Frontiers in Geofluids

By Craig Manning

Frontiers in Geofluids is a collection of invited papers chosen to highlight recent developments in our understanding of geological fluids in different parts of the Earth, and published to mark the first ten years of publication of the journal Geofluids. The scope of the volume ranges from the fundamental properties of fluids and the phase relationships of fluids encountered in nature, to case studies of the role of fluids in natural processes. New developments in analytical and theoretical approaches to understanding fluid compositions, fluid properties, and geological fluid dynamics across a wide range of environments are included. A recurrent theme of research published in Geofluids is the way in which similar approaches can be applied to geological fluids in very different settings and this is reflected in the diverse range of applications of fluid studies that are included here. They include deep groundwater flow, hydrocarbons in faulted sedimentary basins, hydrothermal ores, and multiphase flow in mid-ocean ridge systems. Other topics covered are geothermal waters, crustal metamorphism, and fluids in magmatic systems.

The book will be of great interest to researchers and students interested in crustal and mantle fluids of all sorts.

• Language: English
• Publisher: Wiley
• Release date: Mar 10, 2011
• ISBN: 9781444394894

Book preview

INTRODUCTION

Frontiers in geofluids: introduction

This set of papers was originally published electronically as a special double issue of Geofluids to mark the tenth anniversary of the launch of the journal. For this volume, we sought to bring together a collection of papers spanning a range of topics to which the role of fluids in the Earth is central. Geofluids was founded to help emphasise the common ground between fluid processes that take place in different geological settings, and to provide an outlet for research that considers the interactions of chemical and physical processes. While we cannot pretend to provide a comprehensive coverage of all the important recent advances, we are delighted to have been able to bring together some excellent and wide-ranging new science that continues in this tradition.

The first four articles all concern our fundamental theoretical and experimental understanding of essentially aqueous fluids. Liebscher provides an overview of the properties of water-rich fluid systems and how these are affected by solutes, while noting the remaining limitations in the experimental database. Dolejs and Manning present the first comprehensive study to produce a more flexible alternative to the HKF model for aqueous electrolytes, better suited to the range of compositions and conditions encountered in nature, while Sherman shows how modern computational power means that some fundamental problems in natural fluid chemistry can be addressed from first principles using quantum chemistry and molecular dynamics. In the final article in this section, Newton and Manning review recent experimental results for lower crustal conditions and present new data to quantify the importance of dissolved salts for the solubility of the major rock-forming elements, Si and Al, and for a range of important Ca-minerals.

The second group of articles relate to a specific geological setting where fluid processes are of the highest importance: sedimentary basins. Parnell provides a concise review of the use of hydrocarbon fluid inclusions to understand the evolution of reservoirs through time and the relationships between fluid stages and mineral cements. He shows in particular how this approach has contributed to understanding the oil charge history of the North Sea and UK Atlantic margin. Hanor and Mercer describe the behaviour of saline waters and their distribution in the Gulf of Mexico, and show how salinity differences arising through salt dissolution can dictate flow patterns. They also explore the likely impacts of salt on the potential of the region as a source of methane hydrates. The article by Manzocchi, Childs and Walsh reviews how faults affect the flow of fluids, in particular hydrocarbons, in siliciclastic basins, and also comment on the extent to which current industry practice for evaluating the effects of faults is actually grounded in science.

A third group of article deals with fluid processes in oceanic settings. Saffer has modelled the lateral variations along the Nankai margin of Japan and shown that large scale variations along strike in the taper angle of the accretionary wedge can be linked back to lithological variations from more turbidite-rich sequences to mudrocks. The lithology affects the development of fluid overpressure and the draining of the subduction zone fault, which in turn influences the overall geometry of the wedge. The interplay between permeability, heat flow and discharge characteristics at mid-ocean ridges is explored by Driesner. His results support some findings from terrestrial geothermal systems: high temperature discharges, and the highest fluid salinities, may be associated with low fluid fluxes, while large discharges at relatively low temperatures may in fact dominate the removal of heat. Fisher and Harris take three specific examples of mid-ocean ridge settings to explore the controls on heat loss. The relative importance of conductive heat loss is variable, and specific features of the basement geology can serve to target fluid flow and hence heat loss. Hydrothermal vents are also of likely significance for both abiotic and metabolic organosynthesis and this is explored by Shock and Canovas. Different patterns of mixing of seawater with different hydrothermal fluids can lead to different evolutionary paths, but in general, the mixing favours formation of organic compounds from inorganic reactants. Hence, microbes could produce components of biomolecules simply by catalysis of reactions that are already energetically favoured.

A fourth group of articles deals with the continental crust. Ingebritsen and Manning present a crustal-scale overview of permeability and argue that while there is a power-law relation between permeability and depth in tec-tonically active continental crust, some regions exhibit markedly higher permeabilities, probably as transients, while stable crust may decay to lower permeability. The specific issue of the relationship of hydrologic response to earthquake activity is discussed by Wang and Manga. They demonstrate that, in the intermediate and far-field, changes in groundwater flow are linked to changes in permeability which arise in response to cyclic deformation and oscillatory flow. The relationships between faulting and flow at depth is explored by Cox, who shows how fluid pressure and stress influence failure modes and hence the styles of permeability enhancement and vein development, in both mineralized and unmineralized systems. Fluid flow coupled to deformation often introduces water into high grade crystalline basement rocks which undergo retrogression. Yardley, Harlov and Heinrich present the results of experiments designed to measure the rate at which high grade rocks undergo retrogression under mid- to lower-crustal conditions, and conclude that water infiltrated along fine cracks is likely to be rapidly consumed. The article by Bucher and Stober addresses deep groundwaters found in crystalline basement rocks today by tunnelling and drilling. They argue that in areas of high relief such as the Alps, such waters are of relatively low TDS because of flushing by meteoric water, whereas much more saline brines may evolve where the hydraulic gradients are less. Migration of fluids can lead to mineralogical and chemical changes (metasomatism) in a wide variety of crustal settings, and Putnis and Austreheim explore some diverse examples of metasomatism on a range of scales. They are able to demonstrate that, while aqueous fluids partly act as a catalyst to permit minerals to react, they can also influence the course of the reaction through a thermodynamic role.

The final section comprises two interdisciplinary articles that deal with ore deposits and draw on a range of aspects of fluids. Boiron, Cathelineau and Richard review the fluid systems that give rise to ore deposits near the unconformity between sedimentary basins and their underlying crystalline basement. They contrast the Proterozoic unconformity uranium deposits with younger base metal deposits that develop in similar settings, and conclude that there are many similarities in both the nature of the fluids and the flow patterns that give rise to mineralization. Kamenetsky and Kamenetsky evaluate fluid processes at the other temperature extreme of ore formation, associated with magmatism. They present evidence from inclusions to document the development of immiscibility as magmas cool, and evaluate the importance of immiscibility for magma chamber processes, including degassing and the partitioning of metals.

Although we have grouped the articles for convenience, we believe that the true value of the collection arises from the basic new data presented, from the insight into the interactions between physical and chemical processes, and from the opportunity they provide to take ideas developed in a field where particular types of observation or measurement may be possible to understand processes in different settings or at different times, where different types of data may be available.

G. GARVEN¹, C. E. MANNING ² and B. W. D. YARDLEY³

¹Department of Geology, Tufts University, Medford, MA,

USA; ²Department of Earth and Space Sciences, University of California, Los Angeles, CA, USA;

³School of Earth and Environment, University of Leeds, Leeds, UK

Corresponding author: B. W. D. Yardley

School of Earth and Environment,

University of Leeds,

Leeds LS2 9JT, UK.

Email: B. W. D. [e-mail].

Tel: [phone]. Fax: [phone].

