Low-temperature fabrication of morphology-controllable Cu₂O for electrochemical CO₂ reduction

Abstract

Cu₂O has been successfully synthesized in different morphologies/sizes (nanoparticles and octahedrons) via a low-temperature chemical reduction method. Trapping metal ions in an ice cube and letting them slowly melt in a reducing agent solution is the simplest way to control the nanostructure. Enhancement of charge transfer and transportation of ions by Cu₂O nanoparticles was shown by cyclic voltammetry and electrochemical impedance spectroscopy measurements. In addition, nanoparticles exhibited higher current densities, the lowest onset potential, and the Tafel slope than others. The Cu₂O electrocatalyst (nanoparticles) demonstrated the Faraday efficiencies (FEs) of CO, CH₄, and C₂H₆ up to 11.90, 76.61, and 1.87%, respectively, at −0.30 V versus reference hydrogen electrode, which was relatively higher FEs than other morphologies/sizes. It is mainly attributed to nano-sized, more active sites and oxygen vacancy. In addition, it demonstrated stability over 11 h without any decay of current density. The mechanism related to morphology tuning and electrochemical CO₂ reduction reaction was explained. This work provides a possible way to fabricate the different morphologies/sizes of Cu₂O at low-temperature chemical reduction methods for obtaining the CO, CH_4, and C₂H₆ products from CO₂

Introduction

The world's energy consumption is extremely dependent upon fossil fuels, which causes energy shortages and environmental problem issues via CO₂ gas emission [1,2,3,4,5]. Researchers focus on capturing, storing, and converting CO₂ to solve these issues. Among them, CO₂ conversion is the most prominent route because it can transmute CO₂ into fuels and useful chemicals [6]. The electrochemical CO₂ reduction reaction (CO₂RR) is one of the most promising strategies in CO₂ conversion technique because of green technology and controllable reduction potential/reaction temperature under ambient conditions [7]. Recently, Cu-based electrocatalysts have been widely applied for CO₂RR because of several advantages such as low cost, intermediate binding energy for adsorbed CO (*CO), high yield of multi-carbon products, low overpotential, different oxidation states, non-toxic, competing hydrogen evolution reaction, and their stability [8, 9].

Copper oxide (Cu₂O) is considered a superior electrocatalyst for CO₂RR among Cu-based materials because of its ability to trap the subsurface oxygen, kinetically favorable, enhancement of *CO binding, and higher C1/C2 selectivity [10]. Various strategies (facet, defect, morphology, and heterostructure) have been applied to enhance the electrochemical CO₂RR performances. Among these strategies, morphology-controlled is a perfect way to increase the electrochemical CO₂RR for obtaining C1/C2 products because it can control selectivity and stability [11]. Different synthesis techniques (electrochemical deposition, chemical deposition, chemical reduction, and thermal treatment) have recently been used [12]. However, low-temperature fabrication of morphology-controllable Cu₂O for electrochemical CO₂ reduction has not been reported yet. The low-temperature fabrication technique via ice melting can effectively control the release of reactants, nucleation, and agglomeration of particles in aqueous solution [13]. So, tuning the morphology (shape/size) by low-temperature technique is the perfect option to boost the electrochemical CO₂RR performances of Cu₂O towards producing useful fuels [14].

In the present work, different morphologies/sizes of Cu₂O were synthesized by a simple low-temperature chemical reduction method via ice melting using a strong reducing agent. The variation of precursor concentrations has been applied to control the morphologies/sizes of Cu₂O. The samples were characterized by X-ray diffraction (XRD), field-emission scanning electron microscopy (FESEM), transmission electron microscopy (TEM), and X-ray photoelectron spectroscopy (XPS) measurements. The electrochemical measurements (cyclic voltammetry, electrochemical impedance spectroscopy, linear sweep voltammetry, and chronoamperometry) of samples were obtained by using an H-type cell for electrochemical CO₂RR. Tafel plots, detection of gaseous products, Faradic efficiencies (FEs), and possible mechanism of CO₂RR were explained.

