Simulated STM images of reconstructed Pt(111) [Narasimhan and Pushpa, Phys. Rev. B

In our group, we use theoretical techniques to explore novel physics and chemistry at the nanoscale. Our main interest is in examining how properties (structural, mechanical, electronic, magnetic and chemical) change upon lowering dimensionality and/or reducing size. While we are mainly driven by intellectual curiosity, some of the problems we work on have technological relevance for pressing problems such as clean energy and device miniaturization.

To address such issues in a precise way, the main tool used in our group is quantum mechanical density functional theory. We perform first-principles calculations, i.e., with no input other than atomic numbers and masses, we are able to compute the structure and properties of a wide range of systems and materials.

Some examples of problems of recent and current interest:

 Mixing and magnetism on surfaces:

From ancient times, it has been known that mixing two or more metals to form an alloy can result in materials with superior properties. However, many combinations of metals don't form alloys in three-dimensional bulk systems. Is this still true in two-dimensional surface systems? We have shown that many combinations of bulk-immiscible metals can form strain-stabilized surface alloys. By assembling and analyzing a large database of ab initio results, we have improved the understanding of the factors that govern mixing in two dimensions.

The 'Rational Design' of Catalysts:

While industrial processes are strongly reliant on the use of catalysts, most commercial catalysts have been developed by a process of trial and error. Nowadays, however, there is an effort to gain a greater theoretical understanding of how catalysts operate, and then use this understanding to design better and/or cheaper catalysts. One of the processes we have studied is NO dissociation; this is important for reducing NO in automobile exhaust. We have examined how good rhodium and platinum surfaces and nanoparticles are at adsorbing NO and breaking the N-O bond. Some of the factors we have scrutinized are the effect of changing coordination number, and the role of the oxide support. We have postulated that a simple quantity called the 'effective coordination number' may serve as a good indicator of adsorption energies and dissociation barriers.

Size-Dependent Properties of Nanosystems:

Until recently, it was believed that a really small object would always melt at a lower temperature, and be more soft and floppy, than a large one. But is this always true? Using density functional theory and density functional perturbation theory, we have shown the existence of scaling relations connecting bond stiffness, bond length and coordination number. These can be used as a framework to explain recent experimental and theoretical results that suggest that for some elements, small clusters melt at a higher temperature than the bulk.

Gas storage:

Many cities in India have mandated that public transport vehicles should run on natural gas rather than petrol, in a bid to reduce pollution; India also has more natural gas resources than petroleum. However, as occasional newspaper reports make clear, the current way of storing compressed natural gas in cylinders is unsafe. Further, the cylinders used are heavy, not very adaptable to vehicular space constraints, and the pressurization can be expensive. Adsorptive storage of natural gas, if achievable economically, presents an attractive alternative. Activated carbons are cheap and work as a good 'sponge' to soak up methane (the main constituent of natural gas); however it has not been very clear why this is so. By performing ab initio density functional calculations on a variety of defective and/or chemically functionalized graphene systems we are developing strategies to increase the adsorptive capacity of carbon.