Geochimica et Cosmochimica Acfa Vol. 53, pp. 797-804 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Copyright 0 1989 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Pergamon Press pk. F’rinted in zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA U.S.A. 00 I6-7037/89/$3.00 + .oO Sample preparation and aging effects on the dissolution rate and surface composition of diopside CARRICK M. EGGLESTON’, MICHAEL F. HOCHELLA, JR.’ and GEORGE A. PARKS’ ‘Department of Applied Earth Sciences, Stanford University, Stanford, CA 94305, U.S.A. ZDepartment of Geology and Center for Materials Research, Stanford University, Stanford, CA 94305, U.S.A zyxwvutsrqponml (ReceivedMarch I, 1988; accepted in revisedform February 1, 1989) Abstract-This study shows that the initialdissolution rate of ground diopside powders depends on sample preparation history, in general agreement with the findings of previous studies. However, our attempts to reproduce our own dissolution rate experiments were unsuccessful. Instead, we have found that the initial dissolution rate of all diopside powders used in this study decrease with the time elapsed since the sample was ground. This aging effect, measured from samples stored in air and at room temperature, was followed over a period of several months. These results suggest that diopside surfaces created by grinding gradually relax toward a less-reactive state under these conditions. In addition, aging also affects the surface composition (measured using X-ray photoelectron spectroscopy) of the pretreated ground diopside powders. Surface reaction controlled dissolution of silicate minerals is a complicated function of variables both internal and external to the dissolving crystal. The internal variables (e.g. dislocations, structural distortion, microcracks, surface composition, etc.) are affected by mechanical or chemical pretreatments. Various mechanisms by which pretreatments and aging may affect these variables and thereby affect initial dissolution rates and surface composition are discussed. INTRODUCIION THE AQUEOUS DISSOLUTION of silicate minerals has been a subject of active inquiry for many years (e.g. AAGAARD and HELGESON, 1982; HELGESON et al., 1984; VELBEL, 1986; HOCHELLA et al., 1988, and references therein). In most cases the rateof silicate dissolution is controlled by reactions at the solid/solution interface (LAGACHE, 1965; BERNER, 1978; HOLDREN and BERNER, 1979; DJBBLE and TILLER, 198 1). Dissolution rates and mechanisms should therefore be sen- sitive to the structural and compositional nature, on a mo- lecular scale, of the dissolving surface. Dissolution reaction rates have indeedprovento be a complicated functionof many variables, some having to do with the dissolution en- vironment (e.g. the solventand its contents, temperature, pressure, and stirring or flow rate) and others inherent to the crystal (e.g. crystallographic orientation of the dissolving sur- face, impurities within the crystal structure, and other defects such as microcracks, dislocations, etc.). Sample preparation affects the latter “internal factors” (SANGWAL, 1987). Most mineral hydrolysis experiments have used powders produced by grinding, and although artifacts of grinding (and treatments used to remove them)have widely recognized ef- fects on observed dissolution rates for a variety of minerals (HOLDREN and BERNER, 1979; RIMSTIDT and BARNES, 1980; SCHOTT et al., 1981; PETROVICH, 198 la,b; BoLDYREV, 1979), particularlyduring the initial stages of dissolution, our un- derstanding of the nature of these artifacts and their relation to dissolution rate is incomplete. In this regard, the distinction between rate and mechanism may be important in experimental practice. When studying the rate of dissolution reactions that may be occurring in nature, for example, it is important to remove artifacts of sample preparation so that only the “natural”dissolution rateis measured. However, the effects of various pretreat- ments and handling procedures, rather than posing problems, may offer valuable clues as to the dissolution mechanism. The purpose of this paper is to briefly examine the effect of sample preparation and handling on the surface compo- sitionand initialdissolution rate of diopside. In particular, our attempts to reproduce dissolution rate experiments, when replicate experiments were separated by several months, were unsuccessful, suggesting that diopside surfaces created by grinding may gradually relax toward a less reactive state dur- ing storage in laboratory air at room temperature. X-ray photoelectron spectroscopy (XPS) shows that surface com- position also changes with aging in some cases. Several authors have reported an aging effect in non-geo- chemical cases (e.g. TOMPKINS, 1976). The rate of thermal decomposition of lead oxalate depends upon sample age (YEREMIN, 1979).BOLDYREV (1975) attributes this to a de- creasein the number of dislocationsthought to be sites of preferential reaction. SANGWAL (1982) finds that etch pits formed during aqueous dissolution of MgO are smaller at aged than at fresh dislocations, and SANGWAL (1987) reviews literature which reports both increases and decreases in the reactivityof aged dislocations,depending on the nature of impurities segregated along them. COTTRELLand BILBY (1949)have proposeda strain aginglaw for dislocations. An- other purpose of this paper, therefore, is to examine several mechanisms by which aging mightaffect the macroscopic dissolution rate of diopside. METHODS Diopside (CaMgS&O,) was selected for this study because it has been previously studied using XPS, SEM, and resonant nuclear re- action (RNR) techniques in addition to conventional hydrolysis rate experiments (SCHOTT et nl., 198 I; PETIT et al., 1987). Clear diopside was obtained from Pitcaim, New York. Electron microprobe analysis of diopside from this location by sCH0’I-r et al. (198 1) showed near- end-member diopside with 2% of M 1 sites occupied by Fe’+. Optical and SEM examination of our samples revealed minorcalcite and amphibole impurities in amounts too small to be detected by powder X-ray diffraction and showed that the sample preparations, with one exception, removed these impurities (see below).