V21A-0588 0800h
Effect of fluid composition on monazite solubility, coarsening and crystal size distribution at 1.0 GPa and $1000\deg$C
The dependence of monazite aqueous solubility and coarsening rate on fluid composition was evaluated by performing experiments in a piston cylinder apparatus at 1.0 GPa and $1000\deg$C. Single crystal weight loss measurements in H2O $\pm$ HCl, NaOH show that the aqueous solubility of monazite increases when the deviation of pH from neutrality at $25\deg$C increases above six units. Monazite is amphoteric, with a U-shaped solubility curve demonstrating high solubility in acidic and alkaline fluids, similar to ambient conditions. Results of growth experiments show that the rate of monazite coarsening in quartzite + fluid is also strongly sensitive to fluid composition. Addition of 2 molal NaCl to H2O completely halted growth, presumably by ion adsorption on monazite surfaces. No modification of the initial crystal size distribution was evident. The strong dependence of monazite coarsening rate on fluid composition suggests that the size of crystals produced by metamorphic coarsening may not be a reliable indicator of duration of metamorphism. Results also suggest that adsorption of ions on mineral surfaces may be an important phenomenon even at high temperatures. Although measurable growth by Ostwald ripening $\pm$ coalescence occurs in 2 m HCl, and monazite solubility is higher in 2 m HCl than in pure H2O, the limited data do not conclusively show a higher growth rate in 2 m HCl than in H2O. Monazite coarsening in the presence of aqueous fluids will produce dateable metamorphic rims unless growth is inhibited by dissolved salts. In the growth experiments the volumetric phase proportion of monazite is constant, so CSD plots measured at different times rotate around a point at small size to form a CSD fan. For growth processes such as Ostwald ripening and coalescence that do not affect the volumetric phase proportion of a phase, CSD slopes and intercepts are not independent, so CSD diagrams cannot be used to characterize the growth process.
http://www.cas.vanderbilt.edu/ayers/research.htm
V21A-0589 0800h
Sr Mobility and Isotope Homogenization During Metamorphic Overprint - an Example From the Cycladic Blueschist Belt
Diversity of age clusters obtained from regional metamorphic areas are very often interpreted as the result of intermittent wide spread infiltration of fluids to have been the cause of "resetting" and "rejuvenation" of age values. On white mica out of impure marbles und carbonate bearing mica schists from the Cycladic blueschist belt (South Evia, Greece) we determine ages varying between 20 and 50 Ma. The fabrics of rocks of the main geological units on southern Evia (Ochi and Styra) were shaped by penetrative deformation under retrograde isothermal (350 $\pm$ $50\deg$C) metamorphic condition during uplift. The age variation even have been detected within the dimension of a thin section challenging the idea of a pervasive fluid which reorganize und homogenize the isotopic Sr composition of the mineral assemblage. Therefore, we measured the Sr isotope composition of Sr rich and Rb poor minerals (mostly calcite and albite, n = 180). Within the dimension of thin sections the Sr isotopic composition is quite uniform (e.g. $^{87}$Sr/$^{86}$Sr = 0.70772 $\pm$ 1). The scatter of Sr composition increases within one lithology (e.g. $^{87}$Sr/$^{86}$Sr = 0.70771 $\pm$ 16). These values found in impure marbles and in calcite poor mica schists refer to the Sr isotope composition of seawater during Permo-Mesozoic time, the period the sedimentation occurred. $^{87}$Sr/$^{86}$Sr values of about 0.704 from calcites in metabasites implying an origin from a MORB source corroborate the obvious preservation of original Sr composition beyond metamorphic overprint. Extensional cracks originated during uplift were filled with epidote which reveals the same Sr isotope composition as calcite. Sr mobilization and replacement were restricted to adjacent minerals. The Sr composition of minerals was acquired very early in the history of the rock probably during the sedimentary stage. Short-term fluid infiltration did not induce complete Sr isotope homogenization on a regional scale. Sr mobilization only occurred locally and is definitely not the reason for varying ages of metamophic minerals in the Styra and Ochi units on South Evia.
