Temperature-Dependent Pathways in Carbon Dioxide Electroreduction

doi:10.21203/rs.3.rs-4925085/v1

Download PDF

Article

Temperature-Dependent Pathways in Carbon Dioxide Electroreduction

https://doi.org/10.21203/rs.3.rs-4925085/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Temperature affects both the thermodynamics of intermediate adsorption and the kinetics of elementary reactions. Despite its extensive study in thermocatalysis, temperature effect is typically overlooked in electrocatalysis. This study investigates how electrolyte temperature influences CO₂ electroreduction over Cu catalysts. Theoretical calculations reveal the significant impact of temperature on *CO and *H intermediate adsorption thermodynamics, water microenvironment at the electrode surface, and the electron density and covalent property of the C–O bond in the *CH–COH intermediate, crucial for the reaction pathways. The theoretical calculations are strongly verified by experimental results over different Cu catalysts. Faradaic efficiency (FE) toward multicarbon (C₂₊) products is favored at low temperatures. Cu nanorod electrode could achieve a FE_C2+ value of 90.1% with a current density of ~ 400 mA cm^− 2 at − 3°C. FE_C2H4 and FE_C2H5OH show opposite trends with decreasing temperature. The FE_C2H4/FE_C2H5OH ratio can decrease from 1.86 at 40°C to 0.98 at − 3°C.

Introduction

Electrochemical CO₂ reduction reaction (CO₂RR) into high-value products stands as one of the most promising strategies for mitigating CO₂ emissions through the utilization of renewable electricity^1–2. CO₂RR is a complex process involving multiple reaction pathways that harvest a diverse array of chemical products^3–4. However, the simultaneous occurrence of various CO₂RR routes alongside the hydrogen evolution reaction (HER) can diminish the selectivity toward desired products^5–8. The adsorption behavior of carbonous intermediates and the intricate water microenvironment at the electrode surface are pivotal factors for influencing these reaction pathways, thereby dictating the distribution of products^9–12. By far, researchers have developed a wide range of electrode materials and electrolytes tailored to finely control intermediate adsorption and the water microenvironment on the electrode surface^13–16. These advancements hold significant promise for steering the CO₂RR pathway toward desired product with enhanced efficiency and selectivity.

The adsorption or dispersion of intermediates, as well as the water microenvironment, are significantly influenced by temperature since they are thermodynamically controlled^17–19. For instance, both C₂H₄ and C₂H₅OH share the same precursor *CH–COH, leading to their simultaneous production^20–23. The kinetics of their distinct reduction pathways can be influenced by temperature, offering a feasible means to control the ratio of C₂H₄ to C₂H₅OH. Hence, adjusting the temperature of the electrolyte to regulate both thermodynamic and kinetics processes emerges as a potent method for steering the CO₂RR pathway. Consequently, a comprehensive investigation into the relationship between performance and temperature is crucial, providing invaluable insights and guiding significance for optimizing CO₂RR performance^{4, 24}.

CO₂RR experiments are typically conducted at room temperature, which can vary, for example from − 3°C to 40°C, depending on seasons and regions. The environmental temperature, typically indicated by the electrolyte temperature, can significantly influence the performance of CO₂RR, yet it is often ignored in CO₂RR studies^25–28. In this study, we systematically investigated the impact of temperature on CO₂RR performance. We initiated our study with theoretical calculations, including density functional theory (DFT) and molecular dynamics (MD) simulations, to explore the impact of temperature on intermediate adsorption and kinetics of elementary reactions in CO₂RR. Subsequently, Cu catalysts were synthesized and employed for CO₂RR at various temperatures. The theoretical findings aligned well with experimental observations, indicating that lower temperatures favor C₂₊ production and promote the formation of C₂H₅OH over C₂H₄. For instance, a Faradaic efficiency toward multicarbon products (FE_C2+) of 90.1% was achieved with a current density of ~ 400 mA cm^− 2 at − 1.3 V vs RHE over a Cu nanorod (Cu-NR) electrode at − 3°C. Moreover, the FE_C2H4/FE_C2H5OH ratio decreases gradually from 1.86 to 0.98 in 1 M KOH as the temperature decreases from 40°C to − 3°C. Further characterizations, including in situ surface-enhanced infrared absorption spectroscopy (ATR-SEIRAS), in situ Raman spectroscopy and electrochemical analysis, provide a comprehensive understanding of the temperature effect on CO₂RR performance.

Physical sciences/Chemistry/Catalysis/Electrocatalysis

Physical sciences/Chemistry/Electrochemistry/Electrocatalysis

ectrochemical CO₂ reduction reaction (CO₂RR) into high-value products stands as one of the most promising strategies for mitigating CO₂ emissions through the utilization of renewable electricity^1-2. CO₂RR is a complex process involving multiple reaction pathways that harvest a diverse array of chemical products^3-4. However, the simultaneous occurrence of various CO₂RR routes alongside the hydrogen evolution reaction (HER) can diminish the selectivity toward desired products^5-8. The adsorption behavior of carbonous intermediates and the intricate water microenvironment at the electrode surface are pivotal factors for influencing these reaction pathways, thereby dictating the distribution of products^9-12. By far, researchers have developed a wide range of electrode materials and electrolytes tailored to finely control intermediate adsorption and the water microenvironment on the electrode surface^13-16. These advancements hold significant promise for steering the CO₂RR pathway toward desired product with enhanced efficiency and selectivity.

The adsorption or dispersion of intermediates, as well as the water microenvironment, are significantly influenced by temperature since they are thermodynamically controlled^17-19. For instance, both C₂H₄ and C₂H₅OH share the same precursor *CH–COH, leading to their simultaneous production^20-23. The kinetics of their distinct reduction pathways can be influenced by temperature, offering a feasible means to control the ratio of C₂H₄ to C₂H₅OH. Hence, adjusting the temperature of the electrolyte to regulate both thermodynamic and kinetics processes emerges as a potent method for steering the CO₂RR pathway. Consequently, a comprehensive investigation into the relationship between performance and temperature is crucial, providing invaluable insights and guiding significance for optimizing CO₂RR performance^{4, 24}.