Experimental

Catalyst synthesis

The chemicals used were of analytical grade and were used without further purification. 0.1 mm thickness copper foil was purchased from Merck, Germany. Copper nitrate trihydrate [Cu(NO₃)₂.3H₂O] was purchased from Acro’s Organic, Poland. Sodium borohydride (NaBH₄) was purchased from Spectrum Chemical MFG Corp, California, and ethanol [C₂H₅OH, (95% pure)] was purchased from VWR Chemicals, Canada. The samples were synthesized using a low-temperature chemical reduction method using one-pot synthesis. The main advantage of this method is that it enables the tuning of particles' morphology by slowly releasing Cu ions in the solution because of the slow release of metal ions during the melting of the ice cube compared to other solution-based methods [13]. It was carried out by forming an ice cube of Cu(NO₃)₂.3H₂O solution. Cu(NO₃)₂.3H₂O was weighed and dissolved in 100 mL of distilled water. Different concentrations of Cu(NO₃)₂.3H₂O (0.02 M, 0.1 M, and 0.5 M) were used for the synthesis. During synthesis, the pH of the solution was 4.4. The as-prepared ice cubes were added to NaBH₄ solution (0.01 M) in 100 mL ethanol. The sample was collected after 60 min and washed with distilled water and ethanol several times to remove the impurities. The obtained powders were dried and ground using mortar and piston. The catalysts were named Cu₂O-1, Cu₂O-2, and Cu₂O-3 for 0.02 M, 0.1 M, and 0.5 M concentrations of Cu salt, respectively. The schematic illustration of Cu₂O synthesis is shown in Fig. S1.

Material characterization

Powder X-ray diffraction (XRD) patterns were recorded in Rigaku Miniflex 600 (2θ: 10–80°, step: 0.02, and continuous: 1°/min) diffractometer. The morphology of samples was obtained by field emission scanning electron microscopy (FESEM, JEOL, JSM-IT800). Transmission electron microscopy (TEM), high-resolution transmission electron microscopy (HRTEM), and selected area diffraction patterns (SAED) images were recorded by JEOL 2100 PLUS TEM acquired at 120 kV. The X-ray photoelectron spectroscopy (XPS) was carried out on thermo-scientific ESCALAB™ XI (200 eV and Al Kα). Gas chromatography (GC) (SRI 8610C) analyzed the gaseous products.

Electrochemical measurements

The electrochemical measurements were performed in an H-type cell with a three-electrode system. The H-type cell consists of platinum (as a counter electrode), Ag/AgCl electrode (as a reference electrode stored in 3 M KCl solution), and Cu₂O (as a working electrode) in 0.5 M KHCO₃ solution. The electrode was prepared by mixing the 0.05 g of sample, and 50 µL Nafion in 500 µL ethanol. The mixture was sonicated for 60 min. The highly dispersed mixture was deposited on the 1 cm × 1 cm copper foil and left to dry for 12 h in an oven at 60 °C. The cyclic voltammetry (CV) of the prepared materials was measured that consists of −0.6 V to 0.6 V potential window with a scan rate between 40 mV/s and 100 mV/s. Similarly, the conductivity of the electrode was studied by electrochemical impedance spectroscopy (EIS) between 0.1 Hz and 100,000 Hz. Likewise, the linear sweep voltammetry (LSV) was studied between 0 and 0.1 V versus the reversible hydrogen electrode (RHE) at the scan rate of 10 mV/s. The following equation was used for the calculation of RHE:

$$ E_{RHE} = E_{Ag/AgCl} + 0.197 + 0.059 pH\;\left( {0.{\text{5M KHCO}}_{{3}} \sim { 8}.{5}} \right) $$

where E_Ag/AgCl represents the potential against the reference electrode, and 0.197 denotes the standard potential of Ag/AgCl at 25 ºC [15].