REVIEW

Aqueous fluids at elevated pressure and temperature

A. LIEBSCHER

Centre for CO2 Storage, Helmholtz Centre Potsdam, German Research Centre for Geosciences GFZ, Telegrafenberg, Potsdam, Germany

ABSTRACT

The general major component composition of aqueous fluids at elevated pressure and temperature conditions can be represented by H2O, different non-polar gases like CO2 and different dissolved metal halides like NaCl or CaCl2. At high pressure, the mutual solubility of H2O and silicate melts increases and also silicates may form essential components of aqueous fluids. Given the huge range of P–T–x regimes in crust and mantle, aqueous fluids at elevated pressure and temperature are highly variable in composition and exhibit specific physicochemical properties. This paper reviews principal phase relations in one- and two-component fluid systems, phase relations and properties of binary and ternary fluid systems, properties of pure H2O at elevated P–T conditions, and aqueous fluids in H2O–silicate systems at high pressure and temperature. At metamorphic conditions, even the physicochemical properties of pure water substantially differ from those at ambient conditions. Under typical mid- to lower-crustal metamorphic conditions, the density of pure H2O is ρH2O = 0.6–1.0gcm−3, the ion product Kw = 10−7.5 to approximately 10−12.5, the dielectric constant ε = 8–25, and the viscosity η = 0.0001–0.0002 Pa sec compared to ρH2O = l.0gcm−3, Kw = 10−14, ε = 78 and η = 0.001 Pa sec at ambient conditions. Adding dissolved metal halides and non-polar gases to H2O significantly enlarges the pressure–temperature range, where different aqueous fluids may co-exist and leads to potential two-phase fluid conditions under must mid- to lower-crustal P–T conditions. As a result of the increased mutual solubility between aqueous fluids and silicate melts at high pressure, the differences between fluid and melt vanishes and the distinction between fluid and melt becomes obsolete. Both are completely miscible at pressures above the respective critical curve giving rise to so-called supercritical fluids. These supercritical fluids combine comparably low viscosity with high solute contents and are very effective metasomatising agents within the mantle wedge above subduction zones.

Key words: fluid–fluid interactions, fluid phase relations, fluid properties, fluid systems, metamorphic fluids, supercritical melts/fluids

Received 23 July 2009; accepted 14 March 2010

Corresponding author: Axel Liebscher, Centre for CO2 Storage, Helmholtz Centre Potsdam, German Research

Centre for Geosciences GFZ, Telegrafenberg, D-14473 Potsdam, Germany.

Email: [e-mail]. Tel: [phone]. Fax: [phone].

Geofluids (2010) 10, 3–19

INTRODUCTION

Aqueous fluids play a fundamental role in the geochemical evolution of the Earth. By transport and recycling of volatile components and different solutes from the atmosphere and hydrosphere through the solid interior of the Earth and back to the surface they chemically link the different spheres of the Earth on the local, regional and also global scale. As a mobile phase, they can transport heat very efficiently by convection and contribute to the local and regional heat budget and heat distribution. The highly variable P–T–x regimes at the surface, in the crust and in the mantle generate equally variable, characteristic fluids with specific compositions and specific physical properties like density and viscosity. Aqueous fluids at elevated pressure and temperature conditions may form by infiltration of meteoric waters in geothermal and basinal systems (e.g., Hanor 1994; Arnórsson et al. 2007), by diagenetic reactions with connate fluids (Hanor 1994), by infiltration of seawater in oceanic hydrothermal systems (e.g., German & Von Damm 2003; Foustoukos & Seyfried 2007), by prograde metamorphic dehydration and decarbonation reactions as response to changes in P–T–x conditions (e.g., Yardley & Graham 2002; W. Heinrich 2007), and by liberation from crystallizing magmas (e.g., Cline & Bodnar 1994; C.A. Heinrich 2007). Due to the different fluid sources and different geologic environments with notably different P–T–x regimes, the composition of these aqueous fluids is highly variable. Meteoric fluids are almost pure water, which may gain considerable amounts of dissolved salts during diagenetic reactions. The salinity of aqueous basinal fluids ranges over five orders of magnitude from a few ppm in shallow meteoric regimes to more than approximately 40 wt% in evaporite-rich basins (Hanor 1994). In oceanic hydrothermal systems, the dominant fluid source is seawater, which can be modelled as a 3.3 wt% aqueous NaCl equivalent solution (NaCleq; Bischoff & Rosenbauer 1984). However, by interaction with the oceanic crust, the salinity of the generated hydrothermal fluids may range from only approximately 0.2 up to approximately 7.3 wt% NaCl (e.g., Oosting & Von Damm 1996; Lüders et al. 2002). Prograde metamorphic dehydration reactions form highly variable saline fluids that may even reach salt saturation (Fig. 1B; Yardley & Graham 2002). However, the salinity of metamorphic fluids depends only little on metamorphic grade but is strongly linked to the protolith’s original setting. Metamorphism of oceanic or accretionary prism protoliths generally forms fluids with salinity below approximately 6 wt% NaCleq, whereas metamorphism of rocks from shallow marine or continental margin origin forms fluids that span the complete range from almost salt free up to approximately 60 wt% NaCleq (Fig. 1B; most salinity data for metamorphic fluids come from fluid inclusions and are estimated from the melting temperature of ice. They are usually given as ‘equivalent’ concentration of NaCl or NaCleq, referring to the salinity of a NaCl solution that would yield the same melting temperature of ice as the measured fluid inclusion). Metamorphic decarbonation reactions, triggerd by changes in pressure and temperature or by an infiltrating fluid itself, potentially supply CO2 to the system, force the fluids to un-mix and generate aquo-carbonic fluids co-existing with variable saline H2O–salt fluids (e.g., Trommsdorff et al. 1985; Skippen & Trommsdorff 1986; Trommsdorff & Skippen 1986; W. Heinrich 1993, 2007). H2O–salt fluids are also reported from fluid inclusions in diamonds from the upper mantle (Navon et al. 1988; Izraeli et al. 2001). But at high pressure, also the solubility of silicates in aqueous fluids notably increases and silicates may form important solutes at these conditions. This may even lead to complete miscibility between aqueous fluids and silicate melts (e.g., Shen & Keppler 1997; Hack et al. 2007) and individual components continuously change their character from low, typical solute-like concentrations to high, typical solvent-like concentrations; finally, any major element may become an essential phase component of the fluid.

This review aims at presenting some of the fundamental topics of aqueous fluids at elevated pressure and temperature. Given the vast number of different geological environments in which fluids evolve, each of which is unique in space, time and physicochemical properties, such a review is necessarily incomplete and the reader is kindly referred to the references given for more and detailed information on specific aspects of aqueous fluids. The review first describes the principal phase relations in one- and two-component fluid systems. Then, phase relations and properties of binary and ternary fluid systems, mostly based on experimental studies, are summarized followed by a review of the properties of pure H2O at elevated P–T conditions. Finally, aqueous fluids in H2O–silicate systems at high pressure and temperature are discussed.

Fig. 1. (A) Schematic drawing showing the different geological fluid systems in crust and mantle. Modified from Liebscher & Heinrich (2007). (B) Compilation of salinity data for metamorphic fluids. Dark grey fields represent protoliths of shallow marine or continental margin origin, white boxes those from oceanic or accretionary prism origin. Hatched boxes indicate high-pressure metamorphic rocks. Redrawn and modified from Yardley & Graham (2002). Lighter grey fields indicate compositional ranges for fluids from magmatic, metamorphic, oceanic and basinal systems as compiled by Kesler (2005).

missing image file

PRINCIPAL PHASE RELATIONS IN ONE- AND TWO-COMPONENT FLUID SYSTEMS

Aqueous geological fluids are rarely pure water at elevated P–T conditions and represent mixtures with several additional major components like salts (here used in the restricted sense of dissolved metal halides), non-polar gases and rock components like silica. In these multicomponent systems, phase relations become more complicated, fluid miscibility and immiscibility play an important role and concentrations of individual components may continuously change from solute-like to solvent-like or vice versa. In the following, the principal phase relations in two-component model fluid systems are described starting from the one-component case as exemplified for H2O (Fig. 2A).

