Data generation in materials science is often limited by the time it takes to perform experiments or simulations. To facilitate the exploration and characterization of complex materials, we have developed the Bayesian Optimization Structure Search (BOSS) code. BOSS uses an active learning technique that strategically samples the parameter space of material-science tasks be it experimental or computational. BOSS proposes new data acquisition points for maximum knowledge gain, balancing exploitation with exploration. I will demonstrate BOSS' smart and efficient data strategy for two examples...
Ab initio wavefunction-based quantum chemistry represents a tantalizing alternative to density functional theory for problems in materials science, due to the former's ability to achieve high accuracy with systematic improvability. I will give an overview of our group's research efforts in this direction, describing theoretical and methodological developments at the periodic Hartree-Fock and post-Hartree-Fock levels (especially perturbation theory and coupled-cluster theory) and connections to popular Green's function-based techniques such as the random-phase...
Entanglement lies at the root of quantum theory. It is a remarkable resource that is generally believed to diminish when entangled systems interact with their environment. On the contrary, we find that engaging a system with its environment increases its ability to retain entanglement. The maximum rate of retaining entanglement is given by the quantum channel capacity. We counter-intuitively boost the quantum capacity of a channel by allowing it to leak almost all quantum information to...
Across the world and across disciplines, numbers reveal that the term “alt-ac” – referring to positions within higher education and research alternative to the professoriate – is a misnomer. Permanent academic jobs are, in fact, the “alt-ac”. In this talk, I’ll share my (happy) experience going from a computational chemistry lab to my current career on the “other side” of scientific publishing, and explores roles for STEM Ph.D.s in the publishing industry.
In the American economic system, competition is a critical driver of performance and innovation. The same can be said for materials physics. My group focuses on studying a variety of strongly correlated quantum systems, where the competition between charge, spin and orbital degrees of freedom can lead to novel or enhanced properties. It is this sensitivity that makes these materials useful for devices. A good device has a measured property (such as resistance or magnetization) that changes dramatically with an external stimulus (such as current, temperature or magnetic field). Competition is a valuable strategy for creating this interplay of parameters. Magnetic competition in magnetic systems, on the other hand, has often been seen as a hindrance. While it typically decreases the overall net magnetization, I will show that it can be utilized to generate novel phenomena useful for devices, such as giant negative magnetization and enhanced magnetization at small applied fields. While much research on magnetism utilizes large fields to strengthen the net magnetization, most devices will need to utilize small fields. While my group also collaborates on a wide range of other systems (such as topological insulators, delafossites and transition edge sensors), much of our focus has been to grow high-quality films and understand the interfacial interactions in magnetic and magnetoelectric layers. I will discuss our first observation of a magnetoelectric dead layer, which motivated our recent interest and successes in magnetic phase competition and then some of the interesting features we have discovered in complex oxide thin films.
Dr. Mikael Kuisma seeks quantitative and qualitative understanding of nanoscale quantum dynamics, such as collective excitations in functionalized noble metal nanoparticles and hot carrier generation with potential applications from microscopy to photovoltaics. He is also a developer of GPAW electronic structure program, which he further utilized to run large scale parallel models of electron dynamics in nanosystems.
Experimental research at the nanoscale continues to challenge our ability to predict the behavior of quantum systems. Advances with lithographically patterned solid-state electronic devices have enabled multiple platforms for the simulation of quantum matter. In particular, semiconductor quantum dots and superconducting qubits provide tools for studying the wealth of physics induced by nonlinearities at the single electron and single microwave-photon level, respectively, and have been separately pursued as enabling technologies for qubits. In recent years, hybrid devices that combine such historically distinct lines of research have received greater attention, whether to enable novel sensing or measurement applications, or to couple small systems of qubits together at long range (e.g. quantum transduction). I will showcase the rich behaviors and phases of quantum matter that coupled quantum dots can exhibit, including a surprising transport mechanism called cotunneling drag, signatures of Kondo physics with emergent symmetry , and non-Fermi liquid states. I will also discuss my work towards fabricating superconducting qubits on silicon-on-insulator substrates for hybrid device applications. The integration of quantum dots and superconducting resonators promises to yield new probes for studying quantum matter, and superconducting qubits are coming of age in their own right for the implementation of many-body spin models.
When atomically thin two-dimensional (2D) materials are layered, they often form incommensurable noncrystalline structures that exhibit long-period moiré patterns when examined by scanning probes. In this presentation we will use graphene and hexagonal boron nitride as examples of gapless and gapped Dirac materials to illustrate how the moire superlattices due to interlayer coupling can alter the materials' intrinsic electronic properties. The derivation of the effective models for these van der Waals materials heterojunctions for arbitrary twist angles can benefit from input obtained from ab initio calculations carried out for commensurate short period crystalline structures. We will discuss how the moire pattern modified electronic structures give rise to a variety of experimentally measurable features including enhanced density of states through van Hove singularities, and flat bands, or to their suppression due to formation of band gaps.