Gibbsite growth kinetics on gibbsite, kaolinite, and muscovite substrates: Atomic force microscopy evidence for epitaxy and an assessment of reactive surface area

New experimental data for gibbsite growth on powdered kaolinite and single crystal muscovite and published data for gibbsite growth on gibbsite powders at 80 degrees C in pH3 solutions show that all growth rates obey the same linear function of saturation slate provided that reactive surface area is evaluated for each mineral substrate. Growth rate (mol m(-2) s(-1)) is expressed by Rate(ppt) = (1.9 +/- 0.2) X 10(-10)\Delta G(r)\/ RT(0.90+/-0.01), which applies to the range of saturation states from Delta G(r) = 0 to 8.8 kJ mol(-1), where Delta G(r) = RT[ln(Q/K)] for the reaction Al3+ + 3H(2)O = Al(OH)(3) + 3H(+), and equilibrium defined as Delta G(r) = 0 was previously determined. Identification of the growth phase as gibbsite was confirmed by rotating anode powder x-ray diffraction. Rates an kaolinite were determined using steady-state measured changes between inlet and outlet solutions in single-pass stirred-flow experiments. Rates on muscovite were determined by measuring the volume of precipitated crystals in images obtained by Tapping Mode(TM) atomic force microscopy (TMAFM). In deriving the single growth rate law, reactive surface area was evaluated for each substrate mineral. Total BET surface area was used for normalizing rates of gibbsite growth onto powdered gibbsite. Eight percent of the BET surface area, representing the approximate amount occupied by the aluminum octahedral sheet exposed at crystal edges, was used for powdered kaolinite. The x - y area of the TMAFM images of the basal surface was used for single crystal muscovite. Linearity of growth rates with saturation state suggests that the dominant nucleation and growth mechanisms are two dimensional. Such mechanisms are supported by observations of the morphologies of gibbsite crystals grown on muscovite at Delta G(r) = 8.8 kJ mol(-1). The morphologies include (1) apparent epitaxial films as determined by hexagonal outlines of edges and thicknesses of 30 to 40 Angstrom, (2) elongate crystals 30 to 40 Angstrom thick aligned with the structure of the distorted Si-tetrahedral sheet of the 2M, muscovite, and (3) micrometer-scale three-dimensional clumps of intergrown crystals. Reactive surface area as defined now for heterogeneous crystal growth in reactive-transport models must be modified to include substrates other than that of the growing mineral and to account for the role of structural and chemical controls on epitaxial nucleation and growth. Copyright (C) 1999 Elsevier Science Ltd.