A 50 mL H-type electrochemical cell was used. Two compartments (anodic and cathodic) were separated by Nafion 117 membrane. Nafion 117 membrane was used after treatment with 1 M H₂SO₄ and water. The volume of electrolyte used was 35 mL. For the saturation of electrolytes, 99.99% pure CO₂ gas was bubbled in a cathodic cell compartment for 60 min with 10 sccm (standard cubic centimeters per minute) with the help of a mass flow controller (MC-100SCCM-D, Alicat Scientific). The outlet of the cathodic cell compartment was connected to GC. GC was calibrated with a standard gas mixture (ARC3). The gaseous products obtained during the reaction were detected by a flame ionization detector (FID). The current–time (it) measurements were performed at a fixed potential of −0.30 V versus RHE (Supplementary information). Gasses from the cell were injected into GC for 400 s, and electrocatalytic CO₂RR was evaluated. The stability of Cu₂O-1 was performed over 11 h.

Results and discussion

Figure 1 revealed the XRD patterns of catalysts. As observed, the peaks were assigned to cuprite Cu₂O (Pn-3m) along with strong (111) orientation (JCPDS: 05-0667) in all samples [16]. A possible reason for the formation of Cu₂O is associated with the reduction of Cu(NO₃)₂.3H₂O in the presence of a reducing agent and dissolved oxygen under ambient conditions [17, 18]. However, a small peak was observed at 38º (2θ), which suggests the presence of a trace amount of CuO (JCPDS: 45-0937) in Cu₂O-1 [18]. It may be related to Cu(I) oxidation during the synthesis [19]. In addition, a very low intense peak was found at 42º (2θ), which indicates the existence of a metallic copper (JCPDS: 04-0836) phase in Cu₂O-3 [20]. The result suggests that a high concentration of precursor favors the small quantity of Cu (0) in the sample. Besides phases, the crystallite size of samples was calculated based on Scherrer’s equation based on (111) diffraction peak. It was found that the crystallite size of Cu₂O-1(1.77 nm) is smaller than the Cu₂O-2 (2.1 nm) and Cu₂O-3 (2.4 nm). Cu₂O-1 and Cu₂O-2 revealed the formation of nanoparticles (Fig. 2a, b). The nanoparticles were slightly agglomerated. However, Cu₂O-3 presented the octahedron morphology (Fig. 2c). The size of the octahedron consists of 0.5–1.5 μm. Furthermore, The FESEM elemental mapping and spectrum indicated the uniform distribution of Cu and O elements in the samples (Fig. S2).

Figure 1

XRD patterns of samples.

Figure 2

FESEM images of samples. a Cu₂O-1, b Cu₂O-2, and c Cu₂O-3. (Scale bar: 0.5 μm).

The mechanism related to the controlled synthesis of different morphologies/sizes was presented in Fig. 3. First, Cu₂O crystal nuclei were formed by the reaction between the salt solution of Cu(NO₃)₂.3H₂O with NaBH₄ via the nucleation process. In the case of Cu₂O-1, the lower concentration of Cu(NO₃)₂.3H₂O enhances the reduction rate that can generate multiple single nuclei. Also, the crystal nucleation rate is higher than the crystal's growth in Cu₂O-1. Due to this reason, a smaller size of nanoparticles was formed. However, the slight increase in precursor concentration lowers the reduction rate, and few nuclei may be produced in Cu₂O-2. In addition, these nuclei may collide with other nuclei to generate large size of particles [21]. Furthermore, at high concentrations, the precursor reduces the crystal nucleation rate than the crystal growth, which can increase the particle size in Cu₂O-3.

Figure 3

A schematic illustration of the formation mechanism of different sizes/morphologies of samples.

