotential. The electrochemical performance of Co₃O₄ doped with different Mo content was tested to explore its influence on catalytic ability. As shown in Fig. S8 (Supporting information), the electrochemical ability of Co₃O₄ for HMF oxidation has been rapidly improved with the doping of Mo. What's more, Mo-Co₃O₄ displayed the most excellent electrochemical ability. As shown in Fig. 2b that Mo-Co₃O₄ only need 1.39 V to reach 30 mA/cm², which was 146 mV smaller than that of Co₃O₄. The current density of Mo-Co₃O₄ was about two times higher than that of Co₃O₄ at 1.50 V indicating the greatly improved electrochemical ability of Co₃O₄ after the introduction of Mo-heteroatoms. The electrocatalytic performance of Mo-Co₃O₄ compared with other HMF electro-oxidation catalysts has been provided in Table S3, which displayed excellent catalytic ability. In order to exclude the influence of the number of exposed active sites on electrochemical ability, the electrochemical active surface areas (ECSA) was calculated according to the electrical double-layer capacitor (C_dl). The CV plots of Co₃O₄ and Mo-Co₃O₄ at different scan rates were shown in Fig. S9 (Supporting information), whose C_dl were 23.92 and 26.27 mF/cm², respectively. After normalized by the ECSA, Mo-Co₃O₄ also exhibited better electrochemical ability than that of Co₃O₄, suggesting the enhanced intrinsic ability (Fig. S10 in Supporting information). HPLC was used to explore the reaction pathway and detect products of HMFOR on Mo-Co₃O₄. In usual, HMF oxidation takes two pathways. In path 1, the aldehyde group of HMF was prior to be oxidized into carboxyl group forming 5-hydromethyl-2-furancarboxylic acid (HMFCA). For path 2, the hydromethyl group was firstly oxidized into aldehyde groups forming 2,5-diformylfuran (DFF). Both HMFCA and DFF can be oxidized into 2-furancarboxylic acid (FFCA). Finally, FDCA could be obtained by the oxidation of FFCA (Fig. 2c) [35]. The potentiostatic electrolysis at 1.40 V in 5 mmol/L HMF was carried out to detect intermediate products of HMF oxidation using Mo-Co₃O₄ electrode. The concentration of HMF, DFF, HMFCA, FFCA and FDCA standard substances was quantified by HPLC (Figs. S11-S15 in Supporting information). As shown in Fig. 2d and Fig. S16 (Supporting information), DFF was hardly to be detected and the concentration of HMFCA was higher, indicating that the dominated pathway of HMF oxidation was path 1. Mo-Co₃O₄ exhibited a high FDCA yield (yield_FDCA) (95%) and faradaic efficiency (92%). To evaluate the electrochemical stability of Mo-Co₃O₄, four successive electrolysis cycles was performed. In Fig. 2e, the yield_FDCA and Faradaic efficiency were almost unchanged, suggesting a good durability of Mo-Co₃O₄. The prior oxidation of aldehyde group on Mo-Co₃O₄ was further evidenced by DFT calculation. The reaction models and reaction free energy were shown in Figs. 2f and g and Fig. S17 (Supporting information). In path 1, both OH⁻ and HMF are firstly adsorbed on the Mo-Co₃O₄ surface. Then, OH⁻ was attend to capture H⁺ from the aldehyde of HMF to form H₂O. Another OH⁻ was combined with dehydrated aldehyde to form carboxylic acid. For path 2, the adsorbed OH⁻ was combined with H⁺ of hydroxy to form H₂O following by desorption. As shown in Fig. 2f, the free energy barrier of dehydration of path 2 (1.69 eV) was higher than that of dehydration of path 1 (1.10 eV), suggesting aldehyde was preferential than hydroxymethyl to be oxidized on Mo-Co₃O₄, which was consistent with HPLC results [36].

