Journal of Petrology Advance Access originally published online on March 8, 2007
Journal of Petrology 2007 48(4):807-828; doi:10.1093/petrology/egm002
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Liquidus Equilibria in the System K2O–Na2O–Al2O3–SiO2–F2O–1–H2O to 100 MPa: II. Differentiation Paths of Fluorosilicic Magmas in Hydrous Systems
*Department of Earth and Planetary Sciences, Mcgill University, Montreal, QC H3A 2A7, Canada
RECEIVED OCTOBER 7, 2005; ACCEPTED JANUARY 9, 2007
 |
ABSTRACT |
|---|
TOP
ABSTRACT
INTRODUCTIONWe investigated phase equilibria in the six-component system Na2O–K2O–Al2O3–SiO2–F2O–1–H2O at 100 MPa to characterize differentiation paths of natural fluorine-bearing granitic and rhyolitic magmas. Topaz and cryolite are stable saturating solid phases in calcium-poor systems. At 100 MPa the maximum solidus depression and fluorine solubility in evolving silicic melts are controlled by the eutectics haplogranite–cryolite–H2O at 640°C and
4 wt % F, and haplogranite–topaz–H2O at 640°C and 2 wt % F. Topaz and cryolite form a binary peralkaline eutectic at 660°C, 100 MPa and fluid saturation. The low-temperature nature of this invariant point causes displacement of multiphase eutectics with quartz and alkali feldspar towards the topaz–cryolite join and enables the silicate liquidus and cotectic surfaces to extend to very high fluorine concentrations (more than 30 wt % F) for weakly peraluminous and subaluminous compositions. The differentiation of fluorine-bearing magmas follows two distinct paths of fluorine behavior, depending on whether additional minerals buffer the alkali/alumina ratio in the melt. In systems with micas or aluminosilicates that buffer the activity of alumina, magmatic crystallization will reach either topaz or cryolite saturation and the system solidifies at low fluorine concentration. In leucogranitic suites precipitating quartz and feldspar only, the liquid line of descent will reach topaz or cryolite but fluorine will continue to increase until the quaternary eutectic with two fluorine-bearing solid phases is reached at 540°C, 100 MPa and aqueous-fluid saturation. The maximum water solubility in the haplogranitic melts increases with the fluorine content and reaches 12· 5 ± 0· 5 wt % H2O at the quartz–cryolite–topaz eutectic composition. A continuous transition between hydrous fluorosilicate melts and solute-rich aqueous fluids is not documented by this study. Our experimental results are applicable to leucocratic fluorosilicic magmas. In multicomponent systems, however, the presence of calcium may severely limit enrichment of fluorine by crystallization of fluorite. KEY WORDS: granite; rhyolite; topaz; cryolite; magmatic differentiation
|
INTRODUCTION |
|---|
In natural, fluorine-bearing silicic magmas, H2O is an important volatile constituent (Thomas & Klemm, 1997
; Thomas et al., 2005). Fluorine and water in silicate melts exert similar effects, and (1) depress the melting temperature (Manning, 1981; Pichavant et al., 1987; Webster et al., 1987; Weidner & Martin, 1987), (2) decrease melt density (Dingwell et al., 1993; Knoche et al., 1995), (3) decrease melt viscosity (Dingwell et al., 1985; Baker & Vaillancourt, 1995; Giordano et al., 2004) and (4) increase element diffusivity (Baker & Bossànyi, 1994). These factors are capable of extending the differentiation of fluorine-bearing hydrous granites towards low-temperature mobile residual liquids whose petrogenetic significance is not yet known. To understand the liquid lines of descent, we need to know the stability of fluorine-bearing solid phases, the miscibility gaps between silicate and fluoride melts and the fluorine solubility in silicic melts.
Several unclear geochemical features of natural fluorine-bearing magmatic rocks require experimental investigation. The rock sequence granite/rhyolite–topaz granite/ongonite–quartz topazite is characterized by a gentle decrease in SiO2 concentrations as a result of the expansion of the quartz stability field (Korzhinskiy, 1959, 1960; Manning et al., 1980; Kogarko & Krigman, 1981; Manning, 1981) and, in addition, by alkali depletion. Topaz rhyolites and ongonites become K-poor (less than 3· 5 wt % K2O), transitional topaz trondhjemites are K-depleted (0· 4 wt % K2O, Kortemeier & Burt, 1988) and quartz topazites are alkali-free (0· 1–0· 5 wt % Na2O + K2O, e.g. Zhu & Liu, 1990; Johnston & Chappell, 1992). The alkali loss can be attributed to separation of an immiscible alkali–fluoride melt or exsolution of alkali–halide fluids (Kortemeier & Burt, 1988). The absence of alkali feldspars in quartz topazites has been explained by the existence of a peritectic transition albite + melt = quartz + topaz + cryolite/chiolite (Kovalenko & Kovalenko, 1976; Kogarko & Krigman, 1981). However, there is no alkali loss in this equilibrium and thus the origin of quartz topazites remains unexplained.
In a companion paper to this experimental study (Dolej & Baker, 2007) we investigated melting equilibria in the quaternary system silica–albite–topaz–cryolite under anhydrous conditions. The silica–albite–topaz–cryolite system contains an extensive fluoride–silicate liquid miscibility gap that spans cryolite and silica liquidus volumes at temperatures above 960°C. Differentiation paths of natural fluorine-bearing magmas, however, do not reach liquid–liquid immiscibility but saturate with solid cryolite and/or topaz. Under anhydrous conditions levels of fluorine enrichment are strongly dependent on the melt alkali/aluminum ratio in the melt. In subaluminous compositions at 100 MPa and 740°C, fluorine concentration may be as high as 30 wt %.
Here we study melting equilibria in fluorosilicate systems under hydrous conditions. First we discuss the effect of H2O on the sections cryolite–topaz and quartz–Cry53Tp47. We then show the effect of alkali/aluminum ratio on the maximum fluorine solubilities in quartz–albitic and granitic melts. Finally we discuss how the presence of micas or other phases that can buffer alumina activity has an effect on the fluorine content of granitic melts.
|
EXPERIMENTAL METHODS |
|---|
All experiments were performed at 100 MPa in cold-seal pressure vessels (<850°C) or rapid-quench TZM pressure vessels (>850°C) using argon as pressure medium. Starting materials were synthetic glasses and natural mineral phases (Tables 1 and 2). Thirty-four base mixes in the albite–K-feldspar–quartz–topaz–cryolite system were prepared by careful weighing of constituents in the desired proportions and mixing in an agate mortar for 1 h (Table 3). Capsules were prepared from seamless gold or platinum tubing, distilled and deionized water was loaded with a microsyringe and covered with the starting powder. Loaded capsules were crimped and welded with an arc welder while partially submerged in a cold-water bath. Random checks of capsules by piercing and estimating water content by loss during heating revealed no H2O loss during welding within weighing precision (0· 02 mg); the total weight loss during welding is 0· 04–0· 08 mg in both anhydrous and hydrous runs and is attributed to metal loss. The weighed-in H2O contents are accurate to 0· 1 wt %. At the end of each experiment, the cold-seal vessel was placed in an air jet and quenched at 150°C/min whereas the runs in the TZM pressure vessels were quenched by free fall into the cooling collar at 100°C/s. Recovered capsules were weighed to check for leakage, opened immediately and studied by optical microscopy and electron microprobe.
