Assessing elevated pressure impact on photoelectrochemical water splitting via multiphysics modeling

Abstract

Photoelectrochemical (PEC) water splitting is a promising approach for sustainable hydrogen production. Previous studies have focused on devices operated at atmospheric pressure, although most applications require hydrogen delivered at elevated pressure. Here, we address this critical gap by investigating the implications of operating PEC water splitting directly at elevated pressure. We evaluate the benefits and penalties associated with elevated pressure operation by developing a multiphysics model that incorporates empirical data and direct experimental observations. Our analysis reveals that the operating pressure influences bubble characteristics, product gas crossover, bubble-induced optical losses, and concentration overpotential, which are crucial for the overall device performance. We identify an optimum pressure range of 6-8 bar for minimizing losses and achieving efficient PEC water splitting. This finding provides valuable insights for the design and practical implementation of PEC water splitting devices, and the approach can be extended to other gas-producing (photo)electrochemical systems. Overall, our study demonstrates the importance of elevated pressure in PEC water splitting, enhancing the efficiency and applicability of green hydrogen generation.

Fig. 1. Comparison of bubble properties obtained from calculation and measured values in the literature.

a O₂ bubble diameter ($D_{{\mathrm{O}}_2}$) and (b) O₂ number density ($N_{{\mathrm{O}}_2}$) as calculated from Eq. 1 vs. measured values in the literature. The same plots for H₂ bubble diameter and number density are shown in (c) and (d). The goodness of the fits is evidenced from the $R^2$ values. We attribute the relatively low R² value for H₂ bubble densities to the rather dispersed datasets; Note that the data for oxygen is distributed across two orders of magnitude, while the hydrogen data spans three orders of magnitude.

Fig. 2. Pressure-dependent bubble characteristics.

a O₂ bubble formation efficiency (${\eta }_{\mathrm{bub}}$) vs. operating pressure. Comparison of the O₂ bubble formation efficiency as determined from our experiments and calculated using Eq. 1. The error bars represent ±20% of the measured values, while the values are the average of 500 image frames. (b–d) Calculated pressure-dependent O₂ and H₂ bubble characteristics. b Average bubble diameters (D_bub), (c) number density of bubbles (N_bub), and (d) derived bubble formation efficiency (η_bub) as a function of the operating pressure. The current density and the electrolyte velocity are kept constant at 10 mA cm⁻² and 3 cm s⁻¹, respectively.

Fig. 3. Bubble plume distribution at different operating pressures and device tilt angles.

Colormaps of the simulated gas volume fraction (O₂) generated from the anode (left boundary of the domain) in a device tilted at θ = 90° (a–e) and 30° (f–j) and operated at various pressures. (a) and (f) are at 1 bar, (b) and (g) are at 2 bar, (c) and (h) are at 4 bar, (d) and (i) are at 6 bar, (e) and (j) are at 10 bar. The current density is 10 mA cm^-2, and the average inlet velocity of the electrolyte is set as 3 cm s^-1.

Fig. 4. Multiphysics simulation results.

a Simulated molar flux of dissolved products and O₂ bubbles at the device outlet under operating pressures of 1 bar (solid) and 10 bar (dashed). b Simulated O₂ bubble molar flux at the device outlet under various operating pressures. c Simulated bubble-plume-induced optical loss (f_{opt. loss}) as a function of the operating pressure. The average velocity at electrolyte inlet is set as 3 cm s^-1, current density is 10 mA cm^-2, device tilt angle is ${30}^{\circ }$ in (a) and (b), while $\theta ={90}^{\circ }$ in (c).

Fig. 5. Influence of operating pressure to the simulated local pH distribution and concentration overpotentials.

a Colormaps of pH in the cell at different operating pressures, i.e., 1 bar vs. 10 bar. b pH gradient near the electrodes at different operating pressures, c pH gradient induced voltage loss (V_{pH grad}) vs. pressure. The current density is 10 mA cm^-2, the inlet electrolyte velocity is 3 cm s^-1, the device tilt angle is ${90}^{\circ }$. The electrolyte is 2 M KP_i (pH 7.2).

Fig. 6. Assessment of the benefits and penalties of operating PEC water splitting cells at elevated pressure.

a Schematic illustration of the two scenarios considered in our analysis. The two branches in the image represent different options for performing PEC water splitting, either at (i) atmospheric pressure or ii) at elevated pressure $p$. The collected hydrogen from both operation modes are fed into mechanical compressors in order to reach the pressure of 700 bar at the final storage tank. b Energy losses associated with operating PEC water splitting cells at various pressures. A 1 m² PEC water splitting device generating 1 kg of H₂ is considered. The gray horizontal dashed line indicates the lower heating value (LHV) of 1 kg of H₂ for a comparison. In panel (b), $f_{\mathrm{swc}}$ is the specific work required for the compression of H₂ from collected to 700 bar, $f_{\mathrm{col}}$ is the crossover loss induced by the gaseous products, $f_{\mathrm{cop}}$ denotes the pH gradient induced voltage loss, $f_{\mathrm{opl}}$ is the optical loss induced by bubble scattering effect, $f_{\mathrm{tcv}}$ is the thermodynamic cell voltage loss associated with pressure elevation.