M. Kochmuller, EXPERIMENTALLY DETERMINED FE-MG EXCHANGE BETWEEN SYNTHETIC STAUROLITEAND GARNET IN THE SYSTEM MGO-FEO-AL2O3-SIO2-H2O, Lithos, 41(1-3), 1997, pp. 185-212
The Fe-Mg exchange between synthetic staurolite and garnet was determi
ned experimentally as a function of pressure, temperature and bulk com
position in the range of 5-25 kbar/600-750 degrees C on the basis of t
he reaction 1/3 pyrope + 1/4 Fe-staurolite = 1/3 almandine + 1/4 Mg-st
aurolite (Eq. (1)). The results of the exchange runs clearly indicate
that the Fe-Mg fractionation between staurolite and garnet does not de
pend much on pressure but on temperature and bulk composition. The com
positional data on coexisting grt(ss) and st(ss), as determined by mic
roprobe, were processed by a nonlinear least-squares procedure using d
ifferent mixing models for staurolite solid solutions and the asymmetr
ic solution model for garnet (Hackler and Wood, 1989). In all the mode
ls, staurolite solid solutions show large negative deviation from idea
lity. The experimental results are best described by a temperature-dep
endent symmetric mixing model with W-s(st) = -79 J/K (one-site model).
Such a large negative W-S-value, however, is rather unrealistic, espe
cially considering, that only a relatively small temperature interval
has been investigated. Therefore a symmetric mixing model with W-G(St)
= (-11539 +/- 830) J (one-site model) is recommended for use in therm
odynamic calculations. The expression for the derived geothermometer i
s based on the temperature-dependent mixing parameters, because these
describe the Fe-Mg fractionation between staurolite and garnet very we
ll. The geothermometer is valid for staurolites with x(Fe) > 0.6 and i
s T = [-18.02 - 0.044 P + RT ln gamma(Fe)(grt) - ln gamma(Mg)(grt) ((-87.78) + 0.057P)*(x(Fe)(st))(2)-(X-Mg(st))(2)) + 12.55(x(Ca)(grt)
+ x(Mn)(grt))]/[-R10(-3)*ln K-D - 0.021 + ((-0.079)*((x(Fe)(st))(2)-(
x(Mg)(st))(2)))], where pressure (P) is in kbar and temperature in deg
rees C. (C) 1997 Elsevier Science B.V.