The detailed structural information of the Cu₂O-1 sample was shown by TEM, HRTEM, and high-angle annular dark-field scanning (HAADF) TEM images with elemental mapping (Fig. 4). The TEM images suggested the existence of nanoparticles in Cu₂O-1. The rough surface with several steps and kinks on the surface would provide a more active site for catalytic reaction. According to HRTEM image, the interplanar spacing of crystalline was found to be 0.246 nm, corresponding to Cu₂O (111) plane (Fig. 4b). The fringe spacing agreed with Cu₂O (JCPDS: 05–0667), revealing good agreement with XRD patterns (Fig. 1). The uniform distribution of copper and oxygen was observed in TEM imaging (Fig. 4c).

Figure 4

a TEM, b HRTEM and c HAADF-TEM images with EDX mapping of Cu₂O-1.

The chemical states and elemental composition of Cu₂O-1, Cu₂O-2, and Cu₂O-3 were measured via XPS spectra (Fig. 5). The Cu 2p spectra of samples were deconvoluted into 2p_3/2 (Cu₂O-1: 935.79 eV, Cu₂O-2: 935.37 eV, and Cu₂O-3: 934.42 eV), 2p_3/2 satellite (Cu₂O-1: 943.77 eV, Cu₂O-2: 943.80 eV, and Cu₂O-3: 943.60 eV), 2p_1/2, (Cu₂O-1: 955.61 eV, Cu₂O-2: 955.42 eV, and Cu₂O-3: 954.12 eV), and 2p_1/2 satellite (Cu₂O-1: 963.61 eV, Cu₂O-2: 963.66 eV, and Cu₂O-3: 962.74 eV). It suggests the existence of Cu⁺ in samples (Fig. 5a–c) [22,23,24]. According to Fig. 5d–f, O 1s spectra demonstrated three distinct fitted peaks which were lattice oxygen (Cu₂O-1: 529.00 eV, Cu₂O-2: 528.81 eV, and Cu₂O-3: 529.17 eV), chemisorbed oxygen (Cu₂O-1: 530.07 eV, Cu₂O-2: 530.41 eV, and Cu₂O-3: 530.96 eV), and oxygen vacancies (Cu₂O-1: 532.51 eV, Cu₂O-2: 532.04 eV, and Cu₂O-3: 532.61 eV) [25, 26]. This result also provides a higher oxygen vacancy in Cu₂O-1 nanoparticles. Furthermore, the survey spectra showed the presence of Cu and O in samples.

Figure 5

XPS analysis of Cu₂O-1 (a Cu 2p, d O1s, and g survey), Cu₂O-2 (b Cu 2p, e O1s, and h survey), and Cu₂O-3 (c Cu 2p, f O1s, and i survey).

The electrochemical characterization of morphology controlled Cu₂O was performed by CV analysis that can determine the oxidation and reduction behavior of the electrocatalyst. Figure 6a–c represent the CV curves of Cu₂O-1, Cu₂O-2, and Cu₂O-3 in the potential range -0.6 to 0.6 V vs Ag/AgCl at the various scan rates (40 mV/s, 60 mV/s, 80 mV/s, and 100 mV/s), respectively. Each of the unsymmetrical CV profiles revealed the anodic (Cu₂O-1: 0.11 V, Cu₂O-2: 0.084 V, and Cu₂O-3: 0.089) and cathodic peaks (Cu₂O-1: −0.37 V, Cu₂O-2: −0.39 V, and Cu₂O-3: −0.37 V), indicating the redox reaction of Cu₂O. It can be attributed Cu(I)/Cu(II) redox reaction [18]. The oxidation and reduction peaks were shifted towards the positive and negative potential with increased scan rates, respectively. It may be associated with internal resistance over the electrode and charge diffusion polarization in the electrodes [27]. Also, the increase in scan rate promoted the current response, which suggests the kinetic of the interfacial oxidation and reduction reactions and the rapid rate of the ionic/electronic responses [28]. Moreover, Cu₂O-2 demonstrated lower current density than Cu₂O-1 and Cu₂O-3 (Figs. 6a–c, and S3). It indicates that nanoparticles and octahedron morphologies may provide more active sites for electrocatalytic CO₂RR. The CV performance Cu-foil was evaluated to eliminate the interference from the substrate (Fig. S4). The current density of Cu-foil is observed to be lower than the Cu₂O samples. The electrochemical performance of catalysts was further explored by EIS measurements; the results are presented in Fig. 6d. The catalysts have a semicircle and straight line in high-frequency and low-frequency regions. Compared to Cu₂O-2 and Cu₂O-1, Cu₂O-3 revealed a smaller semicircle diameter. In addition, an equivalent circuit model is shown in Fig. S5. The estimated value of solution resistance, charge transfer resistance, Warburg coefficient, and electric double-layer capacitance are shown in Table S1. The results suggest that Cu₂O-1 and Cu₂O-3 demonstrated lower charge transfer resistance than Cu₂O-2. It suggests the smaller charge-transfer resistance in nanoparticles and octahedrons. The superior charge transfer efficiency and good conductivity of these catalysts can enhance the electrocatalytic CO₂RR.