V21A-0590 0800h
Field Based Constraints on Reaction Rates in the Crust
Modern research in plate boundary processes involving metamorphism frequently employs complex physical models. Such models require some quantification (or assumption) of the rate at which metamorphic reactions, or chemical exchange, proceed in natural systems. Here, a compilation of available quantitative field-based constraints on high temperature reaction rates will be presented. These include quantifications based on isotopic exchange, porphyroblast and reaction corona growth models, geochronology, and textural analysis. Additionally, natural strain rates provide an important upper bound on simultaneous reaction rates by virtue of a direct mechanistic link between reaction and strain that applies in most situations within the deforming crust. These data show that reaction rates attending regional metamorphism are 4-7 orders of magnitude slower than most laboratory-based predictions. A general rate law for regional metamorphic reactions has been derived which best describes these field-based data: log10(Rnet) = .0029T-9.6$\pm$1, where Rnet is the net reaction rate in g/cm2/yr and T is temperature (C) (Baxter 2003, JGSL). Reaction rates attending contact metamorphism differ from laboratory-based predictions by less than 2 orders of magnitude, and are in closest agreement at higher temperatures. Regional metamorphic reaction rates may be limited by comparatively lesser (or transient) availability of aqueous fluid in the intergranular medium, slower heat input, and smaller deviations from equilibrium. Implications of slow natural metamorphic reaction rates may include a delay in the completion of metamorphic reactions which release (or take in) volatiles, and transform the mineralogy of the crust in dynamic plate boundary settings such as subduction zones.
V21A-0591 0800h
Transient Chemical Equilibration Between Dehydration Fluid and Host Rock: an Efficient way to Mobilize Insoluble Elements
During prograde metamorphism, the fluid released by dehydration reactions is out of equilibrium with respect to the host rock. On its way to chemical equilibration with the rock, the released fluid will get saturated with respect to successive mineral phases (thus likely to crystallize) as predicted by reaction path modeling (Helgeson 1968, 1979). In order to quantify to which extent elements can be mobilized during fluid-rock equilibration and stored in secondary crystallizations, we coupled thermodynamic modeling of reaction path with an experimental study. We carried out closed system isothermal experiments on two simple chemical systems (Al$_{2}$O$_{3}$-SiO$_{2}$-H$_{2}$O and K$_{2}$O-Al$_{2}$O$_{3}$-SiO$_{2}$-H$_{2}$O): ground natural minerals (quartz, kyanite $\pm$ muscovite) and pure water were encapsulated and placed in an Internally Heated Pressure Vessel, at 7 kbar and $350\deg$C or $550\deg$C. We used a tube-in-tube set-up: initial minerals were confined into an inner capsule, which was perforated so that only fluid could circulate towards the external tube. In the external tube, we observed secondary crystallization of: at $350\deg$C, diaspore, kaolinite and pyrophyllite in the ASH system and diaspore, kaolinite and muscovite in the KASH system; at $550\deg$C, kyanite and quartz in the ASH system and Al-rich phase and muscovite in the KASH system. The nature and amount of these secondary phases are in agreement with the sequence predicted by reaction path calculations. In view of its very low solubility in pure water, aluminum was often considered as an immobile element. However, the volume of Al-rich crystallizations in these experiments shows that Al was strongly mobilized during transient fluid-mineral equilibration. Calculated amount of aluminum that crystallizes along the reaction path is 1000 times greater than the aqueous Al concentration in the fluid itself. Thus in natural rocks, transient equilibration between dehydration fluid and host rock is a potential efficient mechanism for mobilizing insoluble elements such as Al. This amount of Al potentially mobilized could contribute to aluminosilicate crystallization in metamorphic veins. Fluid-rock ratios calculated from the abundance of aluminosilicates in veins, considering Al as immobile, should take into account this potential mobilization. Moreover, mass transfer in the experimental closed system was probably largely effected by diffusion. Diffusion thus deserves consideration as a potential mass transfer mechanism in natural rocks. Finally, the tube-in-tube set-up is a way to explore reaction path beyond the P-T range of thermodynamic data application.
