Textural and compositional complexities resulting ...

6 downloads 0 Views 8MB Size Report
Nappe, Norway (reprinted from Smit et al., 2008 with permission from John Wiley and Sons). 631 ...... Brantley, S.L., Evans, B., Hickman, S.H., Crerar, D.A., 1990.
Earth-Science Reviews 150 (2015) 628–651

Contents lists available at ScienceDirect

Earth-Science Reviews journal homepage: www.elsevier.com/locate/earscirev

Textural and compositional complexities resulting from coupled dissolution–reprecipitation reactions in geomaterials Alexander Altree-Williams a, Allan Pring b, Yung Ngothai a, Joël Brugger c,⁎ a b c

School of Chemical Engineering, University of Adelaide, North Terrace, Adelaide, SA 5000, Australia School of Chemical and Physical Sciences, Flinders University, Bedford Park, SA 5042, Australia School of Earth, Atmosphere and the Environment, Monash University, Clayton, VIC 3800, Australia

a r t i c l e

i n f o

Article history: Received 12 June 2015 Received in revised form 21 August 2015 Accepted 31 August 2015 Available online 4 September 2015 Keywords: Dissolution and precipitation Fluid-driven mineral transformations Interface chemistry Ore petrography Pseudomorphism Reactive transport Reaction mechanism Textural evolution

a b s t r a c t Texture encompasses ‘the overall appearance a rock has because of the size, shape, and arrangement of its constituent mineral grains’. Textural observations are crucial for deciphering the origin and geological history of rocks and their constituting minerals. In metamorphic and hydrothermal settings, textural observations hence serve to reconstruct the P,T path and the compositions and origins of the parent fluids. Over the past 13 years, a number of studies have emphasized the role of ‘coupled dissolution reprecipitation reactions’ (CDR) in geological systems. In these fluid-driven reactions, the replacement of one phase by another occurs via coupling between the dissolution of the parent and the precipitation of the product. In this paper we review the diversity of textures that arise from the CDR mechanism. The great diversity of textures relates to the diversity of mechanisms responsible for the coupling between dissolution and precipitation. Key parameters defining textures include volume change, the rate-limiting process, and the local composition at the mineral–fluid interface. In many of the reviewed examples, reaction mechanisms, rather than intensive properties such as P–T history, control the textures in the products, and far-from-equilibrium or local equilibriums at the mineral–fluid interface play a key role in controlling the final textures and mineral assemblages. These processes can also lead to the scavenging of trace elements from hydrothermal fluids. Because by nature CDR reactions are interface-controlled, many of the products are metastable, which further drives the reactions. These subsequent reactions can add to the textural complexity, or on the contrary obscure the original reaction mechanism. This review emphasizes the need to improve our understanding of reaction mechanisms, especially in systems containing even minor amounts of fluids (ore systems; metasomatic and metamorphic systems). Such a process-driven understanding is vital to supporting the petrological interpretation of textures. © 2015 Elsevier B.V. All rights reserved.

Contents 1.

2.

3. 4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1. Textures and their interpretation. . . . . . . . . . . . . . . 1.2. Coupled dissolution–reprecipitation (CDR) reaction mechanism. 1.3. The physics and chemistry of CDR reaction mechanism. . . . . 1.3.1. Reaction drivers . . . . . . . . . . . . . . . . . . 1.3.2. Mass transfer . . . . . . . . . . . . . . . . . . . 1.3.3. Mineral solubility and dissolution rates . . . . . . . . 1.3.4. Nucleation and growth . . . . . . . . . . . . . . . Length scale of replacement . . . . . . . . . . . . . . . . . . . . 2.1. Preservation of crystallographic information. . . . . . . . . . 2.2. Nanoscale to macroscale replacement . . . . . . . . . . . . Volume change . . . . . . . . . . . . . . . . . . . . . . . . . . Fluid pathways and porosity . . . . . . . . . . . . . . . . . . . . 4.1. ‘Homogeneous’ reaction-induced porosity. . . . . . . . . . .

⁎ Corresponding author. E-mail address: [email protected] (J. Brugger).

http://dx.doi.org/10.1016/j.earscirev.2015.08.013 0012-8252/© 2015 Elsevier B.V. All rights reserved.

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

629 629 629 630 630 630 630 632 633 634 635 636 637 637

629

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

4.2. Reaction-induced fracturing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Grain boundary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4. Cleavage and twin boundaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Nature of the end products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Dynamic evolution of textures, porosity, and composition during CDR . . . . . . . . . . . . . . . . . . . . . . . 6.1. Complex compositional zoning in solid solutions: local equilibrium revealed . . . . . . . . . . . . . . . . . 6.2. Changes in mineralogical composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3. Complex textures via competition among solid-state diffusion and CDR . . . . . . . . . . . . . . . . . . . 6.4. Ripening by surface energy minimization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.5. Formation of passivating layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7. Reaction fronts as microreactors and scavenging of trace and minor elements. . . . . . . . . . . . . . . . . . . . 8. Discussion: the interpretation of textures in CDR reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.1. Complex textures can develop via CDR reactions as a result of non-equilibrium processes and local equilibrium . 8.2. The importance of post-CDR reactions on textural evolution and preservation . . . . . . . . . . . . . . . . 8.3. Perspectives and future directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

638 638 640 640 641 641 643 643 644 645 645 647 647 647 648 648 648

1. Introduction

1.2. Coupled dissolution–reprecipitation (CDR) reaction mechanism

1.1. Textures and their interpretation

When a fluid contacts a mineral with which it is undersaturated, the mineral begins to dissolve. An interfacial layer can then become supersaturated with respect to a more stable product phase, which can nucleate and grow at the surface of the parent phase (Fig. 1a). This can lead to the pseudomorphic (isovolumetric) replacement of the parent phase by the product phase (Fig. 1b), assuming that the dissolution of the parent phase and the precipitation of the product are coupled in both space and time (Brugger et al., 2010; Putnis, 2009; Qian et al., 2010). Several key identifying features of such pseudomorphic replacements (also named ‘interface coupled dissolution reprecipitation reactions’, ICDR) have been highlighted by Putnis (2009) and can be seen in Fig. 1, including:

The differences among rocks (including unconsolidated sediments) are based on two main features: the mineral assemblage and the texture. According to Skinner et al. (1999), texture means ‘the overall appearance a rock has because of the size, shape, and arrangement of its constituent mineral grains’. Textural observations are key for deciphering the origin and geological history of rocks and their constituting minerals. Following the seminal paper by Putnis (2002) emphasizing the role of ‘coupled dissolution reprecipitation reactions’ (CDR) in geological systems, a large amount of work has been done to understand the molecular-level controls on these fluid-mediated reactions and their significance for a large range of geological conditions and processes, ranging from weathering, to ore deposit formation and associated alteration, to metamorphism and metasomatism (Hellmann et al., 2003, 2012; Jamtveit et al., 2011; Oliver et al., 2004; Putnis and Austrheim, 2010, 2013). The recognition of the importance of CDR reactions in natural systems also has inspired a range of applications in material sciences (Brugger et al., 2010; Hellmann et al., 2015; Putnis, 2014; Reboul et al., 2012, 2015). Previous reviews have emphasized the catalyzing effect of the fluid phase on the replacement process, the importance of the mineral–fluid interface in controlling the nature of the replacement, and the wide range of conditions and mineral systems governed by the CDR reaction mechanism (e.g., Putnis, 2002, 2009; Ruiz-Agudo et al., 2014). The interpretation of the changes in mineralogy and composition associated with reactive fluid flow relies on understanding the relationship between the mineral products (textures) and the evolving fluids (Bickle and Baker, 1990). CDR reactions can result in a remarkable variety of textures, and can also affect the nature of the resulting mineral association in often unexpected ways. The aim of this paper is to review the diversity of textures that arise from the fluid-mediated coupled dissolution reprecipitation mechanism. We attempt to identify the parameters that control the development of particular textures, and illustrate the role of the reaction mechanism (rather than intensive properties such as P–T history) in controlling the nature and textures of mineral assemblages. This review builds upon the recent review by Ruiz-Agudo et al. (2014) by emphasizing examples from ore-forming environments and illustrating the role of the mineral–fluid interface in stabilizing the formation of metastable product phases and the scavenging of trace elements from hydrothermal fluids.

I. close coupling between the dissolution and precipitation processes which enables preservation of external morphology; II. a sharp interface between the parent and product that exhibits no diffusional profile; III. a permeable porosity generated within the product phase; and IV. transfer of crystallographic information from the parent to the product when an epitaxial relationship exists.

Local controls will influence the equilibrium compositions of both the solid and fluid at the interface, and the product phase is often observed to evolve both compositionally and texturally until an essentially homogeneous product forms (Fig. 1c). Fluid-mediated mineral replacements are prominent at the relatively low temperatures associated with environmental and upper crustal conditions. Solid-state diffusion (SSD) is related to the availability of defects within the crystal structure. Crystal defects take the form of either extrinsic impurities, the number being independent of temperature, or intrinsic vacancies, the number of which increases exponentially with increasing temperature (Lee, 1993; Putnis, 1992). Thus, at high temperatures the number of defects increases, and kinetic energy increases the probability of overcoming activation energy barriers, such that atomic diffusion within the crystal becomes the dominant mechanism for re-equilibration processes. On the other hand, at low temperatures SSD is retarded by the relatively limited number of defects in most materials, and the CDR mechanism provides a kinetically more favorable re-equilibration pathway (Putnis, 2002; Rubie, 1986; Zhao et al., 2013). The CDR mechanism acts over a wide range of conditions when a fluid interacts with a rock, but is particularly common at low

630

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

1.3. The physics and chemistry of CDR reaction mechanism 1.3.1. Reaction drivers At a fundamental level, minimization of the Gibbs free energy drives mineral replacement processes in response to changes in system parameters such as T, P, stress and local solution chemistry. Changes in internal energy (U) are caused by changes in both thermal and non-thermal energy contributions through the First Law, dU ¼ δQ þ

X

δW;

ð1Þ

where δQ is the heat absorbed/released by the system (equal to TdS for a reversible system), and ∑δW is the total work done on the system by non-thermal contributions. The summation of work terms carries the key non-thermal energy contributions to fluid-mineral systems including mechanical work, − PdV; surface work, γdA; chemical work, μdn; and strain work, σdε. The Gibbs free energy is related to enthalpy, entropy and internal energy by Eq. (2), G ≡ H−TS ¼ U þ PV−TS;

ð2Þ

where the enthalpy and internal energy are related through H = U + PV. Taking the differential of Eq. (2) and inserting Eq. (1) and its associated work terms for a system of i components and j surfaces, the total differential for the Gibbs free energy can be written as (Parks, 1990). dG ¼ −TdS þ VdP þ

X

μ dni i i

þ

X

j

γ j dA j þ

X

kl

σ kl dϵkl ;

ð3Þ

where μi is the chemical potential of species i; γj is the surface energy of interface j; and σkl is the stress tensor associated with strain εkl (Alberty, 2001). At constant T and P, the major drivers for mineral replacements are the chemical, surface and strain work terms (Eq. (3)). The actual driving force will depend on the system and its position relative to equilibrium, but in general the surface work term becomes increasingly important as the system approaches equilibrium or when the particle size is small (Carlson, 1999; Lifshitz and Slyozov, 1961; Voorhees, 1985).

Fig. 1. Principles of coupled dissolution reprecipitation reactions and the main physicochemical controls on texture development during pseudomorphic mineral replacement. (a) Initiation of dissolution of parent mineral (BX) and nucleation of daughter product at the surface. Circles represent the nuclei. (b) Reaction progress inwards. (c) Final product, in stable or metastable equilibrium with the surrounding solution.

temperature when diffusional mechanisms are less favorable. Prominent examples can be found in association with the alteration of ore minerals, including weathering in platinum group minerals (Fig. 2a and b), replacement of chromite (FeCr 2 O 4 ) by stichtite (Mg 6 Cr 2 [(OH) 16 CO 3 ]·4H 2 O; Fig. 2c), or the transformation of pentlandite ((Fe,Ni)9 S8 ) to violarite ((Ni,Fe) 3 S4) in the cementation zone (Fig. 2d). These reactions can also happen under diagenetic or mild hydrothermal conditions, e.g. the pentlandite to violarite reaction; calcite (CaCO 3 ) replacing anhydrite (CaSO 4 ) in dolomite reservoirs (Fig. 2e); and aragonite (CaCO3 ) replacing magnesite (MgCO 3 ; Fig. 2f). The ability of the fluid to penetrate these natural samples is seen by the fact that the product mineral forms along cracks and cleavages, as well as from the rim to core progression of sharp replacement fronts within individual grains.

1.3.2. Mass transfer Transport of solutes to and away from the reaction front is required for sustaining the reaction (Fig. 1b). A key feature of many CDR reactions is that secondary porosity is created as a result of the reaction itself. The generation of an interconnected porosity facilitates the total re-equilibration of a crystal without the need for volume diffusion (Putnis and John, 2010). Solute transport usually occurs via diffusion rather than advection, given the small-scale nature of the porosity. Chemical potential gradients develop between the replacement interface and the “bulk” fluid, driving the diffusion of species to and from the replacement front (Zhang, 2010). 1.3.3. Mineral solubility and dissolution rates The dissolution rate of a mineral in the presence of a fluid is an important factor in CDR replacements, since a mineral needs to dissolve to leave space and provide chemical components for the formation of a different mineral or mineral assemblage. Many empirical equations describe the kinetics of mineral dissolution (e.g., Avrami equation). Physically-based formalisms account explicitly for the fact that dissolution rates depend upon the specific surface area of the dissolving mineral and the chemistry of the solution. One of the simplest such equations is: " ! # rk ¼ As kþ 1−Q K ;

ð4Þ

where rk is the reaction rate (mol s− 1 g− 1), positive for dissolution; As is the specific surface area of the mineral (cm2 g− 1), k+ is the rate

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

631

Fig. 2. Natural examples of CDR reactions. (a and b) Backscatter electron images of platinum group minerals from Umtshingwe River, Zimbabwe (from Oberthür et al., 2013). (a) Sperrylite (PtAs2) grain with rim of pure Pt and (b) corroded cooperite (PtS) replaced by Pt (light gray). (c) Backscatter electron image of stichtite (Mg6Cr2(OH)16[CO3]·4H2O) replacing chromite (FeCr2O4). (Tasmania, Australia). Stichtite is the gray fibrous mineral (reprinted from Melchiorre et al., 2014 with permission from Elsevier). (d) Violarite replacing pentlandite along cracks and cleavages (reprinted from Misra and Fleet, 1974). (e) Calcite (Cal) replacing anhydrite (Anh) cement in dolomite-hosts (Dol), Khuff formation, Abu Dhabi (reprinted from Worden et al., 2000 with permission from the Society for Sedimentary Geology). (f) Aragonite (Arg) replacing magnesite (Mgs) along magnesite cleavages in eclogites from Jæren Nappe, Norway (reprinted from Smit et al., 2008 with permission from John Wiley and Sons).

constant (mol cm− 2 s− 1), and Q and K are the activity product and equilibrium constant for the dissolution reaction, respectively. In general, the temperature dependence of the rate constant follows the Arrhenius law (e.g., Lasaga, 1984, 1998), which relates the reaction rate (k+) to the activation energy EA (J mol− 1) and the reaction temperature T (K): kþ ¼ Ae−EA =ðRTÞ ;

ð5Þ

A is a pre-exponential factor (mol cm−2 s−1) and R is the gas constant (8.3143 J K−1 mol−1). The Q/K ratio depends on T, P, and solution parameters such as pH, PCO2, and salinity (Rimstidt, 1997). It has been established that it is the local composition at the replacement interface that controls the replacement process by governing the saturation states of the parent and product phases (e.g., King et al., 2010; Ruiz-Agudo et al., 2014). CDR replacements are aided by the solubility differences between the parent and product phases under the given set of replacement conditions. It is the relative solubility of the two

632

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

phases that is critical, as even limited dissolution of the parent can lead to the interfacial fluid becoming supersaturated with a secondary phase (Hövelmann et al., 2012; Putnis et al., 2005; Putnis, 2009; Ruiz-Agudo et al., 2014). Hence, even in systems where the parent is essentially insoluble, e.g. fluorapatite (Ca5(PO4)3F) replacing hydroxylapatite (Ca5(PO4)3(OH)) (Rendón-Angeles et al., 2000), replacement via CDR occurs due to the role of the interfacial control of mineral saturation states. The stability of solid solutions is of interest, since solid solutions are widespread throughout geological systems. Equilibrium compositions in solid solution – aqueous solution (SSAS) systems are represented by the total solubility product, ΣΠ, defined as the sum of the partial solubility products of the endmembers (Lippmann, 1980). For a binary solid solution (B,C)A, the total solubility product is expressed as (Prieto, 2009): "$ % $ %#$ % ΣΠ ¼ Bnþ þ Cnþ An‐ ;

ð6Þ

ΣΠ ¼ KBA γBA XBA þ KCA γCA XCA ;

ð7Þ

where [Bn+], [Cn+], and [An−] represent activities of aqueous ions. Construction of SSAS phase diagrams take advantage of the fact that Eq. (6) can be expressed in terms of both solid-phase composition (solidus) and aqueous-phase composition (solutus). For a given solid solution with endmember solid fractions XBA and XCA, the solidus is defined as:

where KBA, KCA are solubility products, and γBA, γCA are activity coefficients for the subscripted endmembers. For a system with aqueous activities XB,aq and XC,aq, the solutus is defined as: 1 ½Bnþ & ; with XB; aq ¼ nþ ; XC; aq XC;aq XB;aq ½B & þ ½Cn' & þ KBA γBA KCA γCA ½Cn' & ¼ nþ ½B & þ ½Cn' &

ΣΠ ¼

ð8Þ

Eqs. (7) and (8) can be used to qualitatively interpret reaction pathways in systems that include binary SSAS (Glynn et al., 1990; Lippmann, 1980; Prieto, 2009). 1.3.4. Nucleation and growth CDR reactions involve temporal and spatial couplings between dissolution and precipitation. Precipitation involves the formation of nuclei followed by the growth of the product. The interaction of nucleation and growth mechanisms strongly influences the nature of CDR replacement textures. Local supersaturation at the replacement interface controls the nature of the nucleation and growth of the product phase (Fig. 3). The initial formation of the product phase at the interface will depend on the ability of the product to nucleate at the surface of the parent (Fig. 1a). An activation energy barrier exists that limits product nucleation and defines a threshold supersaturation — a kinetic property specific to a given set of conditions (Harlov et al., 2011; Prieto et al., 1993;

Fig. 3. Relationship between nucleation and growth rates and the degree of supersaturation (adapted from Sunagawa, 1981). Transitional supersaturations (σ* and σ**) defined the supersaturation regions where different growth mechanisms control product formation. The regions where nucleation or growth dominates the formation of the product phase are defined by the threshold supersaturations (σEp and σHet) which depend on the energy barrier to nucleation (Steefel and Van Cappellen, 1990).

