Study of indium-defect interactions in diamond using two-dimensional conversion-electron emission channelling
Doyle, BP Storbeck, EJ Wahl, Ulrich Connell, SH Sellschop, JPF #
Iop publishing ltd
Journal of physics-condensed matter vol:12 issue:1 pages:67-78
Channelling has, since its inception, proven to be a valuable tool in locating the geometric position of atoms in the crystal lattice. Allied with powerful theoretical models, it can yield detailed information on the positions that these impurities occupy. In-111, a radioactive isotope with a conveniently shea half-life, is an often-used probe of heavy-atom doping of materials. Previous work has centred on the lattice location of In-111 implanted in type IIa diamond. Theoretical calculations on this 'pure' system have also recently been made. We have performed the first studies of In-111 implanted into various carefully selected, defect-rich diamond systems and obtained fractions for the sites occupied. The defect systems investigated include nitrogen in various configurations, boron, hydrogen and vacancies. The use of two-dimensional conversion-electron emission channelling (CEEC) has enabled the system to be studied in greater detail than with conventional one-dimensional CEEC. Coupled with the acquisition of the CEEC spectra for all the major channelling axes, this has yielded a comprehensive data set. The spectra are consistent with a pure substitutional fraction as well as another fraction, approximately 0.45 Angstrom from the substitutional along a (111) direction. Previous measurements observed these two components together as substitutional or 'near-substitutional'. The data have been compared to simulated CEEC spectra and earlier quantum chemical calculations. The pure substitutional fraction is indicated to be in a defect-free configuration while the component displaced away from substitutional involves most probably the divacancy and another nearby defect. The results show no dependence on impurity type, even after annealing. If indium complexation with these defects does occur it is shown not to measurably affect the channelled spectra and thus the projected lattice location of the In-111 probe. The origin of the random fraction measured in previous studies is proposed to be partially due to In in different multi-vacancy complexes. Taken together, the data indicate the importance of vacancies (complexes) in the final configuration for In after implantation in diamond.