  Figure 2

  

  Figure 2.  (a) LSV plots of Co₃O₄ and Mo-Co₃O₄ in 1.0 mol/L KOH with or without 50 mmol/L HMF without iR compensation. (b) The potential at 30 mA/cm² and current density at 1.50 V on both Co₃O₄ and Mo-Co₃O₄. (c) The reaction pathway of HMF oxidation; (d) The concentration of DFF, HMFCA, FFCA, FDCA and HMF during HMF oxidation. (e) The yield_FDCA (%) and Faradic efficiency (%) of Mo-Co₃O₄ under four successive electrolysis cycles. (f) Reaction free energy barrier for the oxidation of aldehyde and hydroxyl groups on Mo-Co₃O₄. (g) Structure models for the dehydration of hydroxyl and aldehyde oxidation on Mo-Co₃O₄.

  DownLoad: Full-Size Img PowerPoint

  In order to investigate the adsorption ability of Co₃O₄ and Mo-Co₃O₄, TPD was performed. Since C=O group was rich in electron, the adsorption ability of Lewis acidic sites at NH₃ atmosphere was shown in Fig. S18 (Supporting information) [37]. It can be seen clearly that Mo-Co₃O₄ exhibited higher desorption temperature than Co₃O₄ suggesting the adsorption ability has been strengthen with Mo⁶⁺-doping. Open-circuit potential (OCP) was recorded to evaluate the absorb ability in Helmholtz layer [38]. As displayed in Fig. S19 (Supporting information), the OCP of Co₃O₄ was 0.06 V after injecting 50 mmol/L HMF. Mo-Co₃O₄ exhibited a more significant drop of OCP (0.11 V) indicating a strong surface adsorption of HMF. The electrochemical abilities of Co₃O₄ and Mo-Co₃O₄ in furfuraldehyde and furfuryl alcohol were tested to explore the adsorption ability towards different groups of HMF (Fig. S20 in Supporting information). As summarized in Fig. S21 (Supporting information), Mo-Co₃O₄ displayed the promoted oxidation of hydroxymethyl and aldehyde groups simultaneously [39]. What is more, the current density of hydroxymethyl oxidation was increased more obviously than that of aldehyde groups on Mo-Co₃O₄ electrode demonstrating the oxidation of hydroxymethyl seriously limited the oxidation process of HMFOR. With the introduction of Mo-heteroatoms, hydroxymethyl oxidation was accelerated [40]. In situ Bode-phase plots was performed to explore the reaction kinetics at electrode/electrolyte interface. It has been widely accepted that the signal at high frequency (10²–10³ Hz) was attributed to the electron transfer between electrocatalyst inner and electrode interface while the peak at low frequency (10⁻²–10² Hz) was corresponding to charge transfer during electrocatalytic process [41]. In Fig. 3a, Co₃O₄ displayed a peak at low frequency revealing the electrocatalytic process was dominated by charge transfer. With the increase of potential, the peak became sharper and lower indicating that charge transfer was accelerated. Moreover, the tendency of peak on Co₃O₄ electrode was gently, which may be limited by poor electronic conductivity. For Mo-Co₃O₄ (Fig. 3b), it showed similar signal with Co₃O₄ demonstrating the electrocatalytic process of both Mo-Co₃O₄ and Co₃O₄ was dominated by charge transfer. The tendency of peak was sharp suggesting the electronic conductivity of Co₃O₄ was improved with the doping of Mo-heteroatoms. For comparison, the Bode-angle plots at 1.50 V of Mo-Co₃O₄ and Co₃O₄ were shown in Fig. 3c. It is obvious that the peak of Mo-Co₃O₄ was lower and sharper than that of Co₃O₄ demonstrating that more electrons were participated in electrochemical reaction on Mo-Co₃O₄ electrode [42]. The inset was their corresponding nyquisty plots. The smaller of semicircle diameter, the smaller of charge transfer resistance. It can be clearly found that Mo-Co₃O₄ displayed faster reaction kinetics than Co₃O₄. DFT calculation was carried out to explore the origin of superior electrocatalytic ability towards HMFOR. It has been widely accepted that the electronic structure of electrocatalyst plays a significantly role on the adsorption/desorption with reaction intermediates. The differential charge density of Mo-Co₃O₄ was calculated and shown in Fig. 3d. Due to the introduction of high valance of Mo⁶⁺, the electron was transferred from Mo to Co, resulting in assembled charge at surrounding cobalt atom. The Bader charge analysis results showed that the number of electron transferred from Mo to Co was 1.81, which was accordance with the XPS results in Fig. 1e. The modulated electronic structure may have a great effect on adsorption with HMF on Mo-Co₃O₄ electrode [43]. The optimal adsorption structure of HMF molecule on Mo-Co₃O₄ and Co₃O₄ were shown in Fig. 3d. It is clearly can be seen that the adsorption site in both Mo-Co₃O₄ and Co₃O₄ were same. However, the adsorption energy of HMF (∆E_HMF) was −2.10 eV, which was smaller than that of Co₃O₄ (−1.99 eV), which revealed that the adsorption of HMF molecule was more favorable on Mo-Co₃O₄. The enhanced adsorption ability with HMF may result from the modulated electron structure with the introduction of Mo⁶⁺. From the electrochemical LSV carves, it can be seen that the current density of HMFOR on Mo-Co₃O₄ electrode was began to increase at 1.15 V, which was smaller than the potential of catalyst oxidation, indicating the HMFOR was a direct oxidation process [36]. It has been reported previously that the adsorption of OH⁻ plays a crucial role on the reaction kinetics of HMFOR [44]. The free energy barrier of OH⁻ on Co₃O₄ and Mo-Co₃O₄ was calculated and shown in Fig. 3e. It has been found that Mo-Co₃O₄ can adsorb OH⁻ much more easily than Co₃O₄. Above all, the reaction kinetic of Co₃O₄ for HMFOR has been accelerated with the introduction of Mo-heteroatoms. The superior electrocatalytic ability was mainly attributed to enhanced adsorption ability towards HMF molecule owing to modulated electronic structure of surrounding cobalt atoms.