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

Putnis et al., 1995; Putnis and Mauthe, 2001; Scherer, 1999; Stack et al., 2014; Steefel and Van Cappellen, 1990). The threshold supersaturation, represented in Fig. 3 by σEp for epitaxial nucleation and σHet for heterogeneous nucleation, defines the point at which the rate of nucleation increases sharply, and also plays an important role in controlling the initial growth mechanism (García-Ruiz et al., 2007; Prieto et al., 1993). Above the threshold value new product formation is dominated by nucleation, while below it crystal growth defines new product formation (Putnis et al., 1995; Steefel and Van Cappellen, 1990). The formation of the product by nucleation provides sites for the subsequent growth of the product phase. The mechanism by which the nuclei grow will depend on the degree of supersaturation and the availability of attachment sites such as kinks and steps (Otálora and García-Ruiz, 2014; Sunagawa, 1981). The growth mechanisms are defined by transitional supersaturations (σ* and σ**, Fig. 3), which are related to the smoothness of the solid–liquid interface. At high supersaturations (σ N σ**), the solid–liquid interface is rough and continuous 3D growth of island-like structures will occur (Fig. 4a). High availability of kink sites aids the concurrent growth of multiple product layers, leading to a dendritic-like morphology (Hövelmann et al., 2014; Otálora and García-Ruiz, 2014). Sunagawa (1981) notes that below σ** the solid–liquid interface becomes smooth, switching the growth mechanism to 2D nucleation growth or spiral growth. As the supersaturation decreases (σ* b σ b σ**), the availability of kink sites decreases and 2D nucleation growth begins leading to each product layer completing before the next one forms (Fig. 4b). This mechanism is closely linked to epitaxial nucleation, as on completion of each layer new nuclei must form for

633

growth to continue, and it is therefore favored by high deposition rates (Markov, 2003). In the case of very low supersaturations (σ b σ*), when the system is close to equilibrium, spiral growth will be the main mechanism by which the product forms. The low degree of supersaturation limits both nucleation events and the number of available kink sites. However, growth is facilitated by dislocations, which can form by mechanical deformation or around impurities (Otálora and García-Ruiz, 2014), and provide steps for attachment (Fig. 4c). Spiral growth can continue at these low supersaturations as the spiral advancement of the step supplies non-vanishing attachment sites removing the need for nucleation of a new layer (Frank, 1949). 2. Length scale of replacement The length scale of replacement can vary from the molecular (unit cell) level up to several meters, and is an important element of the mineral replacement texture (Qian et al., 2010; Xia et al., 2009a). Fig. 5 illustrates the variety of textures obtained during the replacement of pentlandite by violarite. In all cases, the replacement preserved the crystallographic orientation of the parent pentlandite. However, the nature of the replacement textures differs depending on changes of reaction parameters. For example, at pH 1–6, the replacement is pseudomorphic (Fig. 5b-d), with no significant gap at the reaction interface (Fig. 5e). The reaction progresses along the rim of the grain and along cracks within the grain (Fig. 5c). In contrast, at pH b 1, the reaction is no longer pseudomorphic, and the grains show a cauliflower-like morphology (Fig. 5f). Additionally, a large gap develops at the reaction interface (Fig. 5i). The violarite appears to grow on the outside of the grain, and large

Fig. 4. Representations of the major growth mechanisms (adapted from Otálora and García-Ruiz, 2014 with permission of The Royal Society of Chemistry). (a) At high supersaturations the growth of 3D islands is favored by the high availability of attachment sites (e.g. kinks and steps). Attaching molecules are more strongly coupled to the initially formed product relative to the parent resulting in 3D island growth. (b) 2D nucleation growth favored by topotactic replacement. Nucleation is subsequently followed by attachment to the steps resulting in layerby-layer growth. Associated with high nucleation rates as each layer requires a new nuclei to form. (c) Spiral growth dominates at low supersaturations when limited nucleation events occur. Dislocations provide permanent attachment sites for the product to form.

634

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

Fig. 5. Comparison of the textures associated with different rate-limiting processes in the replacement of pentlandite by violarite. The images are reprinted from Xia et al. (2009a) with the permission of Elsevier, except for (d) which is reprinted from Xia et al. (2008), copyright 2008 American Chemical Society. (a) The initial unreacted pentlandite grain. (b–e) Dissolutionlimited textures of the replacement of pentlandite by violarite. (b) External dimensions are preserved by the replacing violarite. (c) Violarite forms along cracks within the pentlandite grain as well as at the exterior. (d) BSE image of a fully reacted grain illustrating porosity within the violarite product. (e) The tight coupling of the dissolution and precipitation processes results in a sharp replacement interface. (f–i) Precipitation-limited textures of the replacement of pentlandite by violarite. (f) Exterior of the reacted grains exhibits violarite overgrowths. (g) Violarite forms only at the exterior of the grain rather than along fractures as well. (h) Atoll-like textures develop such that violarite surrounds a large void as pentlandite dissolves. (i) Loose coupling between the dissolution and precipitation processes results in a large gap developing at the reaction interface.

amounts of dissolution occurred along cracks within the pentlandite (Fig. 5g and h). 2.1. Preservation of crystallographic information The preservation of crystallographic orientation across the replacement interface is a commonly observed phenomenon in pseudomorphic replacements (e.g., Álvarez-Lloret et al., 2010; Raufaste et al., 2011; Xia et al., 2009a, 2009b). Originally, such a preservation was considered as strong evidence for solid-state diffusion driven

reaction mechanism, since the dissolution step destroys any crystallographic information. However, the preservation of crystallographic orientation via CDR reactions is a result of structurally controlled topotaxy between the two participating solid phases. Spry (1969) defines topotaxy as the structural relationship between the parent and product mineral during replacement. This topotactic relation has been constrained to a mismatch of lattice parameters of less than 14% when growing from solution (Frank and Van der Merwe, 1949a, 1949b; Kelly and Groves, 1970; Markov, 2003). For example, the lattice mismatch between pentlandite and violarite

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

is approximately 6% based on the lattice parameters measured by Xia et al. (2008). Preservation of crystallographic information is tightly controlled by the nucleation and growth of the product phase. Topotactic relations reduce the threshold supersaturation (σEp in Fig. 3; Otálora and García-Ruiz, 2014; Prieto et al., 1993) by minimizing strain energy associated with lattice misfit (Raufaste et al., 2011; Spry, 1969). The lower supersaturation reduces the driving force for growth, shifting the mechanism to 2D nucleation growth and driving layer-by-layer growth (Fig. 3; Markov, 2003). On the other hand, a higher threshold supersaturation (σHet; Fig. 3) favors heterogeneous nucleation and 3D growth preventing a “continuous” product layer forming at the interface. The difference in growth mechanism is evident when comparing the replacement of pentlandite by violarite (Fig. 5c) and calaverite (AuTe2) by gold (Fig. 7a). Growth of worm-like gold is the result of heterogeneous nucleation and subsequent 3D growth, while the limited misfit between pentlandite and violarite facilitates epitaxial nucleation and 2D layer-by-layer growth. Topotactic relations between the parent and product thus facilitate the preservation of crystallographic information by controlling nucleation and growth of the product. 2.2. Nanoscale to macroscale replacement Although unit-cell scale information is preserved by such epitaxial growth, the length scale of replacement depends on the subsequent formation of new product after the initial nuclei formation. In order to obtain nm-scale preservation of the textures of the initial mineral, there must exist a tight spatial coupling between dissolution and precipitation, i.e. dissolution and precipitation must happen close to each other. Additionally, nm-scale preservation requires nucleation rather than growth to be the dominant means by which the product forms. This implies fast nucleation of the daughter mineral near the dissolving surface (Fernández-Díaz et al., 2009). In the case of the replacement of pentlandite by violarite at mildly acidic pH (Fig. 5b-e), internal textural details of pentlandite by violarite are preserved down to ~20 nm along with crystallographic orientation (Xia et al., 2009a); this suggests that the size of the individual violarite crystals and the distance of transport is ≤20 nm, as a result of localized supersaturation limiting the region of nucleation and subsequent growth (Dunkel and Putnis, 2014). Xia et al. (2008) determined the violarite crystallite size to be 18 ± 2 nm,

635

supporting this suggestion. In contrast, at very low pH, nucleation of violarite near the surface of the dissolving pentlandite becomes unlikely as a result of increased pentlandite dissolution rate; once nucleation of violarite is achieved, growth will occur mainly by aggregation on the existing nuclei. This results ultimately in atoll-like textures, with a rim of porous violarite surrounding the dissolving pentlandite. In this case, the only preserved information is the outside shape of the grain and its crystallographic orientation (Fig. 5g and h). Atoll-like textures have also been formed via the replacement of pyrrhotite (Fe1 − xS) by marcasite (FeS2), where pyrrhotite dissolves completely leaving a rim of marcasite (Qian et al., 2011; their Fig. 5a). Rather than relying on a microscopic description of the complex interactions among nucleation, growth, and mass transport, Xia et al. (2009a) recognized that the length scale of replacement can also be related to the rate-limiting step between dissolution of the parent and precipitation of the product. Nanometer-scale replacement requires dissolution to be the rate-limiting step, i.e. precipitation is much faster than dissolution. In contrast, in systems where precipitation is rate-limiting, significant mass transport will occur, leading to poor preservation of fine textures. This mass transport can occur over a wide range of scales, from the μm-scale as in Fig. 5f-i or in the replacement of magnetite (Fe3O4) by marcasite (Qian et al., 2010), to the complete uncoupling of anhydrite dissolution and gypsum (CaSO4·2H2O) precipitation as seen in the Naica megacrystals, characterized by extremely slow precipitation rates compared to dissolution (García-Ruiz et al., 2007; Otálora and García-Ruiz, 2014). Hence, the length scale of replacement provides a useful textural guide for establishing qualitative information regarding kinetic processes involved in CDR replacements. A consequence of larger length scales of replacement (e.g. μm-scale) is often the destruction of the parent phase textures. Sibley (1982) observed that during non-pseudomorphic dolomitization, when limited nucleation sites are available, the nuclei preserve the orientation of the parent but subsequent growth destroys the textures associated with host regions with different orientations. Preservation of parent orientation by initial nuclei is evident in the experimental dolomitization of marble (Etschmann et al., 2014), where EBSD images highlight the relationship between calcite and dolomite (CaMg(CO3)2) orientations (Fig. 6). Dolomite nucleation is difficult (e.g., Land, 1998), and as such the precipitation-limited replacement leads to growth dominating the textural development of the product. The destruction of the parent

Fig. 6. EBSD data for dolomite replacing calcitic marble (reprinted from Etschmann et al., 2014 with kind permission from Springer Science and Business Media). Inverse pole figures for (a) calcite, and (b) dolomite. (c) Relative orientation of dolomite replacing calcite (grains represented by red line). Crosses in the calcite grain represent six reference orientations. The scale runs from white (0°) to the color, e.g. red (10°). Orientations greater than 10° are represented in gray scale. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

636

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

textures by the replacing dolomite can be related to the greater rate of parent dissolution and the limited nucleation events. 3. Volume change In the absence of a fluid phase, the reaction volume change in a closed system is a simple parameter that has a first order relevance (e.g., PV work term in Eq. (3); Lindgren, 1912, 1918; Seward and Barnes, 1997). Based on the volume changes given in Table 1, this simple interpretation of volume change would severely limit the viability of numerous replacement systems. The presence of a fluid phase complicates the supposedly trivial volume balance, since elements can be added to, or removed from, the mineral assemblage as solutes. Thus, one must consider the relative solubilities of the parent and product phase and the mobility of species in solution when determining volume changes during solution-mediated mineral replacements (Pollok et al., 2011; Putnis and Mezger, 2004; Putnis, 2009; Zhao et al., 2009). Fig. 7 highlights the limitation of only considering molar volumes when interpreting replacement textures. Molar volume changes during mineral replacements cover a wide range of both positive and negative values (Table 1). Negative molar volume changes, such as seen during the replacement of calaverite by gold (Fig. 7a; Zhao et al., 2009), are associated with porosity in the product phase. However, in systems exhibiting minor decreases in molar volume (Fig. 7b), the volume of reaction-produced porosity is larger than can be explained by the change in molar volume. In the case of patch perthitization (ΔVm ~ − 0.6%) porosities of up to 4.75% in natural samples (Parsons and Lee, 2009; Walker et al., 1995), and 11% in experimental samples (Norberg et al., 2013) have been observed. This porosity is related to the increased solubility of cryptoperthite as a result of stored strain energy associated with coherent exsolution. Thus, a larger volume of cryptoperthite is lost to solution compared to the patch perthite that precipitates. A question remains as to replacements involving substantial increases in volume, particularly when they display textural features associated with dissolution-limited ICDR replacements such as a sharp replacement front and porous replacement product (Fig. 7c).

Li et al. (2015) and Zhao et al. (2014a) observed that the replacement of hematite (Fe2O3) by chalcopyrite (CuFeS2; ΔVm ~ + 196%; Table 1) results in both a porous chalcopyrite that replaces the hematite in a pseudomorphic manner, and a chalcopyrite overgrowth that grows from solution. Hence, it is clear that such a large volume increase is accommodated through the transport of species away from the replacement interface rather than via reaction-induced fracturing as has been observed in other systems exhibiting a volume increase (Jamtveit et al., 2009; Plümper et al., 2012; C.V. Putnis et al. 2007). Texturally, this results in the simultaneous formation of pseudomorphic textures and heterogeneous nucleation. Mobilized species are transported away from the replacement reaction interface, potentially facilitating further replacements elsewhere within a mineral assemblage or, if nucleation sites become available, precipitating away from the interface. The fact that porosity was still observed in the replaced zone and that a chalcopyrite overgrowth formed may be the result of kinetic factors such as the inverse relationship between pore size and threshold supersaturation (Putnis et al., 1995) and the growth mechanism of the product (Hövelmann et al., 2014). The examples considered in Fig. 7 and Table 1 highlight the complexities associated with determining volume change and its impact on replacement textures. Changes in molar volume play a role in the extent of the volume change, but it is the solubility of the two phases that determines the sign of the volume change (Putnis and Austrheim, 2013; Ruiz-Agudo et al., 2014). The solubility of each phase is a function of the grain size, fluid composition, temperature and pressure, among other variables, and hence will likely evolve as the replacement continues (Ruiz-Agudo et al., 2014). Pollok et al. (2011) defined the change in volume by considering not only molar volumes but the relative solubilities of the parent and product: ΔV ¼ 100 (

& ' np Vm;p −nd Vm;d ; nd Vm;d

ð9Þ

where np and nd are the number of moles of product precipitated and parent dissolved and Vm,p and Vm,d are the molar volumes of the precipitating and dissolving phases. A negative volume change will in general lead to the generation of porosity during pseudomorphic

Table 1 Some experimental CDR reactions and the associated calculated molar volume changes. Reaction

Conservative ion

ΔVm

Observed replacement features

Calaverite + 2O2(aq) + 2H2O gold + 2H2TeO3(aq) Gypsum + CO32−(aq) calcite + SO42−(aq)

Au Ca

−79.58% −50.66%

Ilmenite + 2 H+ rutile + Fe2+ + H2O

Ti

−41.53%

2Pyrrhotite + 14 H+ + 7/2O2(aq) 9pyrite + 7Fe2+(aq) + 7H2O 2Pyrrhotite + 14 H+ + 7/2O2(aq) 9marcasite + 7Fe2+(aq) + 7H2O Pentlandite + 3 H+ + 1.875O2(aq) violarite + 1.5Ni2+ + 0.75Fe2O3 + 1.5H2O Cryptoperthite 0.54Ab96Or04 + 0.46Or87Ab17 (patch perthite)