In any one-component system, the three phase states solid (in case of H2O ‘ice’), liquid and vapour (in case of H2O ‘steam’) have identical and fixed composition. All three co-exist at the invariant triple point from which the three univariant solid–liquid, solid–vapour and liquid–vapour equilibria emanate (Fig. 2A). In case of H2O, the triple point is at 0.00061 MPa/0.01 °C. In one-component systems, co-existence of different fluid phases is exclusively restricted to the liquid–vapour equilibrium. However, although identical in composition, liquid and vapour differ in their physical properties like density, viscosity and electric permittivity. These properties show abrupt, discontinuous changes at the first-order liquid–vapour phase transition along the liquid–vapour equilibrium. The differences in physical properties, however, diminish along the liquid–vapour equilibrium towards higher temperature and pressure and finally disappear at the critical point. The critical point of H2O is at 373.95°C/22.06 MPa (Wagner & Pruß 2002). Above critical temperature and pressure, neither changes in temperature nor pressure induce any phase transition and the physical properties of the homogeneous single phase fluid continuously change in response to changes in P and T.

Addition of a second component to a one-component system adds composition as an additional degree of freedom and co-existing fluids not only differ in their physical properties but also in composition. Consequently, the invariant triple point, the univariant liquid–vapour equilibrium and the critical point of the one-component system turn into a univariant solid–liquid–vapour equilibrium, a divariant liquid–vapour field and a critical curve in the two-component system. The P–T range in which two fluids coexist may thereby greatly expand (Fig. 2B). The principal phase relations of two-component fluid systems and how they apply to different H2O–salt systems are shown in Fig. 3. For an in-depth presentation and discussion of the different phase topologies the reader is referred to Ravich (1974) and Valyashko (1990, 2004) and references given therein. A discussion of the principal phase topologies in the context of liquid–liquid immiscibility and the magmatic-hydrothermal transition is given by Veksler (2004). The overall phase topology of H2O–salt systems is determined by the pressure and temperature conditions of the univariant salt–liquid–vapour equilibrium, which emanates from the triple point of the salt endmember systems, relative to the pressure-temperature conditions of the critical curve. In so-called Type 1-systems, with a high and pro-grade solubility of the salt in H2O, the univariant salt–liquid–vapour equilibrium does not intersect the critical curve, which is continuous over the entire P–T–x space and connects the critical points of the pure systems (Fig. 3A, B). Between the univariant salt–liquid–vapour equilibrium and the critical curve a two-phase fluid volume forms in P–T–x space. The univariant salt–liquid–vapour equilibrium may intersect the univariant ice–liquid–vapour equilibrium, which emanates from the triple point of pure H2O, forming an invariant point where liquid, vapour, ice, and salt co-exist (Fig. 3A). In so-called Type 2-systems, with a low solubility of the salt in H2O, the univariant salt–liquid–vapour equilibrium intersects the critical curve at the lower and upper critical endpoints (Fig. 3C). These general Type 1 and 2 phase topologies may get more complicated due to liquid-liquid immiscibility. Such liquid-liquid immiscibility gives rise to a number of principal phase topologies, which are presented and discussed in detail by Valyashko (1990, 2004). Not all of these principal phase topologies, however, have yet been found or are relevant for geological fluid systems. The most simple Type 1a system represents the binaries between H2O and the highly soluble salts like NaCl (e.g., Sourirajan & Kennedy 1962), KCl (Tkachenko & Shmulovich 1992; Dubois et al. 1994), CaCl2 (Tkachenko & Shmulovich 1992; Bischoff et al. 1996) but also NaOH (Urusova 1974) (Fig. 3A). Here, the critical curve is continuous over the entire P–T–x space and no liquid–liquid immiscibility occurs. The effects of liquid-liquid immiscibility are exemplified by Type 1d (Fig. 3B). Here, two separate critical curves form. One critical curve emanates from the critical point of the more volatile component (H2O in Fig. 3B) and shows critical behaviour between vapour and liquid 1, the other critical curve emanates from the critical point of the less volatile component (‘salt’ in Fig. 3B) and shows critical behaviour between liquid 1 and 2. H2O–salt systems of Type 1d include exotic salts like KH2PO4 (Marshall et al. 1981) and are not relevant for geological fluids. The simple Type 2a topology without liquid-liquid immiscibility is representative for binaries between H2O and sparingly soluble second components like SiO2, Al2O3 and other silicates, different sulphates and carbonates (Fig. 3C). Owing to the intersection of the univariant liquid–vapour–salt equilibrium with the critical curve at the lower and upper critical endpoints, lower and uppesegments of the two-phase fluid volume form. In systems in which silicates are the additional second component, the high temperature part of the liquid–vapour–salt equilibrium is identical to the wet solidus and describes the silicate’s water saturated melting behaviour. As the upper liquid–vapour–salt equilibrium or wet solidus terminates at the upper critical endpoint, no discrete melting reaction is possible at pressure above the upper critical endpoint (see below).

Fig. 2. (A) Phase diagram for pure H2O showing the stability fields of the different phase states of H2O. Co-existing fluid phases are restricted to the liquid–vapour (‘l + v’) curve. Above the critical point, only one homogeneous single phase fluid exists. (B) P–T projection of the critical curves for the systems H2O–CO2, H2O–CH4, H2O–CaCl2 and H2O–NaCl. Hatched sides for of the critical curves mark direction of fluid un-mixing. Based on data by Takenouchi & Kennedy (1964) for H2O–CO2, Welsch (1973) for H2O–CH4, Shmulovich et al. (1995), Bischoff et al. (1996) and Zhang & Frantz (1989) for H2O–CaCl2, and Driesner & Heinrich (2007) for H2O–NaCl. Liquid–vapour curve for H2O from Wagner & Pruß (2002).

missing image file

Fig. 3. Schematic drawings of phase relations in water–salt systems of (A) Type 1a, (B) Type 1d and (C) Type 2a as P–T–x presentation and projections onto the P–T and P–x planes. Assignment of different salts to the different systems after Valyashko (2004). Drawn and modified according to Valyashko (1990, 2004).

missing image file

BINARY AND TERNARY FLUID SYSTEMS

The phase relations in binary and ternary fluid systems are simplified representations of natural fluids but form the framework for the more realistic higher component systems. The most important binary systems for aqueous fluids are the H2O–non-polar gas systems H2O–CO2 and H2O–CH4 and the H2O–salt systems H2O–NaCl, H2O–CaCl2 and H2O–KCl. Other components like the non-polar gas N2 and salts like LiCl, SrCl2 or MgCl2 are normally present only at the minor or even trace element level. Combining these binary systems yields the important ternary fluid systems H2O–NaCl–CO2, H2O–CaCl2–CO2 and H2O–NaCl–CH4.

The principal phase relations in the systems H2O–CO2 and H2O–CH4 resemble each other and can be taken as proxies for other H2O–non-polar gas systems (Fig. 4A, B). Starting from the critical point of pure water, their critical curves initially extend towards lower temperature with only minor pressure dependence, then pass through temperature minima and extend to high pressure with only minor temperature increase. The temperature minima occur at 155–190 MPa and approximately 265°C in the H2O–CO2 system (Takenouchi & Kennedy 1964) and approximately 100 MPa and 353°C in the H2O–CH4 system (Welsch 1973). At temperatures below the critical curves, both systems show pronounced opening of the two-phase fluid region towards lower temperature, reflecting the very low solubility of non-polar solutes in water at low temperatures. A detailed and thorough discussion of phase relations within the H2O–CO2 system that also treats the important formation of clathrates at low temperatures is given by Diamond (2001).

