Simulations of Multi-atom Vacancies in Diamond

  • PDF / 227,751 Bytes
  • 6 Pages / 612 x 792 pts (letter) Page_size
  • 58 Downloads / 221 Views

DOWNLOAD

REPORT


0978-GG13-15

Simulations of Multi-atom Vacancies in Diamond Istvan Laszlo1, Miklos Kertesz2, and Yury Gogotsi3 1 Institute of Physics, Budapest Technical University, Budapest, H-1111, Hungary 2 Chemistry Department, Georgetown University, 37th and O Streets, NW, Washington, DC, 20057-1227 3 Department of Mateirals Science and Engineering, Drexel University, Philadelphia, PA, 19104

ABSTRACT Multi-atom vacancies and pores in diamond-structured carbon phases play an important role in carbon sieves and carbon based storage. The size and shape of pores have a profound effect on the energetics of adsorptive storage. We are modeling large numbers of vacancy configurations with a combination of ab initio density functional theory (DFT), tight binding DFT and simpler methods based on Brenner-potentials and modified Brenner potentials. The more accurate calculations serve as the basis of the parametrization which is then used in lattice Monte Carlo simulations on more complex vacancies. The results will be put in the context of SiC-derived porous carbon materials with the purpose to explore basic questions of energetics in porous carbons. INTRODUCTION With the recent development in the production of porous carbons with controllable pore size [1] the basic question arises: what are the stable pore sizes and shapes in porous carbons? The size and shape of pores have a profound effect on the energetics of adsorptive storage.[2] While many studies addressed porous graphitic carbons, pores in diamond received little attention. However, porous sp3-bonded carbon, such as sintered nanodiamond [3], have been receiving increasing attention. In this report, we describe our initial work in answering this question by using atomistic computational tools. Modeling structures with large multiatomic voids has to wrestle with the extremely large number of degrees of freedom that arises as soon as the number of missing atoms reaches ~10-15. Even the modest aim of obtaining qualitative descriptions is severely hampered by this problem. We start with the modest question: which are the most stable structures for a void with n contiguously missing atoms, Vn, in a diamond lattice? We describe an algorithm that generates a large yet manageable number of stable (low energy) vacancy structures for each n, starting from selected Vn-1 vacancy structures based on their low energy. After discussing this algorithm, we describe the methods used for the energy calculations. This is followed by a discussion of the vacancy structures with n values of up to 48. Vacancy clusters will be represented by the missing atoms from the bulk, as illustrated for the chair-like cyclohexane-skeleton shown in Figure 1. This V6 vacancy cluster is an example of the sub-group of structures that contain the minimum number of dangling bonds for a given n, which we call “adamantane-like” vacancy clusters. These types appear intuitively as most stable, at least for small and medium n values and these have been considered as likely vacancy clusters in the silicon vacancy literature