ESTIMATING CRYSTAL SHAPE TO ESTIMATE PLAGIOCLASE WEATHERING RATES IN NATURALLY WEATHERED BASALT FROM HAWAI'I

Plagioclase is one of the most abundant silicate minerals in Earth's crust. Weathering of plagioclase is an important process which has been the subject of many studies. This study estimates natural rates of plagioclase weathering by measuring the size and shape of cavities formed by dissolution of plagioclase in a weathered rind on a Hawai'ian basalt flow of known age. From the petrographic measurements, the volume and initial surface area of each crystal can be quantitatively determined; time is constrained by the known age of the flow. This study employs the method developed by Morgan and Jerram (2006, J. Volcanol. Geotherm. Res., v. 154, p.1-7) to estimate the three-dimensional shape of tabular plagioclase crystals from scanning electron microscope (SEM) backscattered images of (two-dimensional) sections through crystal-mold cavities in a polished thin section. Measurements of 361 crystal-mold cavities are most consistent with a 1:6:10 ratio of the three dimensions of the plagioclase crystals. For this shape, the apparent ratios of short:long dimensions of slices through plagioclase crystals in various orientations are nearly identical to the true ratio if the apparent ratio is ~1:9. The number of moles of plagioclase removed is estimated from the volume of the four largest cavities left by complete removal of plagioclase crystals of appropriate alignment. Plagioclase composition was determined by the Michel-Levy method and energy dispersive X-ray spectroscopy. The initial geometric surface area of plagioclase is estimated from the long dimensions and the inferred 1:6:10 ratio of crystal dimensions. The age of basalt flow (3400 yr) is the maximum time for complete dissolution of the plagioclase crystals.

The base-ten logarithm of the rate of plagioclase weathering (in units of moles per square meter per second) estimated by this approach is ~-12. Based as it is on the largest cavities, representing complete removal of the largest crystals, this rate is the most similar to the actual rate; all smaller plagioclases were removed before the largest one was. The estimated rate represents the minimum possible rate, because it is likely that even the largest plagioclases were completely removed from the weathered rind before the full 3400 years since flow emplacement had elapsed.