Most aqueous fluids contain important amounts of dissolved salts (see above), and water–salt systems therefore have attracted much interest. The H2O–NaCl system has been studied among others by Keevil (1942), Sourirajan & Kennedy (1962), Khaibullin & Borisov (1966), Urusova (1975), Bischoff & Rosenbauer (1984), Bodnar et al. (1985), Chou (1987), Bischoff et al. (1986), Rosenbauer & Bischoff (1987), Bischoff & Rosenbauer (1988), Bischoff & Pitzer (1989), Bischoff (1991) and Shmulovich et al. (1995). Based on the available experimental data, Driesner & Heinrich (2007) derived the most recent representation of the phase topology in the H2O–NaCl system up to 1000°C/220 MPa, and xNaCl = 0–1.0, where x is mole fraction. Their data form the basis for the H2O–NaCl phase relations shown in Fig. 4C. The critical curve in the H2O–NaCl system monotonously extends from the critical point of pure H2O towards higher pressure and salinity with increasing temperature, at least within reasonable crustal P–T conditions. Beside salinity also the density increases along the critical curve with increasing pressure and temperature (Fig. 4D; Urusova 1975; Chou 1987; Bischoff 1991). The data show that at elevated P–T conditions even the vapour in H2O–salt systems is considerably dense and notably differ from vapour at ambient conditions. For the role of dense vapour as an ore-forming fluid of its own the reader is referred to C.A. Heinrich (2007; and references given therein). The system H2O–KCl has been studied by Keevil (1942), Hovey et al. (1990), Dubois et al. (1994) and Shmulovich et al. (1995). The critical curve in the H2O–KCl system as well as the corresponding phase relations closely resemble those of the H2O–NaCl system and the phase relations of natural KCl dominated fluids may be approximated by the H2O–NaCl system (Fig. 4E). However, the liquid–vapour–salt equilibrium is at slightly lower pressure in the H2O–KCl than in the H2O–NaCl system. H2O–alkaline earth salt systems were studied among others by Zhang & Frantz (1989), Tkachenko & Shmulovich (1992), Shmulovich et al. (1995), and Bischoff et al. (1996) for H2O–CaCl2 and Urusova & Valyashko (1983, 1984) and Shmulovich et al. (1995) for H2O–MgCl2. The data indicate that the critical curves in H2O–alkaline earth salt systems are at higher pressure than in the H2O–alkaline salt systems, whereas the liquid–vapour–salt equilibrium is at lower pressure. In H2O–alkaline earth salt systems the P–T range where a low salinity vapour co-exists with high salinity liquid therefore expands compared to H2O–alkaline salt systems.

Fig. 4. Phase relations in the binary systems (A) H2O–CO2, (B) H2O–CH4, and (C) H2O–NaCl; (D) densities in the system H2O–NaCl; (E) P-T projections of the critical curves and liquid + vapour + salt equilibria in the systems H2O–NaCl, H2O–CaCl2 and H2O–KCl showing the limits for co-existing fluids; and (F) hydrolysis reaction in the system H2O–CaCl2. Thin lines in (A), (B), (C), and (E) are isotherms with temperature given in °C. Based on data in (A) from Takenouchi & Kennedy (1964), in (B) from Welsch (1973), in (C) from Shmulovich et al. (1995) and Driesner & Heinrich (2007), in (D) from Urusova (1975), Chou (1987) and Bischoff (1991), in (E) from Keevil (1942), Sourirajan & Kennedy (1962), Ketsko et al. (1984), Chou(1987), Zhang & Frantz (1989), Hovey et al. (1990), Shmulovich et al. (1995), Bischoff et al. (1996), and in (F) from Bischoff et al. (1996).

missing image file

Fluid un-mixing into low salinity vapour and high salinity liquid in water-salt systems not only influences salinity and density of the co-existing fluid phases but may also change the availability of ligands and the acid or basic character of the co-existing fluids. In H2O–salt systems hydrolysis occurs according to the equilibrium

missing image file

At ambient conditions, the equilibrium constant is such that reactant activities are significantly greater than those of the products and the amounts of HCl and missing image file in the fluid are negligible. At elevated P–T conditions, however, the equilibrium constant may change in such a way that product activities are significantly greater than those of the reactants leading to notable amounts of HCl and missing image file . In case of co-existing fluids, HCl preferentially fractionates into the vapour whereas missing image file fractionates into the liquid. This gives rise to an HCl-enriched, potentially acidic vapour and a co-existing missing image file -enriched, potentially basic liquid. This effect of fluid un-mixing has been experimentally studied by Bischoff et al. (1996), Vakulenko et al. (1989) and Shmulovich et al. (2002). Bischoff et al. (1996) analysed experimentally phase separated fluids in the system H2O–CaCl2 for Ca and Cl. Below 25.0 MPa at 400°C and 58.0 MPa at 500°C they observed an increasing ion imbalance between Ca²+ and Cl− in the vapour with Cl− concentrations being notably higher than Ca²+ concentrations (Fig. 4F). This observation reflects formation and preferential fractionation of HCl into the vapour. The amount of HCl produced by this reaction is remarkable and may reach 0.1 mol kg−1 in the vapour.

The phase relations of the two binary H2O–non-polar gas and H2O–salt subsystems form the framework for the phase relations within the ternary H2O–salt-non-polar gas systems. The third binary subsystem salt-non-polar gas can safely be assumed as immiscible at all geologic relevant P–T conditions. All ternary H2O–salt–non-polar gas systems share some common features. Increasing salt concentrations decrease the solubility of the non-polar gas in H2O–salt mixtures ('salting out'; Fig. 5A) and salt concentrations in H2O–non-polar gas mixtures are generally low. The P–T–x range of fluid immiscibility therefore greatly expands and extends to conditions where the binary subsystems may already be completely miscible. Fluid immiscibility may therefore prevail over the entire crustal P–T range (Heinrich et al. 2004). The H2O–NaCl–CO2 system is the most relevant ternary system and has been studied, among others, by Takenouchi & Kennedy (1965), Gehrig et al. (1986), Kotel'nikov & Kotel'nikova (1991), Joyce & Holloway (1993), Shmulovich & Plyasunova (1993), Gibert et al. (1998), Shmulovich & Graham (1999, 2004), Schmidt & Bodnar (2000) and Anovitz et al. (2004). The H2O–CaCl2–CO2 ternary has been experimentally determined by Zhang & Frantz (1989), Plyasunova & Shmulo-vich (1991), Shmulovich & Plyasunova (1993), and Shmulovich & Graham (2004) and the H2O–NaCl–CH4 ternary by Krader (1985) and Lamb et al. (1996, 2002). The principal phase relations that occur in the H2O–NaCl–CO2 but also in the other H2O–salt–non-polar gas systems are exemplified in Fig. 5B based on the data by Anovitz et al. (2004) for 500°C/50 MPa. At these conditions, the H2O–NaCl binary shows liquid–vapour immiscibility and six principal stability fields can be distinguished: (i) At low NaCl concentrations, single phase H2O–CO2 vapour is stable; (ii) at H2O rich compositions, the NaCl poor H2O–CO2 vapour co-exists with a CO2 poor H2O–NaCl liquid, defining a vapour + liquid two-phase field; (iii) at higher NaCl concentrations, a single phase field of H2O–NaCl liquid with only minor amounts of CO2 forms; (iv)–(vi) at high NaCl concentrations the fluids are salt saturated and, depending on the H2O/CO2 ratio, liquid + salt, liquid + vapour + salt or vapour + salt co-exist. With increasing pressure, the miscibility gap within the H2O–NaCl binary closes and the single phase fluid field extend from salt-saturated CO2-poor H2O–NaCl fluids through ternary fluids at H2O–dominated conditions to NaCl poor H2O–CO2 fluids (Fig. 5C). The available data clearly show that the solubility of NaCl in H2O–CO2 vapour is very low and substantially decreases with increasing pressure and temperature. For instance, at 500°C, e.g., xNaCl in H2O–CO2 vapour decreases from about 0.074 at 50 MPa to <0.025 at 500 MPa (Shmulovich & Plyasunova 1993; Anovitz et al. 2004), whereas at 200 MPa a temperature increase from 600 to 800°C decreases xNaCl from about 0.052 to >0.032 (Kotel'nikov & Kotel'nikova 1991). In the ternary systems H2O–NaCl–CH4 and H2O–CaCl2–CO2 the solubility of the salt in H2O–non-polar gas vapour is even less and the field of co-existing fluids greatly expands (Fig. 5C; see also Liebscher 2007). The progression of the phase relations in ternary systems as function of T and x and the effects of opening and closure of two-phase fluid fields are shown for the system H2O–NaCl–CH4 at 50 MPa in Fig. 5D based on the data by Krader (1985). Remembering that increasing pressure favours fluid immiscibility in H2O–salt-non-polar gas systems, it is clear, that single phase fluids are restricted to very narrow P–T–x regimes and that most parts of the geological relevant P–T–x space are characterized by co-existing fluids with different physicochemical properties.

