BCCMS  /  Groups  /  CMS  /  People - Bremen  /  M. Lorke
English  /  Deutsch

Dr. Michael Lorke

Fields of Interest

Many-Particle Physics
Optoelectronic Properties of Semiconductors 
Semiconductor Laser Theory 
Density Functional Theory 
Quantum Field Theory

PublicationsGoogle Scholar

 

Research

I work on a crossroad between computational material science and theoretical physics. A focus area is the microscopic description of novel materials for opto-electronic applications such as lasers, LEDs, single photon sources, and solar cells. I utilize mainly atomistic methods such as density functional theory and many-body perturbation theory to determine the electronic structure. Additionally, I recently developed a novel exchange-correlation potential, that fulfills the linearity requirements of the exact functional and allows forstructural relaxation. Ground state properties as determined by these methods can either be analyzed separately and be compared to experimental results or be thestarting point for multi-scale investigations of the non-equilibrium properties, usingSchwinger-Keldysh Green’s function or cluster expansion techniques. This approach allows to treat both fundamental physical processes like quantum optical effects or carrier relaxation and dephasing processes, as well as directly application relevant effects like turn-on delay, laser modulation or charge transfer dynamics. The direct multi-scale connection between state-of-the-art band structure analysis and modern non-equilibrium methods allows for a fully material realistic description of, .e.g., the carrier dynamics in two-dimensional materials or the modulation properties of nano-lasers.

Carrier multiplication in van der Waals layered transition metal dichalcogenides

Carrier multiplication (CM) is a process in which high-energy free carriers relax by generation of additional electron-hole pairs rather than by heat dissipation. CM is promising disruptive improvements in photovoltaic energy conversion and light detection technologies. Current state-of-the-art nanomaterials including quantum dots and carbon nanotubes have demonstrated CM, but are not satisfactory owing to high-energy-loss and inherent difficulties with carrier extraction. Here, we report CM in van der Waals (vdW) MoTe2 and WSe2 films, and find characteristics, commencing close to the energy conservation limit and reaching up to 99% CM conversion efficiency with the standard model. This is demonstrated by ultrafast optical spectroscopy with independent approaches, photo-induced absorption, photo-induced bleach, and carrier population dynamics. 

Defect calculations with hybrid functionals in layered compounds and in slab models

Layered materials are presently under intense study, and most applications require knowledge about their defects. It has been shown earlier that the screened hybrid functional of Heyd, Scuseria, and Ernzerhof (HSE), with parameters tuned to reproduce the relative position of the band edges and to satisfy the generalized Koopmans’ theorem in the given material, is capable of providing defect properties very accurately in traditional bulk semiconductors. This success is thought to be connected to the proper description of electronic screening. In this paper we investigate whether such a functional can be optimized for layered compounds, such as GaSe and hexagonal BN (hBN).While we find this to be possible in the bulk materials, as expected, the optimal parameters change significantly in the monolayers (ML), due to the effect of the surface on the electronic screening. For ML GaSe (thickness ∼5Å), where the dielectric constant does not change much within the layer, a reoptimization is possible, while for the atomically thin ML hBN the optimization criteria can only be met approximately. In other words, the accuracy which can be achieved for the electronic structure of defects by a HSE functional will always be less in atomically thin monolayers than in the bulk built from such layers. We also show that the accuracy of HSE functionals, with parameters optimized for the bulk in traditional (nonlayered) bulk semiconductors, also diminishes in surface calculations, even for thick slabs, if the dielectric constant (ε ∞) is low. Generally, the accuracy of bulk and surface calculation by any functional can be very different.