Welcome to Ab Initio Materials Simulations (AIMS) at Duke!

The Ab Initio Materials Simulations group at Duke University is headed by
Dr. Volker Blum, Associate Professor, Department of Mechanical Engineering and Materials Science.

Welcome to the AIMS Group at Duke!

Our focus is to predict the properties of real materials - known ones and those that are not yet known - using computers and the basic mathematical "first principles" that govern the world around us: the laws of quantum mechanics and "multiscale models" based on it. We create the infrastructure to do so, and we work on specific high-value materials problems - either on our own, or with other groups (theory and experiment) around the world.

The picture shows the short version of Schrödinger's Equation, and a cartoon of a graphene film grown on an SiC substrate, which we addressed in Physical Review Letters 111, 065502 (2013), Physical Review Letters 114, 106804 (2015), and in ACS Nano (2016), doi:10.1021/acsnano.6b02402.

In addition to work on inorganic materials, the group also has a significant investment in molecular structure and spectroscopy, most recently focused on molecules for magnetic resonance with sensitivities enhanced by hyperpolarization, together with the groups of Warren Warren and Thomas Theis at Duke University.

The group's work is based on the internationally used and developed "FHI-aims" code for molecular and materials simulations, of which Volker Blum is coordinator and lead developer. This is an accurate, all-electron electronic structure code that scales efficiently to system sizes up to thousands of atoms, and to computers with (ten)thousands of CPU cores.

An offshoot of this work is the massively parallel dense eigenvalue solver library ELPA, developed by a consortium and used by other electronic structure codes around the globe.

In a recent, large collaborative study published in Science, led by Kurt Lejaeghere and Stefaan Cottenier (Ghent University), FHI-aims emerged among the most accurate available codes for a broad variety of systems. See here for more details.

Within a recently granted NSF SI2-SSI infrastructure project "ELSI" (Electronic Structure Infrastructure), we are now taking this work to the next level for accurate, efficient, scalable electronic structure theory.

And in a recent high-throughput approach to derive chemical trends, we found a very interesting correlation between calcium and lead. Amino acids, the building blocks of proteins, really do not offer much chemical selectivity between the two, and this may have implications for the long-term toxicity of lead. (See Figure 5 of this paper for the binding energy trends of these ions to amino acids.)

Much of our vision is summarized in Volker's New Faculty Lecture, available on Duke's Youtube channel by following the link.


If you are interested in undergraduate, graduate or post-doctoral research experience in an exciting field embedded within an international community of colleagues, contact us. If you are excited about computational science applied to materials research and if you are comfortable with computational methods and tools, we would love to hear from you!

Volker Blum • Department of Mechanical Engineering and Materials Science
Office: 1111 Hudson Hall

Postal: 144 Hudson Hall, Box 90300 • Duke University • Durham, NC 27708 • USA
Phone: +1 (919) 660 5279 • E-mail volker.blum <at> duke.edu