Fig. 5. (A) Salting-out effect in the system H2O–NaCl–CO2 at 150 and 250°C, and phase relations in the ternary systems (B) H2O–NaCl–CO2, (C) H2O–NaCl–CO2 and H2O–CaCl2–CO2 and (D) H2O–NaCl–CH4. Thin lines in (B) and dashed lines in (C) are only schematic to indicate principal phase relations. Data in (A) from Takenouchi & Kennedy (1964, 1965), in (B) from Anovitz et al. (2004), and in (C) from Kotel’nikov & Kotel’nikova (1991) and Zhang & Frantz (1989). (D) Redrawn and modified after Krader (1985).

Numerous equations of state (EOS) exist to describe the pressure–temperature–volume relations in the different fluid systems at elevated pressure–temperature conditions. For a thorough review of the different EOS and the underlying fundamentals, the reader is referred to the reviews by Ferry & Baumgartner (1987) and Gottschalk (2007) and to Prausnitz et al. (1999) and Sengers et al. (2000). EOS for H2O–non-polar gas systems are given among others by Kerrick & Jacobs (1981) and Duan & Zhang (2006) for H2O–CO2, Jacobs & Kerrick (1981) and Duan et al. (1992a,b) for H2O–CO2–CH4, and Saxena & Fei (1987, 1988), Belonoshko & Saxena (1992), Duan et al. (1992c, 1996), Shi & Saxena (1992) and Churakov & Gottschalk (2003a,b) for H2O–CO2-CH4 plus additional phase components. H2O–salt systems are covered by Anderko & Pitzer (1993a,b) for H2O–NaCl–KCl, Jiang & Pitzer (1996) for H2O–CaCl2 and Duan et al. (2006) for H2O–CaCl2 and H2O–MgCl2. EOS for ternary H2O–salt–non–polar gas systems are restricted to those by Duan et al. (1995, 2003) for H2O–NaCl–CO2 and H2O–NaCl–CO2–CH4. According to the authors, some of the EOS for H2O–non-polar gas systems are valid to rather high pressure and temperature conditions, e.g., Duan & Zhang (2006) up to 2300°C/10 GPa and Churakov & Gottschalk (2003a,b) up to 1730°C/10 GPa. Unfortunately, those that include salts are applicable only up to mid-crustal pressure conditions of maximum 500 MPa (e.g., Duan et al. 1995, 2003).

PROPERTIES OF PURE H2O AT ELEVATED P-T CONDITIONS

For the evaluation of the properties of aqueous fluids at elevated pressure and temperature conditions, knowledge of the fundamental properties density, dissociation behaviour, static dielectric constant and dynamic viscosity of pure water at elevated pressure and temperature conditions is of primary importance. The available experimental data as well as formulations for the calculation of these properties at elevated pressure and temperature are summarized below and shown in Fig. 6. Emphasis is given to metamorphic P-T conditions (grey fields in Fig. 6), which are here somewhat arbitrarily defined as P > 100 MPa and T > 150°C. Metamorphic conditions are limited towards low temperature and high pressure by the lowest possible geotherm, which was assumed to 5 MPa/°C. The shown boundaries of the different metamorphic facies are from Spear (1993). Some of the reviewed studies report the respective properties as a function of temperature and density while others use temperature and pressure. In Fig. 6, all properties are plotted as a function of temperature with isobars showing the pressure dependence. For data that were presented as function of temperature and density in the original papers density was converted to pressure by applying the IAPWS formulation for H2O by Wagner & Pruß (2002). For near-critical behaviour of H2O and aqueous systems, see Anisimov et al. (2004).

Density

At ambient conditions, liquid water has a density of approximately 1 g cm−3. With increasing pressure and temperature along the liquid–vapour equilibrium, the density contrast between liquid water and steam diminishes and at the critical point both approach the critical density of 0.322 g cm−3. Numerous EOS and molecular dynamics simulations describe the PVT properties of pure H2O. Among others, these include those by Kerrick & Jacobs (1981), Halbach & Chatterjee (1982), Saul & Wagner (1989), Belonoshko & Saxena (1991) and Wagner & Pruß (2002). At elevated P–T conditions, the density of H2O ranges from approximately 0.3 g cm−3 up to >1.2 g cm−3 (Fig. 6A; calculated with the IAPWS formulation by Wagner & Pruß 2002). Only at low pressure the density may be < 0.3 g cm−3. Increasing temperature generally decreases density whereas increasing pressure has the opposite effect. In low-P/high-T settings as they occur in oceanic hydrothermal systems, volcanic systems and metamorphic systems with very high geothermal gradients H2O densities are below approximately 0.6 g cm−3. Under typical crustal metamorphic conditions, however, the density of H2O ranges between approximately 0.6 and 1.0 g cm−3. Only at rather high pressures, as in subduction zones or within the upper mantle, are conditions such that H2O densities exceed 1.2 g cm−3.

Fig. 6. Density (A), ion product (B), static dielectric constant (C) and dynamic viscosity (D) of pure water as function of temperature. Thin lines are isobars with pressure given in MPa. Shadowed fields represent range of metamorphic conditions with 100 MPa arbitrarily chosen as lower P-limit. Thick grey lines define the different metamorphic facies where SGr = sub-greenschist, Gr = greenschist, B = blueschist, EA = epidote-amphibolite, A = amphibolites, G = granulite and E = eclogite facies (after Spear 1993). Data in (A) from Wagner & Pruß (2002), other data sources given in the Figure. Stippled lines in (D) are eye-drawn through the data points of Abramson (2007).

