ABSTRACT
Recent developments in organic solar cells show interesting power conversion efficiencies. However, with the use of organic semiconductors and bulk heterojunction cells, many new concepts have to be introduced to understand their characteristics. Only few models investigate these new concepts, and most of them are one-dimensional only. In this work, we present a two-dimensional model based on solving the drift-diffusion equations. The model describes the generation of excitons in the donor phase of the active layer and their diffusion towards an interface between the two separate acceptor and donor domains. Then, when the exciton reaches the interface, it forms a charge transfer state which can split into free charges due to the internal potential. Finally, these free charges are transported toward the electrodes within their respective domains (electrons in acceptor domain, holes in donor domain) before being extracted. In this model, we can follow the distribution of each species and link it to the physical processes taken into account. Using the finite element method to solve the equations of the model, we simulate the effect of the bulk heterojunction morphology on photocurrent curves. We concentrate on the morphology parameters such as the mean acceptor/donor domain sizes and the roughness of,the interface between the donor and acceptor domains. Results are discussed in relation with experimental observations.
Subject(s)
Computer-Aided Design , Electric Power Supplies , Models, Chemical , Nanostructures/chemistry , Nanostructures/radiation effects , Organic Chemicals/chemistry , Solar Energy , Computer Simulation , Equipment Design , Equipment Failure Analysis , Nanostructures/ultrastructure , Organic Chemicals/radiation effects , Oxidation-ReductionABSTRACT
This paper presents a realistic model that explicitly takes into account the electrostatic coupling between the nanocrystals of a disordered layer constituting the floating gate of a non-volatile memory. A statistical study of the neighborhood of a given nanocrystal is carried out, leading to the mean number of neighboring nanocrystals as a function of the radius of the central nanocrystal. We show that the empty neighborhood of every nanocrystal can be represented by an equivalent torus ring in the previous model of a single nanocrystal. Then the effects of charged nanocrystals are taken into account by an appropriate rigid shift of the energy levels of the central nanocrystal. The proposed model is validated by statistical comparisons with exact 3D computations, and the influence of the electrostatic coupling is analyzed and discussed.
Subject(s)
Computer Storage Devices , Nanostructures/chemistry , Nanostructures/ultrastructure , Nanotechnology/instrumentation , Signal Processing, Computer-Assisted/instrumentation , Computer Simulation , Equipment Design , Equipment Failure Analysis , Models, Chemical , Particle Size , Static ElectricityABSTRACT
In recent years, a lot of attention has been paid to carbon nanotube (CNT) networks and their applications to electronic devices. Many studies concentrate on the percolation threshold and the characterization of the conduction in such materials. Nevertheless, no theoretical study has yet attempted to characterize the CNT features inside finite size CNT networks. We present a theoretical approach based on geometrical and statistical considerations. We demonstrate the possibility of explicitly determining some relations existing between two neighbor CNTs and their contact efficiency in random networks of identical CNTs. We calculate the contact probability of rigid identical CNTs and we obtain a probability of 0.2027, which turns out to be independent of the CNT density. Based on this probability, we establish also the dependence of the number of contacts per CNT as a function of the CNT density. All the theoretical results are validated by very good agreement with Monte Carlo simulations.