  Figure 3

  

  Figure 3.  In situ Bode-phase plots of (a) Co₃O₄ and (b) Mo-Co₃O₄ at different potentials in HMF. (c) The Bode-phase plots and Nyquist plots Co₃O₄ and Mo-Co₃O₄ at 1.50 V. (d) Side-view of differential charge density of Mo-Co₃O₄ (top), where the yellow and cyan represent electron accumulation and depletion with an isosurface value of 0.01 e/ų, respectively. The calculated adsorption energies and optimized structure of HMF on Mo-Co₃O₄ (left) and Co₃O₄ (right). (e) The energy barrier of OH* adsorption on Co₃O₄ and Mo-Co₃O₄.

  DownLoad: Full-Size Img PowerPoint

  Post characterizations of Mo-Co₃O₄ after four successive electrolysis cycles were carried out to explore the evolution of crystalline structure and electronic environment. It can be seen from Fig. S22 (Supporting information) that spinel Co₃O₄ structure was kept after electrolysis cycles. As shown in Fig. S23 (Supporting information), it can be clearly seen that after electrolysis cycles Mo-Co₃O₄ still kept nanowires morphology. As shown in Fig. 4a, the Raman spectra of Mo-Co₃O₄ after four successive electrolysis cycles also displayed a Co₃O₄ vibration mode. HRTEM was performed to index the crystalline structure. The measured lattice fringes of 2.85 and 4.67 Å, corresponding to (220) and (111) crystal face of Co₃O₄, respectively, which was consistent with its FFT image (Fig. 4b inset) along [110] zone axis. XPS was applied to explore the surface valance state of Co and Mo after electrolysis. As shown in Fig. 4c, the ratio of Co³⁺/Co²⁺ was increased to 1.50 indicating the possible formation of CoOOH active species [21]. For Mo 3d (Fig. 4d), it still was in Mo⁶⁺ state. The O 1s XPS spectra of Mo-Co₃O₄ after electrolysis cycles was shown in Fig. S24 (Supporting information), it is obvious that the peak intensity at around 531.0 eV, which was attributed to metal oxyhydroxides [34], was relatively higher than that in Mo-Co₃O₄. Above all, Mo-Co₃O₄ still kept its nanowire morphology and spinel Co₃O₄ structure after electrolysis cycles. CoOOH species may formed at surface and worked as active sites for HMF electro-oxidation.

