A theoretical analysis of optical absorption limits and performance of tandem devices and series interconnected architectures for solar hydrogen production

Photo-driven catalytic (PDC) water splitting, using either photoelectrochemical cells (PEC-cells), PV-electrolyzers, or some hybrid system in-between, has attracted a lot of attention. In single-cell device architectures for solar hydrogen production, based on single band gap photoabsorbers, there is a fundamental efficiency problem originating from the energy distribution of the solar spectrum and the thermodynamic and kinetic requirements for splitting water. The minimum band gap for a single-junction device in order to withhold unbiased overall water splitting is considered to be at least 2.0 eV. This is far from the 1.35 eV which is the optimal band gap of a semiconductor for maximum power conversion of light in the solar spectrum. This discrepancy has been termed as the solar spectrum mismatch problem (the SSM-problem). The standard solution to this problem is to construct tandem devices, whereas an alternative is to interconnect several one band gap cells in series, side by side. Both approaches enable the use of low energy photons in the solar spectrum while still providing a sufficiently high photopotential for driving the full reaction, without seriously compromising with the area efficiency. In this paper, the tandem and serial architectures for handling the SSM-problem are analyzed and compared. The analysis is focused towards differences in the limits of optical absorption, the optimal number of optical. absorbers, and their corresponding band gaps. Taking losses due to charge carrier separation and catalysis into account, the maximum STH-efficiency for a series interconnected solar splitting device was found to be 24.6%, compared to 32.0% for an optimum tandem device at 1 Sun (air mass 1.5, 1000 W m(-2)). This can be compared with the maximum efficiency of 18.0% for an ideal single band gap photoabsorber in single junction device. The analysis shows that the maximum STH efficiency limits for series interconnected architectures for unassisted solar water splitting are not particularly far behind the more commonly studied tandem devices. They could then be an interesting alternative given the simplicity and versatility of series interconnected device architectures. The analysis also compares how tandem devices and series interconnected devices can differ in terms of charge carrier separation, charge carrier transport, catalysis, overall efficiency, device architecture, and expected cost. (C) 2015 Elsevier B.V. All rights reserved.