Figure 6

CV (a Cu₂O-1, b Cu₂O-2, and c a Cu₂O-3) and EIS plot of samples.

To better elucidate the effect of Cu₂O morphologies on the electrocatalytic CO₂RR performance, LSV tests were performed for the Cu₂O-1, Cu₂O-2, and Cu₂O-3 catalysts at a sweep rate of 20 mV/s in CO₂-saturated 0.5 M KHCO₃ (Fig. 7a). As observed, Cu₂O-1 exhibited higher current densities and lowest onset potential than Cu₂O-2, and Cu₂O-3 catalysts. The possible reason may be associated with good electrical conductivity that can accelerate the transfer of electrons during the electrocatalytic CO₂RR. Also, Cu₂O-1 has the smallest current at a voltage of −0.25 V versus RHE. The possible reason may relate to variation in the electrochemical reaction rate [29]. To investigate the reason for the greater CO₂RR performance of the Cu₂O-1 electrode than others, Tafel slopes were evaluated (Fig. 7b). The Tafel slopes of the Cu₂O-1, Cu₂O-2, and Cu₂O-3 catalysts revealed 94.62, 117.81, and 101.83 mV/dec, respectively. It suggests that Cu₂O-1 exhibited the lowest Tafel slope. These results also indicate that the surface of Cu₂O-1 nanoparticles makes the electrocatalytic CO₂RR easier than others by enhancing the initial transfer rate of electrons to the CO₂ molecules.

Figure 7

a LSV, b Tafel slope, c FEs of the electrodes, and d stability test of Cu₂O-1 for CO₂ reduction over 11 h.

CO₂ electroreduction using three catalysts was conducted in a 0.5 M KHCO₃ under applied cathode potential to investigate the catalytic activities towards the product distribution. The chronoamperometry measurements of Cu₂O samples at −0.3 V versus RHE for 400 s were shown in Fig. S6. It indicates that Cu₂O-1 nanoparticles revealed higher current density than others. In addition, the Cu-foil showed lower current density than other Cu₂O samples (Fig. S7). The calculation of FEs was provided in supplementary information and Table S2. FEs of CO (Cu₂O-1: 11.90%, Cu₂O-2: 9.76%, and Cu₂O-3: 12.1%), CH₄ (Cu₂O-1: 76.61%, Cu₂O-2: 65.11%, and Cu₂O-3: 74.63%), and C₂H₆ (Cu₂O-1: 1.87%, Cu₂O-2: 2.39%, and 2.60%) were obtained during electrocatalytic CO₂RR samples (Fig. 7c). These results suggested that CH₄ and CO/C₂H₆ were major and minor products, respectively.