V21A-0592 0800h
Crack Healing in Quartz: Influence of Crack Morphology and pOH$^{-}$
Crack healing in quartz has been investigated by optical microscopy and interferometry of rhombohedral {\it r}-cleavage cracks in polished Brazilian quartz prisms that were hydrothermally annealed. Quartz prisms were pre-cracked at room temperature and then annealed at temperatures T of $250\deg$ and $400\deg$C for 2.4 to 240 hours, fluid pressure P$_{f}$ = 41 MPa (equal to confining pressure P$_{c}$), and varying pOH$^{-}$ (from 5.4 to 1.2 at $250\deg$C for fluids consisting of distilled water and NaOH solutions). Crack morphologies before and after annealing were recorded for each sample in plane light digital images and apertures were determined from interference fringes recorded using transmitted monochromatic light ($\lambda$ = 598 nm). As documented in previous studies ({\it Smith and Evans}, 1984; {\it Brantley} et al., 1990; {\it Beeler and Hickman}, 1996), crack healing of quartz is driven by reductions in surface energy and healing rates appear to be limited by diffusional solute transport; sharply defined crack tips become blunted and break up into fluid-filled tubes and inclusions. However, fluid inclusion geometries are also observed with nonequilibrium shapes that depend on initial surface roughness. Crack healing is significant at $400\deg$C after short run durations (24 hr) with healing rates reaching 10$^{-5}$ mm/s. Crack healing is also observed at T=$250\deg$C, but only for smooth cracks with apertures $<$ 0.6 $\mu$m or for cracks subject to low pOH$^{-}$. The extent of crack healing is sensitive to crack aperture and to hackles formed by fine-scale crack branching during crack growth. Initial crack apertures appear to be governed by the presence of fine particles, often found in the vicinity of hackles, which maintain the separation of crack surfaces. Where rough cracks exhibit healing, hackles are sites of either enhanced or reduced loss of fluid-solid interface depending on slight mismatches and sense of twist of opposing crack surfaces. Hackles of open {\it r}-cleavage cracks are replaced either by (1) healed curvilinear quartz bridges and river patterns surrounded by open fluid-filled crack, or by (2) fluid-filled tubes surrounded by regions of fully healed quartz. For a given temperature, aperture, and anneal time, crack healing is enhanced for samples annealed in NaOH solutions compared with healing for samples annealed in water. Accelerated crack healing rates at low pOH$^{-}$ are interpreted to result from increased rates of diffusive mass transport that depend, in turn, on higher silica concentrations of the fluid.
V21A-0593 0800h
Rheology of fine-grained siliciclastic rocks deforming in the middle crust
Geomechanical models form an essential basis part our quantitative understanding of tectonic processes. In these models, a long-standing problem involves the quantification of the constitutive equations that describe the rheology of the middle crust (7-12 km). A combination of indirect methods has yielded first order descriptions of the rheology of these rocks, but much is still unknown and controversial. Constraints on rock rheology are needed from careful field and observational studies and mechanical modelling. Here we present a new method to quantify the rheology of fine-grained siliciclastic rocks, which are common in the middle crust at around 400 degrees C and deform by solution-precipitation processes. We use a combined structural and numerical analysis of the mullion structures developed in these rocks, to quantify the rheological parameters during flow at geologic strain rates. The method is based on a parameter estimation scheme developed in structural mechanics. Results consistently converge towards a set of rheological properties which are in agreement with observed microstructures and indicate that fine grained siliciclastic rocks in the middle crust have a Newtonian viscous rheology, approximately ten times weaker than wet quartz. Because siliciclastic rocks control the rheology of the middle crust in many sedimentary basins, our results provide new parameters for geodynamic modelling.
