USE OF DEHYDRATION EQUILIBRIA TO ESTIMATE VALUES OF MANTLE H2O-ACTIVITY

Determining values of H₂O activity (aH₂O) for mantle rocks will yield a better understanding of those mantle processes that are controlled, in part, by the availability of H₂O (e.g., melting and deformation).  It is possible to utilize a variety of H₂O-buffering equilibria among end-member components in olivine, two-pyroxenes, amphibole and other phases to obtain estimates of aH₂O.  Examples include:  en + 2parg = 4fo + 2di + 2cats + 2jd + 2H₂O (1), 7en + 4parg = 6fo + 8di + 4sp + 2gl + 2H₂O (2), 7en + 4parg = 4di + 4cats + 10fo + 2gl + 2H₂O (3) and 3en + 4mgts + 4parg = 8cats + 10fo + 2gl + 2H₂O (4) (parg=pargasite, jd = jadeite, gl = glaucophane, sp = spinel, and mgts = Mg-Tschermaks).  A self-consistent thermodynamic database (THERMOCALC, Holland and Powell, 1990) can be used to determine values of H₂O from these equilibria as a function of pressure and temperature (P-T).

We used a mantle peridotite assemblage, which includes amphibole, from Dish Hill (sample DH101-E, McGuire et al., 1991) to calculate aH₂O using this approach.  The stability of spinel in this sample limits the P of mineral equilibration to ~ 4 to 28 kbar at temperatures based on 2-pyroxene thermometry (830 to 900^oC).  At P = 20 kbar, dehydration equilibria 1 through 4 yield values of aH₂O that range from .001 to .01 when the activities of end-members in natural phases are determined using non-ideal mixing models (e.g., calculated using the software "AX" written by Tim Holland).  If the activity of tschermakite in amphibole is included, additional reactions make it possible to generate an isobaric invariant point at T = 886^oC and aH₂O = .01 for P = 20 kbar.

A value of aH₂O of ~ 0.014 at 20 kbar and 886^oC was obtained for the Dish Hill sample using the equilibrium between iron oxy-component and hydroxy-component, as described in the dehydrogenation /oxidation reaction Fe²⁺ + OH^- = Fe³⁺ + O^2- + 1/2H₂ (Popp et al., 2006).  Thus, amphibole dehydration equilibria and the iron oxy-hydroxy equilibrium yield similar values of aH₂O, whereas two-pyroxene geothermometry and amphibole dehydration equilibria both yield similar T estimates.  This concordance, among these three approaches, indicates that the estimated value of aH₂O for the Dish Hill samples is reliable, with aH₂O approximately 0.01.