Self-dissociation

The dissociation equilibrium of water

is described by the equilibrium constant

or by the ion product

At ambient conditions, the dissociation equilibrium of liquid water is strongly shifted to the left hand side, concentrations and activities of H+ and OH− are low and Kw = 1 × 10−14. With pH = −log aH+ this turns into the well known value of pH = 7 for neutral water. At elevated P–T conditions, however, Kw notably changes. The ion product along the liquid–vapour equilibrium has been experimentally determined among others by Bignold et al. (1971), Fischer & Barnes (1972), MacDonald et al. (1973) and Sweeton et al. (1974). Measurements at elevated pressure and temperature conditions above the critical point of water were performed by Franck (1956) at 350–970 MPa/500–1000°C and by Quist (1970) up to 555 MPa and 800°C. Extreme conditions have been studied by Holzapfel & Franck (1966) at 4.5–10.0 GPa/600–1000°C. Formulations for the change of Kw with pressure and temperature are given by Marshall & Franck (1981) up to 1000°C/1000 MPa, Pitzer (1983) for low densities up to approximately 0.1 g cm−3, Tanger & Pitzer (1989) up to 500 MPa/2000°C, and Bandura & Lvov (2006) up to 1.0 GPa/1000°C. Kw as function of pressure and temperature calculated by the formulations of Marshall & Franck (1981) and Tanger & Pitzer (1989) together with selected data by Franck (1956) and Quist (1970) are shown in Fig. 6B. The Marshall & Franck (1981) and Tanger & Pitzer (1989) data agree well for the pressure range between 100 and 500 MPa and reproduce the available experimental data by Quist (1970). However, at pressures below approximately 100 MPa, both formulations diverge and the formulation by Marshall & Franck (1981) predicts notably higher values for Kw than the Pitzer (1983) and Tanger & Pitzer (1989) formulations. In this low pressure, low density region the formulations of Pitzer (1983) and Tanger & Pitzer (1989) are probably more reliable (see discussion in Tanger & Pitzer 1989). Within the pressure range 200–1000 MPa, temperature has only a minor effect on Kw, whereas Kw generally increases with increasing pressure. Under normal mid- to lower-crustal P–T conditions, Kw ranges between approximately 10−7.5 and approximately 10−12.5. However, at very high pressures of the eclogite facies, Kw may even be substantially higher than 10−7.5. In line with this, Holzapfel & Franck (1966) determined Kw = 10−2.8 to 10−1.2 at approximately 7–9 GPa/500–1000°C.

Dielectric constant

The dielectric constant is of primary importance for the evaluation of the solvent properties of water. The attractive or repulsive force between any point charges q1 and q2 in a medium, e.g., ions in aqueous solution, that are separated by distance r is given by

with ε being the electric permittivity of the medium. The electric permittivity of the vacuum is typically designated ε0 and the dimensionless ratio ε/ε0 then yields the static relative permittivity or static dielectric constant for any medium. A high dielectric constant corresponds to a high resistance of the medium to the transmission of an electric field. At ambient pressure and 25°C liquid water has a notably high static dielectric constant of 78.4 (Fernández et al. 1995). This high static dielectric constant makes liquid water a good solvent for charged species at ambient conditions as it minimizes the electrostatic forces between the dissolved ions and prevents them to combine to crystals. However, with increasing pressure and temperature, the static dielectric constant of liquid water substantially decreases. Fernández et al. (1995) presented a database for the static dielectric constant of water and steam. It extends up to a temperature of 873 K and a pressure of 1189 MPa and covers the data available at that time. The static dielectric constant along the liquid–vapour equilibrium has been studied by Oshry (1949), Svistunov (1975), Lukashov (1981), Muchailov (1988) and Mulev et al. (1994). Along the liquid–vapour equilibrium, ε of liquid water continuously decreases from approximately 55 at 100°C to only approximately 9.7 at 370°C, whereas ε of co-existing steam only slightly increases from approximately 1.03 at 150°C to approximately 3.02 at 370°C. Above the critical point, ε has been measured by Heger (1969) and Heger et al. (1980) at 25–500 MPa/400–550°C, Lukashov et al. (1975) at 23–58 MPa/400–600°C, Golubev (1978) at 23–39 MPa/420–510°C, Lukashov (1981) at 24–580 MPa/400–600°C, and Deul (1984) at 30–300 MPa/400°C. Based on the available experimental data, formulations and calculated values for ε up to high P–T conditions are given by Quist & Marshall (1965) up to 1.55 GPa/800°C, Bradley & Pitzer (1979) up to 100 MPa/350°C, Pitzer (1983) up to 880 MPa/927°C, Franck et al. (1990a,b) up to 1.55 GPa/1000°C, Shock et al. (1992) up to 500 MPa/ 1000°C, Wasserman et al. (1995), and Fernández et al. (1997) up to 1.2 GPa/600°C. Fig. 6C reviews the experimental data by Heger (1969) and Deul (1984) together with isobars calculated by formulations of Shock et al. (1992) and Fernández et al. (1997). Also shown are values for the static dielectric constant for selected P–T conditions as calculated by formulations of Pitzer (1983) and Franck et al. (1990a,b). For temperatures up to 1000°C and pressures up to 500 MPa, i.e. the P-range largely covered by experimental data, the different formulations agree quite well. However, at pressures above 500 MPa, the different formulations diverge. At 800°C, e.g., Fernández et al. (1997) predict a static dielectric constant of approximately 14 at 1.0 GPa, whereas Franck et al. (1990b) predict ε ∼ 14 only at notably higher pressure of 1.55 GPa. At slightly higher temperature of 930°C, Pitzer (1983) predicts ε ∼ 11 at 885 MPa, roughly 120 MPa below the prediction of Fernández et al. (1997). This indicates a higher pressure dependence of ε in Pitzer (1983) compared to Fernández et al. (1997) and Franck et al. (1990a,b). Despite these differences at high P conditions, the static dielectric constant of H2O generally decreases with increasing temperature but increases only slightly with increasing pressure. Because the effect of temperature is more pronounced than that of pressure, ε decreases with increasing P–T conditions and range between approximately 8 and 25 at normal crustal P–T conditions. Only at low temperature of greenschist to sub-greenschist facies conditions or notably high pressure of blueschist to eclog-ite facies conditions ε may exceed 25. But give the restricted P–T range covered by experimental data, any calculation of the static dielectric constant at pressures notably in excess of 500 MPa has to be done with great caution.

Viscosity

The transport properties of water strongly depend on its dynamic viscosity η, which also plays an important role for diffusion. Several studies have addressed the dynamic viscosity of water at low to moderate temperature and pressure (see Watson et al. (1980) for a review of available experiments at that time). The dynamic viscosity of water at elevated P–T conditions has been studied by Dudziak & Franck (1966) up to 350 MPa and 560°C and by Abramson (2007) up to 6 GPa and 300°C. Based on the available data, Watson et al. (1980) and Sengers & Kamgar-Parsi (1984) derived representative equations for η (Fig. 6D). At ambient conditions, liquid water has a dynamic viscosity of η = 0.001 Pa sec (Sengers & Kamgar-Parsi 1984). Along the liquid–vapour equilibrium, the viscosity of liquid water substantially decreases whereas that of steam only slightly increases, so that the critical dynamic viscosity of H2O is approximately 3.9 × 10−5 Pa sec. At temperature conditions above approximately 400°C, the data suggest that at isobaric conditions the temperature effect on the dynamic viscosity is only minor whereas increasing pressure increases η. Except for low temperatures of sub-greenschist facies conditions, H2O has dynamic viscosities at normal crustal P–T conditions between 0.0001 and 0.0002 Pa sec. Although no data are available for upper blueschist and eclogite facies, the data suggest that at these high pressure conditions the dynamic viscosity of H2O may range up to 0.0003 Pa sec (Fig. 6D).