V21A-0594 0800h
Geochemical constraints on the origin of serpentinization of oceanic mantle
The lower seismic zone of double seismic zones in subducting oceanic lithosphere is suggested to be a result of serpentine or chlorite dehydration in the lithospheric mantle (Hacker et al., 2003). However, the mechanism by which oceanic lithospheric mantle is serpentinized is unclear. One way is through hydrothermal circulation where the lithospheric mantle represents part of the circuit through which seawater passes and then returns to the ocean. Another way is to inject seawater into the lithospheric mantle through fractures in the overlying crust without having a return path of water to the ocean. The two mechanisms differ in that the former is an open system process whereas the latter is a closed system process in which the mantle serves as a sponge for water. Identifying the dominant process is important. For example, if the mantle is part of a hydrothermal circulation cell, the interaction of seawater with the mantle will influence the composition of seawater. This also has important implications for the heat flow out of seafloor. On the other hand, if serpentinization occurs by a closed system process, there will be no influence on seawater composition. Previous studies have suggested that serpentinization of ophiolite bodies was an isochemical process, hence closed system, but it was not clear in these studies whether serpentinization occurred in situ in the oceanic lithosphere. To better understand serpentinization processes in the oceanic lithosphere, we investigated a continuous transition zone of relatively unaltered harzburgite to completely serpentinized harzburgite in the Feather River Ophiolite in northern California. These samples are highly enriched in Na, K, Rb, Cs, U, and Sr, which strongly suggests that serpentinization occurred while the oceanic lithosphere was beneath the ocean. All samples (n=19) have Al2O3 contents ranging from 0.6 to 2.5 wt.% and have extremely depleted light rare-earth element abundances, indicating that these samples are cpx-free harzburgites, which have experienced roughly 20 to 35% melt extraction. The degree of serpentinization was quantified using the concentration of magnetite, a by-product of serpentinization. The lack of antigorite suggests that serpentinization occurred at temperatures lower than 300 C. By comparing Cr and Cr/Al systematics to that predicted from theoretical partial melting calculations and empirical relationships in unaltered peridotite xenoliths, it is shown that Cr and Al are immobile. Al content was thus used to determine the composition of the protolith, which allows us to estimate the amount of depletion/enrichment of a given element by processes other than melt depletion. Most of the harzburgites show no evidence for mantle metasomatism as evidenced by extreme depletions in LREE elements. Consistent with previous studies, we find no depletions in Mg, Fe, or Ca. As seawater is undersaturated in Mg-bearing minerals, an open system process would yield progressive depletion of Mg as is seen in abyssal peridotites, which have been weathered by seawater at the bottom of the seafloor (e.g., Snow et al. 1995). Collectively, this suggests that, except for the addition of seawater and its constituents, serpentinization of the Feather River Ophiolite, was a closed system process. By combining these observations with the results of our field mapping project, we suggest that serpentinization of the lithospheric mantle occurs by local introduction of seawater through fractures extending from the crust and into the mantle. We find no evidence that serpentinized zones in oceanic lithospheric mantle represents an extremely deep hydrothermal circulation cell.
