Organic photodetectors (OPDs) are promising candidates for ultra-violet (UV), visible, near-infrared (NIR) image sensors due to their ability to provide low dark current, high detectivity, flexibility and large area sensing, which can compete with the conventional inorganic photodetectors. Moreover, due to narrow absorption spectrum of organic materials, OPDs need no color filter system and can be used for compact, lightweight, and high resolution full color complementary metal-oxide-?semiconductor (CMOS) image sensors for digital still camera, camcorders, and so on.
For these reasons, the OPDs are getting more interest and research into OPDs are growing. However, the detectivity of the OPDs are still lower than that of the silicon based photodetectors. Because the dark current density is the major factor contributing to the detectivity, it should be decreased to increase the detectivity of the OPDs. To decrease the dark current density, the origin of the dark current density must be identified. In addition, it is highly desirable to develop a model describing the dark current. Futhermore, the quantitative description and prediction of color selectivity of OPDs are needed.
Firstly, we fabricated green selective transparent OPDs with high detectivity of 4.1E12 cm Hz1/2 /W at a reverse bias of -1 V by adopting indium zinc oxide (IZO) as top electrode and using 1,4,5,8,9,11-hexaazatriphenylene-hexacarbonitrile (HATCN) as the buffer layer. HATCN was used to protect the active organic layers from sputter damage during the deposition of the IZO top electrode. The detectivity of the OPDs using 220 nm thick HATCN was comparable to that of the reported conventional OPDs with highly reflective metal electrodes. Furthermore, the color selectivity of the OPDs as a function of the thickness of HATCN as an optical spacer are described quantitatively using optical simulation. The green selective transparent OPD using the 220 nm thick HATCN layer showed the transparency of 26% and 63% in the blue region and the red region, respectively.
Secondly, a theoretical model is presented to describe the dark current density in planar heterojunction (PHJ) OPDs, which is a combination of the Richardson-Schottky thermionic emission for injected current density, a Shockley diode equation for thermally generated current density and a term for leakage current. The theoretical model not only describes the experimental dark current densities very well obtained from five different combinations of donor/acceptor molecules confirming the validity of the model, but also provides the quantitative analysis of the contribution from different origins. The model can be utilized to estimate the dark current density of OPDs as functions of the injection barriers and the energy difference between the highest occupied molecular orbital (HOMO) level of the donor and the lowest unoccupied molecular orbital (LUMO) level of the acceptor at a certain applied bias and establish the selection criteria of a donor material for a given acceptor or vice versa. The theory predicted that the injection barriers and the energy difference should be larger than 1.32 eV and 0.8 eV, respectively, to obtain the dark current density lower than 1E-10 A cm-2 at a reverse bias of -3 V.
Lastly, a theoretical model has been developed to describe the dark current density of bulk heterojunction (BHJ) OPDs considering the extraction efficiency and the interfacial area between the donor and the acceptor. Sub-phthalocyanine (SubPc):C70 and copper phthalocyanine (CuPc):C70 based BHJ OPDs showed different major origins of the dark current density, which explain the different behavior of the dark current density with varying the composition of the codeposition layer. The dependence of the thermally generated current density on the composition of the codeposition layer for the CuPc:C70 based BHJ OPDs is interpreted in terms of the extraction efficiency and the interfacial area in the BHJ layer between the donor and the acceptor.