H2O–SILICATE SYSTEMS AT HIGH P–T CONDITIONS

Aqueous fluids and silicate melts are the two most important mobile phases in the Earth crust and mantle. H2O–silicate systems typically belong to Type 2 systems (see above), which are characterized by a discontinuous critical curve and lower and upper segments of the two-phase fluid volume that terminate at the lower and upper critical end-points (see Fig. 7C). Up to moderate pressure and temperature conditions the solubility of silicates in aqueous fluids and of H2O in silicate melts is typically low and both phases fundamentally differ in their physicochemical properties. With increasing pressure, however, the solubility of H2O in silicate melts and of silicates in aqueous fluids increases and the differences between both phases finally vanish at the critical curve above which aqueous fluids and silicate melts are completely miscible giving rise to so-called ‘supercritical’ fluids. Here, only some key aspects of H2O–silicate systems at high pressure and temperature conditions are reviewed. A thorough review of this topic is given by Hack et al. (2007). Kennedy et al. (1962) provided the first experimental evidence for complete miscibil-ity between aqueous fluids and silicate melts at high pressure and temperature within the system H2O–SiO2 and proposed an upper critical endpoint at 970 MPa/1080°C with a composition of approximately 25 wt% H2O/75 wt% SiO2. This has been confirmed by Newton & Manning (2008). Shen & Keppler (1997) directly observed complete miscibility between aqueous fluids and silicate melts in the system H2O–albite and determined the location of the critical curve (Fig. 7A). Further experiments by Stalder et al. (2000) at near wet solidus conditions then located the upper critical endpoint in the system H2O–albite at approximately 1.5 GPa/700°C (Fig. 7A). The resulting phase relations within the system H2O–albite up to 1.7 GPa and T > 500°C based on the data by Paillat et al. (1992), Shen & Keppler (1997) and Stalder et al. (2000) are shown in Fig. 7B. At pressure below approximately 1.5 GPa, the critical curve is at higher temperature than the wet solidus, which is defined as onset of melting by the reaction albite + vapour = liquid. However, with increasing pressure, the critical curve shifts to lower temperature, the liquid + vapour two-phase field narrows and at approximately 1.5 GPa the critical curve intersects the wet solidus at about 700°C. Above approximately 1.5 GPa no discrete melting occurs but albite continuously dissolves in the fluid with increasing temperature giving rise to a ntinuous range from solute poor, almost pure H2Oat low temperature to H2O poor silicate dominated liquids at high temperature. Addition of fluorine, boron and sodium to the H2O–albite system shifts the location of the critical curve to even lower pressure and temperature (Sowerby & Keppler 2002). Critical curves for the systems H2O–nepheline and H2O–jadeite have then been determined by Bureau & Keppler (1999). Termination of the wet solidus at the upper critical endpoint, however, not only occurs in simple H2O–mineral systems but is also observed in H2O–rock systems. Kessel et al. (2005a,b) studied the system H2O–potassium-free basalt at 4–6 GPa/700–1400°C. They observed eutectic and peritectic melting at pressures of 4 and 5 GPa, respectively, but no discrete melting at 6 GPa (Fig. 7C). At 6 GPa, the aqueous fluid continuously increases its solute content with increasing temperature suggesting a critical endpoint in this system between 5 and 6 GPa/1050°C. Complete miscibility between aqueous fluids and silicate melts in the systems H2O–Ca-bearing granite and H2O–haplogranite was observed by Bureau & Keppler (1999). The corresponding critical curves are located at 1.34 GPa/1003°C to 204 GPa/735°C in the H2O–haplogranite system and at slightly higher pressure and temperature conditions of 1.61 GPa/898°Cto 2.08 GPa/820°C in the H2O–Ca-bearing granite system. However, no critical endpoints, which determine the termination of the wet solidus, are given for the systems H2O–Ca-bearing granite and H2O–haplogranite in Bureau & Keppler (1999). One important aspect of H2O–silicate systems at pressure conditions above the critical endpoint is the combination of high solute contents in the fluids with comparably low viscosities. Audétat & Keppler (2004) performed a seminal experimental study on the viscosity of high-pressure silicate rich aqueous fluids in the systems H2O–albite, H2O–nepheline and H2O–pectolite. In the H2O–albite system, fluid viscosity linearly increases with increasing solute content from approximately 3.5 × 10−4 Pa sec for 7 mol% solutes to approximately 4 × 10⁰ Pa sec for 67 mol% solutes (Fig. 7D). The viscosity of dry and hydrous albite melts with up to 3 wt% H2O is notably higher and ranges between approximately 10⁹ and 10¹¹.⁵ Pa sec (Romano et al. 2001). The data therefore indicate an exponentially decrease of viscosity with increasing H2O content for <20 wt% H2O (Audétat & Keppler 2004). Given their comparably low viscosity with high solute content, aqueous fluids above the upper critical endpoint potentially form very effective metasomatising agents within the mantle wedge.

Fig. 7. Water-silicate systems at high pressure and temperature. (A) P–T diagram showing location of the wet and dry solidus, critical curve and upper critical endpoint in the system H2O–albite. Wet solidus, filled and empty circles from Stalder et al. (2000), critical curve and diamonds from Shen & Keppler (1997) and dry solidus from Boyd & England (1963). (B) Phase relations within the system albite–H2O up to 1.7 GPa and T > 500°C showing the termination of the water saturated solidus and the critical curve at the upper critical endpoint. Drawn and modified after Paillat et al. (1992), Shen & Keppler (1997) and Stalder et al. (2000) (ab = albite, v = vapour, l = liquid, scf = supercritical fluid). (C) Phase relations within the system eclogite–H2O at 4–6 GPa and T > 600°C based on the data of Kessel et al. (2005a). Critical curve, dashed and stippled lines are hypothetical additions to the data given in Kessel et al. (2005a) to clarify phase relations (v = vapour, l = liquid, scf = supercritical fluid). (D) Arrhenius plot showing the viscosity of H2O–albite solutions as function of temperature and water content at high pressures >1.0 GPa. Data from Audétat & Keppler (2004), stippled lines calculated according to their Eqn (1).

SUMMARY

This paper reviews some of the fundamental physical and chemical properties of aqueous fluids at elevated pressure and temperature conditions. For certain geological environments these physicochemical fluid properties are fairly well known. This is, at least partly, true for low pressure and temperature systems like geothermal systems, basinal systems, and oceanic hydrothermal systems, in which fluids can be sampled and studied directly or which are easily investigated by experimental methods. At higher, typical metamorphic pressure and temperature conditions, the principal phase relations in model fluid systems are known. However, there are important gaps in knowledge of aqueous fluids at elevated pressure and temperature. First, even for pure H2O, most data for important properties like ion product, static dielectric constant, and viscosity are restricted to pressures below 1.0 GPa. Also, some of the available formulations for these properties substantially diverge outside the pressure–temperature range covered by the experimental data. Second, while the principal phase relations in binary and ternary aqueous fluid systems are well known, experimental data precisely addressing the composition of co-existing fluid phases are rare. In addition, data on trace element and isotope fractionation behaviour between co-existing aqueous fluids are rare. Knowledge of these fractionation behaviours are a prerequisite for modelling geochemical cycles in fluid mediated or fluid dominated systems. Finally, for high pressure and temperature of the lower crust and upper mantle, the knowledge of physical fluid properties like viscosity and density in ‘supercritical’ H2O–silicate systems is rather limited. However, for modelling the role of these ‘supercritical’ H2O–silicate fluids as metasomatizing agent for crust-mantle interactions within subduction zones, these data are essential.

ACKNOWLEDGMENTS

I thank the editors for inviting me to contribute to this Geofluids anniversary volume. This paper greatly benefitted from careful and critical reviews by two anonymous reviewers. Helpful editorial handling by C. Manning is gratefully acknowledged.

REFERENCES

Abramson EH (2007) Viscosity of water measured to pressures of 6 GPa and temperature of 300°C. Physical Review E, 76, 051203.

Anderko A, Pitzer KS (1993a) Equation-of-state representation of phase equilibria and volumetric properties of the system NaCl–H2O above 573 K. Geochimica et Cosmochimica Acta, 57, 1657–80.

Anderko A, Pitzer KS (1993b) Phase equilibria and volumetric properties of the systems KCl–H2O and NaCl–KCl–H2O above 573 K: equation of state representation. Geochimica et Cosmochimica Acta, 57, 4885–97.

Anisimov MA, Sengers JV, Levelt Sengers JMH (2004) Near-critical behaviour of aqueous systems. In: Aqueous Systems at Elevated Temperatures and Pressures (eds Palmer DA, Fernández-Prini R, Harvey AH), pp. 29–71. Elsevier, London.

Anovitz LM, Labotka TC, Blencoe JG, Horita J (2004) Experimental determination of the activity-composition relations and phase equilibria of H2O–CO2-NaCl fluids at 500°C, 500 bars. Geochimica et Cosmochimica Acta, 68, 3557–67.

Arnórsson S, Stefánsson A, Bjarnason JÖ (2007) Fluid–fluid interactions in geothermal systems. Reviews in Mineralogy and Geochemistry, 65, 259–312.

