RELATIONSHIP BETWEEN SURFACE-STRUCTURE, GROWTH-MECHANISM, AND TRACE-ELEMENT INCORPORATION IN CALCITE

Crystal growth and coprecipitation experiments demonstrate the manner in which surface structure and, in turn, crystal structure influence g rowth mechanism and trace element incorporation in calcite. Dominant { 10 ($) over bar 14} faces grow by the spiral mechanism, producing asym metric polygonized growth hillocks comprised of two pairs of nonequiva lent vicinal faces. Trace elements Mg, Mn, and Sr are differentially i ncorporated into structurally distinct growth steps that comprise the nonequivalent vicinal faces. The resulting trace element distributions represent intrasectoral zoning patterns and are found to be consisten t with face symmetry. Lateral spreading rates are also different for n onequivalent growth steps at a given degree of supersaturation, and th e rate anisotropy is dependent on the Ca2+:CO32- ratio in the growth s olution. A rounding transition associated with changes in kink site de nsity occurs preferentially on only one pair of equivalent growth step s, further demonstrating the importance of step-specific kinetics and affinities on {10 ($) over bar 14} faces. Growth on other forms, inclu ding {01 ($) over bar 12}, {11 ($) over bar 20}, and {0001}, does not result in differential partitioning of trace elements and intrasectora l zoning, which is consistent with each of their surface symmetries an d allowed growth mechanisms. However, sectoral zoning of Mg, Mn, and S r occurs between these nonequivalent sectors, as well as {10 ($) over bar 14}. We present a model detailing the geometry and coordination of elementary kink sites to explain both the differential incorporation and the rate anisotropy between nonequivalent growth steps on individu al {10 ($) over bar 14} faces. The model, constrained by face symmetry , accounts for affinities of different trace elements among four struc turally distinct kink sites at which incorporation is preferred. The m odel also explains the observed differences in growth step velocity, a s well as the in situ observations of growth step velocities by atomic force microscopy. Trace element incorporation on the dominant {10 ($) over bar 14} faces of calcite is controlled by the detailed structure of the interface, which varies spatially on a face, as well as with e xternal conditions. Consequently, our observed trace element distribut ions violate equilibrium partitioning, and it is likely that many trac e element distributions in natural carbonates also may not reflect equ ilibrium.