Quantum-Chemical Investigation of the Metal-Insulator Transition in realistic Low-Dimensional Systems

• Applicant: Professorin Dr. Beate Paulus
• Subject Area: Theoretical Chemistry: Electronic Structure, Dynamics, Simulation
• Term: from 2012 to 2016
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 208496859

This joint project aims to improve our understanding of the metal-insulator transition (MIT) in low-dimensional systems through theoretical investigations for systems like graphene nanoislands and one-dimensional atomic chains. The topic is at the borderline of microelectronics and molecular electronics, where details of chemical composition of a system under study can have a huge influence on the existence (or absence) of a MIT, and on its sensitivity to external perturbations. Our aim is to analyze and understand the change in the conducting behavior in systems of technological relevance. We further intend to provide support for the design of new materials formed by a combination of graphene and metals. In order to have a realistic description of the system, we solve the full electronic Schrödinger equation with quantum-chemical methods, ranging from full configuration interaction (FCI) via various approximate multi-reference methods to semiempirical methods for very large systems (expertise of the Evangelisti group). To make feasible the application of systematically improvable wavefunction-based correlation methods for large or even periodic systems, we will apply the method of increments. This special version of a local correlation method expands the correlation energy of the system in terms of increments from a single center (one-center term), from a pair of centers (two-center term), and so on, and can be made to converge reasonably quickly even for bulk metals (expertise of the Paulus group). The MIT can be characterized by different indicators, such as energy gap, electric polarizability, fluctuation of the position operator (localizability), and the particle-hole entropy. All these indicators can be, and will be, calculated in an ab initio manner and compared for different systems and different types of MITs.

• DFG Programme: Research Grants
• International Connection: France
• Participating Persons: Professor Stefano Evangelisti; Dr. Antonio Monari