CO₂RR experiments are typically conducted at room temperature, which can vary, for example from −3 °C to 40 °C, depending on seasons and regions. The environmental temperature, typically indicated by the electrolyte temperature, can significantly influence the performance of CO₂RR, yet it is often ignored in CO₂RR studies^25-28. In this study, we systematically investigated the impact of temperature on CO₂RR performance. We initiated our study with theoretical calculations, including density functional theory (DFT) and molecular dynamics (MD) simulations, to explore the impact of temperature on intermediate adsorption and kinetics of elementary reactions in CO₂RR. Subsequently, Cu catalysts were synthesized and employed for CO₂RR at various temperatures. The theoretical findings aligned well with experimental observations, indicating that lower temperatures favor C₂₊ production and promote the formation of C₂H₅OH over C₂H₄. For instance, a Faradaic efficiency toward multicarbon products (FE_C2+) of 90.1% was achieved with a current density of ~ 400 mA cm⁻² at −1.3 V vs RHE over a Cu nanorod (Cu-NR) electrode at −3 °C. Moreover, the FE_C2H4/FE_C2H5OH ratio decreases gradually from 1.86 to 0.98 in 1 M KOH as the temperature decreases from 40 °C to −3 °C. Further characterizations, including in situ surface-enhanced infrared absorption spectroscopy (ATR-SEIRAS), in situ Raman spectroscopy and electrochemical analysis, provide a comprehensive understanding of the temperature effect on CO₂RR performance.

DFT calculations. DFT calculations can elucidate the adsorption thermodynamics of intermediates and predict reaction mechanisms of CO₂RR^29–30. This study initiated with DFT calculations to explore the temperature effect on the adsorption Gibbs free energy of intermediates and reaction pathways, modeling the Gibbs free energy for the adsorption of intermediates and potential reaction steps based on a Cu (111) surface. *CO and *H are key intermediates in CO₂RR and HER, respectively^31–32. The competitive adsorption of *H against *CO frequently results in significant HER, consequently diminishing the selectivity of CO₂RR. The adsorption configurations and Gibbs free energy profiles of *CO and *H [ΔG_ad(CO) and ΔG_ad(H)] on Cu(111) are depicted in Supplementary Fig. 2. As the temperature decreases, ΔG_ad(CO) becomes increasingly negative, favoring C–C coupling, while ΔG_ad(H) gradually approaches zero, inhibiting HER.

AIMD simulations. Further AIMD studies were conducted to investigate the impact of temperature on the spontaneous diffusion path of *CO and *H intermediates on the Cu(111) surface. *CO and *H intermediates are predominantly localized at the threefold site as the temperature drops (Fig. 1a–b). Notably, at higher temperatures, the diffusion of *CO and *H intermediates intensifies on the Cu(111) surface, with spontaneous diffusion observed between two threefold sites. The radial distribution functions (RDFs) between intermediates and the Cu(111) surface reveal a sharp peak centered at 0.3 ~ 0.4 Å at 0 ^oC and − 40 ^oC (Fig. 1c), indicating the average diffusion radius of *CO and *H intermediates around the initial position. When the temperature increases to 40 ^oC, a broad peak is observed at 0.8 ~ 2.0 Å, signifying spontaneous diffusion at high temperatures. The statistical AIMD results corroborates the static results obtained from ΔG_ad values.

Since water is the key proton donor, decreasing the supply of water is an effective method for suppressing HER and enhancing the reduction of CO₂ to C₂₊ products³⁵. Solutions containing K⁺ ions are typically used as electrolytes. An AIMD model incorporating explicit water molecules was further employed to investigate the influence of temperature on the coordination water number of K⁺. The coordination water numbers of K⁺ are found to be 5.52, 4.64, and 4.12 at − 40°C, 0°C, and 40°C, respectively, indicating a continuous increase in water activity with rising temperature. The peak intensity of RDFs between coordination water and K⁺ becomes more pronounced with increasing temperature, suggesting that water thermal motion intensifies and drifts away from K⁺ at higher temperatures (Fig. 1e).

MD simulations. The concentration of ions can influence water microenvironment, thereby impacting the kinetics of C–C coupling^33–34. Surface OH⁻ concentration is closely linked to the selectivity of C₂₊ products during CO₂RR. MD simulations were conducted to explore the influence of temperature on the distribution of interfacial OH⁻ and CO on the Cu(111) surface. As the temperature decreases, OH⁻ ions migrate closer to the Cu surface (Fig. 1f), resulting in an increased local concentration of OH⁻. This promotes C–C coupling and suppresses HER. Additionally, as temperature increases, CO also exhibits a tendency to migrate to the Cu(111) surface, thereby enhancing the probability of C–C coupling.