Audétat A, Keppler H (2004) Viscosity of fluids in subduction zones. Science, 303, 513–6.

Bandura AV, Lvov SN (2006) The ionization constant of water over wide ranges of temperature and density. Journal of Physical and Chemical Reference Data, 35, 15–30.

Belonoshko AB, Saxena SK (1991) A molecular dynamics study of the pressure–volume–temperature properties of supercritical fluids: I. H2O. Geochimica et Cosmochimica Acta, 55, 381–8.

Belonoshko AB, Saxena SK (1992) A unified equation of state for fluids of C–H–O–N–S–Ar composition and their mixtures up to high temperatures and pressures. Geochimica et Cosmochimica Acta, 56, 3611–36.

Bignold GJ, Brewer AD, Hearn B (1971) Specific conductivity and ionic product of water between 50 and 271°C. Transactions of the Faraday Society, 67, 2419–30.

Bischoff JL (1991) Densities of liquids and vapors in boiling NaCl-H2O solutions: a PVTx summary from 300° to 500°C. American Journal of Science, 291, 309–38.

Bischoff JL, Pitzer KS (1989) Liquid–vapor relations for the system NaCl–H2O: summary of the P–T–x surface from 300° to 500°C. American Journal of Science, 289, 217–48.

Bischoff JL, Rosenbauer RJ (1984) The critical point and two-phase boundary of seawater, 200–500°C. Earth and Planetary Science Letters, 68, 172–80.

Bischoff JL, Rosenbauer RJ (1988) Liquid–vapor relations in the critical region of the system NaCl–H2O from 380 to 415°C: a refined determination of the critical point and two-phase boundary of seawater. Geochimica et Cosmochimica Acta, 52, 2121–6.

Bischoff JL, Rosenbauer RJ, Pitzer KS (1986) The system NaCl–H2O: relations of vapour–liquid near the critical temperature of water and of vapour–liquid–halite from 300 to 500°C. Geochimica et Cosmochimica Acta, 50, 1437–44.

Bischoff JL, Rosenbauer RJ, Fournier RO (1996) The generation of HCl in the system CaCl2–H2O: vapor–liquid relations from 380-500°C. Geochimica et Cosmochimica Acta, 60, 7–16.

Bodnar RJ, Burnham CW, Sterner SM (1985) Synthetic fluid inclusions in natural quartz. III. Determination of phase equilibrium properties in the System H2O–NaCl to 1000°C and 1500 bars. Geochimica et Cosmochimica Acta, 49, 1861–73.

Boyd FR, England JL (1963) Effect of pressure on melting of diopside, CaMgSi2O6, and albite, NaAlSi3O8, in the range up to 50 kbar. Journal of Geophysical Research, 68, 311–23.

Bradley DJ, Pitzer KS (1979) Thermodynamics of electrolytes. 12. Dielectric properties of water and Debye-Hückel parameters to 350°C and 1 kbar. Journal of Physical Chemistry, 83, 1599–603.

Bureau H, Keppler H (1999) Complete miscibility between silicate melts and hydrous fluids in the upper mantle: experimental evidence and geochemical implications. Earth and Planetary Science Letters, 165, 187–96.

Chou I-M (1987) Phase relations in the system NaCl–KCl–H2O.III: solubilities of halite in vapor-saturated liquids above 445°C and redetermination of phase equilibrium properties in the system NaCl–H2O to 1000°C and 1500 bars. Geochimica et Cosmochimica Acta, 51, 1965–75.

Churakov SS, Gottschalk M (2003a) Pertubation theory based equation of state for polar molecular fluids: I. Pure fluids. Geochimica et Cosmochimica Acta, 67, 2397–414.

Churakov SS, Gottschalk M (2003b) Pertubation theory based equation of state for polar molecular fluids: II. Fluid mixtures. Geochimica et Cosmochimica Acta, 67, 2415–25.

Cline JS, Bodnar RJ (1994) Direct evolution of brine from a crystallizing silicic melt at the Questa, New Mexico, molybdenum deposit. Economic Geology, 89, 1780–802.

Deul R (1984) Dielektrizitatskonstante und Dichte von Wasser-Benzol-Mischungen bis 450°C und 3000 bar. PhD Dissertation, University of Karlsruhe, Germany.

Diamond LW (2001) Review of the systematic of CO2-H2O fluid inclusions. Lithos, 55, 69–99.

Driesner T, Heinrich CA (2007) The System H2O–NaCl. I. Correlation formulae for phase relations in pressure-temperature-composition space from 0 to 1000°C, 0 to 5000 bar, and 0 to 1 XNaCl. Geochimica et Cosmochimica Acta, 71, 4880–901.

Duan Z, Zhang Z (2006) Equation of state of the H2O, CO2, and H2O–CO2 systems up to 10 GPa and 2573,15 K: molecular dynamics simulations with ab initio potential surface. Geochimica et Cosmochimica Acta, 70, 2311–24.

Duan Z, Moller N, Weare JH (1992a) An equation of state (EOS) for CH4–CO2–H2O system: I. Pure systems from 0 to 1000°C and 0 to 8000 bar. Geochimica et Cosmochimica Acta, 56, 2605–17.

Duan Z, Moller N, Weare JH (1992b) An equation of state for the CH4–CO2–H2O system: II. Mixtures from 50 to 1000°C and 0 to 1000 bar. Geochimica et Cosmochimica Acta, 56, 2619–31.

Duan Z, Moller N, Weare JH (1992c) Molecular dynamics simulations of PVT properties of geological fluids and a equation of state of non-polar and weakly polar gases up to 2000 K and 20,000 bar. Geochimica et Cosmochimica Acta, 56, 3839–45.

Duan Z, Moller N, Weare JH (1995) Equation of state for the NaCl–H2O–CO2 system: prediction of phase equilibria and volumetric properties. Geochimica et Cosmochimica Acta, 59, 2869–82.

Duan Z, Moller N, Weare JH (1996) A general equation of state for supercritical fluid mixtures and molecular dynamics simulation of mixture PVTX properties. Geochimica et Cosmochimica Acta, 60, 1209–16.

Duan Z, Moller N, Weare JH (2003) Equations of state for the NaCl–H2O–CH4 system and the NaCl–H2O–CO2–CH4 system:

phase equilibria and volumetric properties above 573 K. Geochimica et Cosmochimica Acta, 67, 671–80.

Duan Z, Moller N, Weare JH (2006) A high temperature equation of state for the H2O–CaCl2 and H2O–MgCl2 systems. Geo-chimica et Cosmochimica Acta, 70, 3765–77.

Dubois M, Weisbrod A, Shtuka A (1994) Experimental determination of the two-phase (liquid and vapour) region in water-alkali chloride binary mixtures at 500° and 600°C using synthetic fluid inclusions. Chemical Geology, 115, 227–38.

Dudziak KH, Franck EU (1966) Measurements of water viscosity up to 560°C and 3500 bars. Berichte der Bunsengesellschaft für Physikalische Chemie, 70, 1120–8.

Fernández DP, Mulev Y, Goodwin ARH, Levelt Sengers JMH (1995) A database for the static dielectric constant of water and steam. Journal of Physical and Chemical Reference Data, 24, 33–69.

Fernández DP, Goodwin ARH, Lemmon EW, Levelt Sengers JMH, Williams RC (1997) A formulation for the static permittivity of water and steam at temperatures from 238 K to 873 K at pressures up to 1200 MPa, including derivatives and Debye-Huckel coefficients. Journal of Physical and Chemical Reference Data, 26, 1125–66.

Ferry JM, Baumgartner L (1987) Thermodynamic models of molecular fluids at the elevated pressures and temperatures of crustal metamorphism. Reviews in Mineralogy, 17, 323–66.

Fischer JR, Barnes HL