|
|
|
Attainment of equilibrium is facilitated by the presence of H2O and fluorine. No evidence for disequilibrium was encountered in experiments on the hydrous quartz–albite and haplogranite joins in 7 day runs. This is in agreement with attainment of equilibrium in the volatile-bearing systems after 4 days (Candela & Holland, 1984
; Williams et al., 1997; Frank et al., 2003). For further details of experimental techniques and run-product descriptions, the reader is referred to the first part (Dolej & Baker, 2007). |
TERMINOLOGY |
|---|
Abbreviations for all phases are summarized in Table 1. The terms liquid, fluid and vapor are used in accordance with Stalder et al. (2000
) and Wyllie & Ryabchikov (2000): liquid represents silicate, fluorosilicate or fluoride melt without or with a limited amount of dissolved H2O (less than 15 wt % in this study); vapor is a low-density aqueous phase with a small amount of solutes (less than 20 wt %); fluid is a general term used for a H2O-dominated phase with low to high solute concentration.
The term ‘melt aluminosity’ is a synonym for the aluminum/alkali cation ratio and is used to describe the relative variations of this ratio. The terms peralkaline, subaluminous and peraluminous are used as defined by Shand (1927) and Holtz et al. (1992); in the abbreviation Al/(Na + K) we use molar proportions.
For divariant fields and trivariant volumes, we use standard labeling (e.g. L + cry). For univariant curves and invariant points, we use the notation of Greig et al. (1955). For example, L (tp) indicates a phase boundary between L and L + tp fields. Similarly, L + V (cry + tp) is an invariant point between four fields: L + V, L + V + cry + tp, L + V + cry and L + V + tp. Phases reported in square brackets are present in all fields of the phase diagram; for example, [+V] indicates vapor-saturated conditions. The phase-diagram descriptions refer to the practical number of components (e.g. haplogranite–topaz binary, rather than haplogranite–topaz pseudobinary or quartz–albite–K-feldspar–topaz–H2O quinary).
|
THE TOPAZ–CRYOLITE–H2O SYSTEM |
|---|
Peralkaline to peraluminous silicic magmas saturate with topaz and/or cryolite (Dolej
& Baker, 2004, 2007). Therefore, we studied the topaz–cryolite binary system with 10 wt % H2O (Table 4; Fig. 1). At 100 MPa, it is characterized by simple eutectic behavior with eutectic point cry + tp + L (V) at 660°C and cation Al/Na 0· 7. The eutectic temperature in fluid-saturated conditions is depressed by 110°C at 100 MPa relative to the anhydrous system and the cation Al/Na ratio decreases by at least 0· 25 (Dolej & Baker, 2007; Fig. 1). The occurrence of the three-phase field cry + tp + L (Fig. 1) confirms that the melt in the 10 wt % section is vapor-undersaturated (see Koster van Groos & Wyllie, 1968), and this implies that water solubility in the topaz–cryolite melt is greater than 10 wt % H2O at 100 MPa.
|
|
The eutectic temperature of the cryolite–topaz–H2O system is 60°C lower and the H2O solubility is at least three times by weight higher than that of the haplogranite–H2O system at 100 MPa (Tuttle & Bowen, 1958
; Burnham, 1997; Holtz et al., 2001). This suggests that the topaz–cryolite–haplogranite–H2O eutectic may be displaced to very high fluorine concentrations near the topaz–cryolite side and water solubility may significantly increase in residual fluorogranitic melts. |
THE QUARTZ–TOPAZ–CRYOLITE–H2O SYSTEM |
|---|
The liquidus relations in the quartz–topaz–cryolite system under hydrous conditions (Table 5) provide a general illustration of the liquid line of descent and describe variations in SiO2 and F concentrations in residual fluorosilicate liquids. Figure 2 presents the isobaric section from SiO2 to Cry53Tp47 at 10 wt % H2O, which is the subaluminous section through the system. The pseudobinary solidus is a piercing point cry + V (qz + tp + L) and the sequence of liquidus fields indicates that the location of ternary eutectic departs from the join to weakly peraluminous conditions. This is in contrast to the weakly peralkaline composition of the quartz–topaz–cryolite eutectic under anhydrous conditions (Dolej
& Baker, 2007). In addition, the eutectic composition is more SiO2-rich and F-poor than at anhydrous conditions (Fig. 2; Dolej & Baker, 2007, fig. 14).
|
|
The residual liquids in the quartz–topaz–cryolite system have compositions close to nepheline with fluorine. The location of the nepheline–F2O–1 join is marked by TCQ-3 (Fig. 2, Table 3), and we have determined H2O solubility for this composition at 100 MPa. Melts of this composition cannot be quenched owing to their very high fluorine content (29· 6 wt % F) and the water solubility must be estimated from the temperature–X(H2O) section (Fig. 3). The location of the vapor saturation, i.e. L (V) univariant curve, is given by the inflection on the topaz and topaz + cryolite liquidus curves. This defines maximum H2O content in the melt 12· 5 ± 0· 5 wt % and this value represents a threefold increase by weight, in comparison with the fluorine-free haplogranitic minimum at the same pressure (Burnham, 1975
; Holtz et al., 2001). This measurement also agrees with the vapor undersaturation of the cryolite–topaz cotectic with 10 wt % H2O (Fig. 1). These observations are in agreement with the increase in water solubility in granitic melts with increasing fluorine contents described by Holtz et al. (1993) and Webster & Rebbert (1998). Those workers determined experimentally an increase by 0· 5 and 0· 8 wt % H2O for each wt % F at 200 MPa, but only for fluorine concentrations less than 5 wt % in the melt.
|
Our experimental results suggest that the common presence of fluorine and H2O in the melt does not result in F– and OH– site competition, which would lead to a decrease in water solubility with increasing fluorine concentration. Rather, additional fluorine in the melt promotes incorporation of hydroxyl species and/or molecular H2O. This effect is possibly explained by decreasing ion polarizability when fluorine is added to silicate melts (Duffy, 1989
) and this promotes hydroxylation of network modifiers and aluminosilicate tetrahedra. Increase in H2O solubility in fluorosilicate melts can also be described by Lewis acid–base interactions (see London, 1987). Addition of F2O–1 (strong Lewis acid; Duffy, 1989) leads to strong short-range order with low-field strength cations (Lewis bases) forming alkali–F and (alkali,Al)–F complexes. High-field strength cations and highly polarizable oxygen atoms remain associated (Si–O; Schaller et al., 1992; Zeng & Stebbins, 2000). The added H2O (a weak Lewis base) is expected to more extensively form alkali–OH bonds and replace bridging oxygens by hydroxyl groups than it does in fluorine-free compositions (Oglesby & Stebbins, 2000; Schmidt et al., 2000). Such mechanism of increasing proportion of hydroxyl species in the melt can explain the increase in the H2O solubility observed in experiments. |
THE ALBITE–QUARTZ–TOPAZ–CRYOLITE–H2O SYSTEM |
|---|
In the quinary albite–quartz–topaz–cryolite–H2O system, we experimentally studied the pseudobinary join that connects the quartz–albite eutectic, Qz41Ab59 (800°C, 100 MPa and H2O saturation; Tuttle & Bowen, 1958
) with the subaluminous topaz–cryolite composition, Cry53Tp47 (Table 6). The resulting temperature–composition section at 100 MPa is presented in Fig. 4.
