Page 47 - Multifunctional Photocatalytic Materials for Energy

36 Multifunctional Photocatalytic Materials for Energy

3.3.4 BiVO 4 -based materials

In spite of the attention dedicated to various photoelectrode materials (encompassing
TiO 2 , Fe 2 O 3 , WO 3 , ZnO) for PEC water oxidation during the past few decades, none
of these binary oxides fulfills all the requirements necessary for an eventual tech-
nology implementation. As a consequence, ternary semiconductor oxide photoanodes
have also attracted great interest [15]. Most of these systems, with a general for-
mula ABO 4 , possess Vis light activity, stability in aqueous electrolytes, and low cost.
Among them, monoclinic bismuth vanadate (BiVO 4 , E G = 2.4 eV) [15,24], an n-type
semiconductor, is particularly promising since it can yield a maximum theoretical
−2
photocurrent of ≈7.5 mA × cm at 1.23 V versus RHE [9,23,65,119]. In fact, it can
absorb Vis light and possesses band edges appropriately positioned for OER (see
Fig. 3.2) [10,24]. Nevertheless, rapid charge carrier recombination processes and in-
efficient charge transport [10,65,112] are responsible for photocurrent density values
−2
lower than 1.0 mA × cm at 1.23 V versus RHE. This problem, along with the high
onset potential, has prompted morphology tailoring or functionalization/doping (for
instance, with Cr, Mo, Nb, …) of this material in order to improve the efficiency of the
OER reaction [12,15,24]. So far, the slow hole transfer to the solution remains a major
bottleneck, requiring BiVO 4 modification with OECs, such as CoPi [23,65]. A re-
cent work looked at the liquid phase preparation of BiVO 4 films [65], whose porosity
was tailored by a proper choice of ex-situ annealing treatments. SEM micrographs
evidenced the formation of compact structures for the “dense” BiVO 4 (Fig. 3.10A and
B), whereas the corresponding “porous” system presented a network of interconnected
nanoaggregates (Fig. 3.10C and D). In both cases, XRD analyses revealed the sole
presence of monoclinic BiVO 4 .
Film photoactivity was evaluated in PEC water splitting, and the photocurrent/po-
tential curves obtained under chopped illumination (Fig. 3.10E) revealed in both cases
a certain extent of charge recombination, as indicated by the current spikes upon light
ignition, followed by an exponential decay. Under continuous illumination, the behav-
ior of the two systems was very similar up to ≈1 V versus RHE, and for higher po-
tentials, dense BiVO 4 performed better (Fig. 3.10E, inset). This difference, amplified
upon chopped illumination (Fig. 3.10E), was mainly related to a higher photocarrier
recombination rate in porous BiVO 4 , a phenomenon attributed to the higher content of
grain boundary defects. Additional analyses showed that the charge transfer kinetics in
the dense material is about 3-fold faster than in the porous film [65].
In an attempt to enhance the photoactivity of dense BiVO 4 systems, functional-
ization with CoPi was carried out and yielded a ≈6-fold photocurrent increase at
1.23 V, confirming the CoPi role as an effective water oxidation catalyst (Fig. 3.10F).
Nevertheless, the above-mentioned exponential decay immediately after illumination
turn-on was still observed [10], suggesting that the recombination phenomena were not
completely suppressed. Hence, the development of alternative routes to BiVO 4 films
with improved PEC performances remains an open challenge. In this context, other
efforts have been dedicated to the preparation and chemical modification of BiVO 4 sys-
−2
tems [9,65,120,121], and some of these studies yielded J values up to 2.16 mA × cm
even in natural seawater [30], an important result for real-world utilization. Other