APPLICATION OF CRYSTALLOGRAPHIC DATABASES TO MATERIALS

J. Faber, R. Jenkins and R. L. Snyder

The International Centre for Diffraction Data (ICDD), 12 Campus Boulevard, Newtown Square, PA 19073

There are a number of databases available to the analytical diffraction community. Two of the more important of these are the Powder Diffraction File (PDF) maintained by the International Centre for Diffraction Data (ICDD), and the Inorganic Crystal Structure Database (ICSD) maintained by Fachsinformationzentrum (FIZ, Karlsruhe). In application, the PDF has been used as an indispensable tool in phase identification and identification of unknowns. The ICSD database has extensive and explicit reference to the structures of compounds: atomic coordinates, space group and even thermal vibration parameters. These databases are often used as independent sources of information. However, little thought has been given on how to exploit the combined properties of structural database tools. A recently completed agreement between ICDD and FIZ provides a first step in complementary use of the PDF and the ICSD databases. The focus of this paper (as indicated below) is to examine ways of exploiting the combined properties of both databases [1].

In 1996, there were approximately 76,000 entries in the PDF and approximately 43,000 entries in the ICSD database. The ICSD database is being used to calculate entries in the PDF. Thus, to derive d-spacing and peak intensity data requires the synthesis of full diffraction patterns, i.e., we use the structural data in the ICSD database and then add instrumental resolution information. The combined data from PDF and ICSD can be effectively used in many ways. For example, we can calculate PDF data for an ideally random crystal distribution in the absence of preferred orientation. Again, we can use systematic studies of intermediate members in solid solutions series to help produce reliable quantitative phase analyses. In some cases, we can study how solid solution properties vary with composition and extensive properties (e.g., T, P, H), thus gaining insight in bonding behavior. In this paper, we will show a number of examples that focus on the results obtained from "mining" multiple databases.

1. J. Faber, R. Jenkins and R. L. Snyder, invited talk at the 5^th Japan International Society for the Advancement of Material and Processing Engineering (SAMPE) Symposium and Exhibition, Tokyo, Japan, October 28-31, 1997. A preliminary account of this work was given in the session on Innovations in Materials Design.