  Figure 4

  

  Figure 4.  (a) Raman spectra, (b) HRTEM, (c) Co 2p XPS spectra and (d) Mo 3d XPS spectra of Mo-Co₃O₄ after four successive electrolysis cycles.

  DownLoad: Full-Size Img PowerPoint

  In summary, to further enhance the electrocatalytic HMFOR activity of Co₃O₄, we have doped high-valance Mo⁶⁺ on its surface to regulate electronic structure and strength absorption ability. With the doping of Mo⁶⁺, the content of Co³⁺ has been decreased and interaction of metal-oxygen bond has been strength. Electrochemical results demonstrated that Mo-Co₃O₄ exhibited more outstanding HMFOR ability and faster reaction kinetics than Co₃O₄. DFT calculations revealed that the electron was assembled at surrounding cobalt site leading to an enhanced adsorb with HMF after Mo⁶⁺-doping. Post-characterization demonstrated that Mo-Co₃O₄ after four successive electrolysis cycles was kept its original crystalline structure revealing good structure stability. XPS results proved the formation of CoOOH species, which may work as active sites. This work provides a valuable reference for designing efficient heteroatom-doped electrocatalysts for electrochemical HMFOR.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  This work was supported by National Natural Science Foundation of China (Nos. 92061201, 21825106, 22102155 and 32072304), the China Postdoctoral Science Foundation (Nos. 2021M692909 and 2022T150587), the Program for Innovative Research Team (in Science and Technology) in Universities of Henan Province and Zhengzhou University (No. 19IRSTHN022) and the Key Scientific and Technological Project of Henan Province (No. 2021102210027).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2022.107810.

  
  
• 1. [1]

     S. Pacala, R. Socolow, Science 305 (2004) 968–972. doi: 10.1126/science.1100103

  2. [2]

     J. Wang, W. Cui, Q. Liu, et al., Adv. Mater. 28 (2016) 215–230. doi: 10.1002/adma.201502696

  3. [3]

     N. Jiang, B. You, R. Boonstra, et al., ACS Energy Lett. 1 (2016) 386–390. doi: 10.1021/acsenergylett.6b00214

  4. [4]

     H. Zhou, Z. Li, L. Ma, H. Duan, Chem. Commun. 58 (2022) 897–907. doi: 10.1039/D1CC06254A

  5. [5]

     C. Xu, E. Paone, D. Rodríguez-Padrón, R. Luque, F. Mauriello, Chem. Soc. Rev. 49 (2020) 4273–4306. doi: 10.1039/D0CS00041H

  6. [6]

     Y. Zhao, M. Cai, J. Xian, Y. Sun, G. Li, J. Mater. Chem. 9 (2021) 20164–20183. doi: 10.1039/D1TA04981J

  7. [7]

     G. Yang, Y. Jiao, H. Yan, et al., Adv. Mater. 32 (2020) 2000455. doi: 10.1002/adma.202000455

  8. [8]

     Y. Lu, T. Liu, C.L. Dong, et al., Adv. Mater. 34 (2022) 2107185. doi: 10.1002/adma.202107185

  9. [9]

     M. Lu, M. Zhang, C.G. Liu, et al., Angew. Chem. Int. Ed. 60 (2021) 4864–4871. doi: 10.1002/anie.202011722

  10. [10]

      M.J. Kang, H. Park, J. Jegal, et al., Appl. Catal. B: Environ. 242 (2019) 85–91. doi: 10.1016/j.apcatb.2018.09.087

  11. [11]

      M. Li, Y. Deng, G. Wu, et al., Aggregate 2 (2021) e17.

  12. [12]

      H.L. Chen, Q.T. Huang, S.R. Hu, et al., Chin. J. Struct. Chem. 39 (2020) 1035–1043.