|
|
Addition of topaz and cryolite causes depression of both quartz and albite liquidi, with quartz exhibiting a steeper dT/dx drop than albite (Fig. 4). This effect probably stems from a decrease in the bulk SiO2 content (by adding topaz and cryolite) and from NaAl–F short-range order in the melt structure (Manning et al., 1980
; Schaller et al., 1992; Zeng & Stebbins, 2000). This depression of both quartz and albite liquidi is in contrast to the findings of Wyllie & Tuttle (1961) and Wyllie (1979), who observed decreasing liquidus temperature of albite and increasing liquidus temperature of quartz, with addition of fluorine. This difference is a result of adding fluorine in the form of HF, i.e. comparing two distinct phase diagram sections. Addition of hydrofluoric acid to the Qz41Ab59 eutectic composition reaches composition Qz73· 4Cry14· 2Tp12· 4, which has also been investigated in this study. It plots at 73· 4 wt % SiO2 and 26· 6 wt % Cry53Tp47 (Fig. 2) and the quartz liquidus temperature has increased to approximately 1000°C. This is in agreement with the extended liquidus trend of Wyllie & Tuttle (1961) and Wyllie (1979).
The Ab59Qz41–Cry53Tp47 section illustrates that crystallization temperatures are depressed from the fluid-saturated albite–quartz eutectic at 800°C (Tuttle & Bowen, 1958) through piercing points L + V (ab + cry) and L + V (cry + tp) at 600–610°C and to the quinary albite–quartz–topaz–cryolite eutectic at 580°C (100 MPa, H2O saturation; Fig. 4). The cryolite and topaz liquidus curves intersect at the L + V (cry + tp) piercing point and the sequence of stability fields implies that the cryolite–topaz cotectic curve passes from peralkaline to peraluminous space. Below the L + V (ab + cry) and ab + L + V (qz + cry) piercing points, residual melts have a peraluminous composition.
|
THE HAPLOGRANITE–TOPAZ–CRYOLITE–H2O SYSTEM |
|---|
Addition of K2O to the previous system completes the senary composition space Na2O–K2O–Al2O3–SiO2–F2O–1–H2O necessary for the description of haplogranitic melts and full interpretation of the liquid lines of descent of natural silicic magmas. This system was studied in four temperature–composition sections and two isothermal sections through the haplogranite (Qz38Ab33Or29)–topaz–cryolite–H2O space. The compositions of the starting mixes are presented in Fig. 5, and experimental results are listed in Table 7.
|
|
The temperature–composition sections for the two limiting binaries: haplogranite–topaz and haplogranite–cryolite at 10 wt % H2O, are shown in Fig. 6. The fluorine-free haplogranitic minimum occurs at 720°C, 100 MPa and H2O saturation (Tuttle & Bowen, 1958
). In the haplogranite–topaz pseudobinary, the solubility of topaz in the H2O-saturated haplogranitic melt is low, corresponding to less than 2 wt % F below 700°C. In the haplogranite–cryolite pseudobinary, the solubility of cryolite is higher, about 4 wt % F. The pseudobinary eutectics are located at 640°C (100 MPa and H2O saturation, Fig. 6), i.e. both fluorine-bearing minerals cause a solidus depression of 80°C, relative to the H2O-saturated haplogranite minimum (Tuttle & Bowen, 1958). These eutectics and saturation limits represent limiting cases for strongly peralkaline or strongly peraluminous granitic melts, respectively.
|
Addition of either topaz or cryolite to the haplogranite system causes distinct depressions of quartz and feldspar liquidi, i.e. topaz and cryolite affect stabilities of quartz and feldspar differently (Fig. 6). In the haplogranite–cryolite pseudobinary, the alkali-feldspar liquidus is more depressed than that of quartz. Such increase in the activity of quartz compared with alkali feldspar is probably a result of strongly positive deviations from mixing in the silica–cryolite binary, manifested by liquid–liquid immiscibility (Dolej
& Baker, 2007). In the haplogranite–topaz pseudobinary, the effect is reversed, i.e. the quartz liquidus is more depressed than that of feldspar. This means that topaz does not cause, but rather suppresses positive deviations from ideal mixing in the melt. Such behavior is in agreement with the disappearance of liquid–liquid immiscibility in the silica–cryolite system and with the strong quartz liquidus depression when topaz is added (Dolej & Baker, 2007). In addition, the greater depression of the quartz liquidus (Fig. 6a) is promoted by a decrease in bulk SiO2 content upon addition of topaz. The phase relations along the subaluminous haplogranite–Cry53Tp47 join at 10 wt % H2O (Fig. 7) are very similar to those in the quartz–albite-cryolite–topaz system (Fig. 4). Addition of topaz and cryolite causes a depression in granite crystallization temperatures to the quaternary eutectic at 540°C (100 MPa and H2O saturation). Intersection of the topaz and cryolite liquidus curves defines the L + V (cry + tp) piercing point, whose presence indicates that melt compositions change from peralkaline to peraluminous along the topaz–cryolite cotectic. The L + V (cry + tp) piercing point is located at significantly higher fluorine concentration (11 wt % F) than individual solubilities of fluorine at topaz or cryolite saturation (Fig. 6). This means that fluorine solubility is much higher in subaluminous melts than in peralkaline or peraluminous systems. That is, the melt alkali/aluminum ratio has a significant effect on fluorine solubility.
|
The effect of the melt aluminum/alkali ratio on fluorine solubility can be interpreted from the topology of cryolite and topaz saturation surfaces. The saturation isotherms, L (cry) and L (tp), at 800°C and 100 MPa are shown in Fig. 8. Under both anhydrous and hydrous conditions the liquid field is strongly elongate and it extends to very high fluorine concentrations. The cryolite and topaz saturation isotherms follow a course similar to alkali/aluminum ratio isopleths. In this system, fluorine concentration in the melt and the alumina saturation index are not independent and are dictated by the location of the liquid on the cotectic curve. With progressive fractionation, fluorine contents in the melt increase and the alkali/aluminum ratio is constrained by the cryolite and topaz saturation surfaces to fall within a narrow range.
|
The symmetric location of cryolite and topaz liquidus isotherms (Fig. 8) around the subaluminous isopleth has implications for the speciation of fluorine in the melt structure. It suggests short-range order between alkali, aluminum and fluorine, where Na:Al
1. The relevant melt species is NaAlF4 and its existence in fluorine-bearing aluminosilicate melts has been confirmed by spectroscopic investigations (Zeng & Stebbins, 2000).