Based on the hydrocarbon products, the possible mechanisms have been proposed. At first, the CO₂ molecules are adsorbed and activated on the crystalline plane of Cu₂O (111). Then, it is quickly converted into *COOH via multiple proton-coupled electron transfer (PCET) processes. In addition, this process coupled with *COOH intermediate to produce *CO. In addition, Cu₂O (111) tends to provide stable adsorption of *CO. The adsorption of *CO shows superior stability on the catalyst surface, making C1 and C2 products possible. The desorption of *CO from the active site to form CO [30, 31]. Furthermore, *CO may convert into *CHO, *CH₂O and *CH₃O intermediates via PCET to generate CH₄. Also, *CHO is converted into *CH₃ via *CH₂ formation by PCET. Then, dimerization of CH₃ produces the C₂H₆ [9, 32, 33].

The high selectivity of Cu₂O for CH₄ may be attributed to (111) crystalline plane that can provide active sites. Also, abundant hydroxyl groups tend to tune the stability of intermediates for CH₄ formation by hydrogen bonds [23]. The low selectivity of C₂H₆ may relate to the kinetic barrier for the coupling, which can effectively influence the degree of the adsorbed CO hydrogenation. Such type of kinetic barrier decreases with an increase in the degree of the surface-bound CO hydrogenation that may favor C₂H₆ [34]. Also, the adsorbed CO is an intermediate state of C1 and C2 products. Due to this reason, low selectivity of Cu₂O for CO was observed [35]. Cu₂O-1 demonstrated higher electrocatalytic CO₂RR efficiency compared to others. It consists of nanosized among all samples. It consists of more active sites, oxygen vacancy, high flexibility, and rapid oxygen diffusion along the dense boundary between the nanoparticles [35]. Also, the co-existence of Cu⁺ and Cu²⁺ ions can provide active sites and more favorable free energy for the production of *CO intermediates that facilitate the generation of C–C bonds for C₂₊ formation [36]. Besides producing CO, CH₄, and C₂H₆, stability is important for electrocatalysts towards CO₂RR efficiencies. According to the stability test of Cu₂O-1, there was no decay of current densities over an 11 h period, indicating significant stability (Fig. 7d). Furthermore, the stability of the catalyst was investigated using XRD and FESEM analyses. Figure S8 presented the XRD patterns of Cu₂O-1 after electrocatalytic stability after 11 h. It was observed that the XRD of the used catalyst is well matched with the fresh catalyst. In addition, the FESEM image of the used catalyst was like that of the unused catalyst (Fig. S9). These characterization techniques suggest the excellent stability of the catalyst for electrocatalytic CO₂RR. Table 1 shows a comparison of Cu₂O with previously reported Cu₂O-based electrode materials. Our results are comparable to those of previous literature.

Table 1 Comparison of the catalytic performance of synthesized Cu₂O with other Cu₂O-based catalysts

Conclusion

In summary, various morphologies/sizes of Cu₂O were synthesized by the chemical reduction method at low temperatures by changing the concentration of metal ions. XRD suggested the trace amount of CuO and Cu phases at low and high concentrations of precursors, respectively, in the Cu₂O catalyst. FESEM provided evidence of different sizes/morphologies. Elemental mapping demonstrated the uniform distribution of Cu and O in samples. The evidence of the Cu₂O (111) plane was proved by the HRTEM image. The electrochemical characterization (CV, EIS, LSV, Tafel plot, and chronoamperometry curves) was carried out for electrochemical CO2RR performance on H-type cells. Cu₂O nanoparticles revealed higher electrocatalytic CO₂RR efficacy than others (bigger particles and octahedron). In addition, Cu₂O catalysts tend to provide CH₄ (major) and minor (CO and C₂H₆) products at −0.3 V versus RHE. The possible reasons for the enhancement of electrocatalytic efficiency of Cu₂O nanoparticles are the presence of a greater number of active sites, oxygen vacancy, good electrical conductivity, the ability of rapid electron transfer, and a trace amount of Cu²⁺ phase. In conclusion, tuning Cu₂O morphology using low temperature chemical reduction method is a great way to make efficient catalysts for generating CO, CH₄, and C₂H₆ via electrocatalytically CO₂RR.