This CAREER award supports theoretical and computational research and education seeking to predict the properties of new materials where strong electronic correlations play an important role. There is a firm understanding of simple materials such as noble metals and semiconductors and many of their properties can be predicted by electronic structure methods based on Density Functional Theory. In materials with partially occupied d and f shells, strong Coulomb repulsion tends to localize electrons leading to unusual phenomena such as high temperature superconductivity, colossal magnetoresistance, anomalous optical and dc conductivities and large thermoelectric coefficients. These phenomena result from collective correlated behavior of electrons and are not captured by present day electronic structure methods. The sensitivity of strongly correlated materials to small changes in control parameters resulting in large responses makes their study challenging, and the prospects for their application particularly exciting.
The PI will focus on connecting experiments on correlated electron materials with theoretical modeling using many-body computational methods combined with Density Functional Theory. Encouraged by recent advances in many-body methods, such as Dynamical Mean Field Theory and its cluster extensions, the PI aims to develop ab initio computational methods to compute various spectroscopies including optical spectroscopy, transport properties and photoemission spectroscopy of strongly correlated materials. The PI will develop new algorithms and computational tools to study the competition between the Kondo effect and magnetism, and the interplay between magnetism and superconductivity at finite temperature. These methods will be enhanced to study correlation effects at interfaces of strongly correlated materials, for example an interface of a Mott insulator and a ferroelectric.
This project will promote teaching, training and learning via intensive integration of undergraduate and graduate students into the research effort. The PI will develop a graduate and undergraduate course on computational physics as well as yearly two-week summer research programs for high school students. Materials from course development will be made available to the broader community through the internet and a book ?Computational Methods and Simulations in Condensed Matter Physics? being written by the PI. The summer research program will bring high-school students into today's world of materials science and engineering with special emphasis on targeting students who are members of traditionally underrepresented groups in science. This outreach effort aims to nurture an appreciation of modern computational materials science in high-school and undergraduate classrooms with a view towards creating a more scientifically literate general public.
NON-TECHNICAL SUMMARY: This CAREER award supports theoretical and computational research and education that will develop new theoretical and computational tools to predict properties of complex materials and new artificially structured materials. The research focuses on a class of materials in which electrons interact strongly with each other giving rise to correlations in their motions and unusual properties that lie outside the standard textbook paradigms. The PI?s approach builds on the successes of current powerful theories, like density functional theory, and adds advances from the quantum mechanical theory of systems containing many particles. The hybrid approach is more powerful than either approach alone and will be used to study complex materials, many displaying a fierce competition at the level of electrons to become magnets, unusual insulators, superconductors, and more. The research contributes to the discovery of new materials with unusual properties that illuminate the fundamental nature of materials and matter and that may form the foundations of future technologies. The research may be important step toward being able to design new materials with desired properties using computers and starting only from knowledge of the identity of the constituent atoms. Computational approaches will be implemented to allow materials exploration by non-experts and will be made freely available to the Internet community.
This project will promote teaching, training and learning via intensive integration of undergraduate and graduate students into the research effort. The PI will develop a graduate and undergraduate course on computational physics as well as yearly two-week summer research programs for high school students. Materials from course development will be made available to the broader community through the internet and a book ?Computational Methods and Simulations in Condensed Matter Physics? being written by the PI. The summer research program will bring high-school students into today's world of materials science and engineering with special emphasis on targeting students who are members of traditionally underrepresented groups in science. This outreach effort aims to nurture an appreciation of modern computational materials science in high-school and undergraduate classrooms with a view towards creating a more scientifically literate general public.
The research enabled by this grant lead to development of new computational tool, which predicts physical properties of materials using computer simulations only. Density Functional Theory is a powerful tool for predicting properties of weakly correlated materials from first principles, but it fails to explain why electrons in transition metal oxydes, and many intermetallic compounds containg f electrons, tend to localize, forming very narrow bands or even cease to conduct all together, in so called Mott insulators. The new computational method based on combination of Dynamical Men Field Theory and Density Functional theory corrects this deficiency of Density Functional Theory, and enables precise first principle simulations of many materials which could not be addressed before, including heavy fermions, and high temperature superconductors. A few examples of how successful are predictions of this theory are shown in the accompanied figures. Figure 1a shows theoretical prediction for ordered magnetic moments across many classes of iron superconductors. Density functional theory severely overestimates the size of the magnetic moments in most of high-temperature superconductors, while the new method is in very good agreement with experiment. Fig.1b shows theoretical mass enhancement in iron superconductors, and compares the prediction to effective mass measured in optics and ARPES experiment. This work was published in Nature Materials 10, 932–935 (2011). The agreement is very good. Fig.2 shows optical conductivity of representative iron superconductor BaFe2As2. Panel a shows theoretical prediction, and panel b the experimental measurements. Panel c shows predicted anisotropy of optical conductivity, which at the time of publication, was not known experimentally. A few months later experiment confirmed the theoretical prediction, which was published in Nature Physics 7, 294–297 (2011). Figure 3 shows theoretical prediction for local density of states in heavy fermion material URu2Si2, as measured by scanning tunneling microscopy. This compound was studies for over 30 years, because it undergoes a second order phase transition into on ordered states, which remains unknown. The order parameter of this state was named "hidden order" and remains mysterious to date.In Nature Physics 5, 796 - 799 (2009) we used the theoretical simulation and proposed that order parameter is a hexadecapole order, a high rank multipole order of uranium orbitals. Recently this state was confirmed by Raman spectroscopy in arXiv:1410.6398.