The field of silicate liquids at high fluorine concentrations forms a narrow prismatic wedge with decreasing temperature (Fig. 9). Its boundaries are the cryolite and topaz liquidus surfaces, respectively. These surfaces constrain the alkali/aluminum ratio in the melt to a progressively narrower range with decreasing temperature. Finally, the liquid [+vapor] volume closes at a subaluminous composition, Al/(Na + K) = 1, at an invariant point L + V (cry + tp) at 590°C. The sequence of phases at this point implies that the pseudoternary haplogranite–cryolite–topaz eutectic is located at less than 30 wt % cryolite + topaz.
|
Near-solidus crystallization of fluorine-rich granitic melts is illustrated in two isothermal–isobaric sections of the central portion of the haplogranite–cryolite–topaz–H2O system. At 580°C, 100 MPa and 10 wt % H2O (Fig. 10a) the presence of the L [+V] field indicates that the liquid compositions still remain within the ternary plane and exhibit no quartz or feldspar enrichment compared with the hydrous haplogranitic minimum. The liquid field is located between 6 and 10 wt % F and has a narrow span of aluminum/alkali cation ratios, 0· 94–1· 08. The L [+V] field closes at 550°C (Fig. 10b) and is replaced by the fsp + L [+V] field, indicating the departure of melt composition towards quartz-rich compositions. The eutectic melt composition is located in the quartz–feldspar–cryolite–topaz tetrahedron, more specifically in its quartz–haplogranite–cryolite–topaz subspace. The fsp + L [+V] field closes as an invariant point fsp + L (cry + qz + tp) [+V] at 3· 6 wt % F and a weakly peraluminous composition (cation Al/(Na + K) = 1· 05). This invariant point is a piercing point on the ternary plane of a tie-line connecting the eutectic melt composition in the quartz–feldspar–cryolite–topaz tetrahedron with the feldspar apex. By chemography, the eutectic melt composition is bracketed between G90Cry4· 7Tp5· 3 [by weight, 3· 6 wt % F, Al/(Na + K) = 1· 05] and Qz83· 9Cry7· 6Tp8· 5 [5· 9 wt % F, Al/(Na + K) = 1· 18].
|
, locations of starting compositions.The experimental design imposes some constraints on interpretation of the results. As the amount of residual melt in the experimental charges decreases with decreasing temperature, the melt becomes fluid-saturated (the total H2O content in the system is constant) and the fluid/melt ratio increases. As a result of incongruent dissolution of aluminosilicates in aqueous fluid (Manning, 1981
; Dingwell, 1985; Webster, 1990), the composition of the near-eutectic melt departs from its projected position on the phase diagram. On the basis of our preliminary partitioning experiments and results of thermodynamic calculations, it is expected that the eutectic melt becomes slightly depleted in SiO2 and F and that its aluminum/alkali ratio increases. |
PETROLOGICAL IMPLICATIONS |
|---|
The behavior of fluorine in silicate melts (see also Manning, 1981
; Webster, 1990; Mysen et al., 2004) stands in remarkable contrast to the effects of other volatile elements (Cl, S, B, P; Carroll & Webster, 1994). The very high solubility of fluorine in silicate melts (Koster van Groos & Wyllie, 1968; Webster, 1990; Carroll & Webster, 1994) is a consequence of fluorine–oxygen substitution in the melt structure (Mysen et al., 2004) because of the similarity of ionic radii of fluorine (1· 29 Å) and oxygen (1· 35 Å; Shannon, 1976). Formation of fluorosilicate and fluoroaluminate tetrahedral and octahedral complexes (Schaller et al., 1992; Zeng & Stebbins, 2000; Mysen et al., 2004) is responsible for a decrease in the activities of silicate melt components (Manning et al., 1980; London, 1987). Therefore, liquidus and solidus temperatures are depressed and compositional shifts of haplogranite minima occur (Manning, 1981). Upon fluid saturation, incongruent fluid–melt partitioning is responsible for selectively sequestering elements from residual melts (Dingwell, 1985). In addition, the presence of rock-forming elements such as calcium or lithium may stabilize new fluorine-bearing phases, e.g. fluorite (Dolej & Baker, 2006) or lithium micas.
Differentiation mechanisms of leucocratic silicic melts
The solidus temperatures of hydrous albitic and haplogranitic melts decrease with increasing fluorine contents to less than 600–630°C at 275 and 100 MPa, respectively (Wyllie & Tuttle, 1961; Koster van Groos & Wyllie, 1968; Manning, 1981). Experimental studies on natural fluorine-bearing silicic compositions with 0· 9–1· 2 wt % F demonstrate that solidus temperatures range between 675 and 500°C at 100–150 MPa and aqueous-fluid saturation (Webster et al., 1987; Weidner & Martin, 1987; Xiong et al., 2002). Our experimental determination of a solidus temperature of 540°C in the haplogranite–topaz–cryolite–H2O system at 100 MPa falls within this range of solidus temperatures. It is noteworthy that differences among previous studies must partly be due to variable fluorine concentrations in the system and in the residual melt.
Silicate–fluoride liquid–liquid immiscibility (Kogarko & Krigman, 1981; Veksler et al., 2005; Dolej & Baker, 2007) does not propagate to the low-temperature fluorosilicate systems studied here. In fluorine-bearing hydrous silicic systems, cryolite and topaz are the saturating solid phases. The low eutectic temperature in the hydrous topaz–cryolite join (660°C) causes displacement of ternary and quaternary eutectics towards this join and thus enables the silicate-precipitating surfaces to extend to elevated concentrations of fluorine in residual melts. The relevant invariant points at 100 MPa and aqueous-fluid saturation are quartz–topaz–cryolite at 590°C, quartz–albite–topaz–cryolite quaternary eutectic at 580°C and quartz–alkali feldspar–topaz–cryolite quaternary eutectic at 540°C. Phase relations in these systems define differentiation paths of Li-, Ca- and Fe-poor fluorine-bearing granites, rhyolites, ongonites and their differentiates (quartz topazites, xianghualingites, elvans). We compare natural whole-rock compositions with experimental liquidus relations in the schematic Jänecke projection (Jänecke, 1906) on the quartz saturation surface (Fig. 11). Fluorine-bearing natural rocks are moderately to strongly peraluminous, whereas peralkaline types are nearly absent. Fluorine-bearing granites and ongonites cluster close to the feldspar–aluminosilicate (mica, andalusite)–topaz [+quartz] cotectic curves and represent magmatic liquids. The scatter most probably reflects effects of additional minor components on the phase relations and/or variable accumulation of crystallizing solids. On the other hand, compositions of quartz topazites and xianghualingites plot on the topaz [+quartz] surface, consistent with their biminerallic assemblage; importantly, topazites do not appear to represent liquid compositions at reasonable temperatures. We propose that natural occurrences of fine-grained quartz topazites with magmatic flow banding and trapped xenoliths are crystal assemblages produced by alkali-bearing melts (see Kortemeier & Burt, 1988). Coarse-grained and miarolitic quartz topazites and topaz silexites can be interpreted as products of the disequilibrium crystallization of pegmatite-forming melts after volatile loss or were affected by hydrothermal alteration (Birch, 1984; Kleeman, 1985; Johnston & Chappell, 1992; see Hervig et al., 1987) and are not comparable with experimental data. Importantly, the fluoride–silicate liquid–liquid immiscibility is located at very high fluorine contents, beyond the feldspar–topaz–cryolite [+quartz] eutectic, and is not approached by any whole-rock compositions (Fig. 11).
|
Effects of aluminum/alkali ratio
In the haplogranite–topaz–cryolite system with H2O, the individual solubilities of topaz or cryolite in their pseudobinaries are very low,
2 and 4 wt % F, respectively (Fig. 6). The solubilities of both phases, however, rapidly increase in the central subaluminous portion of the pseudoternary haplogranite–topaz–cryolite system (Fig. 8). When the isopleths of aluminum/alkali ratio and fluorine concentrations in the melt are imposed on this system (Fig. 5) these two variables are not independent in topaz- or cryolite-precipitating melts. That is, the fluorine concentration in the melt at topaz or cryolite saturation is not unique, but it strongly depends on the alumina/alkali ratio of the silicate melt. Granitic and rhyolitic magmas evolve by crystal fractionation along the quartz–feldspar cotectic surface and become enriched in fluorine. Once the melt is saturated in topaz or cryolite and, if a(Al2O3) is not buffered by other solid phases, the residual melt will evolve to higher fluorine concentrations along the quartz–feldspar–topaz or quartz–feldspar–cryolite cotectic. These cotectics dictate the alkali/aluminum ratio of the melt, which will converge to a subaluminous value. All melts completely crystallize at the quaternary eutectic, where saturation with a second fluorine-bearing mineral occurs.