From the discussion above, it is evident that temperature influences the adsorption of *CO and *H intermediates, as well as the interfacial microenvironment, which in turn regulates C–C coupling and HER. However, C–C coupling generating different C₂₊ products follows distinct pathways, resulting in diverse selectivity toward a single product. Specifically, C₂H₄ and C₂H₅OH are usually generated simultaneously due to the presence of the same *CH–COH intermediate, which can either undergo hydrogenation via the Eley-Rideal mechanism and dehydration to form *C–CH (C₂H₄ pathway), or hydrogenation via the Langmuir-Hinshelwood mechanism to produce *CH–CHOH (C₂H₅OH pathway) (Fig. 2a and Supplementary Fig. 6). The energy barriers calculated for these two elementary reactions on the Cu(111) surface are 1.54 and 0.75 eV, respectively (Fig. 2b). Thereby, starting from crucial branching intermediate *CH–COH, the energy barrier difference between *CH–C formation and *CH–CHOH formation is positive, with a value of 0.79 eV. According to Arrhenius equation $\:\raisebox{1ex}{${k}_{CH-C}$}\!\left/\:\!\raisebox{-1ex}{${k}_{CH-CHOH}$}\right.=a{e}^{-\frac{\varDelta\:{E}_{TS}}{RT}}$, $\:\raisebox{1ex}{${k}_{CH-C}$}\!\left/\:\!\raisebox{-1ex}{${k}_{CH-CHOH}$}\right.$is negatively correlated with temperature change. Consequently, the formation of C₂H₄ and C₂H₅OH exhibits a reverse trend: high temperatures favor C₂H₄ formation, whereas low temperatures are conducive to producing C₂H₅OH. An AIMD model was utilized to examine the impact of temperature on the bond vibration of the *CH–COH intermediate. The C–O bond on the Cu(111) surface shows a more pronounced response to temperature compared to the C–C and O–H bonds, as indicated by the lowest σ value of the normal distribution fitting (Supplementary Fig. 7). The electronic properties, including projected crystal orbital Hamilton population (pCOHP), electron localization function (ELF) and charge density difference (CDF), were employed to analyze the vibration of the C–O bond in *CH–COH. When changing the C–O bond length to 0.94, 0.97, 1.00, 1.03, and 1.06 times longer than its original length, the corresponding integrated pCOHP (IpCOHP) values are − 9.49, − 8.88, − 8.31, − 7.79, and − 7.33 eV, respectively. The value of IpCOHP becomes less negative as the C–O bond elongates, indicating an increase in the antibonding state of the C–O bond, resulting in the significant activation C–O bond of the *CH–COH intermediate (Fig. 2c–e). As the C–O bond stretches, the electron density and covalent property of C–O gradually decrease, as revealed by the ELF patterns (Supplementary Figs. 8–9). AIMD results indicate that higher temperatures cause the more obvious stretching of the C–O bond in *CH–COH, resulting in decreased electron density and covalent property of the C–O bond. This, in turn, favors the activation of the C–O bond in *CH–COH for hydrogenation to *C–CH and subsequently to C₂H₄. In summary, high temperatures promote the production of C₂H₄ over C₂H₅OH.

CO ₂ RR at various temperatures. CO₂RR experiments were carried out using a flow-cell reactor in 1 M KOH, with the temperature of the electrolyte being regulated by a temperature controller. The catalyst employed in the study was Cu-NR catalyst, predominantly oriented along the (111) crystal plane and existed in the Cu(0) state, consistent with the Cu(111) model used for thermotical calculations. Relevant characterizations, such as the X-ray diffraction (XRD) pattern, scanning electron microscopy (SEM) image, transmission electron microscope (TEM) image, X-ray Photoelectron Spectroscopy (XPS) spectra and X-ray absorption near edge structure (XANES) spectrum of the Cu electrode are presented in (Supplementary Figs. 10–13). When temperatures are varied, significant variations in product distributions are observed, yet there is relatively small variation in total current density (Fig. 3a–c and Supplementary Fig. 14). For example, at − 1.3 V vs RHE, FE_C2+ increases from 73.6–85.3% as the temperature drops from 40°C to − 3°C (Fig. 3d). Notably, the increased FE_C2+ originates from the inhibited HER and enhanced C–C coupling when decreasing temperature (Fig. 3e–f). Moreover, as the temperature decreases, the concentration of FE_C2H4 gradually diminishes, while that of FE_C2H5OH evidently rises. Consequently, the FE_C2H4/FE_C2H5OH ratio exhibits a change from 1.86 to 0.98 when transitioning catholyte temperature from 40°C to − 3°C (Fig. 3d), and the primary product of CO₂RR shifts from C₂H₄ to C₂H₅OH.

To further investigate the relationship between temperature and CO₂RR selectivity, we conducted experiments using different electrolytes and catalysts. Cu-NR electrode was employed for CO₂RR in 3 M KOH across various temperatures. When decreasing the electrolyte temperature from 20°C to − 3°C, FE_C2+ value increases from 83.1–90.1% over the Cu catalyst electrode at − 0.95 V vs RHE, with the FE_C2H4/FE_C2H5OH ratio varying from 1.44 to 0.96 (Supplementary Fig. 15a–b). Additionally, commercial Cu nanoparticles (Cu-NP) with the dominant (111) crystal plane (Supplementary Figs. 10–12) were used as catalysts for CO₂RR across a temperature range of − 3°C to 20°C using 1 M KOH (Supplementary Fig. 15c–d). A similar trend that lower temperatures favor C–C coupling and C₂H₅OH formation is observed over Cu-NP electrode.

Variations of CO ₂ RR intermediates at various temperatures. In situ ATR-SEIRAS spectra during CO₂RR over Cu-NR electrode at varying temperatures and potentials were performed to investigate the influence of temperature on water structure and adsorbed intermediates. The OH stretching of water within the range of 3000 ~ 3800 cm^− 1 can be divided into isolated water (~ 3600 cm^− 1), weakly hydrogen-bonded water (~ 3450 cm^− 1) and strongly hydrogen-bonded water (~ 3250 cm^− 1) after Gaussian fitting (Fig. 4a–c)³⁵. The proportion of isolated water decreases as the potential becomes increasingly negative, and this variation is more pronounced at lower temperatures (Fig. 4d). The band center shift is negligible with potential at 40°C and 20°C (Fig. 4a–b). However, at − 3°C, the band center of OH stretching notably shifts to lower wavenumbers at more negative potentials (Fig. 4c), indicating the formation of a robust hydrogen bond network. This strong hydrogen bond network at low temperature leads to reduced water activity, effectively suppressing HER. The peaks around 1210 and 1340 cm^− 1 are identified as *OCCOH and *OCCHO, respectively (Fig. 4f–g), representing typical C–C coupling intermediates³³. Additionally, the presence of the *OC₂H₅ intermediate at 1330 cm^− 1 suggests the formation of C₂H₅OH^36–37. Comparatively, at − 3°C, the characteristic peaks corresponding to *OCCOH, *OCCHO, and *OC₂H₅ shift to lower wavenumbers compared to those at 40°C, indicating increased adsorption of these intermediates. This suggests an enhancement in C–C coupling and the C₂H₅OH pathway at lower temperatures³⁸.