S S S K–Na–Si–O

−32.43% −30.78% −16.63% −0.58%

pseudomorphic (Zhao et al., 2009) pseudomorphic (Fernández-Díaz et al., 2009) pseudomorphic with fracture generation (Janssen et al., 2010) pseudomorphic (Qian et al., 2011) pseudomorphic (Qian et al., 2011) pseudomorphic (Xia et al., 2009a) pseudomorphic (Norberg et al., 2013) pseudomorphic with minor overgrowth (Perdikouri et al., 2013) pseudomorphic (Xia et al., 2009b) ICDR with significant overgrowth (Zhao et al., 2014b) pseudomorphic with minor overgrowth (Qian et al., 2011) pseudomorphic (Cruz-Uribe et al., 2014; Kapp et al., 2009) replacement with fracture generation (Kelemen and Hirth, 2012; Plümper et al., 2012) replacement and overgrowth (Qian et al., 2010) replacement and overgrowth (Qian et al., 2010) ICDR with significant overgrowth (Zhao et al., 2014b) pseudomorphic (Schultheiss et al., 2013) ICDR with significant overgrowth (Zhao et al., 2014a)

Aragonite calcite

Ca–C–O

8.12%

Leucite + Na+ + H2O analcime + K+

Al–Si

9.76%

2Chalcopyrite + 3Cu(HS)−2 + 3OH− bornite + Fe(OH)2 + 6HS− + 0.5H2O + 0.25O2(aq)

S

12.54%

2Pyrrhotite + 14H2S(aq) + 7O2(aq) 16pyrite + 14H2O

Fe

15.70%

2Zoisite + rutile + quartz 3anorthite + titanite + H2O

Ti

18.78%

2Forsterite + 3H2O lizardite + brucite

Si

47.83%

Magnetite + 6H2S(aq) + O2(aq) 3pyrite + 6H2O Magnetite + 6H2S(aq) + O2(aq) 3marcasite + 6H2O

Fe Fe

59.80% 63.72%

Chalcopyrite + 4Cu(HS)−2 + 2OH− bornite + 6HS− + 2H2O

Fe

125.1%

Magnesite + (NH4)2H(PO4)(aq) dittmarite + NH3 + CO2

Mg

152.4%

Hematite + 2CuCl−2 + 4H2S(aq) 2chalcopyrite + 4Cl− + 2 H+ + 3H2O

Fe

196.3%

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

637

replacements, allowing continued fluid infiltration towards the replacement interface. Pervasive replacement at the grain scale is thus facilitated by the generation of porosity, which in itself is closely controlled by the change in volume during replacement. Putnis and Austrheim (2013) note that in systems moving towards equilibrium, initial porosity formation is dominated by the difference in solubility. However, closer to equilibrium the difference in molar volume begins to play a greater role. Furthermore, volume changes not only result in the development of porosity but can lead to the generation of microfractures as a result of generated stresses that provide additional pathways for fluid transport (Etschmann et al., 2014). This is discussed in the following section. 4. Fluid pathways and porosity Efficient fluid flow at both macroscopic and microscopic levels is a key requirement for large-scale (m to km) mineral replacement (e.g., dolomitization; serpentinization; hydrothermal alteration; Harlov et al., 2005; Konrad-Schmolke et al., 2011). The patterns of replacement are usually controlled by the major fluid pathways (e.g., flow in fractures; porous flow). An important feature of CDR reactions is the creation of porosity on the mineral grain scale. The interaction between pre-existing flow paths and reaction-enhanced porosity affects fluid flow in complex, poorly understood ways. For example, positive feedback from the creation of porosity via CDR reactions can result in ‘porosity waves’ in metamorphic environments (Connolly and Podladchikov, 2013). Fig. 8a–d illustrates a diverse range of macroscopic textures resulting from the coupling between large-scale fluid flow and CDR reactions, which, by generating reaction-induced porosity, enables atomistic level mineral–fluid interactions through the entirety of the rock. Fig. 8a and b illustrate the coupling between fractured flow and CDR. Fig. 8c is an example of infiltration metasomatism, and Fig. 8d shows the weathering of a fractured, homogeneous rock. Weathering along fractures is commonly observed and is useful evidence for identifying limited fluid events (e.g., Putnis and Austrheim, 2013). All these reactions can be considered as replacement reactions, essentially preserving the volume of the protolith. In this section we illustrate the different grain-scale textures that arise from the different mechanisms that ensure that CDR replacements behave in a self-propagating manner, by generating the porosity required for the reaction to progress at grain-scale. We distinguish four different mechanisms, ranging from the formation of ‘homogeneous’ reaction-induced porosity in the product (e.g., Putnis, 2002, 2009), to reaction-induced microfracturing ahead of the front (e.g., Plümper et al., 2012), to enhancement of flow along grain boundaries and internal structural features of the minerals (cleavage planes and twin boundaries) (Fig. 8e–k). 4.1. ‘Homogeneous’ reaction-induced porosity

Fig. 7. CDR replacements act over a wide range of changes in molar volume. (a) Metallic gold replacing calaverite involves a large decrease in molar volume (ΔVm ~ −80%) that results in abundant porosity (reprinted from Zhao et al., 2009 with permission from Mineralogical Society of America). (b) Coarsening of cryptoperthite (medium gray) to patch perthite (albite dark gray, microlite light gray) is close to isovolumetric (ΔVm ~ −0.6%) but up to 4.5% porosity has been observed in natural samples (reprinted from Parsons and Lee, 2009 with kind permission from Springer Science and Business Media). (c) Sulfidation of hematite is accompanied by a significant volume increase (ΔVm ~ 196%), but porosity is still observed in the replacement zone (reprinted from Li et al., 2015 with permission from Mineralogical Society of America). The dashed yellow lines represent the boundary between replacement chalcopyrite and overgrowth chalcopyrite. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Reaction-induced pores provide an efficient mass transport pathway between the reaction interface and the bulk solution by maintaining fluid contact with the replacement front, which is well illustrated in the replacement of leucite (KAlSi2O6) by analcime (NaAlSi2O6·H2O) (Fig. 8e and f; Xia et al., 2009b). The textures associated with reactioninduced porosity are dependent on the nucleation rate and growth mechanism of the product phase, and are influenced by the overall change in volume during the replacement. A small lattice misfit facilitates the topotactic replacement that results in the development of a highly ordered pore network, with higher leucite solubility accounting for molar volume increase (see Eq. (9)). This ordered and connected porosity thus provides an efficient means by which species can be transported between the exterior of the grain and the replacement front that migrates towards the grain center (e.g., Fig. 2). In the case of non-pseudomorphic replacement, when the precipitation step is ratelimiting, the growth mechanism controls the nature of the porosity.

638

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

Hövelmann et al. (2014) demonstrated the importance of the growth mechanism on the formation of porosity during the replacement of plagioclase ((Na,Ca)(Si,Al)4O8) by cordierite (Mg2Al4Si5O18). The decoupling of plagioclase dissolution and cordierite precipitation results in a non-pseudomorphic replacement. Porosity observed by Hövelmann et al. (2014; their Fig. 1) is the result of cordierite growth under conditions of high supersaturation (e.g., initially greater than σ**, see Fig. 3). Under high supersaturation, fast cordierite growth traps channel-like structures enabling formation and preservation of porosity (Cho and Fawcett, 1986). Subsequently, decreasing degrees of supersaturation shift the growth mechanism and cordierite forms via dislocation-controlled growth (e.g., Hövelmann et al., 2014; their Fig. 3). 4.2. Reaction-induced fracturing Fractures within rocks provide localized regions of high permeability enabling high rates of mass transfer (Brantley et al., 1990). The generation of fractures has long been associated with deformation and deviatoric stresses associated with tectonic processes (Iyer et al., 2008; Putnis and Austrheim, 2010), but a number of studies have recently demonstrated the role of CDR replacements in fracture propagation (Jamtveit et al., 2009; Plümper et al., 2012; C.V. Putnis et al. 2007). Reaction-generated fractures are now recognized as an important factor in enhancing mineral replacements in both natural (Jamtveit et al., 2009; Røyne et al., 2008) and experimental/ engineered (Dunkel and Putnis, 2014; Janssen et al., 2010) systems. Plümper et al. (2012) suggested initial fracture propagation to be a result of stress build up associated with a CDR replacement within surface defects such as etch pits during anisotropic dissolution processes. Kelemen and Hirth (2012) estimated that stresses generated during the serpentinization of olivine ((Mg,Fe)2SiO4) could exceed 300 MPa, sufficient to fracture rocks within the upper 10 km of the Earth's crust. The importance of etch pits to fracture generation was recently demonstrated experimentally for the serpentinization of olivine (Malvoisin et al., 2012) and the replacement of scolecite (CaAl2 Si 3O 10 ·3H2 O) by tobermorite (Ca4.5 Na 1.3Si5.2 AlO 16(OH) 2 ; Dunkel and Putnis, 2014). A key result of reaction-induced fracturing is the initial acceleration of the replacement rate due to the generation of new reactive surface area (Jamtveit and Austrheim, 2010; Majumdar et al., 2014; Røyne et al., 2008). Several serpentinization experiments have demonstrated the importance of new surface area generation as replacement is arrested after a limited time despite a high initial rate of replacement in the absence of fracture formation (Godard et al., 2013; Malvoisin and Brunet, 2014). Jamtveit and Hammer (2012) note that fluids infiltrating the newly generated fractures result in new etch pits forming due to anisotropic dissolution, which facilitates hierarchal fracturing, often resulting in a mesh-like texture (Fig. 8g and h). Original grains are split into subdomains as a result of reaction-induced fracturing, generating new surfaces where replacement can begin, and greatly enhancing the replacement rate. Importantly, reaction-induced fracturing has been observed to occur across a wide range of length scales (Jamtveit and Austrheim, 2010; their Fig. 3). Therefore, fracturing as a direct result of replacement processes is an important factor both in the

development of replacement textures and in the ability of a fluid phase to penetrate a rock. Reaction-induced fracturing also demonstrates the role of local fluid chemistry in defining replacement textures through its control of the volume change. Observation of experimental and natural samples, as well as numerical modeling, has demonstrated a common set of fracture patterns that are dependent on the direction and extent of the volume change (Okamoto and Shimizu, 2015; Ulven et al., 2014a, 2014b). The local fluid plays two important roles. Firstly, it defines the relative solubilities of the parent and product phases and will therefore control the volume change as defined by Eq. (9) (Pollok et al., 2011). Secondly, the degree of supersaturation of the product within the interfacial solution will influence the force of crystallization impacting the stresses generated (Kelemen et al., 2011; Kelemen and Hirth, 2012; Scherer, 1999; Steiger, 2005). This further emphasizes the importance of threshold supersaturation as described in Section 1.3 — higher threshold supersaturation results in higher crystallization force. Volume changes associated with replacement processes generate both tensile and compressional stresses at the replacement front. When such stresses exceed the tensile strength of the parent or product phase, fractures will be generated with a pattern that depends upon the nature of the volume change. When the replacement involves a decrease in volume, as seen during eclogitization of anorthosites (Jamtveit et al., 2000) and the replacement of ilmenite (FeTiO3) by rutile (TiO2; Fig. 8g; Janssen et al., 2010), branch-like fractures form, enabling efficient ingress of infiltrating fluids. On the other hand, the patterns associated with increases in volume will depend on the extent of the volume change. Ulven et al. (2014b) note that the limited volume increase allows the replacement to progress further without generating significant compressional stresses, leading to a build-up of tensile stress within the host that can result in tensile fracturing of the host phase itself. This is the dominant mechanism in the formation of hierarchal fracture patterns (see Fig. 8h). Hierarchical fracturing provides an important means for continued fluid access during serpentinization and carbonation of peridotites (e.g., Iyer et al., 2008; Kelemen and Matter, 2008; Okamoto et al., 2011; Plümper et al., 2012; Power et al., 2013). Compressional stresses that build up in systems undergoing large volume increases will result in spalling due to shear fracturing (Ulven et al., 2014b). This often results in the rounding of replacement textures as seen during spheroidal weathering (e.g., Jamtveit et al., 2011; Røyne et al., 2008; Ulven et al., 2014a) and pervasive hydrothermal alteration (e.g., Fig. 8d). 4.3. Grain boundary Fractures provide localized zones of high mass fluxes. However, large-scale mineral replacements require a fluid phase to pervasively penetrate significant volumes of rock. Fluid-saturated grain boundaries provide such a means for extensive mass transport by providing diffusional shortcuts relative to dry systems (Gardés et al., 2012; Rubie, 1986). A fluid capable of forming an interconnected network along grain boundaries is thus important for regional-scale mineral replacements (Harlov et al., 2005). While numerous experimental studies

Fig. 8. The ability of a fluid to penetrate rock is provided over a range of scales due to mineral and replacement textures. (a–d) Feedback between macroscopic flow paths and CDR reactions result in different topographies for the replacement. (a) Carbonate-potassic alteration associated with gold mineralization at the Sunrise Dam Au deposit is controlled by faults (Brown et al., 2002; Sung et al., 2007, 2009; photo JB); similarly (b) the sulfidation of banded iron formation at Sunrise Dam is controlled by faults, and then the reaction proceeds preferentially along particular layers of the original stratigraphy (reprinted from Qian et al., 2010 with permission from Elsevier). (c) The Neoproterozoic siliciclastic sediments of Hidden Valley, Northern Flinders Ranges were affected by pervasive fluid infiltration from the Paleozoic British Empire Granite, resulting in the replacement of some layers by coarse pink orthoclase (Weisheit et al., 2013; photo JB) (d) Alteration (silica-clay-hematite; Dobson et al., 2003) of Yellowstone rhyolite (640–160 ky) by geothermal fluids at Lower Falls, Yellowstone Grand Canyon. Image height ~ 60 m (photo JB). (e) Porosity within analcime replacing leucite, and (f) leucite twin boundaries provide effective pathways for fluids (reprinted with permission from Xia et al., 2009b. Copyright 2009 American Chemical Society). Reaction-induced fracturing resulting (g) in fish-bone-like textures during rutile replacement of ilmenite (from Janssen et al., 2010), and (h) mesh-like textures associated with olivine serpentinization (reprinted from Plümper et al., 2012), provide new pathways for infiltrating fluids. (i) Calcite grain boundaries provide flow pathways for dolomitizing fluids (reprinted from Etschmann et al., 2014 with kind permission from Springer Science and Business Media). Fluid penetration along cleavage planes during the transformation of (j) rhodonite (MnSiO3) to tephroite (Mn2SiO4) (from Brugger and Gieré, 2000) and (k) chalcopyrite to bornite (reprinted from Zhao et al., 2014b with permission from Mineralogical Society of America).

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

have demonstrated that porosity is vital for complete replacement of single grains (see Putnis, 2009 and references therein), Jonas et al. (2014) established that the rate of mass transport along grain boundaries was up to an order of magnitude greater than mass transport

639

through reaction-induced porosity. Fluids penetrating grain boundaries can greatly enhance mineral replacement rates as fluids are able to surround grains and then migrate towards the grain center via CDR replacement (Fig. 8i).

640

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

4.4. Cleavage and twin boundaries Preferential flow paths leading to irregularly shaped replacement fronts have also been found to result from mass transport through twin boundaries and cleavage planes. Twin boundaries have been seen to develop preferential mass transport paths by acting as oriented grain boundaries (Fig. 8f). Pearce et al. (2013; their Fig. 3) found that siderite (FeCO3) and dolomite replacing calcite nucleated along calcite twin boundaries, with the oriented nucleation of the products emphasizing preferential migration of the infiltrating fluid phase. Similarly, twin boundaries have been observed to play an important role in the replacement of leucite by analcime (Xia et al., 2009b). The irregular nature of the reaction front often observed in CDR replacements (Fig. 8j and k) can be related to cleavage planes acting as favorable fluid pathways. The ability of a fluid to migrate along cleavage planes is particularly important in the case of bornite–digenite solid solution (Cu5FeS4–Cu9S5) replacing chalcopyrite (Zhao et al., 2014b). As the replacement progresses the reaction-induced porosity is destroyed, and the replacement eventually continues along the chalcopyrite cleavage plane resulting in a highly irregular reaction front (Fig. 8k). The highly irregular nature of several replacement systems can be related to anisotropic dissolution along cleavage planes. For example, saw-tooth like replacement fronts have been observed in feldspar replacements (e.g., Niedermeier et al., 2009; Norberg et al., 2011), which has been suggested to be related to the weakness of the Al–O–Si bond relative to Si–O–Si bond within the tetrahedral framework of alkali feldspar (Arvidson et al., 2004; Schott and Oelkers, 1995; Schott et al., 2009; Xiao and Lasaga, 1994). Similarly, a number of investigations have illustrated the anisotropic dissolution of olivine, which is an important process in the fracture generation associated with serpentinization (King et al., 2010, 2014; Peuble et al., 2015). 5. Nature of the end products Mineral assemblages are key information on which to base the reconstruction of the pressure, temperature, and chemical conditions of rock formation. Especially at elevated temperature (e.g. ore-formation; metamorphic petrology), mineral assemblages are interpreted on the basis of equilibrium thermodynamics. One of the fundamental principles of equilibrium thermodynamics is that the mineral assemblage depends only on the pressure, temperature, and composition of the system, but not on the reaction pathway. Yet, CDR reactions appear to provide a large number of examples where this principle does not seem to apply, or applies at a local scale that does not reflect the overall fluid–rock interaction history of the mineral assemblage. Specifically, several experimental studies of CDR replacements have found that the end product that formed was metastable (either intrinsically, or relative to the bulk solution composition) or was not the first possible stable product expected from the composition of the initial bulk solution, yet may become thermodynamically stable as fluid compositions evolve as a result of the replacement reaction. We illustrate the role of the reaction mechanism in controlling the final mineral assemblage using two examples (Fig. 9): the replacement of pentlandite by violarite (Tenailleau et al., 2006; Xia et al., 2009a), and the replacement of calaverite by gold (Okrugin et al., 2014; Zhao et al., 2009). In the case of pentlandite to violarite, thermodynamic modeling suggests that alternative product phases were thermodynamically stable with respect to the system temperature, pressure, and bulk composition. This is illustrated in Fig. 9a, showing that polydymite (Ni3S4), heazlewoodite (Ni3S2), and millerite (NiS) are all supersaturated as violarite forms. Yet, violarite was the only Ni-sulfide observed in hydrothermal experiments conducted over a wide range of temperatures (125–210 °C) and solution compositions (Tenailleau et al., 2006; Xia et al., 2007, 2009a). In experiments conducted at similar temperature, but under dry conditions, Xia et al. (2009a) observed a complex assemblage of phases, more consistent with the predictions from the

