Elastic deformation in a crystal plate where lattice–parameter mismatch is present between adjacent growth sectors. II. Measurement of lattice tilts by synchrotron X–ray reticulography

Abstract

Synchrotron X–ray reticulography measures lattice–plane orientation variations point by point on the crystal with microradian resolution capability. A fine–scale X–ray–absorbing mesh placed close to the crystal in the back–reflected diffracted beam of ‘white’ X–rays splits the beam into an array of microbeams whose direction differences are measured from the mesh–bar shadow shifts that result from controlled changes of the distance between the mesh and the recording high–resolution photographic plate. In the present work, lattice tilts were measured on either side of a growth–sector boundary outcropping at the surface of a single–crystal synthetic diamond cut and polished parallel to (110) in which the displacements and strains due to lattice–parameter mismatch at coherent growth–sector boundaries have been analysed in part I. Attention was concentrated on a crystallographically well–defined planar segment of a boundary separating two sectors, each of uniform, but significantly differing, concentration of substitutional nitrogen impurity, and for which the anisotropic elasticity theory of part I provided quantitative relations between the lattice–parameter relative difference, Δa/ā, and the lattice tilts due to coherency strains at the boundary. This enabled a novel method for determining Δa/ā to be realized, using reticulographic measurements of tilts of near–surface lattice planes alone. The paper explains the reticulographic technique fully. Factors affecting the quality and reliability of reticulographic records are examined in detail. For the growth sectors concerned, the reticulographic measurement Δa/ā = (1.10 ± 0.10) × 10−5 is in accord with an earlier finding by synchrotron X–ray double–crystal diffractometry, slightly revised as recommended in part I, Δa/ā = (1.20 ± 0.07) × 10−5. The closeness of the findings by two quite different methods confirms the conclusion from X–ray topography that the growth–sector boundary was free from misfit dislocations compensating the lattice–parameter mismatch. On mismatch boundaries in crystals too highly absorbing for transmission X–ray topography, such comparisons of reticulographic and diffractometric measurements would reveal information otherwise inaccessible by non–destructive methods.