In situ Raman peaks at approximately 360 and 1890/2060 cm^− 1 are attributed to the frustrated rotation of Cu–CO and C ≡ O stretching, respectively (Fig. 5a–b) ^39–41. The peak at ~ 700 cm^− 1 is assigned to adsorbed *OH species, originating from H₂O dissociation⁴². Due to the Stark effect, these peaks exhibit blue shifts as the potential becomes more negative over the Cu-NR electrode, indicating a stronger Cu–CO interaction⁴³. The Stark tuning slope of Cu–CO increases as temperature varies from 40°C to − 3°C, suggesting that lower electrolyte temperature enhances CO adsorption on the electrode surface (Fig. 5c). The CO stripping potential shifts to a higher value as the temperature decreases (Fig. 5f), providing further confirmation of the strengthened interaction between CO and the electrode surface, in accordance the DFT calculations (Supplementary Fig. 2) ⁴⁴.

The dependence of j_C2H4 and j_C2H5OH on temperature is illustrated in Fig. 5d. Both j_C2H4 and j_C2H5OH exhibit positive correlations with temperature, with C₂H₄ production showing greater sensitivity to temperature changes (Supplementary Fig. 17). Since both C₂H₄ and C₂H₅OH share the same *CH–COH intermediate, decreasing the temperature evidently suppresses the formation of C₂H₄, leading to an increase in FE_C2H5OH. The Tafel slopes for the production of C₂₊ at 40°C, 20°C, and − 3°C are measured to be 114, 117, and 142 mV·dec^− 1 (Fig. 5e), respectively. These values closely align with the theoretical value of 119 mV·dec^− 1, indicating that the rate-determining step is C–C coupling at all investigated temperatures⁴⁵. The electrochemical double-layer capacitance (C_dl) values are 74.6, 44.7 and 25.9 mF·cm^− 2 at 40°C, 20°C, and − 3°C, respectively (Supplementary Fig. 18). The cations exhibit a tendency to migrate from electrolyte to the electrode surface at low temperature (Fig. 1f), leading to the formation of a closely packed electrochemical double layer, consequently resulting in a decrease in the C_dl value at lower temperatures. The partial current density for C₂₊ products (j_C2+) was normalized on the basis of the electrochemically active surface area (ECSA), (Supplementary Fig. 19). A significantly higher j_C2+ is observed at − 3°C, confirming that low temperature is indeed effective in promoting C–C coupling.

In summary, we have initiated an investigation for the influence of environmental temperature on CO₂RR performance through theoretical calculations. As temperature decreases, *CO adsorption is favored while HER is inhibited, indicating a preference for C–C coupling. The increased hydration of K⁺ cations at low temperatures results in the reduced water activity, thereby inhibiting HER. The higher energy barrier for the hydrogenation of *CH–COH to *CH–C compared to *CH–CHOH favors the C₂H₅OH pathway at lower temperatures. In situ ATR-SEIRAS and Raman spectra, along with electrochemical characterizations, further confirm that *CO adsorption, water activity and the hydrogenation pathway of *CH–COH intermediate are highly temperature-dependent, with C–C coupling and C₂H₅OH formation being enhanced at lower temperature. Consequently, FE_C2+ achieves a high value of 90.1% at − 3°C over the Cu-NR electrode, with the main C₂₊ product transitioning from C₂H₄ to C₂H₅OH solely by decreasing the temperature. We believe this study will bring attention to the temperature effect on CO₂RR as well as other electrocatalytic reactions, thereby paving the way for optimizing their efficiency.

DFT simulations. Density functional theory (DFT) method was used to perform all the spin-polarized calculations of these structures, as implemented in the Vienna ab initio simulation package code (VASP)⁴⁶. Exchange and correlation energies were described in the methods of generalized gradient approximation (GGA) in the form proposed by Perdew–Burke–Ernzerhof (PBE) functional^47–48. The projector-augmented wave (PAW) method was used to describe interaction between valance and core electrons⁴⁹. The plane wave energy cutoff was set to be 500 eV. The structure optimization was relaxed until convergence criteria were met with the energy and force of 1×10^− 5 eV and 0.02 eV/Å, respectively. For structure optimizations, the first Brillouin-zone of such a slab sampled with the Monkhorst − Pack mesh with 3×3×1 grids, were used. To obtain more accurate electronic properties, a denser 7×7×1 k-point grids were further employed⁵⁰. The weak dispersion interaction was described by the DFT-D3 method with the standard parameters proposed by Grimme and his coworkers⁵¹. To avoid interactions between two neighboring catalyst monolayers under periodic boundary conditions, a minimum vacuum space of 15 Å was set. Solvent effects were included by using implicit solvent model implemented by VASPsol with a dielectric constant of 80⁵². To confirm the bonding nature of the investigated systems by characterizing covalent bond, the electron localization function (ELF) and Critic2 were further employed^53–54. The Crystal Occupation Hamiltonian Population (COHP) method was also employed to investigate the nature of bonding and anti-bonding states, as implemented by LOBSTER code^55–56.

Searching transition states (TS) were performed by employing improved dimer method implemented in Henkelman’s scripts, where the convergence force was set to be smaller than 0.02 eV/Å^57–58.

Ab initio molecular dynamics (AIMD) simulations were performed with the Nose-Hoover thermostat approach, at the average temperature of − 40, 0 and 40°C, respectively^59–60. The time step in AIMD was set to be 1 fs. For AIMD simulations, a gamma-centered 1×1×1 k-point grid was used. We carried out 10 ps of AIMD simulation to obtain a well-equilibrated system.