thermodynamic model. This example illustrates the key role of the pentlandite substrate during CDR. Violarite and pentlandite share the same arrangement of closed packed S atoms, and the pentlandite surface catalyzes the nucleation of violarite at the expense of the more thermodynamically stable phases. Once violarite forms, no other Ni– Fe-sulfide nucleates, and the final product is violarite. Interestingly, the final product is close to equilibrium with the final experimental solution, and hence there is little or no reason for violarite to react at this stage. In the case of the transformation of calaverite to gold, equilibrium thermodynamics modeling of the experimental system predicts the congruent dissolution of calaverite. The resulting increase in Te concentration in solution leads to the precipitation of tellurite (TeO2; Fig. 9c). All Au present in calaverite can be put into solution in this scenario. In experiments, however, the dissolution of calaverite is coupled with the precipitation of gold near the reaction front (Fig. 9d). In the final assemblage, gold and tellurite coexist as stable products in thermodynamic equilibrium with the solution, with fO2(g) values buffered by the Au(s)-calaverite assemblage (Zhao et al., 2009). The process responsible for Au(s) precipitation at the calaverite surface remains unknown. There is no crystallographic relationship between calaverite and gold, so that either local solution conditions at the reaction front or catalytic effects at the calaverite surface may be responsible. Whatever the precise mechanism is, surface-enhanced nucleation in this example controls the final stable mineral assemblage in this reaction. There are several other examples of surface-enhanced nucleation controlling the final assemblage of CDR replacement reactions in sulfide systems. During the experimental sulfidation of hematite in Cu(I)- and (Cu(I) + Fe(II))-bearing solutions, chalcocite (Cu2S) and bornite, respectively, were the thermodynamically stable phases, yet chalcopyrite was the first phase to form on the hematite surface (Li et al., 2015; Zhao et al., 2014a). In the Fe(II)-rich runs, bornite eventually nucleated, resulting in a rim of bornite surrounding a rim of chalcopyrite (Zhao et al., 2014a; their Fig. 7). Trace elements present in solution but not formally involved in the reaction have recently been shown to control the reaction path in CDR reactions. Li et al. (2015) demonstrated that the addition of U during the sulfidation of hematite in Cu-bearing solutions leads to the formation of pyrite (FeS2), whereas no pyrite was observed in U-absent runs, where chalcopyrite formed instead. This suggests that dissolved U (as uranyl complexes) acts as a catalyst for the formation of pyrite. Subsequently, pyrite was replaced by chalcopyrite, with a thin layer of scavenged U forming at the interface. Qian et al. (2013) showed that the reaction pathway can control the chemistry of the product. Arsenic in cationic form replaces Fe rather than substituting for S in anionic form, as is the case for most natural arsenian pyrite (i.e., ((Fe,As)S2) vs. (Fe(S,As)2), in arsenian pyrite formed experimentally via the replacement of magnetite under mild hydrothermal conditions (T = 125 and 220 °C; Psat) in the presence of S(− II) and various As-containing species. This provides a pathway explaining the occurrence of pyrite containing As in cationic form in some epithermal Au deposits (Deditius et al., 2008). The Fe disulfide polymorph that forms via the replacement of pyrrhotite is closely related to the degree of supersaturation at the interface. Qian et al. (2011; their Fig. 8) identify pH and external S(− II) sources as being critical to the nature of the Fe disulfide phase that forms. Low supersaturation favors the epitaxial nucleation of marcasite as a result of having a S-lattice similar to that of pyrrhotite. The presence of an external S(−II) source significantly increases pyrite (FeS2) supersaturation, where a high degree of supersaturation is critical for pyrite nucleation as previously noted (Harmandas et al., 1998; Rickard et al., 2007). A similar phenomenon may be responsible for the widespread occurrence of dolomite replacing calcite/aragonite during diagenesis. Heterogeneous nucleation of dolomite is extremely unlikely, leading to the so-called ‘dolomite problem’ (e.g., Al-Awadi et al., 2009; Hardie, 1987; Warren, 2000). Etschmann et al. (2014) highlighted the role of grain

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

641

boundaries as acting as microreactors in the dolomitization of calcite. They proposed a three-step model in which initial calcite dissolution facilitates the nucleation and growth of a Mg-poor protodolomite, which is subsequently replaced by stoichiometric dolomite with time. In this model, the proto-dolomite composition evolves towards stoichiometric dolomite via subsequent CDR. This is driven by the higher solubility of the proto-dolomite. Increased exposure time to the fluid facilitates the formation of the stoichiometric dolomite via this step-wise process (Etschmann et al., 2014; Katz and Matthews, 1977). Jonas et al. (2014) found a similar situation occurring in the replacement of calcite by calcium phosphates, in which an intermediate, finely porous β-tricalcium phosphate (β-Ca3(PO4)2) formed first before being replaced by a hydroxylapatite exhibiting coarser porosity. In these two examples the grain boundaries allow an intermediate phase to form first, which, with time, is then replaced by the final product. 6. Dynamic evolution of textures, porosity, and composition during CDR Mineral systems are constantly evolving in response to changing system properties (e.g. T, P, X). Given the dynamic nature of CDR reactions, one can expect a complex evolution as replacement reactions progress. Indeed, complex textures have been observed to develop in experiments, in spite of simple T–P–X evolutions of the bulk systems. The compositional and textural evolution of the reaction products are controlled by local conditions at the reaction front and re-equilibration of the metastable products that initially formed because of kinetic and local conditions. For example, the replacement of calcite by hydroxylapatite in arsenate-bearing solutions results in a highly complex and dynamic As distribution (Fig. 10). Complex textures also evolve where fluid-mediated and solid-state mechanisms are interacting (e.g., patch perthitization, Fig. 11a-c; replacement of sylvanite ((Au,Ag)2Te4) under mild hydrothermal conditions, Fig. 11d, e). Textural evolution can also be evidenced from changes in the nature and distribution of porosity through time, as graphically illustrated by the replacement of KBr(s) by K(Cl,Br)(s) (Fig. 12). 6.1. Complex compositional zoning in solid solutions: local equilibrium revealed

Fig. 9. Reaction path and end products for the pentlandite to violarite (a,b; Xia et al., 2009a) and calaverite to gold (c,d; Zhao et al., 2009) reactions. (a) Equilibrium calculation for the oxidation of pentlandite (40 mg) in 6 g of solution at 200 °C; 9.68 × 10–5 moles oxygen (~3 mg) from air are titrated into the solution. (b) In the experiments, violarite was the only phase forming. (c) Under equilibrium conditions, for the experimental conditions (10 mg calaverite in 15 ml solution, air in the cell (~8.45 × 10–5 moles), 220 °C), the model predicts congruent calaverite dissolution, but (d) in reality the dissolution of calaverite is coupled to the precipitation of gold; this can be modeled assuming that all gold immediately precipitates from solutions.

Solid solutions are extensive in both natural and industrial crystallization processes. Solid solution – aqueous solution interactions on a local scale in binary SSAS systems can be interpreted using Lippmann diagrams (Fig. 13; Bruno et al., 2007; Lippmann, 1980). The composition of a solid solution is closely related to how the local solution composition at the reaction interface evolves (e.g., Majumdar et al., 2014). It is also influenced by the relative solubilities of the end members, with similar end member solubilities defining a larger range of intermediate compositions (Fig. 13; Prieto et al., 1997; Zhang et al., 2011). We examine the sensitivity to changes in local solution chemistry and the role played in controlling the evolution of solid solutions using two examples: K(Cl,Br)(s) replacing KBr(s) in saturated KCl solutions (Putnis and Mezger, 2004; Putnis et al., 2005; Raufaste et al., 2011), and As-bearing hydroxylapatite (Ca10(PO4)6-x(AsO4)x(OH)2) replacing calcite in a mixed arsenate-phosphate solution (Borg et al., 2014). By considering a localized replacement zone, the Lippmann diagram for the KBr–KCl–H2O system can be utilized to follow compositional evolution when a KBr crystal is placed into a saturated KCl solution (Fig. 13a). Upon dissolution of the KBr crystal, the composition of the local fluid becomes richer in Br (dashed arrow in Fig. 13a). The local solution will quickly reach the solutus curve, at which point it is in equilibrium with a Br-rich K(Br,Cl)(s) phase, defined by S1 (Fig. 13a), that precipitates in (near-)equilibrium with the local solution. This equilibrium is only temporary, as the continued transport of species from the bulk will see the local Cl concentration increase. This

642

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

Fig. 10. Calcite replacement by As-bearing apatite. (a) Regions of low and high As concentrations develop in the replaced regions. Quantitative spot analysis results for As are 1.01% (1), 13.15% (2), and 8.61% (3). (b) Heterogeneous As distribution in the apatite rim after 5 h. (c) Homogeneous distribution within a fully replaced grain after 48 h. Adapted from Borg et al. (2014).

shifts the equilibrium composition towards the Cl-rich endmember (solid arrow in Fig. 13a). The product composition continually adjusts towards the KCl endmember as the local solution equilibrates with that of the bulk, moving the composition towards Sf. Kasioptas et al. (2008, 2011) investigated the replacement of calcite by hydroxylapatite via a phosphate-rich solution. Borg et al. (2014) found that the addition of arsenate (AsO34 −) moieties to the solution initially results in a complex zoning pattern of As within the apatite structure (Fig. 10a, b), but with time the zoning disappears and a homogeneous As-bearing hydroxylapatite forms as the final product. By considering the Lippmann diagram (Fig. 13b) for hydroxylapatite and its As-equivalent johnbaumite (Ca5(AsO4)3(OH)), we find a similar situation as with KCl-KBr. Calcite dissolution at the interface provides a source of Ca2+(aq), shifting the local composition (dashed

arrow in Fig. 13b) until it intersects the solutus curve. The difference in solubility favors the uptake of the least soluble endmember and as such As-poor hydroxylapatite will be in equilibrium with the solution, even in As-rich solutions. Initial hydroxylapatite formation increases the local As concentration in the interfacial solution (solid arrow in Fig. 13b). Borg et al. (2014) observed that an As-rich (up to 8 times bulk solution concentration) hydroxylapatite forms as a result of localized increase in the As concentration (Fig. 10a), moving the equilibrium solid composition back towards the hydroxylapatite endmember (dotted arrow, Fig. 13b). As the replacement front migrates through the calcite, a series of As-rich and As-poor zones develop (Fig. 10b). After complete calcite dissolution, homogenization of the As in the hydroxylapatite occurs (Fig. 10c) as the local solution equilibrates with the bulk solution, causing recrystallization of the initial product.

Fig. 11. The interaction of CDR and SSD as illustrated in (a-c) patch perthitization (reprinted from Norberg et al., 2013 with permission from Mineralogical Society of America), and (d-e) dealloying of sylvanite (reprinted from Zhao et al., 2013 with permission from Mineralogical Society of America). (a) A transition zone develops, in which coarsening of cryptoperthite is enhanced by water-assisted diffusion of Na and K, resulting in (b) maintenance of coherency during coarsening. (c) CDR driven by the reduction of strain and surface energy then drives the coarsening process leading to a loss of coherency between albite and sanidine. (d) Products and textures of partially replaced sylvanite highlighting the exsolution of calaverite-I to calaverite-II and phase X (represented by petzite and hessite, which are likely the result of exsolution of phase X during quenching; e.g. Cabri, 1965). (e) Elemental map of a partially replaced sylvanite ([Au,Ag]2Te4) grain. A porous rim of Au-Ag alloy was always observed to form, while a complex set of textures developed internally consisting of calaverite, petzite and hessite.

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

643

Fig. 12. Images of textural healing after KBr is replaced by K(Cl,Br) (reprinted from Putnis et al., 2005 with permission from Mineralogical Society of America). (a) Partially reacted sample exhibiting the K(Cl,Br) reaction rim. (b) SEM image of completely replaced KBr crystal. The interior consists of coarse, channel-like pores, but the porosity at the exterior has started to close. (c) After 24 h pores continue to coarsen. (d) Transmitted light microscope image after 12 days shows continued healing, which result in a transparent K(Cl,Br) crystal with a thick, non-porous exterior.

6.2. Changes in mineralogical composition A complex evolution of the chemical composition of the product has been illustrated in Section 6.1. Here, a number of experimental systems also display a change in mineralogy closely related to the systems evolution towards an equilibrium position via a number of product polymorphs. Such a process is often referred to as Ostwald's step rule, in which the first phase to precipitate may be some stable nano-cluster (Demichelis et al., 2011; Gebauer et al., 2014; Pouget et al., 2009) or a metastable phase that is in the nearest state to the original state (Cardew and Davey, 1985; Morse and Casey, 1988; Threlfall, 2003). Kinetic factors play a crucial role in driving the system towards a stable product phase via a number of less stable (more soluble) amorphous and crystalline polymorphs (Ogino et al., 1987; Rodriguez-Navarro et al., 2015). This is well illustrated in the replacement of gypsum by calcium carbonate, where the first phase observed to form was an amorphous calcium carbonate (acc), followed by vaterite or aragonite, and finally calcite (Fernández-Díaz et al., 2009). The room temperature transformation of the calcium carbonate phases is facilitated by the presence of the fluid phase. The formation of acc is the result of high degrees of supersaturation with respect to all carbonate phases within the interfacial fluid. Studies investigating the transformation of acc to crystalline calcium carbonate have demonstrated that the initial amorphous phase is hydrated and highly disordered (Baltrusaitis and Grassian, 2009; Rodriguez-Blanco et al., 2011; Rodriguez-Navarro et al., 2015). The transformation of the amorphous phase to a crystalline product occurs with the dehydration of the initial hydrated acc. Ihli et al. (2014) determined that in dry systems crystallization of calcite after anhydrous acc is associated with high activation energies (~ 245 kJ mol− 1) and high temperatures are required for a solid state transformation. However, in the presence of water, either as a bulk fluid phase or adsorbed on the surface in a humid environment, the transition can occur at room temperature via a CDR process (Rodriguez-Navarro et al., 2015). The complexities involved in this evolution towards a stable crystalline phase may also be related to other systems such as amorphous titania-anatase/rutile or ferrihydrite

(Fe2O3·0.5H2O)-hematite/goethite (FeO(OH); Ihli et al., 2014), highlighting the importance of microscopic processes in the evolution of product mineralogy. 6.3. Complex textures via competition among solid-state diffusion and CDR Mineral replacements occurring in the presence of a fluid phase will kinetically favor a CDR mechanism rather than solid-state diffusionbased mechanisms in many mineral systems under most crustal conditions (Putnis, 2002; Zhao et al., 2013). Generally speaking diffusion within the mineral lattice is favored at higher temperature where the thermal energy is sufficient to break the activation energy barrier to diffusion (Mehrer, 2007). In cases where the rates of diffusion-based reactions and CDR reactions are in similar range, both mechanisms can interact, resulting in complex textures (Fig. 11). Specifically, several experimental studies found that the interaction of SSD and CDR during fluid-mediated replacement drives multistage replacements via intermediate phases. We illustrate the influence of the interactions between solid-state and CDR mechanisms on the textural and compositional development by considering two examples: perthitization (Norberg et al., 2013), and dealloying of sylvanite (Zhao et al., 2013). Cryptoperthite, i.e., strain-controlled intergrowths of plagioclase and K-feldspar (KAlSi3O8), have been observed to coarsen by up to three orders of magnitude in the presence of a melt or an aqueous fluid (Norberg et al., 2013; Parsons and Lee, 2009; Smith, 1974; Worden et al., 1990). Perthitization is essentially isochemical and coarsening is therefore primarily driven by the strain and surface work terms of Eq. (3). Coherency that develops during cooling from sub-solidus temperatures results in a strain energy (2–4.5 kJ mol−1) between perthite lamellae, which is much greater than the contribution of coherent surface energies, which have been measured to be ~ 10 J mol− 1 (Brown and Parsons, 1993; Lee and Parsons, 1995; Parsons et al., 2013; Putnis, 1992; Robin, 1974). Thus, it is this strain energy that drives the replacement upon contact with an infiltrating fluid. Initial contact with the fluid results in the formation of a transition zone (Fig. 11a, b), in

644

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

Fig. 13. Complex compositional evolution during replacement of solid solution minerals (adapted from Pollok et al., 2011). a) The non-ideal Lippmann diagram for the K(Br,Cl)–H2O SSAS system developed after Glynn et al. (1990), who used the solubility data of Durham et al. (1953) to model the endmember solubility products and non-ideality parameters. The reaction path was developed based on the starting conditions of Putnis and Mezger (2004), with S0 and Sf representing initial and final solid compositions. b) Ideal Lippmann diagram for the hydroxylapatite-johnbaumite-H2O SSAS system developed using solubility products of Zhang et al. (2011). The reaction path follows that described by Borg et al. (2014). Tie-lines connect equilibrium fluid (square) and solid (circle) compositions.

