Effective Visible Light Exploitation by Copper Molybdo-tungstate Photoanodes

Abstract

The need for stable oxide-based semiconductors with a narrow band gap, able to maximize the exploitation of the visible light portion of the solar spectrum, is a challenging issue for photoelectrocatalytic (PEC) applications. In the present work, CuW_1-x Mo _x O₄ (E _g = 2.0 eV for x = 0.5), which exhibits a significantly reduced optical band gap E _g compared with isostructural CuWO₄ (E _g = 2.3 eV), was investigated as a photoactive material for the preparation of photoanodes. CuW_0.5Mo_0.5O₄ electrodes with different thicknesses (80-530 nm), prepared by a simple solution-based process in the form of multilayer films, effectively exhibit visible light photoactivity up to 650 nm (i.e., extended compared with CuWO₄ photoanodes prepared by the same way). Furthermore, the systematic investigation on the effects on photoactivity of the CuW_0.5Mo_0.5O₄ layer thickness evidenced that long-wavelength photons can better be exploited by thicker electrodes. PEC measurements in the presence of NaNO₂, acting as a suitable hole scavenger ensuring enhanced photocurrent generation compared with that of water oxidation while minimizing dark currents, allowed us to elucidate the role that molybdenum incorporation plays on the charge separation efficiency in the bulk and on the charge injection efficiency at the photoanode surface. The adopted Mo for W substitution increases the visible light photoactivity of copper tungstate toward improved exploitation and storage of visible light into chemical energy via photoelectrocatalysis.

Figure 1

(a) Absorption spectra of multilayer (1L–5L) CuW_0.5Mo_0.5O₄ photoelectrodes. Inset: picture of the photoelectrodes. (b) X-ray powder diffraction (XRPD) patterns of 1L, 2L, and 3L CuW_0.5Mo_0.5O₄ electrodes, in the 28–40 degrees 2θ range. The asterisks mark the reflections typical of wolframite.

Figure 2

Top-view SEM images of (a) 1L, (b) 2L, and (c) 3L CuW_0.5Mo_0.5O₄ electrodes. (d) Absorption at 430 nm of multilayer photoanodes vs their thickness. SEM images recorded after PEC tests of (e) 2L and (f) 3L CuW_0.5Mo_0.5O₄ electrodes. The scalebar is 200 nm.

Figure 3

Linear sweep voltammetry (LSV) of (a) 1L–5L CuW_0.5Mo_0.5O₄ electrodes and CuWO₄ electrode in back configuration and (b) 1L and 3L CuW_0.5Mo_0.5O₄ electrodes in back and front configuration; AM 1.5 G solar simulated irradiation, scan rate of 10 mV s^–1. The current in the absence of irradiation is also shown in panel (a) (black dashed line). Incident photon to current efficiency (IPCE) of (c) CuWO₄ monolayer and 1L–5L CuW_0.5Mo_0.5O₄ electrodes at 1.23 V vs RHE and (d) the 3L electrode at different applied potentials.

Figure 4

Chopped chronoamperometry at 1.23 V versus RHE of the 1L–5L CuW_0.5Mo_0.5O₄ electrodes and of the CuWO₄ electrode under irradiation at 650 nm (top panel), 550 nm (middle panel), and 450 nm (bottom panel).

Figure 5

LSV curves recorded with a monolayer CuW_0.5Mo_0.5O₄ electrode in contact with a 0.1 M K₃BO₃ solution, 0.1 M K₃BO₃ solutions containing different hole scavengers at a 0.1 M concentration or 0.1 M Na₂SO₃, under dark (dashed lines) or back side AM 1.5 G irradiation conditions (continuous lines). Scan rate of 10 mV s^–1. All solutions were at pH 9. The photocurrent onset potential was calculated by extrapolation of each LSV line (black dashed lines).

Figure 6

Incident photon to current efficiency (IPCE) of 1L and 5L CuW_0.5Mo_0.5O₄ electrodes at 1.23 V vs RHE in K₃BO₃ both in the absence (void symbols) and in the presence (full symbols) of NaNO₂.

Figure 7

Charge injection efficiency η_inj (continuous lines) and charge separation efficiency η_sep (dashed lines) calculated for the monolayer CuW_0.5Mo_0.5O₄ (fuchsia) and CuWO₄ (light blue) electrodes vs the applied potential.