In multicomponent systems, the presence of other phases (micas, andalusite, cordierite, garnet, amphibole) buffers a(Al2O3) in the melt. For example, an aluminosilicate mineral (andalusite, sillimanite) in the presence of quartz determines a(Al2O3) by the following equilibrium:
|
| (1) |
|
| (2) |
) and Barton (1996). Therefore, if a(Al2O3) in the melt is buffered by additional precipitating mineral phases, the liquid line of descent on the quartz–feldspar cotectic surface will follow a specific isopleth of alkali/alumina ratio in the melt. When the cotectic curve with topaz or cryolite is reached, the assemblage becomes invariant. At this point, the melt completely crystallizes, without further evolving to high fluorine contents and without crystallization of a second fluorine-bearing phase. The first differentiation sequence with no external buffering is applicable to leucogranitic and leucorhyolitic magmas, comparable with highly evolved topaz rhyolites and ongonites, whereas biotite-bearing, two-mica or aluminosilicate-bearing granites will follow a buffered sequence.
Effects of additional components
Additional rock-forming elements (Ca, Mg, Fe, or Li) or volatile constituents (B, P) will affect the differentiation model described above. The fluorine solubilities may be limited by saturation in new fluorine-bearing phases. These may include fluorine-bearing minerals (lithium fluoromicas, viliaumite, fluorite; Burt & London, 1982; Dolej & Baker, 2004, 2006), immiscible fluoride liquids (Kogarko & Krigman, 1981; Veksler, 2004) or fluorine-rich aqueous fluids (Webster, 1990). In natural peralkaline and calc-alkaline magmas, fluorite becomes the stable solid phase (Hogan & Gilbert, 1995; Marshall et al., 1998). Furthermore, Dolej & Baker's (2006) thermodynamic calculations demonstrate that fluorite is also stable in Fe-, Mg- and Ti-bearing silicic rocks. Fluorite buffers fluorine concentrations to low levels, the values of which are determined by the calcium content in the melt (Price et al., 1999; Scaillet & Macdonald, 2004; Dolej & Baker, 2006). The widespread stability and low solubility of fluorite prevents melt enrichment in fluorine concentrations above 0· 5–1 wt % F in most calcium-bearing igneous systems (see Price et al., 1999; Dolej & Baker, 2006).
Fluorine behavior in Li-, B- and P-rich granitic and pegmatitic melts remains, however, much less understood. These suites are Ca-poor (
ern
, 1998; Stilling, 1998) and fluorite stability is suppressed to near-solidus conditions (Webster et al., 1987; Weidner & Martin, 1987). The presence of lithium in many evolved granites (Cuney et al., 1992; Charoy & Noronha, 1996; Förster et al., 1999; ern et al., 2005) stabilizes lithium micas and amblygonite–montebrasite solid solutions that may act as sinks for fluorine during prolonged differentiation (London, 1997) because fluorine preferentially partitions into these mineral phases (Icenhower & London, 1995; London et al., 2001). The amount of precipitating solids is limited by the low amounts of lithium and phosphorus available in the melt. Thus, these minerals are unlikely to inhibit the fluorine enrichment in residual melts. Another effect is the significant depression of crystallization temperatures by lithium. The solidus temperature in the system LiAlSiO4–NaAlSi3O8–SiO2–H2O is lowered to 640°C at 200 MPa (Stewart, 1978), and with addition of Li2B4O7 it further decreases to 500°C at 200 MPa (London, 1986). These depressions are 100 and 240°C relative to the NaAlSi3O8–SiO2–H2O ternary at the same pressure (Tuttle & Bowen, 1958). Similar solidus depressions occur in the feldspar-free but fluorine-bearing systems. In the quartz–trilithionite pseudobinary join, Munoz (1971) determined a solidus temperature of 600°C at 200 MPa and fluid saturation. These observations suggest that lithium, unlike calcium, significantly suppresses the crystallization temperatures in lithium-rich granitic and pegmatitic melts and that the occurrence of Li–F micas does not prevent high enrichment in fluorine in residual melts (Munoz, 1971).
H2O solubility and fluid saturation
Depolymerization of silicate melt by fluorine appears to promote water solubility (Holtz et al., 1993; Webster & Rebbert, 1998). In contrast to the results of Dingwell (1985) and Webster (1990), who documented a decrease or minimal change of the H2O solubility up to 8 wt % F, numerous other studies reported positive correlation between the fluorine content and the H2O solubility in the melt. For example, Holtz et al. (1993) found that at 200 MPa addition of 4· 5 wt % F to synthetic granitic melts increases H2O solubility by 2· 2 wt %. Similarly, Webster & Rebbert (1998) found that addition of 1· 1 wt % F to a natural rhyolite increases water content by 0· 9 wt %. These increases and the presence of melt inclusions in topaz-bearing granites that contain up to 10 wt % H2O (Thomas & Klemm, 1997) are consistent with our experimental results.
The knowledge of H2O solubility in silicate melts is critical for interpreting the timing of fluid saturation. An increase in H2O solubility allows extensive magmatic fractionation, and suppresses saturation with aqueous fluid phase and dispersal of economically important elements. As a consequence, residual magmas attain high fluorine and H2O concentrations and exhibit significant enrichments in lithophile elements (Li, Rb, Cs, Sn, Nb, Ta; Cuney et al., 1992; Webster et al., 1997, 2004). This enrichment is observed in topaz rhyolites and ongonites (
temprok, 1991; Dergachev, 1992) but quartz topazites are remarkably depleted in alkalis (< 1 wt % Na2O + K2O), lithophile elements and ore metals (Kortemeir & Burt, 1988; Johnston & Chappell, 1992). This suggests that saturation in aqueous fluid and sequestration of incompatible elements occurs at the ongonite–topazite transition (see also Birch, 1984; Kortemeier & Burt, 1988; Johnston & Chappell, 1992).