which coarsening begins, but coherency remains between the lamellae. This process has been suggested to be the result of accelerated diffusion of Na and K caused by the diffusion of hydrous species within the transition zone ahead of the CDR replacement front (Brown et al., 1983; Brown and Parsons, 1984; Norberg et al., 2013). Gin et al. (2013; their Fig. 6) observed a similar, albeit at a much finer scale, phenomenon during the corrosion of glass where H and Li counter-diffusion profiles develop 15 nm in front of the B and Na dissolution front, which exhibits a sharp (3 nm) change in profile. At a certain point the strain energy associated with perthite lamellae coherency drives the dissolution of the cryptoperthite leading to the complete loss of coherency and a reduction in strain energy. Fig. 11c highlights the subsequent role the surface work term of Eq. (3) as the size of the patch perthite increases with distance away from the replacement front. This example highlights the multistage nature of perthitization, in which diffusion plays an important role in initial coarsening at the replacement front. In the case of dealloying of sylvanite under mild hydrothermal conditions (Zhao et al., 2013), the interaction of CDR and SSD during replacement results in a complex mineralogical and compositional evolution and complex final textures that belies the simple P,T,X reaction conditions (Fig. 11d). The initial formation of a metastable Agrich, Te-poor calaverite (calaverite-I; (Au0.78Ag0.22)Te1.74) occurs via an ICDR replacement of the parent sylvanite. The composition of the calaverite reflects the Ag- and Te-rich composition of the solution near the dissolving sylvanite surface. Once ~ 30% of sylvanite is replaced by

calaverite-I, calaverite-I starts to break down as a result of a SSD reaction into a non-porous, near-stoichiometric calaverite (calaverite-II; AuTe2) and a Ag-rich phase X. Upon quenching, phase X exsolves into hessite (Ag2Te) + petzite (Ag3AuTe2; Fig. 11d; Zhao et al., 2013). The domination of SSD in the second step is likely related to high metal ion mobility within the petzite-hessite region of the Au–Ag–Te system (Zhao et al., 2013; their Fig. 6), but could also be related to water-assisted diffusion as in the cryptoperthite coarsening example discussed above. Continued interaction with the replacement fluid sees the exsolution products undergo another CDR replacement to a Au–Ag alloy, but with a looser coupling at the interface (Fig. 11e). The final product of the reactions shows complex textures that could easily be mistaken as representing a complex evolution of the hydrothermal conditions (P, T, X), rather than a complex reaction pathway (Fig. 11d, e). 6.4. Ripening by surface energy minimization The generation of porosity and new grain boundaries during CDR replacements is crucial for continued reaction progress. However, the generation of new surfaces (nano-grains and nano-pores) also increases surface energy, further driving the continued equilibration of the product phase (Eq. (3)). For example, the replacement of solid solutions as discussed above (Section 6.2) is accompanied by the concurrent evolution of the product textures. Considering the KCl–KBr–H2O system, the initial replacement of KBr by K(Cl,Br) results in a highly porous product

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

(Fig. 12a). After complete replacement, the K(Cl,Br) product develops a coarse internal porosity while the exterior porosity has closed (Fig. 12b). Continued evolution results in further coarsening of the internal porosity (Fig. 12c), while complete healing of the exterior microporosity results in a transparent rim (Fig. 12d). The coarsening of porosity is driven by the reduction of surface area per unit volume via Ostwald ripening (Ratke and Voorhees, 2002). A similar process was observed by Borg et al. (2014) where tube-like structures develop within the As-apatite formed via replacement of calcite (Fig. 10b). The equilibration of the solid and fluid compositions drives the concurrent destruction of these structures and leads to a reduction in porosity as equilibrium is approached. For a given supersaturation there exists one grain size that is at a state of equilibrium. This grain size is defined by the Gibbs-Kelvin equation (Cole and Chakraborty, 2001; Nielsen, 1964), r) ¼

ð2γvÞ ; ðkT lnσ Þ

ð10Þ

where r* is the critical radius in true equilibrium with the solution, γ is the surface energy, v is the molecular volume, k is Boltzmann's constant, T is the absolute temperature, and σ is the supersaturation. It is evident from Eq. (10) that the degree of supersaturation decreases as the equilibrium grain size increases. Ostwald ripening via the dissolution of small grains and the growth of large grains provides an efficient means by which the interfacial energy within a porous and polycrystalline product can be minimized (Carlson, 2011; Norberg et al., 2011; Putnis, 2009; Raufaste et al., 2011; Voorhees, 1985, 1992). Thus, the ripening of the product via the destruction of excess surface area is a crucial step in the development of a compositionally and texturally stable product(s). Coarsening of replacement textures plays an important role during the albitization of K-feldspar. Norberg et al. (2011; their Fig. 3) observed the formation of two generations of albite (NaAlSi3O8), both of which displayed the characteristic sharp interface associated with ICDR replacements. Initially the replacement was far from equilibrium leading to fine-grained, finely twinned, high porosity albite. A secondary CDR replacement, driven by the reduction of stored strain and surface energy, formed large, periodically twinned albite grains surrounding sizable, yet limited, euhedral pores. Norberg et al. (2011) note the importance of textural evolution in systems that do not develop an extensively connected 3D pore network. Migration of grain boundaries as a result of deformation has been identified as strongly influencing the distribution of fluids within grain boundaries (Schenk and Urai, 2005). Thus, a similar grain boundary migration resulting from the coarsening of grains may act as a crucial means by which fluids maintain contact with the interface in systems that develop limited or isolated porosity. 6.5. Formation of passivating layers Generation of an impermeable (pore-free) reaction layer results in low reaction extents (passivation). For example, thermodynamic considerations suggest that both olivine and wollastonite (CaSiO3) are favorable candidates for carbonation reactions (Guyot et al., 2011; Power et al., 2013). However, the formation of a non-porous silica layer on the olivine surface armors it from further reaction, while a meso-porous silica layer forms on the surface of the wollastonite enabling the carbonation reaction to continue through the grain (Daval et al., 2009, 2011). A similar situation is seen in the uptake of Cd by calcite (Prieto et al., 2003, 2013). In this case the relative solubilities of calcite and otavite (CdCO3) suggest that Cd has a strong tendency to partition into calcite. However, the isostructural nature of the end-member minerals results in a topotactic replacement that develops a product phase that armors the parent from the solution as a result of 2D growth. On the other hand, when aragonite is used, the lack of structural similarity prevents a passivation layer forming as 3D island

645

growth results in the formation of porosity. These examples highlight the need to understand the textural development of the product phases, as it affects reaction pathway and kinetics. The alteration of silicate minerals and glass corrosion in the presence of hydrothermal fluids further demonstrate the importance of textural evolution. In both of these systems the formation of a dynamically restructuring gel-like amorphous silica product plays an important role in controlling mass transport during mineral replacement. Densification and self-organization can lead to the closure of pores, reducing the fluid-assisted transport of species between the bulk and interfacial solutions as volume diffusion through the gel layer becomes increasingly important (Geisler et al., 2015). As the silica gel layer increases in thickness and density, there is a significant drop in corrosion or alteration rate as the parent phase becomes passivated through restricted mass transport (Cailleteau et al., 2008; Gin et al., 2015a, 2015b; Parruzot et al., 2015). Additionally, limited transport through the increasingly dense gel layer will lead to deviation in compositions between interfacial and bulk fluids, but whether complete passivation occurs is still up for debate (Geisler et al., 2015). 7. Reaction fronts as microreactors and scavenging of trace and minor elements Local conditions at the mineral–fluid interface can aid the scavenging of trace elements during mineral replacement. Textural evidence for the coprecipitation of trace amounts of minerals within major phases abound in nature. In some cases, these coprecipitation reactions can control the metal endowment and ore grade distribution in ore deposits. Two recent studies have demonstrated that CDR reaction fronts can provide a favorable microenvironment for the precipitation of minor ore components from solution, i.e. Au and U (Fig. 14; Li et al., 2015; Tooth et al., 2011). The capacity of Bi-melts to scavenge Au from hydrothermal fluids has been recognized across a variety of ore deposits, e.g., NICO deposit, NWT, Canada (Acosta-Góngora et al., 2015); Stormont skarn prospect, north-western Tasmania (Cockerton and Tomkins, 2012); and the Escanaba Trough, Southern Gorda Ridge (Törmänen and Koski, 2005). The formation of the Bi-melt in these systems is the result of the reduction of Bi(III) carried in infiltrating fluids via fluid–rock interaction (Acosta-Góngora et al., 2015; Tooth et al., 2008, 2013; Törmänen and Koski, 2005). The role of the replacement front in facilitating the melt formation has recently been experimentally demonstrated during the replacement of pyrrhotite by magnetite by Tooth et al. (2011). The interfacial conditions where pyrrhotite and magnetite coexist correspond to the region where liquid Bi is stable, i.e. reducing conditions and low sulfur activity (Tooth et al., 2011; their Fig. 11). This was confirmed experimentally as Bi-blebs formed at the replacement interface (Fig. 14a, b). Furthermore, the resultant porosity within the magnetite provided space for the Bi-melt (Fig. 14c), and importantly maintained the contact between the melt and the Au-bearing fluid, allowing for efficient scavenging of Au from solution. Upon cooling, the blebs of native Bi contained abundant inclusions of maldonite (Au2Bi), in textures and associations similar to those found in the Escanaba trough (Törmänen and Koski, 2005; their Fig. 3). The experimental sulfidation of hematite in the presence of Cu- and U-bearing fluids (Li et al., 2015) further highlights the role of the replacement front in scavenging trace components. The presence of U in the solution changes the reaction path. In U-free experiments, chalcopyrite replaced hematite directly (Zhao et al., 2014a). In contrast, the presence of U catalyzed the formation of pyrite, and the formation of pyrite was found to be closely linked to the scavenging of U. The replacement of hematite by pyrite involves the oxidation of the bisulfide found in sodimer found in pyrite. This oxidation is lution (e.g., HS−) into the S2− 2 likely coupled to the reduction of the aqueous uranyl (U(VI)) in solution to UO2 + x(s). The overall replacement of hematite by pyrite is a reduction reaction. Similar to the experiments of Tooth et al. (2011), the

646

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

Fig. 14. (a–c) Bi-blebs forming at the interface during the replacement of pyrrhotite by magnetite (Mt) (reprinted from Tooth et al., 2011, with permission from Elsevier). (a–b) The replacement interface provides appropriate reaction conditions (reducing, low aH2S, catalytic surfaces) for the formation of Bi-melt that can than scavenge Au from undersaturated hydrothermal fluids. (c) Secondary porosity within the magnetite provides space for the Bi-melt. (d–f) Hematite sulfidation in the presence of U results in the formation of pyrite prior to chalcopyrite and the associated scavenging of U (reprinted from Li et al., 2015 with permission from Mineralogical Society of America). (d) A thin U-rich layer forms between pyrite and chalcopyrite, facilitated by appropriate conditions at the interface (reducing). (e) Elemental map illustrating the thin U-rich layer separating pyrite and chalcopyrite. (f) U-rich layer separating pyrite and chalcopyrite as the replacement front moves towards the center of the hematite grain (g) RGB (Fe–U–Cu) image of bornite replaced by siderite along fractures (from Li et al., in press). Uranium was enriched during the replacement reaction along the rim of bornite relicts (BSE image as inset), in a manner analogous to the experiments shown in (d–f). (h) BSE image of an arsenian pyrite from the Sunrise Dam deposit, Western Australia. The primary As zoning is disturbed via a CDR reaction; arsenic in enriched in arsenopyrite. Also note porosity in the replaced area along with inclusions of arsenopyrite. After Sung et al. (2009).

reducing conditions at the interface enabled the formation of uraninite nanocrystals, which form a thin layer between pyrite and chalcopyrite (Fig. 14d–f). Textural evidence for U scavenging during CDR reactions involving sulfide minerals have been reported from the giant Olympic Dam U-Cu–Au–Ag deposit (Li et al., 2015) and (particularly strikingly) from the Moonta IOCG deposit (Fig. 14g). The formation of mineral inclusions during the CDR reaction can also arise from the remobilization of trace elements in the microenvironment of the reaction front. Inclusions associated with CDR

replacements have been observed across a wide spectrum of both natural and experimental systems such as crystalline zircon solid solutions (e.g., Geisler et al., 2007; Rubatto et al., 2008; Tomaschek et al., 2003), xenotime and monazite (e.g., Harlov et al., 2007, 2011; Harlov and Wirth, 2012; Hetherington and Harlov, 2008), apatites (e.g., Harlov et al., 2005; Harlov and Förster, 2003), and sulfides (e.g., Sung et al., 2009; Fig. 14h). The generation of mineral inclusions is the result of the local solution at the replacement interface achieving supersaturation with respect to a secondary phase.

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

The origin of the trace elements will depend on the nature of the system — transport of trace species from external sources in open systems, while mobilization of trace elements within the parent phase is sufficient to generate local supersaturation with respect to a secondary phase under (near) closed system conditions (Harlov et al., 2005, 2011; Norberg et al., 2014). Interfacial porosity developed during CDR replacements facilitates the coprecipitation of supersaturated secondary phases, as exemplified by the formation of hematite rosettes within the interfacial porosity during the alteration of plagioclase by K-feldspar (A. Putnis et al. 2007), or the replacement of arsenian pyrite by pyrite + arsenopyrite (Fig. 14h). Significant experimental work has shown that across a range of solution and glass compositions, CDR is the main mechanism by which the glass corrosion layer forms (Dohmen et al., 2013; Geisler et al., 2010, 2015; Hellmann et al., 2015; Putnis, 2015). Interestingly, non-equilibrium oscillatory banding has been observed to form in the alteration layer during glass corrosion experiments across a wide range of solution conditions, developing textural features similar to that of archeological glasses (Dohmen et al., 2013; Geisler et al., 2010; Verney-Carron et al., 2010). The structure of the experimentally altered product layer consisted of an exterior ‘pristine’ zone which exhibited no pattern formation, but remained at a constant thickness throughout the replacement beyond about 30 μm, followed by a patterned zone with a texturally sharp front with the pristine glass (Geisler et al., 2010; their Fig. 2). The formation of a constant thickness zone on the exterior of the sample that exhibits no zoning suggests that transport of elements released by the congruent dissolution of the glass becomes limiting beyond a certain thickness. This results in the concentration of trace elements at the replacement interface, favoring the formation of secondary phases as is evident in both experimental (Geisler et al., 2010) and archeological (VerneyCarron et al., 2010) samples. 8. Discussion: the interpretation of textures in CDR reactions 8.1. Complex textures can develop via CDR reactions as a result of nonequilibrium processes and local equilibrium This review has illustrated how a diversity of textures can develop during interface-coupled dissolution–reprecipitation reactions. In general, this textural diversity is due to the replacement being controlled by non-equilibrium processes, or by local conditions at the reaction interface, which can be quite different from the ‘bulk’ conditions. In detail, the controls on the reactions are often difficult to predict, involving complex interactions among dissolution kinetics, nucleation and growth of the product, transport of solutes to and from the reaction front, creation and destruction of porosity, and further reaction driven by the metastable nature of the products (e.g., surface energy; chemical gradients). The local interfacial solution controls key processes such as volume change, nucleation rate, and growth mechanism; while in conjunction with the interfacial solution chemistry, the parent phase can act to stabilize metastable phases and/or compositions (Ruiz-Agudo et al., 2014; Xia et al., 2007, 2008). Kinetic factors play a critical role in enabling the formation of metastable products. For example, the prevalence of violarite over other Fe–Ni-sulfides in the cementation zone of many pentlandite-bearing, magmatic, massive sulfide deposits is explained by the similarity of the S-lattices between pentlandite and violarite. This reduces the nucleation barrier for violarite nucleation compared to thermodynamically more stable phases such as heazlewoodite (Fig. 9; Xia et al., 2009a). Pentlandite dissolution becomes the rate-limiting step in the replacement reaction, and pseudomorphic replacement of pentlandite is achieved via epitaxial nucleation of violarite. In addition, positive feedbacks are a common feature explaining the abundance of some key replacement reactions in Nature. For

647

example, in dolomitization the difficulty in nucleating dolomite plays a key role in controlling the evolution of the porosity, by allowing for initial dissolution along grain boundaries, and formation of coarse porosity at the reaction interface; both processes help to sustain the reaction progress (Etschmann et al., 2014). In the case of serpentinization, grain-scale replacement is enabled by the propagation of fractures as a result of stress build up associated with a CDR replacement (Malvoisin et al., 2012). It is evident that, in CDR reactions, complex processes at the interface generally play a greater role than the global T, P and X factors, often resulting in the formation of complex replacement textures even when the overall P, T and X follow a simple evolution. 8.2. The importance of post-CDR reactions on textural evolution and preservation As a result of their formation, either under far-from-equilibrium conditions or near-equilibrium conditions with respect to a microenvironment restricted in space and time, the products of CDR reactions develop excess energy, which commonly drives further textural evolution. In particular, the generation of porosity and new grain boundaries during the precipitation of the product leads to the storage of excess energy in the form of new surface area. As a closed system moves towards equilibrium, the excess surface energy drives textural re-equilibration such that it is reduced through the destruction of excess surface area. This has a number of consequences for the final textures observed in CDR reactions: • The destruction of textural features associated with CDR replacements such as porosity and sharp replacement fronts can disguise evidence of pervasive fluid infiltration events (Putnis and Austrheim, 2013). This textural evolution is driven both by internal (e.g., minimization of excess energy by coarsening) and external processes (e.g., continued interaction between an infiltrating fluid and the secondary mineral phases; Hellmann et al., 2012). Compositional and textural re-equilibration in the presence of a fluid phase is usually extremely fast in terms of geological timescales, which further contributes to masking textural evidence for extensive fluid-mediated mineral alteration. • The destruction of replacement textures can act to passivate the underlying parent phase through destruction of permeable pathways. Closure of porosity through coarsening (minimization of surface energy) can also trap fluid and mineral inclusions that can be the only preserved remnants of such an alteration process (e.g., A. Putnis et al. 2007; Tooth et al., 2011). Hence, textural evolution resulting from the CDR mechanism directly affects the long-term reactive behavior of the products, a key feature for applications such as carbon sequestration (e.g., Daval et al., 2009, 2011) and nuclear waste storage (e.g., Geisler et al., 2010, 2015). • Coarsening processes can also act to facilitate mineral replacement reactions. For example, coarsening of albite after initial albitization of K-feldspar is an important factor in facilitating continued replacement as a result of limited connectivity within the early reaction-generated porosity (Norberg et al., 2011). • Ripening of porosity can trap residual fluids within pores. This, along with secondary mineral inclusions that form in the microenvironment provided by the reaction front, can act as important clues in elucidating conditions prevailing at the time of closure, but needs to be interpreted carefully as the trapped fluids and inclusions do not necessarily represent the bulk fluid responsible for driving the mineral replacement reaction. The preservation of fluid and mineral inclusions after textural ripening thus provides important details regarding fluid-mineral interactions and the role of CDR, if interpreted correctly in the context of a dynamic reaction process.