To simulate the copper electrode surface, a three-layer and 4×4 periodic cell Cu(111) slab was built, in which two bottom layers were fixed and the top layer was allowed to relax. To simulate the real solution environment, aqueous interface models for kinetic calculations contain 33 explicit water molecules, which could maintain the average water density in the bulk regions being around 1.0 g·cm^− 3. One K atom was located in bulk water as a solvated K⁺ in the solution, and a OH group was also located in bulk water to maintain the electrical neutrality of the periodic system.

MD simulations. MD simulations were performed using xTB package⁶¹. The calculated density of this simulated system was about 1.0 g·cm^− 3, closed to realistic solution circumstances. The MD simulation duration was 20 ps, ensuring that the solution system could reach a stable state, as shown in energy variation diagram (Supplementary Figs. 4–5). The simulations were carried out in the constant particle number, volume and temperature (NVT) ensemble with the thermostat set to a constant temperature of − 40, 0 and 40°C, respectively. After the equilibration period of 10 ps, the snapshots were taken at every 50 fs intervals from the simulation trajectories to analyze distribution of free CO/OH⁻ and CO/OH⁻ adsorbed on catalyst surface.

Thermochemistry. The Gibbs free energy difference (∆G) between two neighboring intermediates, named 1 and 2, can be calculated by:

$$\:{\varDelta\:\text{G}}_{21}={\text{G}}_{2}-{\text{G}}_{1}$$

For example, in the reaction *CO₂→*HOCO, where ‘*’ indicates an adsorption site on the catalyst, ΔG was calculated based on the following equation:

$$\:\varDelta\:\text{G}=\text{G}\left(\text{*}\text{H}\text{O}\text{C}\text{O}\right)-\text{G}\left(\text{*}{\text{C}\text{O}}_{2}\right)-\text{G}\left({\text{H}}^{+}/{\text{e}}^{-}\right)\:\left(2\right)$$

For this equation, the chemical potential of the H⁺/e⁻ pair equals to the half value of the chemical potential of the dihydrogen molecule. Given the standard hydrogen electrode conditions, the G(H⁺/e⁻) equals to 1/2G(H₂).

The Gibbs free energy of intermediates can be calculated by employing the computational hydrogen electrode (CHE) model proposed by Nørskov et al. According to the CHE model, the Gibbs free energy of intermediates can be obtained as following Equation:

$$\:G=E+ZPE-TS\:\left(3\right)$$

Adsorbed intermediates were only taken vibrational entropy (S) into account, and the corresponding function is showed in the (4) formula.

$$\:\text{S}=-\text{R}\sum\:_{\text{i}}\text{ln}\left(1-{\text{e}}^{\raisebox{1ex}{${-\text{h}\text{v}}_{\text{i}}$}\!\left/\:\!\raisebox{-1ex}{$\text{k}\text{T}$}\right.}\right)+\text{R}\sum\:_{\text{i}}\frac{{\text{h}\text{v}}_{\text{i}}}{\text{k}\text{T}}\frac{{\text{e}}^{\raisebox{1ex}{${-\text{h}\text{v}}_{\text{i}}$}\!\left/\:\!\raisebox{-1ex}{$\text{k}\text{T}$}\right.}}{\left(1-{\text{e}}^{\raisebox{1ex}{${-\text{h}\text{v}}_{\text{i}}$}\!\left/\:\!\raisebox{-1ex}{$\text{k}\text{T}$}\right.}\right)}\:\:\left(4\right)$$

In the equation, R = 8.314 J·mol^− 1K^− 1, k_B=1.38·10^− 23 J·K^− 1, h = 6.63·10^− 34 J·s, T = 298.15 K; i is the frequency number; ν_i is the vibrational frequency.

Under normal conditions, the values of entropy of free gas molecules in the system should be obtained from NIST database. However, values of entropy of free gas molecules of NIST database is only appropriate for above 0 ℃. Therefore, for our systems, all the values of entropy of free gas molecules and adsorbed gas molecules were obtained from vibration frequency calculation.

The adsorption free energies of *CO₂ and *H on the pure copper surfaces are defined as

$$\:\varDelta\:{\text{G}}_{\text{a}\text{d}}\left({\text{C}\text{O}}_{2}\right)=\text{G}\left(\text{*}{\text{C}\text{O}}_{2}\right)-\text{G}\left(\text{*}\right)-\text{G}\left({\text{C}\text{O}}_{2}\right)\:\left(5\right)$$

$$\:\varDelta\:{\text{G}}_{\text{a}\text{d}}\left(\text{H}\right)=\text{G}\left(*\text{H}\right)-\text{G}\left(\text{*}\right)-\raisebox{1ex}{$1$}\!\left/\:\!\raisebox{-1ex}{$2$}\right.\text{G}\left({\text{H}}_{2}\right)\:\left(6\right)$$

Chemicals. CuSO₄·5H₂O (99.99% metals basis) and C₆H₅O₇Na₃·2H₂O (A. R. grade) were purchased from Aladdin. KOH (A. R. grade), NaOH (A. R. grade), acetone (A. R. grade) and ethanol (A. R. grade) were provided by Sinopharm Chemical Reagent Co., Ltd, China. Cu-NP was purchased from Simga-Aldrich. D₂O (99.9%), gas diffusion electrode (YLS-30) with 10% PTFE and microporous layer, anion exchange membrane (FAA-3-PK-130) and Nickel foil (purity > 99.8%, thickness 0.5 mm) were obtained from Alfa Aesar China Co., Ltd. Both CO₂ and Ar (Beijing Beiwen Gas Chemical Industry Co., Ltd., research grade) has purities of 99.999% and used as received. Aqueous solutions were prepared with deionized water (Millipore 18.2 MΩ cm).

