An Experimental Study of the Reflection of X-Rays from Calcite

Abstract

Per cent., of monochromatic X-rays reflected from calcite, for wave length ranging from.3 \AA{}.8 \AA{} was studied by means of a double X-ray spectrometer. A narrow beam of general radiation fell at an angle upon a crystal designated as $A$ and the reflected beam was allowed to fall upon a second crystal designated $B$. The intensity of the beam from $B$ was then compared, by means of the ionization produced in an ionization chamber with the direct beam from $A$. Per cent., reflection was found surpisingly large when the surfaces of the two crystals were strictly parallel. The per cent., reflection will depend upon the character of the beam reflected and therefore upon crystal A.Significant energy distribution curves were found when the crystal B was rocked through a small angle on either side of parallelism. A very sensitive micrometer motion was provided for rotating crystal $B$ about both a horizontal and also about a vertical axis, necessitated by the extremely narrow energy distribution curves obtained.Three pairs of crystals were investigated. Pair (${A}_{1}\ensuremath{-}{B}_{1}$) where $A$ was a rather imperfect specimen of Montana calcite and ${B}_{1}$ was a clear specimen of Iceland spar. The surface of both crystals were polished. Pair (${A}_{2}\ensuremath{-}{B}_{2}$). Here a clear specimen of Iceland spar was split. The two split surfaces were polished and the crystals mounted so that reflection was from the surfaces that had been contiguous before splitting. Pair (${A}_{3}\ensuremath{-}{B}_{3}$) was another pair obtained by splitting a clear specimen of Iceland spar, but leaving surfaces unpolished. Mounted same as (${A}_{2}\ensuremath{-}{B}_{2}$).The rocking curves vary greatly in width, depending upon the nature of the reflecting surface; 16\ensuremath{''} of arc for width of half maximum when crystal is a good sample and the surface is the natural split surface (not polished). When crystals are not perfect samples and surface polished as much as 57\ensuremath{''} of arc was obtained. For any one pair of crystals the width of the half maximum of these rocking curves is a linear function of the wave length with a definite finite intercept depending upon the condition of the surface. The more perfect the crystal the less does this width depend upon wave-length.Per cent., reflection is a function of wave length, but is less so the more perfect the crystal. For the pair (${A}_{3}\ensuremath{-}{B}_{3}$), it was nearly independent of wave length. It would seem that the reflection from a really perfect crystal would be independent of wave-length and would probably be greater than fifty per cent.