648

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

8.3. Perspectives and future directions Following the influential papers by Putnis (2002, 2009), the role of fluids in the formation of many textures formally interpreted in terms of dry, solid-state diffusion processes is now well established. This review emphasizes the need to improve our understanding of reaction mechanisms, especially in systems containing even minor amounts of fluids (ore systems; metasomatic and metamorphic systems). Such a process-based understanding is key to supporting the petrological interpretation of textures. The interpretation of a texture must rely on a detailed understanding of the reaction mechanisms that resulted in the formation of this texture. The reactions discussed in this review occurred in the presence of a free fluid phase that is driving parent dissolution, mass transport, and product precipitation during mineral replacement. However, the rim-to-core textures characteristic of CDR reactions have also been observed to develop under anhydrous, high temperature conditions (e.g., chromite sulfidation; Ahmad et al., 2015). Indeed, SSD is dominated by volume and grain boundary diffusion, with bulk diffusivities defined by the volume-weighted mean based on the fraction of grains and grain boundaries (Gardés et al., 2011). Even in the absence of a free fluid phase, small amounts of water can greatly enhance grain boundary diffusion and reaction rates (Loring et al., 2011; Wilson and Bish, 2011). Recently, Milke et al. (2013) demonstrated that the presence of limited excess water, beyond grain boundary saturation, was sufficient to replicate reaction rates and textures associated with water-rich systems, and resulted in diffusion of migrating species four to seven orders of magnitude greater than when compared to anhydrous systems (Carlson, 2010; Milke et al., 2009; Shimojuku et al., 2014; Yund, 1997). The influence of water-enhanced diffusion is also well illustrated considering the re-equilibration mechanism of zircons. Hydrothermal alteration of natural metamict zircon occurs as a result of hydrous-assisted diffusion (Geisler et al., 2003), whereas CDR drives the recrystallization of crystalline zircon (Geisler et al., 2007) and Pudoped metamict zircon (Geisler et al., 2005) under hydrothermal conditions. Via the process of water- (or other volatile species such as H2S-, H2-) enhanced diffusion, SSD and CDR define extremes on a continuum defining the dominant transport and kinetic processes, and possibly resulting in very similar textures. Improving our understanding of these processes and how they affect our interpretation of mineral textures in rocks will be a fertile field of research over the next few years. Throughout this review the important role of the reaction interface has been emphasized. It is within the limited space at the interface that complex non-equilibrium processes drive the reaction (e.g., nucleation and growth of the product; dissolution kinetics; solute transport to and from the reaction sites; porosity creation and destruction; Fig. 1). Yet the resulting textures and kinetics can belie this complexity. An illustration of this is provided by some complex mineral replacement reactions that are governed by overall simple, first order kinetic laws (e.g., replacement of pentlandite by violarite; or calaverite by gold; Xia et al., 2009a, Zhao et al., 2009). In general, as a result of the complex interfacial processes and their interactions governing CDR reactions, prediction of how replacement reactions will proceed – and indeed whether they will proceed at all — is limited to empirical studies. It is therefore critical that a clearer picture of what is happening at the replacement interface be obtained in order to develop a solid theoretical framework useful for predicting reaction paths, overall kinetics, and the resulting mineral textures. Only by considering the replacement mechanism in the context of evolving P, T and X conditions can a true understanding of geological and mineralogical evolution be achieved. Acknowledgments The authors acknowledge the financial support from the Australian Research Council (grants DP140102765 and DP1095069). The paper

benefited from suggestions by the editor Arturo Gómez-Tuena and the two anonymous reviewers.

References Acosta-Góngora, P., Gleeson, S.A., Samson, I.M., Ootes, L., Corriveau, L., 2015. Gold refining by bismuth melts in the iron oxide–dominated NICO Au–Co–Bi (±Cu ± W) deposit, NWT, Canada. Econ. Geol. 110, 291–314. Ahmad, S., Rhamdhani, M.A., Pownceby, M.I., Bruckard, W.J., 2015. Sulfidation kinetics of natural chromite ore using H2S gas. Metall. Mater. Trans. B Process Metall. Mater. Process. Sci. 46, 557–567. Al-Awadi, M., Clark, W.J., Moore, W.R., Herron, M., Zhang, T., Zhao, W., Hurley, N., Kho, D., Montaron, B., Sadooni, F., 2009. Dolomite: perspectives on a perplexing mineral. Oilfield Rev. 21, 32–45. Alberty, R.A., 2001. Use of Legendre transforms in chemical thermodynamics. Pure Appl. Chem. 73, 1349–1380. Álvarez-Lloret, P., Rodríguez-Navarro, A.B., Falini, G., Fermani, S., Ortega-Huertas, M., 2010. Crystallographic control of the hydrothermal conversion of calcitic sea urchin spine (Paracentrotus lividus) into apatite. Cryst. Growth Des. 10, 5227–5232. Arvidson, R.S., Beig, M.S., Luttge, A., 2004. Single-crystal plagioclase feldspar dissolution rates measured by vertical scanning interferometry. Am. Mineral. 89, 51–56. Baltrusaitis, J., Grassian, V.H., 2009. Calcite (101¯4) surface in humid environments. Surf. Sci. 603, L99–L104. Bickle, M., Baker, J., 1990. Migration of reaction and isotopic fronts in infiltration zones: assessments of fluid flux in metamorphic terrains. Earth Planet. Sci. Lett. 98, 1–13. Borg, S., Liu, W., Pearce, M., Cleverley, J., MacRae, C., 2014. Complex mineral zoning patterns caused by ultra-local equilibrium at reaction interfaces. Geology 42, 415–418. Brantley, S.L., Evans, B., Hickman, S.H., Crerar, D.A., 1990. Healing of microcracks in quartz: implications for fluid flow. Geology 18, 136–139. Brown, W.L., Parsons, I., 1984. Exsolution and coarsening mechanisms and kinetics in an ordered cryptoperthite series. Contrib. Mineral. Petrol. 86, 3–18. Brown, W.L., Parsons, I., 1993. Storage and release of elastic strain energy: the driving force for low-temperature reactivity and alteration of alkali feldspar. In: Boland, J.N., Fitz Gerald, J.D. (Eds.), Defects and Processes in the Solid State: Geoscience Applications the McLaren Edition. Elsevier, Amsterdam. Brown, W.L., Becker, S.M., Parsons, I., 1983. Cryptoperthites and cooling rate in a layered syenite pluton: a chemical and TEM study. Contrib. Mineral. Petrol. 82, 13–25. Brown, S.M., Groves, D.I., Newton, P.G., 2002. Geological setting and mineralization model for the Cleo gold deposit, Eastern Goldfields Province, Western Australia. Mineral. Deposita 37, 704–721. Brugger, J., Gieré, R., 2000. Origin and distribution of some trace elements in metamorphosed Fe–Mn deposits, Val Ferrera, Eastern Swiss Alps. Can. Mineral. 38, 1075–1101. Brugger, J., McFadden, A., Lenehan, C.E., Etschmann, B., Xia, F., Zhao, J., Pring, A., 2010. A novel route for the synthesis of mesoporous and low-thermal stability materials by coupled dissolution–reprecipitation reactions: mimicking hydrothermal mineral formation. CHIMIA Int. J. Chem. 64, 693–698. Bruno, J., Bosbach, D., Kulik, D., Navrotsky, A., 2007. Chemical Thermodynamics of Solid Solutions of Interest in Nuclear Waste Management: A State-of-the-Art Report. OECD Nuclear Energy Agency, Paris. Cabri, L.J., 1965. Phase relations in the Au–Ag–Te system and their mineralogical significance. Econ. Geol. 60, 1569–1606. Cailleteau, C., Angeli, F., Devreux, F., Gin, S., Jestin, J., Jollivet, P., Spalla, O., 2008. Insight into silicate-glass corrosion mechanisms. Nat. Mater. 7, 978–983. Cardew, P.T., Davey, R.J., 1985. The kinetics of solvent-mediated phase transformations. Proc. R. Soc. A Math. Phys. Eng. Sci. 398, 415–428. Carlson, W.D., 1999. The case against Ostwald ripening of porphyroblasts. Can. Mineral. 37, 403–413. Carlson, W.D., 2010. Dependence of reaction kinetics on H2O activity as inferred from rates of intergranular diffusion of aluminium. J. Metamorph. Geol. 28, 735–752. Carlson, W.D., 2011. Porphyroblast crystallization: linking processes, kinetics, and microstructures. Int. Geol. Rev. 53, 406–445. Cho, M., Fawcett, J.J., 1986. Morphologies and growth mechanisms of synthetic Mgchlorite and cordierite. Am. Mineral. 71, 78–84. Cockerton, A.B.D., Tomkins, A.G., 2012. Insights into the liquid bismuth collector model through analysis of the Bi–Au Stormont skarn prospect, northwest Tasmania. Econ. Geol. 107, 667–682. Cole, D.R., Chakraborty, S., 2001. Rates and mechanisms of isotopic exchange. Rev. Mineral. Geochem. 43, 83–223. Connolly, J.A.D., Podladchikov, Y.Y., 2013. A hydromechanical model for lower crustal fluid flow. In: Harlov, D.E., Austrheim, H. (Eds.), Metasomatism and the Chemical Transformation of Rock. Springer-Verlag, Berlin, pp. 599–658. Cruz-Uribe, A.M., Feineman, M.D., Zack, T., Barth, M., 2014. Metamorphic reaction rates at ∼650–800 °C from diffusion of niobium in rutile. Geochim. Cosmochim. Acta 130, 63–77. Daval, D., Martinez, I., Guigner, J.M., Hellmann, R., Corvisier, J., Findling, N., Dominici, C., Goffé, B., Guyot, F., 2009. Mechanism of wollastonite carbonation deduced from micro- to nanometer length scale observations. Am. Mineral. 94, 1707–1726. Daval, D., Sissmann, O., Menguy, N., Saldi, G.D., Guyot, F., Martinez, I., Corvisier, J., Garcia, B., Machouk, I., Knauss, K.G., Hellmann, R., 2011. Influence of amorphous silica layer formation on the dissolution rate of olivine at 90 °C and elevated pCO2. Chem. Geol. 284, 193–209. Deditius, A.P., Utsunomiya, S., Renock, D., Ewing, R.C., Ramana, C.V., Becker, U., Kesler, S.E., 2008. A proposed new type of arsenian pyrite: composition, nanostructure and geological significance. Geochim. Cosmochim. Acta 72, 2919–2933.

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651 Demichelis, R., Raiteri, P., Gale, J.D., Quigley, D., Gebauer, D., 2011. Stable prenucleation mineral clusters are liquid-like ionic polymers. Nat. Commun. 2, 590. Dobson, P.F., Kneafsey, T.J., Hulen, J., Simmons, A., 2003. Porosity, permeability, and fluid flow in the Yellowstone geothermal system, Wyoming. J. Volcanol. Geotherm. Res. 123, 313–324. Dohmen, L., Lenting, C., Fonseca, R.O.C., Nagel, T., Heuser, A., Geisler, T., Denkler, R., 2013. Pattern formation in silicate glass corrosion zones. Int. J. Appl. Glas. Sci. 4, 357–370. Dunkel, K.G., Putnis, A., 2014. Replacement and ion exchange reactions of scolecite in a high pH aqueous solution. Eur. J. Mineral. 26, 61–69. Durham, G.S., Rock, E.J., Frayn, J.S., 1953. Solid solutions of the alkali halides. I. The systems KBr–KCl–H2O, RbBr–RbCl–H2O and RbBr–KBr–H2O at 25°. J. Am. Chem. Soc. 75, 5792–5794. Etschmann, B., Brugger, J., Pearce, M.A., Ta, C., Brautigan, D., Jung, M., Pring, A., 2014. Grain boundaries as microreactors during reactive fluid flow: experimental dolomitization of a calcite marble. Contrib. Mineral. Petrol. 168, 1045. Fernández-Díaz, L., Pina, C.M., Astilleros, J.M., Sánchez-Pastor, N., 2009. The carbonatation of gypsum: pathways and pseudomorph formation. Am. Mineral. 94, 1223–1234. Frank, F.C., 1949. The influence of dislocations on crystal growth. Discuss. Faraday Soc. 5, 48–54. Frank, F.C., van der Merwe, J.H., 1949a. One-dimensional dislocations. II. Misfitting monolayers and oriented overgrowth. Proc. R. Soc. A Math. Phys. Eng. Sci. 198, 216–225. Frank, F.C., van der Merwe, J.H., 1949b. One-dimensional dislocations. I. Static theory. Proc. R. Soc. A Math. Phys. Eng. Sci. 198, 205–216. García-Ruiz, J.M., Villasuso, R., Ayora, C., Canals, A., Otálora, F., 2007. Formation of natural gypsum megacrystals in Naica, Mexico. Geology 35, 327. Gardés, E., Wunder, B., Marquardt, K., Heinrich, W., 2012. The effect of water on intergranular mass transport: new insights from diffusion-controlled reaction rims in the MgO–SiO2 system. Contrib. Mineral. Petrol. 164, 1–16. Gardés, E., Wunder, B., Wirth, R., Heinrich, W., 2011. Growth of multilayered polycrystalline reaction rims in the MgO–SiO2 system, part I: experiments. Contrib. Mineral. Petrol. 161, 1–12. Gebauer, D., Kellermeier, M., Gale, J.D., Bergström, L., Cölfen, H., 2014. Pre-nucleation clusters as solute precursors in crystallisation. Chem. Soc. Rev. 43, 2348–2371. Geisler, T., Burakov, B., Yagovkina, M., Garbuzov, V., Zamoryanskaya, M., Zirlin, V., Nikolaeva, L., 2005. Structural recovery of self-irradiated natural and 238Pu-doped zircon in an acidic solution at 175 °C. J. Nucl. Mater. 336, 22–30. Geisler, T., Janssen, A., Scheiter, D., Stephan, T., Berndt, J., Putnis, A., 2010. Aqueous corrosion of borosilicate glass under acidic conditions: a new corrosion mechanism. J. NonCryst. Solids 356, 1458–1465. Geisler, T., Nagel, T., Kilburn, M.R., Janssen, A., Icenhower, J.P., Fonseca, R.O.C., Grange, M., Nemchin, A.A., 2015. The mechanism of borosilicate glass corrosion revisited. Geochim. Cosmochim. Acta 158, 112–129. Geisler, T., Pidgeon, R.T., Kurtz, R., van Bronswijk, W., Schleicher, H., 2003. Experimental hydrothermal alteration of partially metamict zircon. Am. Mineral. 88, 1496–1513. Geisler, T., Schaltegger, U., Tomaschek, F., 2007. Re-equilibration of zircon in aqueous fluids and melts. Elements 3, 43–50. Gin, S., Jollivet, P., Fournier, M., Angeli, F., Frugier, P., Charpentier, T., 2015a. Origin and consequences of silicate glass passivation by surface layers. Nat. Commun. 6, 6360. Gin, S., Jollivet, P., Fournier, M., Berthon, C., Wang, Z., Mitroshkov, A., Zhu, Z., Ryan, J.V., 2015b. The fate of silicon during glass corrosion under alkaline conditions: a mechanistic and kinetic study with the international simple glass. Geochim. Cosmochim. Acta 151, 68–85. Gin, S., Ryan, J.V., Schreiber, D.K., Neeway, J., Cabié, M., 2013. Contribution of atom-probe tomography to a better understanding of glass alteration mechanisms: application to a nuclear glass specimen altered 25 years in a granitic environment. Chem. Geol. 349350, 99–109. Glynn, P.D., Reardon, E.J., Plummer, L.N., Busenberg, E., 1990. Reaction paths and equilibrium end-points in solid-solution aqueous-solution systems. Geochim. Cosmochim. Acta 54, 267–282. Godard, M., Luquot, L., Andreani, M., Gouze, P., 2013. Incipient hydration of mantle lithosphere at ridges: a reactive-percolation experiment. Earth Planet. Sci. Lett. 371-372, 92–102. Guyot, F., Daval, D., Dupraz, S., Martinez, I., Ménez, B., Sissmann, O., 2011. CO2 geological storage: the environmental mineralogy perspective. Compt. Rendus Geosci. 343, 246–259. Hardie, L.A., 1987. Dolomitization: a critical view of some current views. J. Sediment. Petrol. 57, 166–183. Harlov, D.E., Förster, H.-J., 2003. Fluid-induced nucleation of (Y + REE)-phosphate minerals within apatite: nature and experiment. Part II. Fluorapatite. Am. Mineral. 88, 1209–1229. Harlov, D.E., Wirth, R., 2012. Experimental incorporation of Th into xenotime at middle to lower crustal P–T utilizing alkali-bearing fluids. Am. Mineral. 97, 641–652. Harlov, D.E., Wirth, R., Förster, H.-J., 2005. An experimental study of dissolution– reprecipitation in fluorapatite: fluid infiltration and the formation of monazite. Contrib. Mineral. Petrol. 150, 268–286. Harlov, D.E., Wirth, R., Hetherington, C.J., 2007. The relative stability of monazite and huttonite at 300–900 C and 200–1000 MPa: metasomatism and the propagation of metastable mineral phases. Am. Mineral. 92, 1652–1664. Harlov, D.E., Wirth, R., Hetherington, C.J., 2011. Fluid-mediated partial alteration in monazite: the role of coupled dissolution–reprecipitation in element redistribution and mass transfer. Contrib. Mineral. Petrol. 162, 329–348. Harmandas, N.G., Fernandez, E.N., Koutsoukos, P.G., 1998. Crystal growth of pyrite in aqueous solutions. Inhibition by organophosphorus compounds. Langmuir 14, 1250–1255. Hellmann, R., Cotte, S., Cadel, E., Malladi, S., Karlsson, L.S., Lozano-Perez, S., Cabié, M., Seyeux, A., 2015. Nanometre-scale evidence for interfacial dissolution– reprecipitation control of silicate glass corrosion. Nat. Mater. 14, 307–311.