Hydrothermal fluids in fluorosilicate systems
With increasing fluorine concentration in the melt, the coexisting aqueous fluid becomes rich in aluminosilicate solutes (Dingwell, 1985; Webster, 1990). The presence of SiO2-rich melt or gel inclusions in quartz topazites and greisens (Eadington & Nashar, 1978; Williamson et al., 1997, 2002) is in agreement with very high solubility of quartz in fluorine-bearing aqueous fluids (Dolej, 2006). Consequently, the solvus between hydrous fluorosilicate melts and solute-rich aqueous fluids contracts with increasing fluorine concentrations. Although a continuous magmatic–hydrothermal transition has been advocated by previous researchers (e.g. London, 1986), we found that fluorine-rich haplogranitic melts had a finite water content under the conditions we studied.
We can draw several important conclusions about fluid–melt partitioning at high fluorine concentrations from phase-diagram topology. Figure 12 is a schematic quaternary projection with the end-members aluminosilicates (Al2O3 + SiO2), alkalis (Na2O + K2O), fluorine (F2O–1) and water (H2O). Phase relations in the tetrahedron define the geometry and tie-line orientation of the liquid (melt)–vapor (fluid) solvus.
|
The front edge of the tetrahedron is the silicate–H2O binary showing immiscibility between hydrous silicate melt (
4· 2 wt % H2O at 100 MPa; Burnham, 1997; Holtz et al., 2001) and an aqueous vapor with very small fraction of aluminosilicate solutes (Eugster & Baumgartner, 1987; Paillat et al., 1992). The miscibility gap between hydrous silicate melts and aqueous fluid shrinks as it propagates into the tetrahedron interior. In peralkaline systems, which plot at the tetrahedron base, the solubility of H2O in the melt increases (Dingwell et al., 1997) and alkali silicates are extensively soluble in the aqueous fluids (Luth & Tuttle, 1969). Complete miscibility between peralkaline silicate melts and aqueous fluids occurs at low pressures (Mustart, 1972); that is, the liquid–vapor solvus closes in the tetrahedron base. In systems containing fluorine, which plot on the front face of the tetrahedron, solubility of aluminosilicates in aqueous fluid increases as a result of the formation of aluminosilicofluoride complexes (Dingwell, 1985; Haselton et al., 1988; Aksyuk & Zhukovskaya, 1998; Tagirov et al., 2002). HF and SiF4, which plot in the central portion of the tetrahedron front face, are low-density fluids at high temperatures (Franck & Spalthoff, 1957; Devyatykh et al., 1999) and form supercritical mixtures with H2O. Therefore, the liquid–vapor gap in the front face has to close before reaching the H2O–HF–SiF4 tie-lines. An additional constraint on the extent of melt–fluid immiscibility arises from the phase equilibria along the alkali fluoride–H2O binary. This join is represented by a line from the left front apex to the centre of the back edge (NaF). If the relationship between alkali fluoride and H2O is supercritical, the melt–fluid gap must completely close within the tetrahedron body. On the other hand, if the alkali fluoride–H2O join is subcritical, there is no continuous miscibility between hydrous fluorosilicate melts and aqueous fluids and the liquid–vapor solvus remains open. The NaF–H2O system exhibits subcritical behavior at less than 400 MPa (Ravich & Valyashko, 1965; Koster van Groos & Wyllie, 1968; Kotelnikova & Kotelnikov, 2002), whereas the KF–H2O system exhibits a continuous transition from molten salt to aqueous vapor with a maximum vapor pressure of 190 MPa (Urusova & Ravich, 1966). This observation implies that the supercritical transition from hydrous fluorosilicate melts to solute-rich aqueous fluids may occur in potassium-rich systems only.
We illustrate petrological applications of these features by projecting the melt–fluid miscibility gap on the silicate–(Na,K)F–H2O plane (Fig. 12b). During magmatic differentiation volatile-bearing magmas will evolve from the lower right apex along the silicate liquidus and eventually reach cotectic with a fluoride solid phase (for example, topaz or cryolite) or will exsolve aqueous fluid. The liquid line of descent depends on the initial F/H2O ratio in the melt. When the melt is fluid-saturated, the composition of the coexisting aqueous vapor is determined by L + V tie-lines between the vapor-saturated silicate liquidus and L + X-present vapors (Fig. 12b). On the vapor side, tie-lines project close to the H2O apex, implying that the fluorine-bearing fluids are not acidic HF-rich or SiF4-dominated solutions, but rather contain alkali–aluminofluoride and silicofluoride complexes. Because the stoichiometry of the predominant aqueous complexes differs from bulk composition of the melt, one can expect moderate departures from congruent partitioning of elements between melt and fluid. This is in agreement with results of previous fluid–melt partitioning and solubility studies (Dingwell, 1985; Haselton et al., 1988; Tagirov et al., 2002).
All crystallization paths converge to the vapor-saturated eutectic with silicate and fluorine-bearing minerals. This invariant point is labeled X + L (V) in Fig. 12b. The composition of the aqueous fluid coexisting with the eutectic hydrous fluorosilicate melts is located at the invariant point on the vapor surface labeled as X + V (L). The continuous transition from volatile-rich silicate melts to solute-rich fluids is expected to appear only at high alkali, high K/Na and/or fluorine concentrations. In natural conditions, formation of residual melts extremely rich in alkalis and fluorine will be inhibited by crystallization of fluoride minerals (topaz, cryolite and villiaumite) and in these systems the continuous melt–fluid transition is unlikely to occur.
|
ACKNOWLEDGEMENTS |
|---|
Both parts of this study represent a portion of the first author's Ph.D. thesis at McGill University, supported by the J. B. Lynch and Carl Reinhardt McGill Major fellowships. We gratefully acknowledge discussions with Miroslav
temprok, John Longhi, Don Burt and Mark Barton. The Theriak-Domino software by Christian de Capitani (University of Basel) was helpful in verifying phase-diagram topologies and mineral–melt thermodynamics. Research costs were covered by the Natural Sciences and Engineering Research Council grants to D.R.B. and by the Geological Society of America and the Society of Economic Geologists student grants to D. D. Bob Loeffler provided topaz crystals from the Topaz Mountain, Utah. Critical reviews by Hanna Nekvasil, Bruno Scaillet, Ilya Veksler, Don Burt and Ron Frost helped to improve the manuscript and are gratefully acknowledged.
*Corresponding author. Present address: Bayerisches Geoinstitut, University of Bayreuth, 95440 Bayreuth, Germany. Telephone: +49-(0)921-553718. Fax: +49-(0)921-553769. E-mail: david.dolejs{at}uni-bayreuth.de
|
REFERENCES |
|---|
Adamczak RL, Mattern JA, Tieckelmann H. (1959) A partial phase study of the system NaF–HF. Journal of Physical Chemistry 63:2063–2065.
Aksyuk AM and Zhukovskaya TN. (1998) The solubility of quartz in aqueous solutions of hydrofluoric acid at temperatures 500–1000°C and pressures of 100–500 MPa. Doklady Earth Sciences 361:745–748.
Baker DR and Bossànyi H. (1994) The combined effect of F and H2O on interdiffusion between peralkaline dacitic and rhyolitic melts. Contributions to Mineralogy and Petrology 117:203–214.[CrossRef][Web of Science]
Baker DR and Vaillancourt J. (1995) The low viscosities of F + H2O-bearing granitic melts and implications for melt extraction and transport. Earth and Planetary Science Letters 132:199–211.[CrossRef][Web of Science]
Barton MD. (1996) In Brown M, Candela PA, Peck DL (Eds.), et al. Granitic magmatism and metallogeny of southwestern North America. The third Hutton symposium on the Origin of granites and related rocks. Geological Society of America, Special Paper 315:261–280.
