Multiscale Engineering, Mathematics and Sciences Group

The focus of the Multiscale Engineering, Mathematics and Sciences group is to develop innovative computational approaches and mathematics to link disparate length and time scales widely encountered in emerging applications in engineering and sciences. A scientific challenge in mechanics and materials, nanotechnology, biological and chemical sciences, and interfacial sciences is to link length scales ranging from Angstroms to centimeters and larger, and time scales ranging from femtoseconds to minutes and longer. 

One of the main goals of this group is to incorporate quantum and atomistic scale information into mesoscale and continuum scales to connect computational results with experimental data. At the mesoscale, the group focuses on developing coarse-grained multiscale methods to preserve important underlying features of the problem such as structure, dynamics, thermodynamics, transport, etc. Mathematical techniques based on information theory, solution of inverse problems, and machine learning are used to develop accurate force-fields to describe molecular interactions at the mesoscale. At the continuum scale, quasi-continuum approaches incorporating interatomic potentials into classical theories are used to describe physical phenomena at larger length and time scales. Deep learning approaches are used to extract physical models and constitutive theories that feed into quasi-continuum theories. 

An important goal of the multiscale group is to connect computational predictions with experiments. Some of the applications that the group is currently working on are fluid mechanics at the nanoscale, solid-liquid interfaces, chemo-opto-electro-mechanically excited nanostructures, novel materials for separations, water desalination, energy storage and conversion, DNA and protein sequencing and connecting the information to disease state, sensing and actuation at micro and nanoscale, catalysis in confined environments, etc.