649

Hellmann, R., Penisson, J.-M., Hervig, R.L., Thomassin, J.H., Abrioux, M.-F., 2003. An EFTEM/ HRTEM high-resolution study of the near surface of labradorite feldspar altered at acid pH: evidence for interfacial dissolution–reprecipitation. Phys. Chem. Miner. 30, 192–197. Hellmann, R., Wirth, R., Daval, D., Barnes, J.-P., Penisson, J.-M., Tisserand, D., Epicier, T., Florin, B., Hervig, R.L., 2012. Unifying natural and laboratory chemical weathering with interfacial dissolution–reprecipitation: a study based on the nanometer-scale chemistry of fluid–silicate interfaces. Chem. Geol. 294-295, 203–216. Hetherington, C.J., Harlov, D.E., 2008. Metasomatic thorite and uraninite inclusions in xenotime and monazite from granitic pegmatites, Hidra anorthosite massif, southwestern Norway: mechanics and fluid chemistry. Am. Mineral. 93, 806–820. Hövelmann, J., Austrheim, H., Putnis, A., 2014. Cordierite formation during the experimental reaction of plagioclase with Mg-rich aqueous solutions. Contrib. Mineral. Petrol. 168, 1063. Hövelmann, J., Putnis, C.V., Ruiz-Agudo, E., Austrheim, H., 2012. Direct nanoscale observations of CO2 sequestration during brucite [Mg(OH)2] dissolution. Environ. Sci. Technol. 46, 5253–5260. Ihli, J., Wong, W.C., Noel, E.H., Kim, Y.-Y., Kulak, A.N., Christenson, H.K., Duer, M.J., Meldrum, F.C., 2014. Dehydration and crystallization of amorphous calcium carbonate in solution and in air. Nat. Commun. 5, 3169. Iyer, K., Jamtveit, B., Mathiesen, J., Malthe-Sørenssen, A., Feder, J., 2008. Reaction-assisted hierarchical fracturing during serpentinization. Earth Planet. Sci. Lett. 267, 503–516. Jamtveit, B., Austrheim, H., 2010. Metamorphism: the role of fluids. Elements 6, 153–158. Jamtveit, B., Hammer, Ø., 2012. Sculpting of rocks by reactive fluids. Geochem. Perspect. 1, 341–480. Jamtveit, B., Austrheim, H., Malthe-Sørenssen, A., 2000. Accelerated hydration of the Earth's deep crust induced by stress perturbations. Nature 408, 75–78. Jamtveit, B., Kobchenko, M., Austrheim, H., Malthe-Sørenssen, A., Røyne, A., Svensen, H., 2011. Porosity evolution and crystallization-driven fragmentation during weathering of andesite. J. Geophys. Res. 116, B12204. Jamtveit, B., Putnis, C.V., Malthe-Sørenssen, A., 2009. Reaction induced fracturing during replacement processes. Contrib. Mineral. Petrol. 157, 127–133. Janssen, A., Putnis, A., Geisler, T., Putnis, C.V., 2010. The experimental replacement of ilmenite by rutile in HCl solutions. Mineral. Mag. 74, 633–644. Jonas, L., John, T., King, H.E., Geisler, T., Putnis, A., 2014. The role of grain boundaries and transient porosity in rocks as fluid pathways for reaction front propagation. Earth Planet. Sci. Lett. 386, 64–74. Kapp, P., Manning, C.E., Tropper, P., 2009. Phase-equilibrium constraints on titanite and rutile activities in mafic epidote amphibolites and geobarometry using titanite–rutile equilibria. J. Metamorph. Geol. 27, 509–521. Kasioptas, A., Geisler, T., Perdikouri, C., Trepmann, C., Gussone, N., Putnis, A., 2011. Polycrystalline apatite synthesized by hydrothermal replacement of calcium carbonates. Geochim. Cosmochim. Acta 75, 3486–3500. Kasioptas, A., Perdikouri, C., Putnis, C.V., Putnis, A., 2008. Pseudomorphic replacement of single calcium carbonate crystals by polycrystalline apatite. Mineral. Mag. 72, 77–80. Katz, A., Matthews, A., 1977. The dolomitization of CaCO3: an experimental study at 252– 295 °C. Geochim. Cosmochim. Acta 41, 297–308. Kelemen, P.B., Hirth, G., 2012. Reaction-driven cracking during retrograde metamorphism: olivine hydration and carbonation. Earth Planet. Sci. Lett. 345-348, 81–89. Kelemen, P.B., Matter, J., 2008. In situ carbonation of peridotite for CO2 storage. Proc. Natl. Acad. Sci. 105, 17295–17300. Kelemen, P.B., Matter, J., Streit, E.E., Rudge, J.F., Curry, W.B., Blusztajn, J., 2011. Rates and mechanisms of mineral carbonation in peridotite: natural processes and recipes for enhanced, in situ CO2 capture and storage. Annu. Rev. Earth Planet. Sci. 39, 545–576. Kelly, A., Groves, G.W., 1970. Crystallography and Crystal Defects. Longman, London. King, H.E., Plümper, O., Putnis, A., 2010. Effect of secondary phase formation on the carbonation of olivine. Environ. Sci. Technol. 44, 6503–6509. King, H.E., Satoh, H., Tsukamoto, K., Putnis, A., 2014. Surface-specific measurements of olivine dissolution by phase-shift interferometry. Am. Mineral. 99, 377–386. Konrad-Schmolke, M., O'Brien, P.J., Zack, T., 2011. Fluid migration above a subducted slab — constraints on amount, pathways and major element mobility from partially overprinted eclogite-facies rocks (Sesia Zone, Western Alps). J. Petrol. 52, 457–486. Land, L.S., 1998. Failure to precipitate dolomite at 25 °C from dilute solution despite 1000Fold oversaturation after 32 years. Aquat. Geochem. 4, 361–368. Lasaga, A.C., 1984. Chemical kinetics of water-rock interactions. J. Geophys. Res. 89, 4009–4025. Lasaga, A.C., 1998. Kinetic Theory in the Earth Sciences. Princeton University Press, New Jersey. Lee, J.K.W., 1993. Problems and progress in the elucidation of U and Pb transport mechanisms in zircon. In: Boland, J.N., Fitz Gerald, J.D. (Eds.), Defects and Processes in the Solid State: Geoscience Applications the McLaren Edition. Elsevier, Amsterdam, pp. 423–446. Lee, M.R., Parsons, I., 1995. Microtextural controls of weathering of perthitic alkali feldspars. Geochim. Cosmochim. Acta 59, 4465–4488. Li, K., Etschmann, B., Rae, N., Reith, F., Ryan, C.G., Kirkham, R., Howard, D., Rosa, D., Zammit, C., Pring, A., Ngothai, Y., Hooker, A., Brugger, J., 2015. Ore petrography using megapixel X-ray imaging: Rapid insights into element distribution and mobilisation in complex Pt and U–Ge–Cu ores. Econ. Geol. (in press). Li, K., Pring, A., Etschmann, B., Macmillan, E., Ngothai, Y., O'Neill, B., Hooker, A., Mosselmans, F., Brugger, J., 2015b. Uranium scavenging during mineral replacement reactions. Am. Mineral. 100, 1728–1735. Lifshitz, I.M., Slyozov, V.V., 1961. The kinetics of precipitation from supersaturated solid solutions. J. Phys. Chem. Solids 19, 35–50. Lindgren, W., 1912. The nature of replacement. Econ. Geol. 7, 521–535. Lindgren, W., 1918. Volume changes in metamorphism. J. Geol. 26, 542–554.

650

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651

Lippmann, F., 1980. Phase diagrams depicting aqueous solubility of binary mineral systems. Neues Jb. Mineral. Abh. 139, 1–25. Loring, J.S., Thompson, C.J., Wang, Z., Joly, A.G., Sklarew, D.S., Schaef, H.T., Ilton, E.S., Rosso, K.M., Felmy, A.R., 2011. In situ infrared spectroscopic study of forsterite carbonation in wet supercritical CO2. Environ. Sci. Technol. 45, 6204–6210. Majumdar, A.S., King, H.E., John, T., Kusebauch, C., Putnis, A., 2014. Pseudomorphic replacement of diopside during interaction with (Ni,Mg)Cl2 aqueous solutions: implications for the Ni-enrichment mechanism in talc- and serpentine-type phases. Chem. Geol. 380, 27–40. Malvoisin, B., Brunet, F., 2014. Water diffusion-transport in a synthetic dunite: consequences for oceanic peridotite serpentinization. Earth Planet. Sci. Lett. 403, 263–272. Malvoisin, B., Brunet, F., Carlut, J., Rouméjon, S., Cannat, M., 2012. Serpentinization of oceanic peridotites: 2. Kinetics and processes of San Carlos olivine hydrothermal alteration. J. Geophys. Res. 117, B04102. Markov, I.V., 2003. Crystal Growth for Beginners: Fundamentals of Nucleation, Crystal Growth and Epitaxy. 2nd edition. World Scientific, River Edge, NJ, USA. Mehrer, H., 2007. Diffusion in Solids Fundamentals, Methods, Materials, Diffusioncontrolled Processes. Springer-Verlag, Berlin. Melchiorre, E.B., Huss, G.R., Lopez, A., 2014. Carbon and hydrogen stable isotope microanalysis and data correction for rare carbonate minerals: case studies for stichtite (Mg6Cr2[(OH)16 |CO3]∙H2O) and malachite (Cu2CO3(OH)2). Chem. Geol. 367, 63–69. Milke, R., Kolzer, K., Koch-Müller, M., Wunder, B., 2009. Orthopyroxene rim growth between olivine and quartz at low temperatures (750–950 °C) and low water concentration. Mineral. Petrol. 97, 223–232. Milke, R., Neusser, G., Kolzer, K., Wunder, B., 2013. Very little water is necessary to make a dry solid silicate system wet. Geology 41, 247–250. Misra, K.C., Fleet, M.E., 1974. Chemical composition and stability of violarite. Econ. Geol. 69, 391–403. Morse, J.W., Casey, W.H., 1988. Ostwald processes and mineral paragenesis in sediments. Am. J. Sci. 288, 537–560. Niedermeier, D.R.D., Putnis, A., Geisler, T., Golla-Schindler, U., Putnis, C.V., 2009. The mechanism of cation and oxygen isotope exchange in alkali feldspars under hydrothermal conditions. Contrib. Mineral. Petrol. 157, 65–76. Nielsen, A.E., 1964. Kinetics of Precipitation. Pergamon Press, Oxford. Norberg, N., Harlov, D., Neusser, G., Wirth, R., Rhede, D., 2014. Element mobilization during feldspar metasomatism: an experimental study. Eur. J. Mineral. 26, 71–82. Norberg, N., Harlov, D., Neusser, G., Wirth, R., Rhede, D., Morales, L., 2013. Experimental development of patch perthite from synthetic cryptoperthite: microstructural evolution and chemical re-equilibration. Am. Mineral. 98, 1429–1441. Norberg, N., Neusser, G., Wirth, R., Harlov, D., 2011. Microstructural evolution during experimental albitization of K-rich alkali feldspar. Contrib. Mineral. Petrol. 162, 531–546. Oberthür, T., Weiser, T.W., Melcher, F., Gast, L., Wöhrl, C., 2013. Detrital platinum-group minerals in rivers draining the Great Dyke, Zimbabwe. Can. Mineral. 51, 197–222. Ogino, T., Suzuki, T., Sawada, K., 1987. The formation and transformation mechanism of calcium carbonate in water. Geochim. Cosmochim. Acta 51, 2757–2767. Okamoto, A., Shimizu, H., 2015. Contrasting fracture patterns induced by volumeincreasing and -decreasing reactions: implications for the progress of metamorphic reactions. Earth Planet. Sci. Lett. 417, 9–18. Okamoto, A., Ogasawara, Y., Ogawa, Y., Tsuchiya, N., 2011. Progress of hydration reactions in olivine–H2O and orthopyroxenite–H2O systems at 250 °C and vapor-saturated pressure. Chem. Geol. 289, 245–255. Okrugin, V.M., Andreeva, E., Etschmann, B., Pring, A., Li, K., Zhao, J., Griffiths, G., Lumpkin, G.R., Triani, G., Brugger, J., 2014. Microporous gold: comparison of textures from nature and experiments. Am. Mineral. 99, 1171–1174. Oliver, N.H.S., Cleverley, J.S., Mark, G., Pollard, P.J., Fu, B., Marshall, L.J., Rubenach, M.J., Williams, P.J., Baker, T., 2004. Modeling the role of sodic alteration in the genesis of iron oxide-copper-gold deposits, Eastern Mount Isa Block, Australia. Econ. Geol. 99, 1145–1179. Otálora, F., García-Ruiz, J.M., 2014. Nucleation and growth of the Naica giant gypsum crystals. Chem. Soc. Rev. 43, 2013–2026. Parks, G.A., 1990. Surface energy and adsorption at mineral/water interfaces: an introduction. Rev. Mineral. Geochem. 23, 133–175. Parruzot, B., Jollivet, P., Rébiscoul, D., Gin, S., 2015. Long-term alteration of basaltic glass: mechanisms and rates. Geochim. Cosmochim. Acta 154, 28–48. Parsons, I., Lee, M.R., 2009. Mutual replacement reactions in alkali feldspars I: microtextures and mechanisms. Contrib. Mineral. Petrol. 157, 641–661. Parsons, I., Fitz Gerald, J.D., Heizler, M.T., Heizler, L.L., Ivanic, T., Lee, M.R., 2013. Eightphase alkali feldspars: low-temperature cryptoperthite, peristerite and multiple replacement reactions in the Klokken intrusion. Contrib. Mineral. Petrol. 165, 931–960. Pearce, M.A., Timms, N.E., Hough, R.M., Cleverley, J.S., 2013. Reaction mechanism for the replacement of calcite by dolomite and siderite: implications for geochemistry, microstructure and porosity evolution during hydrothermal mineralisation. Contrib. Mineral. Petrol. 166, 955–1009. Perdikouri, C., Piazolo, S., Kasioptas, A., Schmidt, B.C., Putnis, A., 2013. Hydrothermal replacement of aragonite by calcite: interplay between replacement, fracturing and growth. Eur. J. Mineral. 25, 123–136. Peuble, S., Andreani, M., Godard, M., Gouze, P., Barou, F., Van de Moortele, B., Mainprice, D., Reynard, B., 2015. Carbonate mineralization in percolated olivine aggregates: linking effects of crystallographic orientation and fluid flow. Am. Mineral. 100, 474–482. Plümper, O., Røyne, A., Magrasó, A., Jamtveit, B., 2012. The interface-scale mechanism of reaction-induced fracturing during serpentinization. Geology 40, 1103–1106. Pollok, K., Putnis, C.V., Putnis, A., 2011. Mineral replacement reactions in solid solutionaqueous solution systems: volume changes, reactions paths and end-points using the example of model salt systems. Am. J. Sci. 311, 211–236.