V21A-0595 0800h
Consequences of adiabatic decompression melting on magmatic channeling instabilities
In many partially molten regions of the earth (e.g. beneath mid-ocean ridges), the solid mantle upwells and melts adiabatically due to decompression. Observations of replacive dunites suggest that the melts react with the upwelling solid to form dissolution ``channels'' which may have widths and spacings of 0.1--100 m or larger (e.g. Braun and Kelemen, G-cubed 2002). Previous work (Spiegelman et. al, 2001) has explored reactive channeling instabilities in a static framework. Here we extend the problem to consider how including solid upwelling may change these results. We explore a simplified 2-D column model for highly reactive systems that includes mantle upwelling and both adiabatic and reactive melting. We present both the linear stability analysis for this solution as well as numerical results for fully non-linear channel formation. Linear analysis provides information on the growth rate and spacing of the channels as a function of two principal parameters, $w_0/W_0$, the melt velocity relative to the solid upwelling velocity and $h/\delta$, the height of the system in compaction lengths. The compaction length $\delta$ is controlled principally by the solid viscosity. In particular, we show that there exists a critical value of both parameters beyond which no channels grow. Using the linear analysis, we produce a phase diagram of numerical solutions of the full non-linear problem and show that they agree well with the predictions of the linear analysis but saturate to a steady state. We then derive an approximate analytic solution for the steady-state that allows us to estimate the strength of channeling as well as the depth and width of the channels as a function of the fundamental parameters. Both the linear analysis and the numerical model suggest that this instability could be a viable mechanism for flow localization in the earth. Moreover, these results suggest that, all else being equal, this mechanism is more likely at fast spreading ridges than slow spreading ridges.
V21A-0596 0800h
The Effects of Strain and Strain Rate on the Spatial Separation/Segregation of Olivine and Orthopyroxene in a Synthetic Harzburgite
The plastic rheology of polyphase aggregates incorporates the grain-matrix deformation of the component phases plus grain- and solid-state phase-boundary sliding. The grain- and phase-boundary sliding acts in kinetic series with the grain-matrix mechanisms; in rheologies involving dislocation deformation, the boundary sliding component is almost never rate-limiting [e.g., Crossman and Ashby, 1975]; on the other hand, for diffusional rheologies, the boundary sliding can be both mechanically dominant and rate-limiting. For a given set of thermodynamic (e.g., temperature, pressure, deviatoric stress) and microstructural (e.g., grain size, lattice-preferred orientation) conditions, the energy-dissipation process for the deforming polyphase aggregate must involve the strain-effected separation of the phases, based upon the relative sliding (effective) viscosities of the grain boundaries and phase boundaries. The periodicity of the phase separation should be a function of the strain rate (or, conversely, for a given set of potentials, the aggregate strain rate will be one manifestation of the phase periodicity). We are engaged in an experimental study of the phase-separation scaling physics involved in the solid-state deformation of harzburgite. Specifically, we have prepared synthetic aggregates consisting of a 50:50 (by weight) mixture of ferromagnesian olivine and orthopyroxene, employing pulverized natural material; the hot-pressed aggregates have a grain size of approximately 5 micrometers. For the conditions employed in our experiments (Griggs molten salt confining-medium apparatus in simple shear ; 17 kb; $1200\,^{\circ}$C; 10$^{-5}$-10$^{-4}$ s$^{-1}$; strains of 3-4), the aggregates deform by boundary diffusional creep, conditions that specifically interrogate the relative viscosity of ol-ol and opx-opx grain boundaries and ol-opx phase boundaries. Backscattered electron imaging is employed to characterize the morphology of phases before and after deformation.
V21A-0597 0800h
Stress-Driven Melt Segregation and Organization in Partially Molten Rocks I: Experimental Observations of Coupled Evolution of Melt Distribution and Rheological Properties
In this series of three papers, we present an experimental study on synthetic partially molten olivine-dominated rocks deformed at high temperatures and pressures, equivalent to several kilometers into the Earth's mantle. During progressive deformation of samples of olivine + MORB, olivine + FeS melt + MORB, and olivine + chromite + MORB, an initially homogeneous melt distribution evolves into well-defined networks of melt-rich bands or channels. These experiments demonstrate the effectiveness of deviatoric stress as a driving force for melt segregation and organization. To explore the dynamics of this process (and ultimately to understand its importance for melt extraction in the Earth and planets), we demonstrate several fundamental observations: (1) With increasing strain (or time), melt progressively segregates and organizes into anastomosing networks of channels not unlike braided streams in two dimensions. (2) The rate of melt-segregation and formation of the melt-rich networks appears to depend on compaction length, a length scale combining several two-phase transport properties and applied stress, to which the compaction length is coupled through stress-dependent viscosity. (3) The characteristic morphology of the melt-rich networks also appears to depend very strongly on applied stress. We demonstrate these phenomena using several statistical descriptions of the melt distribution combined with rheological data. The bulk rheological properties are strongly influenced by the segregation of melt and demonstrate complex relationships between strain partitioning and deformation mechanisms. In the companion papers, we present further aspects of the same data set to provide a global picture of the coupling of melt segregation and strain partitioning into the melt-rich, and thus relatively weak, networks.