Preparation of the Cu-NR electrode. Initially, 1.3 mmol of CuSO₄·5H₂O and 0.91 mmol of C₆H₅O₇Na₃·2H₂O were dissolved into the 40 mL of deionized water with stirring for 15 min at room temperature. 5.3 mmol of NaOH was then added into the mixture and further stirred for 2.5 h. The resultant solution was transferred to autoclave and kept at 160°C for 12 h. When the hydrothermal procedure was finished, the obtained product was washed with deionized water and absolute ethanol for three times, alternately. The obtained product was dried at 60°C for 8 h and then annealed at 400°C for 4 h in air⁶². Subsequently, 10 mg of the as-prepared material and 10 µL of Nafion solution (5 wt %) were added into 1 mL of isopropanol and sonicated for 30 min. The as-prepared ink was then drop-coated on the polytetrafluoroethylene (PTFE)-hydrophobized carbon fiber paper (Toray, YLS-30T GDL) and dried. Finally, the as-prepared electrodes underwent electroreduction in 1 M KOH at -0.5 V vs RHE for 10 minutes, resulting in the formation of Cu-NR electrode.

Material characterizations. The morphology was characterized by SEM TESCAN MIRA LMS and TEM Thermo Fisher Talos F200S G2. XRD was conducted using an X-ray diffractometer (PANalytical Empyrean) with a scan speed of 5^o·min^− 1. XPS analysis was conducted on the Thermo Scientific ESCALab 250Xi (USA) using 200 W monochromatic Al Kα radiation. XANES data were collected at 1W2B station in Beijing Synchrotron Radiation Facility (BSRF) operated at 2.5 GeV with a maximum current of 250 mA.

CO ₂ RR experiments. CO₂RR experiments were conducted using a CHI-660e electrochemical workstation equipped with a high current amplifier CHI-680c in an electrochemical flow-cell consisted of a gas chamber, a cathodic chamber and an anodic chamber. The anion exchange membrane (FumasepFAA-3-PK-130) was used to separate the anodic and cathodic chambers, and a Hg/HgO electrode (1 M KOH electrolyte used as the filling solution) and Nickel foil were used as the reference and counter electrodes, respectively. All potentials were converted to the RHE reference scale using the relation: E_RHE=E_Hg/HgO+0.098 + 0.059×pH. In the electrochemical CO₂RR performance tests, KOH was used as the electrolyte, and the electrolyte was circulated through the cathodic and anodic chambers using peristaltic pumps at a rate of 30 mL·min^− 1. The flow rate of CO₂ gas through the gas chamber was controlled to be 30 sccm using a digital gas flow controller. To control the temperature of electrolyte, the electrolyte was placed in refrigerant at different temperatures.

The gaseous product in the electrochemical experiment was collected into a gas bag and analyzed by gas chromatography (GC, HP 4890D). The liquid products were quantified using nuclear magnetic resonance spectroscopy (¹H NMR). ¹H NMR spectra of freshly acquired samples were collected on a Bruker Avance III 400 HD spectrometer. Dimethyl sulfoxide (DMSO) was used as internal standard.

The Faradaic efficiency (FE) of a product can be calculated by:

$$\:\text{F}\text{E}=\frac{\text{n}}{\raisebox{1ex}{$\text{Q}$}\!\left/\:\!\raisebox{-1ex}{$\text{N}\text{F}$}\right.}\times\:100\text{%}（1）$$

where Q is charge (C), F is Faradaic constant (96485 C·mol^− 1), N is the number of transferred electrons to generate desired products, n is the moles of products. For the H₂, CO, CH₄, C₂H₄, HCOOH, CH₃OH, CH₃COOH, CH₃CH₂OH and n-C₃H₇OH, the N is 2, 2, 8, 12, 2, 6, 8, 12 and 18, respectively.

Electrochemical tests. The variation in the partial current density vs applied potential was obtained via stepped potential electrolysis, and Tafel plots were generated from these data. C_dl was measured by the capacitive current associated with double-layer charging from the scan-rate dependence of cyclic voltammogram (CV). The CV tests were performed in a flow-cell with three electrodes, ranging from 0.25 V to 0.15 V vs RHE. 1 M KOH solution was used as the electrolyte. The scan rates were 10, 20, 30, 40, 60, 80, 100 and 120 mV·s^− 1.

CO stripping experiments were carried out using 0.1 M KHCO₃ as the electrolyte in an H-type cell. Prior to the experiment, all catalyst was electrolyzed at − 0.6 V vs RHE for 5 min to fully remove the oxidation species in 0.1 M Ar-saturated KHCO₃. CO was then introduced into the cell and electrolyzed at − 0.8 V vs RHE for 10 min to obtain CO adsorption at the cathode. Ar was then flowed into the electrolyte to remove residual CO. CV curves were then conducted at a scan rate of 50 mV·s^− 1.

In situ characterizations. In situ ATR-SEIRAS experiments were conducted in a modified electrochemical cell that integrated into a BRUKER VERTEX 70v spectrometer cooled by liquid nitrogen. The catalyst was spread on gold-plated silicon prism. A Pt electrode and an Ag/AgCl electrode were used as counter and reference electrodes, respectively. The 1 M KHCO₃ aqueous solution was used as electrolyte at various temperatures.

In situ Raman experiments were conducted in a flow-cell equipped with a quartz window provided by GaossUnion (Tianjin) Photoelectric Technology Company, utilizing a Horiba LabRAM HR Evolution Raman microscope. A 533 nm excitation laser was used and signals were recorded using a 30 s integration and by averaging two scans. The Cu-NR electrode was used as working electrode. A graphite electrode and a Hg/HgO electrode were used as counter and reference electrodes, respectively. The anion exchange membrane was used to separate counter electrode and working electrode. The circulated 1 M KOH aqueous solution was used as electrolyte at various temperatures.

Competing interests

The authors declare no competing interests.

Author Information

Correspondence and requests for materials should be addressed to X.K. ([email protected]) or B.H. ([email protected]).

