Computational predictions of stable 2D arrays of bidisperse particles

We study computationally the stability of various 2D arrays of bidisperse mixtures of stabilized nanoparticles through a melting simulation employing the Metropolis algorithm for determining surface diffusion. In our previous work [Langmuir 2004, 20, 9408], we studied computationally the stability of bidispersed monolayers of thiol-stabilized gold nanoparticles with a size ratio (sigma) of 0.375. We found that interparticle forces were essential to stabilize the LS (the two-dimensional NaCl analogue) lattice at the experimentally determined surface coverage. In this paper, we extend our study to determine the conditions necessary to form stable LS(2), LS(4), and LS(6) lattices, which have yet to be observed. Using a simple design rule that involves matching the distances between either large-large particles and large-small particles or large-small particles and small-small particles to correspond to the respective potential minima leads to predictions for size ratios that will form each desired lattice, given other parameters characterizing the systems' physical properties. We predict and verify computationally LS(2), LS(4), and LS(6) lattices at relatively low surface coverages. Additional simulations show that the LS, LS(2), and LS(6) lattices are indeed stable structures at their predicted surface coverage, whereas the LS(4) lattice is a metastable structure; however, a modest increase in the surface coverage of the LS(4) lattice converts it to a stable rather than long-lived metastable structure. This study may be used as a guide for experimentalists in their search for these novel structures.