AN EXPERIMENTAL STUDY OF METAMORPHIC NUCLEATION: METHODS AND INITIAL RESULTS

In the study of metamorphic rocks, crystal nucleation remains poorly understood. We have initiated an experimental study of metamorphic nucleation with the aim of establishing the dependence of nucleation rate on thermodynamic driving force.

Our initial focus was on the reaction An+Wo=Grs+Qtz, a reaction simple enough to make the experiments tractable, yet which retains the basic features of reactions studied by others working on crystallization kinetics: aluminous garnet is the nucleating phase and the reaction involves multiple reactants and products. All experiments were carried out in a piston-cylinder apparatus. In order to remove temperature as a variable, runs were isothermal (1100°C), and nucleation was initiated by overstepping the equilibrium pressure (1.38 GPa) by 30 to 520 MPa. The starting material was a synthetic, fine-grained, homogeneous, non-hygroscopic assemblage of An+Wo+Qtz, produced from CaO-Al₂O₃-SiO₂ glass made from chemical reagents. Experimental duration ranged from 4 to 86 hours.

The Grs produced is equant, subhedral, poikiloblastic, often isolated, and, in the best experiments, approximately 10 µm in diameter. However, it invariably contains numerous submicron-size quartz inclusions, preventing imaging of the Grs using X-ray computed tomography. Nucleation rate was therefore determined by the combination of the experimental duration within the Grs+Qtz field, together with a count of Grs crystals per unit area, counted on a BSE image. Most of the nucleation appears to occur rapidly, based upon the similarity of results from ~20- and ~80-hour experiments.

Measured 2-D nucleation rates varied from 31 nuclei/mm²/hr, at a ΔG of 1.48 kJ/mol Grs (ΔP=50 MPa). to 22000 nuclei/mm²/hr, at a ΔG of 15.97 kJ/mol Grs (ΔP=520 MPa). These results confirm the existence of a systematic relationship between nucleation rate and driving force, as well as the presence of a sharp increase in nucleation rate with driving force.