Author Contributions

S.L.: DFT calculations and MD simulation. S.L., Y.Y., J.Y.: syntheses and characterizations of catalysts. S.L., W.Z., M.Z. and H.Q.: CO₂RR experiments and electrochemical characterizations. Y.W., H.W. and X.T.: In situ ATR-SEIRAS and Raman experiments. J.J.: analysis of XAS data. Y.X., X.S., Q.Z. and M. F.: mechanism analysis. X.K. and B.H.: overall design and direction of the project. S.L., X.K. and B.H.: preparation of the manuscript with help from all authors.

Acknowledgments

The work was supported by the National Natural Science Foundation of China (22273108, 22293015, 22121002), Beijing Natural Science Foundation (2222043), CAS Project for Young Scientists in Basic Research (YSBR-050) and Innovation Program of the IHEP (2023000034), Photon Science Center for Carbon Neutrality. The X-ray absorption spectroscopy measurements were performed at Beijing Synchrotron Radiation Facility (BSRF).

Fan M et al (2023) Cationic-group-functionalized electrocatalysts enable stable acidic CO₂ electrolysis. Nat Catal 6:763–772
Ding J et al (2023) A tin-based tandem electrocatalyst for CO₂ reduction to ethanol with 80% selectivity. Nat Energy 8:1386–1394
Yang P, Gao M (2023) Enrichment of reactants and intermediates for electrocatalytic CO₂ reduction. Chem Soc Rev 52:4343–4380
Wang G et al (2021) Electrocatalysis for CO₂ conversion: From fundamentals to value-added products. Chem Soc Rev 50:4993–5061
Ma W et al (2021) Electrocatalytic reduction of CO₂ and CO to multi-carbon compounds over Cu-based catalysts. Chem Soc Rev 50:12897–12914
Ma Z et al (2022) CO₂ electroreduction to multicarbon products in strongly acidic electrolyte via synergistically modulating the local microenvironment. Nat Commun 13:7596
Jiao J et al (2024) Steering the reaction pathway of CO₂ electroreduction by tuning the coordination number of copper catalysts. J Am Chem Soc 146:15917–15925
Yu X et al (2024) Coverage enhancement accelerates acidic CO₂ electrolysis at ampere-level current with high energy and carbon efficiencies. Nat Commun 15:1711
Zeng M et al (2024) Reaction environment regulation for electrocatalytic CO₂ reduction in acids. Angew Chem Int Ed 63:e202404574
He M et al (2022) Selective enhancement of methane formation in electrochemical CO₂ reduction enabled by a Raman-inactive oxygen-containing species on Cu. ACS Catal 12:6036–6046
Timoshenko J et al (2022) Steering the structure and selectivity of CO₂ electroreduction catalysts by potential pulses. Nat Catal 5:259–267
Shi C, Hansen HA, Lausche AC, Nørskov JK (2014) Trends in electrochemical CO₂ reduction activity for open and close-packed metal surfaces. Phys Chem Chem Phys 16:4720–4727
Wang Y et al (2024) Strong hydrogen-bonded interfacial water inhibiting hydrogen evolution kinetics to promote electrochemical CO₂ reduction to C₂₊. ACS Catal 14:3457–3465
Feng J et al (2023) Improving CO₂-to-C₂₊ product electroreduction efficiency via atomic Lanthanide dopant-induced tensile-strained CuO_x catalysts. J Am Chem Soc 145:9857–9866
Ge W et al (2022) Dynamically formed surfactant assembly at the electrified electrode–electrolyte interface boosting CO₂ electroreduction. J Am Chem Soc 144:6613–6622
Deng T et al (2024) Polymer modification strategy to modulate reaction microenvironment for enhanced CO₂ electroreduction to ethylene. Angew Chem Int Ed 63:e202313796
Kaneco S, Iiba K, Ohta K, Mizuno T, Saji A (1998) Electrochemical reduction of CO₂ on Au in KOH + methanol at low temperature. J Electroanal Chem 441:215–220
Kaneco S, Iiba K, Ohta K, Mizuno T, Saji A (1998) Electrochemical reduction of CO₂ at an Ag electrode in KOH-methanol at low temperature. Electrochim Acta 44:573–578
Li K et al (2023) Superhydrophobicity-enabled efficient Electrocatalytic CO₂ reduction at a high temperature. ACS Catal 13:9346–9351
Ouyang Y et al (2023) Selectivity of electrochemical CO₂ reduction toward ethanol and ethylene: the key role of surface-active hydrogen. ACS Catal 13:15448–15456
Cheng T, Xiao H, Goddard WA (2017) Full atomistic reaction mechanism with kinetics for CO reduction on Cu(100) from ab initio molecular dynamics free-energy calculations at 298 K. Proc. Natl. Acad. Sci. U.S.A. 114, 1795–1800
Calle-Vallejo F, Koper MTM (2013) Theoretical considerations on the electroreduction of CO to C₂ species on Cu(100) electrodes. Angew Chem Int Ed 52:7282–7285
Wan M et al (2023) Enhanced CO₂ reactive capture and conversion using aminothiolate ligand-metal interface. J Am Chem Soc 145:26038–26051
Lin L et al (2021) Temperature-dependent CO₂ electroreduction over Fe-N-C and Ni-N-C single-atom catalysts. Angew Chem Int Ed 60:26582–26586
Vos RE et al (2023) How temperature affects the selectivity of the electrochemical CO₂ reduction on copper. ACS Catal 13:8080–8091
Jo SW, Kim JY, Lee MW, Kim Y, Ahn HS (2023) Highly selective reduction of CO₂ to methane induced by subzero depression of the electrode surface temperature. ACS Catal 13:5122–5126
Vos RE, Koper MTM (2022) The effect of temperature on the cation-promoted electrochemical CO₂ reduction on gold. ChemElectroChem 9, e202200239
Ahn ST, Abu-Baker I, Palmore GTR (2017) Electroreduction of CO₂ on polycrystalline copper: Effect of temperature on product selectivity. Catal Today 288:24–29
Vavra J et al (2024) Solution-based Cu⁺ transient species mediate the reconstruction of copper electrocatalysts for CO₂ reduction. Nat Catal 7:89–97