  13. [13]

      X.F. Lu, L.F. Gu, J.W. Wang, et al., Adv. Mater. 29 (2017) 1604437. doi: 10.1002/adma.201604437

  14. [14]

      X. Yang, H. Li, A.Y. Lu, et al., Nano Energy 25 (2016) 42–50. doi: 10.1016/j.nanoen.2016.04.035

  15. [15]

      L. Xu, Q. Jiang, Z. Xiao, et al., Angew. Chem. Int. Ed. 128 (2016) 5363–5367. doi: 10.1002/ange.201600687

  16. [16]

      C. Xie, D. Yan, W. Chen, et al., Mater. Today 31 (2019) 47–68. doi: 10.1016/j.mattod.2019.05.021

  17. [17]

      G. Yang, Y. Jiao, H. Yan, C. Tian, H. Fu, Small Struct. 2 (2021) 2100095. doi: 10.1002/sstr.202100095

  18. [18]

      Y. Wang, M. Qiao, Y. Li, S. Wang, Small 14 (2018) 1800136. doi: 10.1002/smll.201800136

  19. [19]

      R. Li, Y. Guo, H. Chen, et al., ACS Sustain. Chem. Eng. 7 (2019) 11901–11910. doi: 10.1021/acssuschemeng.9b02558

  20. [20]

      S.L. Zhang, B.Y. Guan, X.F. Lu, et al., Adv. Mater. 32 (2020) 2002235. doi: 10.1002/adma.202002235

  21. [21]

      M. Sun, Y. Wang, C. Sun, et al., Chin. Chem. Lett. 33 (2022) 385–389. doi: 10.1016/j.cclet.2021.05.009

  22. [22]

      Y. Lu, T. Liu, C.L. Dong, et al., Adv. Mater. 33 (2021) 2007056. doi: 10.1002/adma.202007056

  23. [23]

      W. Song, Z. Ren, S.Y. Chen, et al., ACS Appl. Mater. Interfaces 8 (2016) 20802–20813. doi: 10.1021/acsami.6b06103

  24. [24]

      Y. Tian, L. Cao, P. Qin, ChemCatChem 11 (2019) 4420–4426. doi: 10.1002/cctc.201900834

  25. [25]

      T. Grewe, X. Deng, H. Tüysüz, Chem. Mater. 26 (2014) 3162–3168. doi: 10.1021/cm5005888

  26. [26]

      L. Huang, D. Chen, G. Luo, et al., Adv. Mater. 31 (2019) 1901439. doi: 10.1002/adma.201901439

  27. [27]

      Q. Hu, Y. Liu, L. Ma, X. Zhang, H. Huang, J. Appl. Electrochem. 48 (2018) 1189–1195. doi: 10.1007/s10800-018-1211-5

  28. [28]

      Y. Xiong, Y. Yang, X. Feng, F.J. DiSalvo, H.D. Abruña, J. Am. Chem. Soc. 141 (2019) 4412–4421. doi: 10.1021/jacs.8b13296

  29. [29]

      C. Guan, W. Xiao, H. Wu, et al., Nano Energy 48 (2018) 73–80. doi: 10.1016/j.nanoen.2018.03.034

  30. [30]

      K. Lu, T. Gu, L. Zhang, et al., Chem. Eng. J. 408 (2021) 127352. doi: 10.1016/j.cej.2020.127352

  31. [31]

      J. Zhao, W. Han, J. Zhang, Z. Tang, Arab. J. Chem. 13 (2020) 4857–4867. doi: 10.1016/j.arabjc.2020.01.014

  32. [32]

      Y. Yang, Y. Ou, Y. Yang, et al., Nanoscale 11 (2019) 23296–23303. doi: 10.1039/C9NR08795H

  33. [33]

      Y. Lu, C.L. Dong, Y.C. Huang, et al., Angew. Chem. Int. Ed. 59 (2020) 19215–19221. doi: 10.1002/anie.202007767

  34. [34]

      Z. Liu, G. Wang, X. Zhu, et al., Angew. Chem. Int. Ed. 59 (2020) 4736–4742. doi: 10.1002/anie.201914245

  35. [35]