Barton MD, Ilchik RP, Marikos MA. (1991) In Kerrick DM (Ed.). Metasomatism. Contact Metamorphism. Mineralogical Society of America, Reviews in Mineralogy 26:321–350.
Birch WD. (1984) Quartz–topaz–loellingite rocks near Eldorado, Victoria. Australian Journal of Earth Sciences 31:269–278.[Web of Science]
Burnham CW. (1975) Water and magmas; a mixing model. Geochimica et Cosmochimica Acta 39:1077–1084.[Web of Science]
Burnham CW. (1997) Magmas and hydrothermal fluids. In Barnes HL (Ed.). Geochemistry of Hydrothermal Ore Deposits(John Wiley, Wiley, New York) pp. 63–123.
Burt DM and London D. (1982) In ern P (Ed.). Subsolidus equilibria. Granitic Pegmatites in Science and Industry. Mineralogical Association of Canada, Short Courses 8:329–346.
Cady GH. (1934) Freezing points and vapor pressures of the system potassium fluoride–hydrogen fluoride. Journal of the American Chemical Society 56:1431–1434.
Candela PA and Holland HD. (1984) The partitioning of copper and molybdenum between silicate melts and aqueous fluids. Geochimica et Cosmochimica Acta 48:373–380.[CrossRef][Web of Science]
Carroll MR and Webster JD. Carroll MR and Holloway JR. (1994) Solubilities of sulfur, noble gases, nitrogen, chlorine, and fluorine in magmas. Volatiles in Magmas. Mineralogical Society of America, Reviews in Mineralogy 30:231–279.
ern P. (1998) Magmatic vs. metamorphic derivations of rare-element granitic pegmatites. Krystalinikum 24:7–36.
ern P, Masau M, Goad BE, Ferreira K. (2005) The Greer lake leucogranite, Manitoba, and the origin of lepidolite-subtype granitic pegmatites. Lithos 80:305–321.[CrossRef][Web of Science]
Charoy B and Noronha F. (1996) Multistage growth of a rare-element, volatile-rich microgranite at Argemela (Portugal). Journal of Petrology 37:73–94.
Cuney M, Marignac C, Weisbroad A. (1992) The Beauvoir topaz–lepidolite albite granite (Massif Central, France): the disseminated magmatic Sn–Li–Ta–Nb–Be mineralization. Economic Geology 87:1766–1794.
Dergachev VB. (1992) Geochemical types of ongonites. Geochemistry International 29:37–47.
Devyatykh GG, Pryakhin DA, Bulanov AD, Balabanov VV. (1999) Phase diagram of silicon tetrafluoride. Doklady Chemistry 364:4–5.
Dingwell DB. (1985) The structure and properties of fluorine-rich silicate melts: implications for granite petrogenesis. In Taylor RP and Strong DF (Eds.). Granite-Related Mineral Deposits: Geology, Petrogenesis and Tectonic Setting(Canadian Institute of Mining, Metallurgy and Petroleum, Halifax, N.S.) pp. 72–81.
Dingwell DB, Scarfe CM, Cronin DJ. (1985) The effect of fluorine on viscosities of melts in the system Na2O–Al2O3–SiO2: implications for phonolites, trachytes and rhyolites. American Mineralogist 70:80–87.[Abstract]
Dingwell DB, Knoche R, Webb SL. (1993) The effect of F on the density of haplogranite melt. American Mineralogist 78:325–330.[Abstract]
Dingwell DB, Holtz F, Behrens H. (1997) The solubility of H2O in peralkaline and peraluminous granitic melts. American Mineralogist 82:434–437.[Abstract]
Dolej D. (2006) Quartz solubility in fluorine-bearing aqueous fluids. Annual Report, Bayerisches Forschungsinstitut für experimentelle Geochemie und Geophysik2005Universität Bayreuth pp. 84–86.
Dolej D and Baker DR. (2004) Thermodynamic analysis of the system Na2O–K2O–CaO–Al2O3–SiO2–H2O–F2O–1: stability of fluorine-bearing minerals in felsic igneous suites. Contributions to Mineralogy and Petrology 146:762–778.[CrossRef][Web of Science]
Dolej D and Baker DR. (2006) Fluorite solubility in hydrous haplogranitic melts at 100 MPa. Chemical Geology 225:40–60.[CrossRef][Web of Science]
Dolej D and Baker DR. (2007) Liquidus equilibria in the system K2O–Na2O–Al2O3–SiO2–F2O–1–H2O to 100 MPa: I. Silicate–fluoride liquid immiscibility in anhydrous systems. Journal of Petrology 48:785–806.
Duffy JA. (1989) A common optical basicity scale for oxide and fluoride glasses. Journal of Non-Crystalline Solids 109:35–39.
Eadington PJ and Nashar B. (1978) Evidence for the magmatic origin of quartz–topaz rocks from the New England batholith, Australia. Contributions to Mineralogy and Petrology 67:433–438.[CrossRef][Web of Science]
Eugster HP and Baumgartner L. Carmichael ISE and Eugster HP. (1987) Mineral solubilities and speciation in supercritical metamorphic fluids. Thermodynamic Modeling of Geological Materials: Minerals, Fluids and Melts. Mineralogical Society of America, Reviews in Mineralogy 17:367–404.
Förster H-J, Tischendorf G, Trumbull RB, Gottesmann B. (1999) Late-collisional granites in the Variscan Erzgebirge, Germany. Journal of Petrology 40:1613–1645.[CrossRef][Web of Science]
Franck EU and Spalthoff W. (1957) Hydrogen fluoride. I. Specific heat, vapor pressure, and density up to 300° and 300 atmospheres. Zeitschrift für Elektrochemie 61:348–357.
Frank MR, Candela PA, Piccoli PM. (2003) Alkali exchange equilibria between a silicate melt and coexisting magmatic volatile phase: an experimental study at 800°C and 100 MPa. Geochimica et Cosmochimica Acta 67:1415–1427.[CrossRef][Web of Science]
Giordano D, Romano C, Dingwell DB, Poe B, Behrens H. (2004) The combined effects of water and fluorine on the viscosity of silicic magmas. Geochimica et Cosmochimica Acta 68:5159–5168.[CrossRef][Web of Science]
Greig JW, Jensen E, Merwin HE. (1955) The system Cu–Fe–S. Carnegie Institution of Washington Yearbook 54:129–134.
Haselton HT Jr, Cygan GL, d'Angelo WM. (1988) Chemistry of aqueous solutions coexisting with fluoride buffers in the system K2O–Al2O3–SiO2–H2O–F2O–1 (1 kbar, 400°–700°C). Economic Geology 83:163–173.
Hervig RL, Kortemeier WT, Burt DM. (1987) Ion-microprobe analyses of Li and B in topaz from different environments. American Mineralogist 72:392–396.[Abstract]
Hogan JP and Gilbert MC. (1995) The A-type Mount Scott granite sheet: importance of crustal magma traps. Journal of Geophysical Research 100:15779–15792.
Holtz F, Johannes W, Pichavant M. (1992) Effect of excess aluminium on phase relations in the system Qz–Ab–Or: experimental investigation at 2 kbar and reduced H2O-activity. European Journal of Mineralogy 4:137–152.
