Computational modeling of nanorod growth.

In this computational study, we used molecular dynamics and the embedded atom method to successfully reproduce the growth of gold nanorod morphologies from starting spherical seeds in the presence of model surfactants. The surfactant model was developed through extensive systematic attempts aimed at inducing nonisotropic nanoparticle growth in strictly isotropic computational growth environments. The aim of this study was to identify key properties of the surfactants which were most important for the successful anisotropic growth of nanorods. The observed surface and collective dynamics of surfactants shed light on the likely growth phenomena of real nanoprods. These phenomena include the initial thermodynamically driven selective adsorption, segregation, and orientation of the surfactant groups on specific crystallographic surfaces of spherical nanoparticle seeds and the kinetic elongation of unstable surfaces due to growth inhibiting surfactants on those surfaces. Interestingly, the model not only reproduced the growth of nearly all known nanorod morphologies when starting from an initial fcc or fivefold seed but also reproduced the experimentally observed failure of nanorod growth when starting from spherical nanoparticles such as the I(h) morphology or morphologies containing a single twinning plane. Nanorod morphologies observed in this work included fivefold nanorods, fcc crystalline nanorods in the [100] direction and [112] directions and the more exotic "dumbell-like" nanorods. Non-nanorod morphologies observed included the I(h) and the nanoprism morphology. Some of the key properties of the most successful surfactants seemed to be suggestive of the important but little understood role played by silver ions in the growth process of real nanorods.