Abstract

Organic heterojunction diodes have been formed by depositing a Langmuir–Blodgett (LB) multilayer of 16-mer polyaniline and stearic acid (1∶1) onto films of poly(3-methylthiophene) electropolymerised onto gold electrodes. The diode was completed by depositing small, circular gold counterelectrodes onto the LB film. The devices thus formed show weak rectification. In forward bias it is argued that all the applied field appears across the LB layer. At high applied fields the resulting current follows the Richardson–Schottky equation indicative of electrode limitation, though it is not clear whether the rate limiting step is electron emission from the gold electrode or hole emission from the polymer into the LB film. In reverse bias it is argued that part of the applied voltage appears across the depletion region formed at the interface between the polymer and the LB film resulting in a smaller reverse current for a given applied voltage. The frequency dependence of capacitance and loss for the diode structure is reminiscent of a Maxwell–Wagner dispersion normally associated with a two-layer structure. However, reasonable agreement between experimental data and theoretical modelling is only possible if a third layer, i.e. the depletion region at the interface of the two organic films, is included in the model. This is confirmed by the reverse-bias capacitance–voltage curve which suggests that the device behaves as a Metal–Insulator–Semiconductor (MIS) structure driven into depletion/inversion. In forward bias, when the LB/polymer interface is in accumulation, the capacitance again decreases as the bias voltage increases. While this effect could be ascribed to the presence of a second depletion layer in the device, it is argued that this behaviour arises from the increase in conductance of the LB multilayer shunting the LB layer capacitance.