      H.G. Cha, K.S. Choi, Nat. Chem. 7 (2015) 328–333. doi: 10.1038/nchem.2194

  36. [36]

      Y. Lu, T. Liu, Y.C. Huang, et al., ACS Catal. 12 (2022) 4242–4251. doi: 10.1021/acscatal.2c00174

  37. [37]

      B.S. Solanki, C.V. Rode, Green Chem. 21 (2019) 6390–6406. doi: 10.1039/C9GC03091C

  38. [38]

      N. Heidary, N. Kornienko, Chem. 56 (2020) 8726–8734.

  39. [39]

      Y. Miao, M. Shao, Chin. J. Catal. 43 (2022) 595–610. doi: 10.1016/S1872-2067(21)63923-2

  40. [40]

      B. Zhou, Y. Li, Y. Zou, et al., Angew. Chem. Int. Ed. 60 (2021) 22908–22914. doi: 10.1002/anie.202109211

  41. [41]

      X. Chao, C.A. Wei, A. Sd, et al., Nano Energy 71 (2020) 104653. doi: 10.1016/j.nanoen.2020.104653

  42. [42]

      D. Zhou, S. Wang, Y. Jia, et al., Angew. Chem. Int. Ed. 58 (2019) 736–740. doi: 10.1002/anie.201809689

  43. [43]

      H. Wang, J. Chen, Y. Lin, et al., Adv. Mater. 33 (2021) 2008422. doi: 10.1002/adma.202008422

  44. [44]

      W. Chen, C. Xie, Y. Wang, et al., Chem. 6 (2020) 2974–2993. doi: 10.1016/j.chempr.2020.07.022

• 

  Scheme 1  An illustration about the core relationship between Mo⁶⁺ doped Co₃O₄, modulated electronic structure, enhanced adsorption ability and its superior electrocatalytic ability towards HMFOR. The blue, pink, red, brown and white balls are represented as Co, Mo, O, C and H, respectively.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  (a) SEM image, (b) HRTEM image and (c) corresponding element mapping image of Mo-Co₃O₄. (d) Raman spectra of Co₃O₄ and Mo-Co₃O₄. (e) Co 2p XPS spectra of Co₃O₄ and Mo-Co₃O₄. (f) Mo 3d XPS spectra of Mo-Co₃O₄.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) LSV plots of Co₃O₄ and Mo-Co₃O₄ in 1.0 mol/L KOH with or without 50 mmol/L HMF without iR compensation. (b) The potential at 30 mA/cm² and current density at 1.50 V on both Co₃O₄ and Mo-Co₃O₄. (c) The reaction pathway of HMF oxidation; (d) The concentration of DFF, HMFCA, FFCA, FDCA and HMF during HMF oxidation. (e) The yield_FDCA (%) and Faradic efficiency (%) of Mo-Co₃O₄ under four successive electrolysis cycles. (f) Reaction free energy barrier for the oxidation of aldehyde and hydroxyl groups on Mo-Co₃O₄. (g) Structure models for the dehydration of hydroxyl and aldehyde oxidation on Mo-Co₃O₄.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  In situ Bode-phase plots of (a) Co₃O₄ and (b) Mo-Co₃O₄ at different potentials in HMF. (c) The Bode-phase plots and Nyquist plots Co₃O₄ and Mo-Co₃O₄ at 1.50 V. (d) Side-view of differential charge density of Mo-Co₃O₄ (top), where the yellow and cyan represent electron accumulation and depletion with an isosurface value of 0.01 e/ų, respectively. The calculated adsorption energies and optimized structure of HMF on Mo-Co₃O₄ (left) and Co₃O₄ (right). (e) The energy barrier of OH* adsorption on Co₃O₄ and Mo-Co₃O₄.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Raman spectra, (b) HRTEM, (c) Co 2p XPS spectra and (d) Mo 3d XPS spectra of Mo-Co₃O₄ after four successive electrolysis cycles.

  下载: 全尺寸图片 幻灯片
•