q 2004 Geological Society of America. For permission to copy, contact Copyright Permissions, GSA, or [email protected]; September 2004; v. 32; no. 9; p. 821–824; doi: 10.1130/G20576.1; 4 figures. 821Metastable prograde mineral reactions in contact aureolesThomas Mu¨llerLukas P. BaumgartnerInstitute of Mineralogy and Geochemistry, University of Lausanne, BFSH2, CH-1015 Lausanne(VD), SwitzerlandC.T. Foster Jr. Department of Geology, University of Iowa, Iowa City, Iowa 52242, USATorsten W. Vennemann Institute of Mineralogy and Geochemistry, University of Lausanne, BFSH2, CH-1015 Lausanne (VD),SwitzerlandFigure 1. Phase diagramof system CaO-MgO-SiO2-H2O-CO2.X(CO2)ismole fraction of CO2inbinary H2O-CO2fluid.Rocks are saturated withcalcite and MgO-richphase (dolomite, brucite,or periclase) and fluid.Note that dashed linemarks metastable reac-tion. Numbers of reactioncorrespond to text. qtz—quartz; tlc—talc; tr—tremolite; cc—calcite;dol—dolomite, di—diopside; fo—forsterite.ABSTRACTExtrapolation of reaction paths and rates of metamorphicmineral growth from experimental to natural systems is compli-cated by a number of factors. Many of these factors are difficultto evaluate for natural systems. A combination of textural model-ing and stable isotope analysis allows for a distinction betweenseveral possible reaction paths for olivine growth in a siliceousdolomite contact aureole. It is suggested that olivine forms directlyfrom dolomite and quartz. The formation of olivine from this meta-stable reaction implies metamorphic crystallization far from equi-librium. Stable and metastable reaction paths predict the texturesobserved (calcite haloes around bladed olivine crystals) well. It ispossible to discriminate between individual reaction paths only onthe basis of the oxygen isotope compositions of the minerals in-volved. Products were found to be in stable isotope equilibrium,but in disequilibrium with the reactants. Only the metastable over-all reaction dolomite 1 quartz → olivine 1 calcite 1 CO2producesno dolomite by local reactions, and hence agrees with the oxygenisotope data. Thus, significant mineral growth occurred far fromequilibrium with respect to the thermodynamically stable reactionsof the system. This amazing finding implies that metamorphism ofcontact aureoles has to be reinterpreted in a more complex, dy-namic fashion, involving metastable reactions and metastable equi-libria as well. The spatial distribution of metamorphic mineral as-semblages in a contact aureole cannot be interpreted as a proxyfor the temporal evolution of a single rock specimen, because eachrock undergoes a different reaction path, depending on tempera-ture, heating rate, and fluid-infiltration rate.Keywords: nonequilibrium mineral growth, reaction path, stable iso-topes, texture model, contact metamorphism.INTRODUCTIONThe occurrence of index minerals in contact aureoles is typicallyexplained as a succession of equilibrium reactions (e.g., Dipple andFerry, 1992; Spear, 1993). The sequence of index minerals in siliceouscarbonates—talc and/or tremolite followed by diopside and/or olivine,and finally periclase—is consistent with the equilibrium phase diagram(Fig. 1). Depending on the rock’s position in the aureole, the sampleis interpreted to have undergone part or all of the reactions given inthe phase diagram. This is a commonly used interpretation, applied tovarious contact aureoles (Kerrick, 1991, and references therein). It hasbeen challenged for these rapid-heating environments by theoreticalcalculations using experimentally determined rate data (e.g., Lasagaand Rye, 1993). Similarly, textural studies (e.g., Holness et al., 1991)have shed doubt on the equilibrium interpretation. Until now the de-termination of reaction paths for the formation of silicates has beenelusive, which makes the application of experimental data to contactaureoles questionable. We demonstrate that reaction paths can be de-termined by combining texture modeling with detailed, grain-scale iso-topic studies. The data presented here have led us to conclude thatolivine production in the Ubehebe Peak contact aureole (Death ValleyNational Park, California) was the consequence of a reaction from themetastable assemblage dolomite 1 quartz. This study presents an ap-proach to distinguish between reaction paths. We propose that the stud-ied aureole is typical, and that metastable reactions are the rule ratherthan the exception in contact aureoles. Hence, the often-implied inter-pretation that the spatial mineral distribution is a proxy for the timeevolution of a sample in the aureole has to be abandoned.GEOLOGIC SETTINGSThe petrology of the Ubehebe Peak contact aureole is well studied(Roselle, 1997). Roselle (1997) estimated pressure-temperature condi-tions for the contact as 1.4–1.7 kbar and 665 8C. Tremolite, forsterite,and periclase were formed in siliceous carbonates in the aureole.Calcite-dolomite solvus thermometry indicates temperatures of 410–440 8C for the tremolite zone and 475–620 8C for the forsterite zone,respectively. Temperatures of 620–665 8C are estimated for the peri-clase zone.Individual lithology horizons can be traced from the contact tovirtually nonmetamorphic regions. The siliceous dolomite sample stud-ied, 02-TUB-48, was collected in the forsterite zone, ;250 m from thecontact of the quartz monzonitic intrusion (Fig. 2). It was chosen be-cause of its simple mineral assemblage of dolomite, calcite, and for-sterite. Calcite is interpreted to be entirely of metamorphic origin, be-cause of its sole occurrence as coronas around olivine, in agreementwith the mostly calcite-free protolith rocks outside the aureole. Cal-culation of the original protolith mineral modes, by using the modalabundance of forsterite and calcite in the investigated sample, resultedin a rock containing 97% dolomite and 3% quartz. These proportionsare in agreement with observed protolith compositions, albeit that thequartz content is somewhat low.ANALYTICAL PROCEDUREThe sample 02-TUB-48 of the forsterite zone was prepared as athick section (300 mm). It was cut into 500 3 500 mm squares byusing a diamond wire saw (130 mm diameter wire), similar to theprocedure described by Kohn et al. (1993). The size of a square waschosen to be smaller than the typical grain size of dolomite crystals(;600–800 mm). Each single square contains only a small number ofgrains. Minerals in individual small squares