Insight into charge carriers recombination in doped titania films using photocurrent measurements

The characterization of the electronic properties of titanium dioxide is a relevant issue in the design of innovative semiconducting systems for catalysis and applications in energetics. Doping can achieve a shrinking of the band gap, but can also create sites in the lattice that can accelerate the rate of electron-hole recombination or, on the other hand, they can trap charge carriers, thus enhancing their lifetime [1]. Within all the electrochemical techniques, chronoamperometric measurements can be used to obtain photocurrent transients, from which the charge recombination rate can be estimated with the final evaluation of the transient time constant (τ) [2]. Photocurrent measurements are well documented, but several different experimental conditions are present in scientific literature, leading to possible difficulties in making comparisons about similar samples. In fact it is possible to work either in absence or in presence of a hole acceptor like hydroxide [3] or chloride [4] species, oxalate [5], methanol [6], and so on. Thus, in a solution of an inert electrolyte as sodium perchlorate in absence of any hole acceptor, the oxidation of oxygen present in water to molecular oxygen can occur. Oxygen produced in this way can affect recombination rate preventing electrons in the conduction band of semiconductor to reach the supporting contact [5]. On the other hand, the presence of a hole acceptor may affect the photocurrent intensity and transients shape because of adsorption phenomena on the semiconductor surface. Thus, a crucial role can be played by some specific semiconductor properties like specific surface area, surface roughness and morphology of the photoanode, which, in turn, is strongly affected by the doping level. In this work we investigate the influence of the atomic ratio of alternatively nitrogen or praseodymium dopant [7] on recombination rates using a spin coated titanium dioxide photoanode in a photoelectrochemical cell, in order to register photocurrent transients [8]. In particular we focus on the influence of the electrolyte type and concentration in affecting reproducibility of this type of analysis. The irradiation sources are both UV and visible lamps. A comparison is made for selected samples by using both types of illumination. Finally we find that, even though a system of this type is more theoretically correct then one in which molecular oxygen can be produced, the inert electrolyte system seems to be better suited for a relative evaluation of recombination rate in samples with different doping level.