Electrocatalysis of Oxygen Reduction Reaction for Electrodeposition of ZnO Thin Films

Abstract

Introduction

Cathodic electrodeposition of zinc oxide (ZnO) thin films employing oxygen reduction reaction (ORR) has been widely studied [1]. The process directly yields highly crystallized thin films at low temperatures. We have also achieved hybrid thin films with organic dye molecules that act as structure directing agents (SDAs) to strongly modify the crystal growth [2]. Highly porous sponge-like crystals were obtained in the presence of eosinY (EY), which exhibited a high performance as a photoelectrode for dye-sensitized solar cells (DSSCs). The added EY was found to catalyse ORR to promote formation of ZnO, rather than hindering it.

The overall electrochemical stoichiometry for formation of ZnO can simply be written as,

Zn²⁺ + 1⁄2 O₂ + 2e^- → ZnO     (1)

The precipitation is triggered by hydroxide formation by ORR as follows,

O₂ + 2H₂O + 4e^- → 4OH^-      (2)

O₂+2H₂O+2e^-→H₂O₂+2OH^-      (3)

They are kinetically limited and should be dependent on the electrodes to be used as substrates. However, when the electrodes are coated with ZnO, the electrode kinetics should become dependent on the surface of the deposit. It is important to give a full understanding how the ORR is influenced in the presence of Zn²⁺ ion and also by the presence of catalytic molecules such as EY. We have carried out hydrodynamic electroanalysis employing rotating ring-disk electrode (RRDE) that allows monitoring of formation of H₂O₂.

Experimental

Steady-state current was monitored at a Pt-Pt RRDE (disk diameter = 5 mm, ω = 1000 rpm) set for -1.0 and +0.6 V vs. Ag/AgCl for the disk and ring electrodes, at which diffusion limited reduction of O₂ and oxidation of H₂O₂ are expected, respectively. A 0.1 M KCl aqueous solution (70ºC) was saturated with O₂, to which ZnCl₂ was injected during the measurement to achieve its concentration of 5 mM. The collection efficiency at the ring electrode (N = 0.264) was determined in a K₃Fe(CN)₆ solution. The proportion of disk (I_D) and ring (I_R) current to their sum can yield number of electrons transferred to a single O₂ molecule (n) and the percentage of H₂O₂ production as Eqs. 4 and 5, respectively [3].

n = 4I_D / (I_D + I_R/N)     (4)

H₂O₂ [%] = (2 I_R/N)/(I_D + I_R/N ) × 100        (5)

Results and discation

Chronoamperograms for I_D and I_R, as well as n and H₂O₂ [%] calculated according to Eqs. 4 and 5 are shown in Fig. 1. Although an excessive current attributable to surface impurities and/or hydrogen evolution is observed in the beginning of the electrolysis, the steady-state current of ca. 0.85 mA measured at the disk almost equals to the diffusion-limited current for full four-electron reduction of oxygen, calculated from Levich equation,

i_d=0.62nFAC^*D^2⁄3ν^-1⁄6ω^1⁄2     (6)

where i_d is the diffusion limited current, n is the number of electrons (4), F is Faraday constant (96,485 C mol^-1), A is the surface area (0.196 cm²), C* is the O₂ concentration at 343 K (5.67 × 10^-7 mol cm^-3), D is the diffusion coefficient of O₂ at 343 K (4.32 × 10^-5 cm² s^-1), ν is the kinematic viscosity (4.14 × 10^-3 cm² s^-1), and ω is the angular speed of rotation (1000 rpm). In the absence of ZnCl₂, almost no current is measured at the ring electrode, so that n calculated from Eq. 4 also approximately stays being 4, indicating diffusion limited complete reduction of oxygen under this condition.

The addition of ZnCl₂ starts off the electrodeposition of ZnO on the electrode surface associated with a decrease of I_D. The steady-state current is halved to ca. 0.4 mA. The decrease of I_D can be interpreted as a slowdown of ORR kinetics by the coverage with ZnO. Incomplete reduction by switching of ORR as that of Eq. 2 to Eq. 3 can half the current as well. I_R actually jumps up on addition of ZnCl₂, indicating formation of H₂O₂. But its magnitude is small to result n = 3.2 from Eq. 6, unmatched with complete change of ORR to two electron reduction as Eq. 3. Although H₂O₂ formation in the presence of Zn²⁺ is evident, the discrepancy of number of electrons expected from halved I_D (2) and I_R (3.2) can be caused by the loss of H₂O₂ before its detection at the ring electrode. In fact, vigorous evolution of gas bubbles (most likely O₂) was observed when the Pt electrode was simply soaked to H₂O₂ solution due to its disproportionation decomposition. We need to quantify H₂O₂ by some other means to calibrate the N value for correct analysis of the ring current.

Fig.1. Chronoamperograms of 5 mM ZnCl₂ added during electrolysis an O₂-saturated KCl solution at disk potential = -1.0 V vs. Ag/AgCl, ring potential = 0.6 V vs. Ag/AgCl (above), and the percentage of H₂O₂ formation and the electrons transfer number (below).