We investigate the electronic structure of a Palladium nanoparticle that is partially embedded in a matrix of silicon carbonitride. From classical molecular dynamics simulations we first obtain a representative atomic structure. This geometry then serves as input to density-functional theory calculations that allow us to access the electronic structure of the combined system of particle and matrix. In order to make the computations feasible, we devise a subsystem strategy for calculating the relevant electronic properties. We analyze the Kohn-Sham density of states and pay particular attention to d-states which are prone to be affected by electronic self-interaction. We find that the density of states close to the Fermi level is dominated by states that originate from the Palladium nanoparticle. The matrix has little direct effect on the electronic structure of the metal. Our results contribute to explaining why silicon carbonitride does not have detrimental effects on the catalytic properties of palladium particles and can serve positively as a stabilizing mechanical support.
MotivationFirst-principles electronic-structure theory 1-13 allows to gain unbiased insights into the microscopic, quantum-mechanical effects that are at the heart of modern nanotechnology [14][15][16][17][18][19][20][21][22][23][24] . However, often full ab-initio studies on systems of experimental relevance remain challenging since the computational effort grows immensely with system size 25 . For instance, in many experiments metal nanoparticles that have a diameter of several nm and contain several hundreds or thousands of atoms are used. For such systems, tens of thousands of electrons have to be taken into account for first-principles studies. Even with modern computers using massive parallelization, geometry optimizations and detailed investigations of the electronic structure of systems of that size are out of reach. Therefore, strategies have to be devised how insights can be gained into large systems despite of these computational limitations. In this work, we take a first step in this direction and investigate a palladium nanoparticle which is partially embedded in a matrix of silicon carbonitride (SiCN).This system is of practical relevance because noble-metal nanoparticles, in particular nanoparticles containing Pd, are effective catalysts for a wide range of chemical reactions [26][27][28][29][30] . In comparison to bulk metals, nanoparticles exhibit a high surface-