V21A-0598 0800h
Stress-Driven Melt Segregation and Organization in Partially Molten Rocks II: Strain Partitioning and Crystallographic Preferred Orientations
Following the previous paper on the evolution of melt distribution in deforming partially molten rocks, here we present olivine crystallographic preferred orientation (CPO) data from the same set of experimental samples. CPO data provide relatively direct information on the mechanisms of deformation in the sample, temporally and spatially averaged. We compare olivine fabrics from deformed samples of olivine alone, olivine plus homogeneously distributed MORB, and olivine plus segregated MORB. The differences from one pattern to the next correspond directly to changes in the melt distribution, implying that the mechanism of melt redistribution is closely coupled to the mechanisms of deformation in the sample. The history of loading to which the sample is subjected (i.e., constant stress, strain rate, or force boundary conditions) also has a significant effect on the CPO, a fundamental observation that will help us understand the interactions between grain scale deformation mechanisms and macroscopic strain partitioning. To explain correspondence between the melt distribution, rheological, and CPO data sets, we propose a cyclic pumping mechanism for the constant re-organization of melt in the sample, considering the mechanical interactions of the melt-poor lenses and the melt-rich bands. The idea for this pumping mechanism derives from calculations of the steady-state dissipation in a partially molten system. The difficult but essential problem is to understand the complex scaling relationships between different aspects of the process so that we can estimate its importance as a melt-extraction mechanism in the Earth's mantle.
V21A-0599 0800h
Stress-Driven Melt Segregation and Organization in Partially Molten Rocks III: Annealing Experiments and Surface Tension-Driven Redistribution of Melt
As discussed in the two previous abstracts in this series, simple shear experiments on synthetic upper mantle-type rock samples reveal the segregation of melt into melt-rich bands separated by melt-depleted lenses. Here, we present new results from experiments designed to understand the driving forces working for and against melt segregation. To better understand the kinetics of surface tension-driven melt redistribution, we first deform samples at similar conditions (starting material, sample size, stress and strain) to produce melt-rich band networks that are statistically similar. Then the load is removed and the samples are statically annealed to allow surface tension to redistribute the melt-rich networks. Three samples of olivine + 20 vol% chromite + 4 vol% MORB were deformed at a confining pressure of 300 MPa and a temperature of 1523 K in simple shear at shear stresses of 20 - 55 MPa to shear strains of 3.5 and then statically annealed for 0, 10, or 100 h at the same P-T conditions. Melt-rich bands are fewer in number and appear more diffuse when compared to the deformed but not annealed samples. Bands with less melt tend to disappear more rapidly than more melt-rich ones. The melt fraction in the melt-rich bands decreased from 0.2 in the quenched sample to 0.1 in the sample annealed for 100 h. After deformation, the melt fraction in the melt-depleted regions are ~0.006; after static annealing for 100 h, this value increases to 0.02. These experiments provide new quantitative constraints on the kinetics of melt migration driven by surface tension. By quantifying this driving force in the same samples in which stress-driven distribution occurred, we learn about the relative kinetics of stress-driven melt segregation. The kinetics of both of these processes must be scaled together to mantle conditions to understand the importance of stress-driven melt segregation in the Earth, and to understand the interaction of this process with melt-rock reaction-driven processes.