Charge Generation and Transport in p-doped Amorphous Organic Semiconductors

Abstract

Electrical doping has been one of the most important and fundamental technologies in organic semiconductors to control the major carrier type and electrical properties of the materials. It can enhance the conductivity of bulk organic layers by increasing the charge density, and also reduce the contact resistance between electrodes and organic layers. Therefore, electrical doping has been widely used in organic electronic devices, e.g. organic light emitting diodes (OLEDs), organic photovoltaic cells (OPVs), and organic thin film transistors (OTFTs) to remove the parasitic resistance in devices and achieve the high efficiency. Despite decades of research on electrical doping and its utilization in organic electronic devices, the fundamental understanding of the doping effect in organic semiconductors is still far from the complete. One is the effect of doping on charge mobility in organic semiconductors. The variation of the charge mobility upon doping in organic semiconductors has been controversial. The main reason for these controversial trends is related to the dependence of the charge mobility on the charge density in the amorphous organic semiconductors. Large number of experimental results were mostly observed from the analysis of current density-voltage (J-V) characteristics of field effect transistors (FETs) or the space-charge-limited-current density (SCLC) in single carrier devices. One problem of these measurements is that the total charge density in the organic semiconductors during the measurements is dominated by the injected carriers via the electrodes over the generated carriers by doping. Therefore, it is difficult to clarify whether the doping effect on charge mobility is originated by doping induced carriers or by injected carriers. To exclude the effect of injected charge density on charge mobility in doped organic semiconductors, we suggest the novel method of analysis in Ohmic current region. Since the charge density induced by doping is much higher than the injected charge density in Ohmic current region, the total charge density in the doped organic layer must be constant to charge density induced by doping. Analysis in Ohmic current region is possible to investigate the charge mobility of doping organic semiconductor with excluding the effect of injected charge density. In order to obtain Ohmic current, Ohmic contact between electrode and organic semiconductor layer should be achieved. Firstly, the formation and mechanism of Ohmic contact without the contact resistance between the electrode and organic semiconductor layer was investigated (Chapter 2). The hole injection efficiency was closed to 100% at the interface of ITO/N,Nʹ-di(naphthalen-1-yl)-N,Nʹ-diphenylbenzidine (NPB) with inserting thin ReO3¬ layer, indicating that a Ohmic contact without contact resistance was formed in the device using ReO3 interlayer. This is the highest hole injection efficiency reported to data for an ITO/NPB contact. In contrast, 1,4,5,8,9,11-hexaazatripheylene hexacarbonitrile (HAT-CN) and MoO3 interlayers gave much lower hole injection efficiencies, 2.3% and 8.9%. The hole injection barrier with ReO3 and MoO3 interfacial layers resulted in similar value of ~0.4 eV to NPB, indicating that the Fermi level is pinned to the polaron energy level of NPB. Even though the hole injection barriers with ReO3 and MoO3 interfacial layers were almost same, the hole injection efficiency with ReO3 interlayer was superior to that with MoO3 interlayer. To resolve this contradiction, the degree of charge generation near the interface of NPB/interlayer was investigated by X-ray photoelectron spectroscopy (XPS) and photoluminescence (PL) quenching measurements. Among the interlayers, ReO3 gave the highest degree of charge generation, and this result was consistent with the trend of hole injection efficiency. We conclude that the formation of Ohmic contact without contact resistance is achieved not only as a result of low hole injection barrier, but also because of the high degree of charge generation at the interface. Secondly, we investigated the effect of doping on hole mobility in ReO3 doped 4,4',4"-tris(N-(2-naphthyl)-N-phenyl-amino)-triphenylamine (2-TNATA). (Chapter 3) We focused on the analysis in Ohmic current region with using the hole-only device with having Mp++pp++M structure, where M is the metal electrode and p++ is the heavily doped organic layer, p is the moderately doped organic layer. The Ohmic contact between the electrode and the doped organic layer was achieved by inserting thin, heavily doped organic layers for both contacts. The J-V characteristics were perfectly followed by Ohmic current and conductivities of doped organic layer were extracted from the slope of linear J-V characteristics. The conductivities increased with increasing the doping concentration and were in the range of 4~9×10−8 S/cm. The hole densities induced by doping were separately determined by Schottky-Mott analysis in metal-insulator-semiconductor (MIS) device, and linearly proportional to the doping concentration. Finally the hole mobilities of the doped organic layers were analyzed from the measured conductivities and hole density within the Drude model, and reduced as the doping concentration increased with values of order 10−6~10−7 cm2/Vs, one to two orders of magnitude lower than that in intrinsic organic films. The reduction of hole mobility was interpreted by the broadening of the Gaussian density-of-states distribution of doped organic layers, since the negatively ionized dopants can be act to Coulomb traps for mobile carriers. Furthermore, this broadening of the Gaussian density-of-states distribution was confirmed by the increasing activation energy with increasing doping concentration. Finally, we analyzed the hole mobilities of various p−doped organic semiconductors possessing different charge generation efficiencies and energetic disorder parameters (Chapter 4). The hole mobilities in the various p−doped organic semiconductors were reduced by orders of magnitude and converged to 10−7~10−6 cm2/Vs at the doping concentration of 5 mol% for all the materials, even though the pristine organic films possess orders of magnitude different mobilities from 10−5 to 10−2 cm2/Vs. Interestingly, further increase of doping concentration above 5 mol% increased or decreased the mobility depending on the energetic disorder parameters of the host materials. These phenomena were understood based on the Coulomb trap depth of the ionized dopants and the relative separation of the Gaussian density-of-state of the intrinsic hosts from that of the traps induced by the ionized dopants.