Pouget, E.M., Bomans, P.H.H., Goos, J.A.C.M., Frederik, P.M., de With, G., Sommerdijk, N.A.J.M., 2009. The initial stages of template-controlled CaCO3 formation revealed by cryo-TEM. Science 323, 1455–1458. Power, I.M., Harrison, A.L., Dipple, G.M., Wilson, S.A., Kelemen, P.B., Hitch, M., Southam, G., 2013. Carbon mineralization: from natural analogues to engineered systems. Rev. Mineral. Geochem. 77, 305–360. Prieto, M., 2009. Thermodynamics of solid solution–aqueous solution systems. Rev. Mineral. Geochem. 70, 47–85. Prieto, M., Astilleros, J.M., Fernández-Díaz, L., 2013. Environmental remediation by crystallization of solid solutions. Elements 9, 195–201. Prieto, M., Cubillas, P., Fernández-Gonzalez, Á., 2003. Uptake of dissolved Cd by biogenic and abiogenic aragonite: a comparison with sorption onto calcite. Geochim. Cosmochim. Acta 67, 3859–3869. Prieto, M., Fernández-González, Á., Putnis, A., Fernández-Díaz, L., 1997. Nucleation, growth, and zoning phenomena in crystallizing (Ba,Sr)CO3, Ba(SO4,CrO4), (Ba,Sr)SO4, and (Cd,Ca)CO3 solid solutions from aqueous solutions. Geochem. J. 61, 3383–3397. Prieto, M., Putnis, A., Fernández-Díaz, L., 1993. Crystallization of solid solutions from aqueous solutions in a porous medium: Zoning in (Ba,Sr)SO4. Geol. Mag. 130, 289–299. Putnis, A., 1992. Introduction to Mineral Sciences. Cambridge University Press, Cambridge. Putnis, A., 2002. Mineral replacement reactions: from macroscopic observations to microscopic mechanisms. Mineral. Mag. 66, 689–708. Putnis, A., 2009. Mineral replacement reactions. Rev. Mineral. Geochem. 70, 87–124. Putnis, A., 2014. Why mineral interfaces matter. Science 343, 1441–1442. Putnis, A., 2015. Glass corrosion: sharpened interface. Nat. Mater. 14, 261–262. Putnis, A., Austrheim, H., 2010. Fluid-induced processes: metasomatism and metamorphism. Geofluids 10, 254–269. Putnis, A., Austrheim, H., 2013. Mechanisms of metasomatism and metamorphism on the local mineral scale: the role of dissolution–reprecipitation during mineral reequilibration. In: Harlov, D.E., Austrheim, H. (Eds.), Metasomatism and the Chemical Transformation of Rock. Springer-Verlag, Berlin. Putnis, A., John, T., 2010. Replacement processes in the Earth's crust. Elements 6, 159–164. Putnis, A., Mauthe, G., 2001. The effect of pore size on cementation in porous rocks. Geofluids 1, 37–41. Putnis, C.V., Mezger, K., 2004. A mechanism of mineral replacement: isotope tracing in the model system KCl–KBr–H2O. Geochim. Cosmochim. Acta 68, 2839–2848. Putnis, A., Hinrichs, R., Putnis, C.V., Golla-Schindler, U., Collins, L.G., 2007. Hematite in porous red-clouded feldspars: evidence of large-scale crustal fluid–rock interaction. Lithos 95, 10–18. Putnis, C.V., Geisler, T., Schmid-Beurmann, P., Stephan, T., Giampaolo, C., 2007. An experimental study of the replacement of leucite by analcime. Am. Mineral. 92, 19–26. Putnis, A., Prieto, M., Fernández-Díaz, L., 1995. Fluid supersaturation and crystallization in porous media. Geol. Mag. 132, 1–13. Putnis, C.V., Tsukamoto, K., Nishimura, Y., 2005. Direct observations of pseudomorphism: compositional and textural evolution at a fluid–solid interface. Am. Mineral. 90, 1909–1912. Qian, G., Brugger, J., Skinner, W.M., Chen, G., Pring, A., 2010. An experimental study of the mechanism of the replacement of magnetite by pyrite up to 300 °C. Geochim. Cosmochim. Acta 74, 5610–5630. Qian, G., Brugger, J., Testemale, D., Skinner, W., Pring, A., 2013. Formation of As(II)-pyrite during experimental replacement of magnetite under hydrothermal conditions. Geochim. Cosmochim. Acta 100, 1–10. Qian, G., Xia, F., Brugger, J., Skinner, W.M., Bei, J., Chen, G., Pring, A., 2011. Replacement of pyrrhotite by pyrite and marcasite under hydrothermal conditions up to 220 °C: an experimental study of reaction textures and mechanisms. Am. Mineral. 96, 1878–1893. Ratke, L., Voorhees, P.W., 2002. Growth and Coarsening: Ostwald Ripening in Material Processing. Springer-Verlag, Berlin. Raufaste, C., Jamtveit, B., John, T., Meakin, P., Dysthe, D.K., 2011. The mechanism of porosity formation during solvent-mediated phase transformations. Proc. R. Soc. A Math. Phys. Eng. Sci. 467, 1408–1426. Reboul, J., Furukawa, S., Horike, N., Tsotsalas, M., Hirai, K., Uehara, H., Kondo, M., Louvain, N., Sakata, O., Kitagawa, S., 2012. Mesoscopic architectures of porous coordination polymers fabricated by pseudomorphic replication. Nat. Mater. 11, 717–723. Reboul, J., Yoshida, K., Furukawa, S., Kitagawa, S., 2015. Reductive coordination replication of V2O5 sacrificial macrostructures into vanadium-based porous coordination polymers. CrystEngComm 17, 323–330. Rendón-Angeles, J.C., Yanagisawa, K., Ishizawa, N., Oishi, S., 2000. Topotaxial conversion of chlorapatite and hydroxyapatite to fluorapatite by hydrothermal ion exchange. Chem. Mater. 12, 2143–2150. Rickard, D., Grimes, S., Butler, I., Oldroyd, A., Davies, K.L., 2007. Botanical constraints on pyrite formation. Chem. Geol. 236, 228–246. Rimstidt, J.D., 1997. Gangue mineral transport and deposition. In: Barnes, H.L. (Ed.), Geochemistry of Hydrothermal Ore Deposits, 3rd ed. John Wiley & Sons, Inc., New York, pp. 487–515. Robin, P.-Y.F., 1974. Stress and strain in cryptoperthite lamellae and the coherent solvus of alkali feldspars. Am. Mineral. 59, 1299–1318. Rodriguez-Blanco, J.D., Shaw, S., Benning, L.G., 2011. The kinetics and mechanisms of amorphous calcium carbonate (ACC) crystallization to calcite, via vaterite. Nanoscale 3, 265–271. Rodriguez-Navarro, C., Kudłacz, K., Cizer, Ö., Ruiz-Agudo, E., 2015. Formation of amorphous calcium carbonate and its transformation into mesostructured calcite. CrystEngComm 17, 58–72. Røyne, A., Jamtveit, B., Mathiesen, J., Malthe-Sørenssen, A., 2008. Controls on rock weathering rates by reaction-induced hierarchical fracturing. Earth Planet. Sci. Lett. 275, 364–369.

A. Altree-Williams et al. / Earth-Science Reviews 150 (2015) 628–651 Rubatto, D., Müntener, O., Barnhoorn, A., Gregory, C., 2008. Dissolution–reprecipitation of zircon at low-temperature, high-pressure conditions (Lanzo Massif, Italy). Am. Mineral. 93, 1519–1529. Rubie, D.C., 1986. The catalysis of mineral reactions by water and restrictions on the presence of aqueous fluid during metamorphism. Mineral. Mag. 50, 399–415. Ruiz-Agudo, E., Putnis, C.V., Putnis, A., 2014. Coupled dissolution and precipitation at mineral–fluid interfaces. Chem. Geol. 383, 132–146. Schenk, O., Urai, J.L., 2005. The migration of fluid-filled grain boundaries in recrystallizing synthetic bischofite: first results of in-situ high-pressure, high-temperature deformation experiments in transmitted light. J. Metamorph. Geol. 23, 695–709. Scherer, G.W., 1999. Crystallization in pores. Cem. Concr. Res. 29, 1347–1358. Schott, J., Oelkers, E.H., 1995. Dissolution and crystallization rates of silicate minerals as a function of chemical affinity. Pure Appl. Chem. 67, 903–910. Schott, J., Pokrovsky, O.S., Oelkers, E.H., 2009. The link between mineral dissolution/precipitation kinetics and solution chemistry. Rev. Mineral. Geochem. 70, 207–258. Schultheiss, S., Sethmann, I., Schlosser, M., Kleebe, H.-J., 2013. Pseudomorphic transformation of Ca/Mg carbonates into phosphates with focus on dolomite conversion. Mineral. Mag. 77, 2725–2737. Seward, T.M., Barnes, H.L., 1997. Metal transport by hydrothermal ore fluids. In: Barnes, H.L. (Ed.), Geochemistry of Hydrothermal Ore Deposits. John Wiley & Sons, Inc., New York, pp. 435–486. Shimojuku, A., Boujibar, A., Yamazaki, D., Yoshino, T., Tomioka, N., Xu, J., 2014. Growth of ringwoodite reaction rims from MgSiO3 perovskite and periclase at 22.5 GPa and 1,800 °C. Phys. Chem. Miner. 41, 555–567. Sibley, D.F., 1982. The origin of common dolomite fabrics: clues from the Pliocene. J. Sediment. Petrol. 52, 1087–1100. Skinner, B.J., Porter, A.C., Botkin, D.B., 1999. The Blue Planet: An Introduction to Earth System Science. 2nd ed. John Wiley & Sons, Inc., New York. Smit, M.A., Bröcker, M., Scherer, E.E., 2008. Aragonite and magnesite in eclogites from the jæren nappe, SW Norway: disequilibrium in the system CaCO3–MgCO3 and petrological implications. J. Metamorph. Geol. 26, 959–979. Smith, J.V., 1974. Feldspar Minerals. Springer-Verlag, Berlin. Spry, A., 1969. Metamorphic Textures. Pergamon Press, Oxford. Stack, A.G., Fernandez-Martinez, A., Allard, L.F., Bañuelos, J.L., Rother, G., Anovitz, L.M., Cole, D.R., Waychunas, G.A., 2014. Pore-size-dependent calcium carbonate precipitation controlled by surface chemistry. Environ. Sci. Technol. 48, 6177–6183. Steefel, C.I., Van Cappellen, P.V., 1990. A new kinetic approach to modeling water–rock interaction: the role of nucleation, precursors, and Ostwald ripening. Geochim. Cosmochim. Acta 54, 2657–2677. Steiger, M., 2005. Crystal growth in porous materials — I: the crystallization pressure of large crystals. J. Cryst. Growth 282, 455–469. Sunagawa, I., 1981. Characteristics of crystal growth in nature as seen from the morphology of mineral crystals. Bull. Mineral. 104, 81–87. Sung, Y.-H., Brugger, J., Ciobanu, C.L., Pring, A., Skinner, W., Nugus, M., 2009. Invisible gold in arsenian pyrite and arsenopyrite from a multistage Archaean gold deposit: Sunrise Dam, Eastern Goldfields Province, Western Australia. Mineral. Deposita 44, 765–791. Sung, Y.-H., Ciobanu, C.L., Pring, A., Brugger, J., Skinner, W., Cook, N.J., Nugus, M., 2007. Tellurides from Sunrise Dam gold deposit, Yilgarn Craton, Western Australia: a new occurrence of nagyágite. Mineral. Petrol. 91, 249–270. Tenailleau, C., Pring, A., Etschmann, B., Brugger, J., Grguric, B., Putnis, A., 2006. Transformation of pentlandite to violarite under mild hydrothermal conditions. Am. Mineral. 91, 706–709. Threlfall, T., 2003. Structure and thermodynamic explanations of Ostwald's rule. Org. Process. Res. Dev. 7, 1017–1027. Tomaschek, F., Kennedy, A.K., Villa, I.M., Lagos, M., Ballhaus, C., 2003. Zircons from Syros, Cyclades, Greece — recrystallization and mobilization of zircon during high-pressure metamorphism. J. Petrol. 44, 1977–2002. Tooth, B., Brugger, J., Ciobanu, C., Liu, W., 2008. Modeling of gold scavenging by bismuth melts coexisting with hydrothermal fluids. Geology 36, 815–818. Tooth, B., Ciobanu, C.L., Green, L., O'Neill, B., Brugger, J., 2011. Bi-melt formation and gold scavenging from hydrothermal fluids: an experimental study. Geochim. Cosmochim. Acta 75, 5423–5443. Tooth, B., Etschmann, B., Pokrovski, G.S., Testemale, D., Hazemann, J.-L., Grundler, P.V., Brugger, J., 2013. Bismuth speciation in hydrothermal fluids: an X-ray absorption spectroscopy and solubility study. Geochim. Cosmochim. Acta 101, 156–172.

651

Törmänen, T.O., Koski, R.A., 2005. Gold enrichment and the Bi–Au association in pyrrhotite-rich massive sulfide deposits, Escanaba trough, Southern Gorda Ridge. Econ. Geol. 100, 1135–1150. Ulven, O.I., Jamtveit, B., Malthe-Sørenssen, A., 2014a. Reaction-driven fracturing of porous rock. J. Geophys. Res. Solid Earth 119, 7473–7486. Ulven, O.I., Storheim, H., Austrheim, H., Malthe-Sørenssen, A., 2014b. Fracture initiation during volume increasing reactions in rocks and applications for CO2 sequestration. Earth Planet. Sci. Lett. 389, 132–142. Verney-Carron, A., Gin, S., Frugier, P., Libourel, G., 2010. Long-term modeling of alterationtransport coupling: application to a fractured Roman glass. Geochim. Cosmochim. Acta 74, 2291–2315. Voorhees, P.W., 1985. The theory of Ostwald ripening. J. Stat. Phys. 38, 231–252. Voorhees, P.W., 1992. Ostwald ripening of two-phase mixtures. Annu. Rev. Mater. Sci. 22, 197–215. Walker, F.D.L., Lee, M.R., Parsons, I., 1995. Micropores and micropermeable texture in alkali feldspars: geochemical and geophysical implications. Mineral. Mag. 59, 505–534. Warren, J., 2000. Dolomite: occurrence, evolution and economically important associations. Earth Sci. Rev. 52, 1–81. Weisheit, A., Bons, P.D., Elburg, M.A., 2013. Long-lived crustal-scale fluid flow: the hydrothermal mega-breccia of Hidden Valley, Mt. Painter Inlier, South Australia. Int. J. Earth Sci. 102, 1219–1236. Wilson, S.A., Bish, D.L., 2011. Formation of gypsum and bassanite by cation exchange reactions in the absence of free-liquid H2O: Implications for Mars. J. Geophys. Res. 116, E09010. Worden, R.H., Smalley, P.C., Cross, M.M., 2000. The influence of rock fabric and mineralogy on thermochemical sulfate reduction: Khuff formation, Abu Dhabi. J. Sediment. Res. 70, 1210–1221. Worden, R.H., Walker, F.D.L., Parsons, I., Brown, W.L., 1990. Development of microporosity, diffusion channels and deuteric coarsening in perthitic alkali feldspars. Contrib. Mineral. Petrol. 104, 507–515. Xia, F., Brugger, J., Chen, G., Ngothai, Y., O'Neill, B., Putnis, A., Pring, A., 2009a. Mechanism and kinetics of pseudomorphic mineral replacement reactions: a case study of the replacement of pentlandite by violarite. Geochim. Cosmochim. Acta 73, 1945–1969. Xia, F., Brugger, J.l., Ngothai, Y., O'Neill, B., Chen, G., Pring, A., 2009b. Three-dimensional ordered arrays of zeolite nanocrystals with uniform size and orientation by a pseudomorphic coupled dissolution–reprecipitation replacement route. Cryst. Growth Des. 9, 4902–4906. Xia, F., Zhou, J., Brugger, J., Ngothai, Y., O'Neill, B., Chen, G., Pring, A., 2008. Novel route to synthesize complex metal sulfides: hydrothermal coupled dissolution– reprecipitation replacement reactions. Chem. Mater. 20, 2809–2817. Xia, F., Zhou, J., Pring, A., Ngothai, Y., O'Neill, B., Brugger, J., Chen, G., Colby, C., 2007. The role of pyrrhotite (Fe7S8) and the sample texture in the hydrothermal transformation of pentlandite ((Fe,Ni)9S8) to violarite ((Ni,Fe)3S4). React. Kinet. Catal. Lett. 92, 257–266. Xiao, Y., Lasaga, A.C., 1994. Ab initio quantum mechanical studies of the kinetics and mechanisms of silicate dissolution: H+ (H30+) catalysis. Geochim. Cosmochim. Acta 58, 5379–5400. Yund, R.A., 1997. Rates of grain boundary diffusion through enstatite and forsterite reaction rims. Contrib. Mineral. Petrol. 126, 224–236. Zhang, Y., 2010. Diffusion in minerals and melts: theoretical background. Rev. Mineral. Geochem. 72, 5–59. Zhang, X., Zhu, Y., Zeng, H., Wang, D., Liu, J., Liu, H., Qian, M., Xu, L., 2011. Dissolution and solubility of the arsenate–phosphate hydroxylapatite solid solution [Ca5(PxAs1-xO4)3(OH)] at 25 °C. Environ. Chem. 8, 133–145. Zhao, J., Brugger, J., Chen, G., Ngothai, Y., Pring, A., 2014a. Experimental study of the formation of chalcopyrite and bornite via the sulfidation of hematite: mineral replacements with a large volume increase. Am. Mineral. 99, 343–354. Zhao, J., Brugger, J., Grundler, P.V., Xia, F., Chen, G., Pring, A., 2009. Mechanism and kinetics of a mineral transformation under hydrothermal conditions: calaverite to metallic gold. Am. Mineral. 94, 1541–1555. Zhao, J., Brugger, J., Ngothai, Y., Pring, A., 2014b. The replacement of chalcopyrite by bornite under hydrothermal conditions. Am. Mineral. 99, 2389–2397. Zhao, J., Brugger, J., Xia, F., Ngothai, Y., Chen, G., Pring, A., 2013. Dissolution-reprecipitation vs. solid-state diffusion: mechanism of mineral transformations in sylvanite, (AuAg)2Te4, under hydrothermal conditions. Am. Mineral. 98, 19–32.