Zhao Q, Martirez JMP, Carter EA (2021) Revisiting understanding of electrochemical CO₂ reduction on Cu(111): Competing proton-coupled electron transfer reaction mechanisms revealed by embedded correlated wavefunction theory. J Am Chem Soc 143:6152–6164
Zheng M et al (2022) Electrocatalytic CO₂-to-C₂₊ with ampere-level current on heteroatom-engineered copper via tuning *CO intermediate coverage. J Am Chem Soc 144:14936–14944
Xie Y et al (2022) High carbon utilization in CO₂ reduction to multi-carbon products in acidic media. Nat Catal 5:564–570
Yang X et al (2024) Cation-induced interfacial hydrophobic microenvironment promotes the C–C coupling in electrochemical CO₂ reduction. J Am Chem Soc 146:5532–5542
Gorthy S et al (2023) Theoretical insights into the effects of KOH concentration and the role of OH^– in the electrocatalytic reduction of CO₂ on Au. ACS Catal 13:12924–12940
Schaefer J, Backus EHG, Nagata Y, Bonn M (2016) Both inter- and intramolecular coupling of O–H groups determine the vibrational response of the water/air interface. J Phys Chem Lett 7:4591–4595
Wang P et al (2022) Boosting electrocatalytic CO₂-to-ethanol production via asymmetric C-C coupling. Nat Commun 13:3754
Sun W et al (2022) V-Doped Cu₂Se hierarchical nanotubes enabling flow-cell CO₂ electroreduction to ethanol with high efficiency and selectivity. Angew Chem Int Ed 34:2207691
Wu Z et al (2022) Identification of Cu(100)/Cu(111) interfaces as superior active sites for CO dimerization during CO₂ electroreduction. J Am Chem Soc 144:259–269
Nguyen TN et al (2022) Catalyst regeneration via chemical oxidation enables long-term electrochemical carbon dioxide reduction. J Am Chem Soc 144:13254–13265
Cao Y et al (2023) Surface hydroxide promotes CO₂ electrolysis to ethylene in acidic conditions. Nat Commun 14:2387
Amirbeigiarab R et al (2023) Atomic-scale surface restructuring of copper electrodes under CO₂ electroreduction conditions. Nat Catal 6:837–846
Li H, Wei P, Gao D, Wang G (2022) In situ Raman spectroscopy studies for electrochemical CO₂ reduction over Cu catalysts. Curr Opin Green Sustain Chem 34:100589
Cui Y et al (2023) N_xC-induced switching of methane and ethylene products’ selectivity from CO₂ electroreduction over Cu catalyst. ACS Catal 13:11625–11633
Wei Z et al (2019) Optimizing alkyne hydrogenation performance of Pd on carbon in situ decorated with oxygen-deficient TiO₂ by integrating the reaction and diffusion. ACS Catal 9:10656–10667
Li J et al (2021) Electrokinetic and in situ spectroscopic investigations of CO electrochemical reduction on copper. Nat Commun 12:3264
Kresse G, Furthmüller J (1996) Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set. Comput Mater Sci 6:15–50
Perdew JP et al (1992) Atoms, molecules, solids, and surfaces: Applications of the generalized gradient approximation for exchange and correlation. Phys Rev B 46:6671–6687
Perdew JP, Wang Y (1992) Accurate and simple analytic representation of the electron-gas correlation energy. Phys Rev B 45:13244–13249
Blöchl PE (1994) Projector augmented-wave method. Phys Rev B 50:17953–17979
Monkhorst HJ, Pack JD (1976) Special points for Brillouin-zone integrations. Phys Rev B 13:5188–5192
Grimme S, Antony J, Ehrlich S, Krieg H (2010) A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J Chem Phys 132:154104
Mathew K, Sundararaman R, Letchworth-Weaver K, Arias TA, Hennig RG (2014) Implicit solvation model for density-functional study of nanocrystal surfaces and reaction pathways. J Chem Phys 140:084106
Savin A, Nesper R, Wengert S, Fässler TF (1997) ELF: The electron localization function. Angew Chem Int Ed 36:1808–1832
Otero-de-la-Roza A, Johnson ER, Luaña V (2014) Critic2: A program for real-space analysis of quantum chemical interactions in solids. Comput Phys Commun 185:1007–1018
Dronskowski R, Bloechl PE (1993) Crystal orbital Hamilton populations (COHP): Energy-resolved visualization of chemical bonding in solids based on density-functional calculations. J Phys Chem 97:8617–8624
Nelson R et al (2020) Local orbital projections, atomic charges, and chemical-bonding analysis from projector-augmented-wave-based density-functional theory. J Comput Chem 41:1931–1940
Henkelman GA, Uberuaga BP, Jónsson H (2000) A climbing image nudged elastic band method for finding saddle points and minimum energy paths. J Chem Phys 113:9901–9904
Henkelman G, Jónsson H (1999) A dimer method for finding saddle points on high dimensional potential surfaces using only first derivatives. J Chem Phys 111:7010–7022
Bussi G, Donadio D, Parrinello M (2007) Canonical sampling through velocity rescaling. J Chem Phys 126:014101
Nosé S (1984) A unified formulation of the constant temperature molecular dynamics methods. J Chem Phys 81:511–519
Bannwarth C et al (2021) Extended tight-binding quantum chemistry methods. WIREs Comput Mol Sci 11:e1493
Chen X et al (2021) Room-temperature NO₂ sensing properties and mechanism of CuO nanorods with Au functionalization. Sens Actuat B-Chem 328:129070

There is NO Competing Interest.

SUPPLEMENTARYINFORMATION.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Temperature-Dependent Pathways in Carbon Dioxide Electroreduction

Status:

Version 1

Abstract

Figures

Introduction

Results

Conclusion

Methods

Declarations

Competing interests

Author Information

Author Contributions

Acknowledgments

References

Additional Declarations

Supplementary Files

Status:

Version 1