Holtz F, Dingwell DB, Behrens H. (1993) Effects of F, B2O3 and P2O5 on the solubility of water in haplogranite melts compared to natural silicate melts. Contributions to Mineralogy and Petrology 113:492–501.[CrossRef][Web of Science]
Holtz F, Johannes W, Tamic N, Behrens H. (2001) Maximum and minimum water contents of granitic melts generated in the crust: a reevaluation and implications. Lithos 56:1–14.[CrossRef][Web of Science]
Icenhower JP and London D. (1995) An experimental study of element partitioning among biotite, muscovite, and coexisting peraluminous silicic melt at 200 MPa (H2O). American Mineralogist 80:1229–1251.[Abstract]
Jänecke E. (1906) Über eine neue Darstellungsform der wässerigen Lösungen zweier und dreier gleichioniger Salze, reziproker Salzpaare und der van't Hoffschen Untersuchungen über ozeanische Salzablagerungen. Zeitschrift für Anorganische Chemie 51:132–157.
Johnston C and Chappell BW. (1992) Topaz-bearing rocks from Mount Gibson, North Queensland, Australia. American Mineralogist 77:303–313.[Abstract]
Joyce DB and Voigt DE. (1994) A phase equilibrium study in the system KAlSi3O8–NaAlSi3O8–SiO2–Al2SiO5–H2O and petrogenetic implications. American Mineralogist 79:504–512.[Abstract]
Kennedy GC, Wasserburg GJ, Heard HC, Newton RC. (1962) The upper three-phase region in the system SiO2–H2O. American Journal of Science 260:501–521.
Kleeman JD. (1985) Origin of disseminated wolframite-bearing quartz–topaz rock at Torrington, New South Wales, Australia. High-Heat Production (HHP) Granites, Hydrothermal Circulation and Ore Genesis(Institution of Mining and Metallurgy, London) pp. 197–201.
Knoche R, Dingwell DB, Webb SL. (1995) Melt densities for leucogranites and granitic pegmatites: partial molar volumes for SiO2, Al2O3, Na2O, K2O, Li2O, Rb2O, Cs2O, MgO, CaO, SrO, BaO, B2O3, P2O5, F2O–1, TiO2, Nb2O5, Ta2O5, and WO3. Geochimica et Cosmochimica Acta 59:4645–4652.[CrossRef][Web of Science]
Kogarko LN and Krigman LD. (1981) Fluorine in Silicate Melts and Magmas(Nauka, Moscow) pp. 124 (inRussian).
Kortemeier WT and Burt DM. (1988) Ongonite and topazite dikes in the flying W ranch area, Tonto basin, Arizona. American Mineralogist 73:507–523.[Abstract]
Korzhinskiy DS. (1959) Acid–base interaction of the components in silicate melts and the direction of cotectic lines. Doklady Akademii Nauk SSSR, Earth Science Sections 128:821–823.
Korzhinskiy DS. (1960) Acidity–alkalinity in magmatic processes. International Geological Congress, Report of the 21st sessionCopenhagen pp. 160–170.
Koster van Groos AF and Wyllie PJ. (1968) Melting relationships in the system NaAlSi3O8–NaF–H2O to 4 kilobars pressure. Journal of Geology 76:50–70.[Web of Science]
Kotelnikova ZA and Kotelnikov AR. (2002) Synthetic NaF-bearing fluid inclusions. Geochemistry International 40:594–600.
Kovalenko VI and Kovalenko NI. (1976) Ongonites: Subvolcanic Analogues of Rare-Metal Li–F Granites(Nauka, Moscow) pp. 127 (in Russian).
London D. (1986) Magmatic–hydrothermal transition in the Tanco rare-element pegmatite: evidence from fluid inclusions and phase-equilibrium experiments. American Mineralogist 71:376–395.[Abstract]
London D. (1987) Internal differentiation of rare-element pegmatites: effects of boron, phosphorus, and fluorine. Geochimica et Cosmochimica Acta 51:403–420.[CrossRef][Web of Science]
London D. (1997) Estimating abundances of volatile and other mobile components in evolved silicic melts through mineral–melt equilibria. Journal of Petrology 38:1691–1706.[CrossRef][Web of Science]
London D, Morgan GB VI, Wolf MB. (2001) Amblygonite–montebrasite solid solutions as monitors of fluorine in evolved granitic and pegmatitic melts. American Mineralogist 86:225–233.
Luth WC and Tuttle OF. (1969) The hydrous vapor phase in equilibrium with granite and granite magmas. In: Igneous and metamorphic geology - A volume in honor of Arie Poldervaart. Geological Society of America, Memoir 115:513–548.
Manning DAC. (1981) The effect of fluorine on liquidus phase relationships in the system Qz–Ab–Or with excess water at 1 kb. Contributions to Mineralogy and Petrology 76:206–215.[CrossRef][Web of Science]
Manning DAC, Hamilton DL, Henderson CMB, Dempsey MJ. (1980) The probable occurrence of interstitial Al in hydrous, F-bearing and F-free aluminosilicate melts. Contributions to Mineralogy and Petrology 75:257–262.
Marshall AS, Hinton RW, Macdonald R. (1998) Phenocrystic fluorite in peralkaline rhyolites, Olkaria, Kenya Rift Valley. Mineralogical Magazine 62:477–486.[Abstract]
Mootz D, Ohms U, Poll W. (1981) Schmelzdiagramm H2O–HF und Strukturen der 1:1- und einer 1:2-Phase. Zeitschrift für Anorganische und Allgemeine Chemie 479:75–83.
Munoz JL. (1971) Hydrothermal stability relations of synthetic lepidolite. American Mineralogist 56:2069–2087.[Web of Science]
Mustart DA. (1972) Phase relations in the peralkaline portion of the system Na2O–Al2O3–SiO2–H2O. Ph.D. thesis Stanford University187.
Mysen BO, Cody GD, Smith A. (2004) Solubility mechanisms of fluorine in peralkaline and meta-aluminous silicate glasses and in melts to magmatic temperatures. Geochimica et Cosmochimica Acta 68:2745–2769.[CrossRef][Web of Science]
Oglesby JV and Stebbins JF. (2000) 29Si CPMAS NMR investigations of silanol-group minerals and hydrous aluminosilicate glasses. American Mineralogist 85:722–731.
Ostrovsky IA. (1966) P–T diagram of the system SiO2–H2O. Geological Journal 5:127–134.
Paillat O, Elphick SC, Brown WL. (1992) The solubility of water in NaAlSi3O8 melts: a re-examination of Ab–H2O phase relationships and critical behavior at high pressures. Contributions to Mineralogy and Petrology 112:490–500.[CrossRef][Web of Science]
Pichavant M, Boher M, Stenger J-F, Aïssa M, Charoy B. (1987) Relations de phase des granites de Beauvoir à 1 et 3 kbar en conditions de saturation en H2O. Géologie de la France 2–3:77–86.
Price JD, Hogan JP, Gilbert MC, London D, Morgan GB VI. (1999) Experimental study of titanite–fluorite equilibria in the A-type Mount Scott granite: implications for assessing F contents of felsic magma. Geology 27:951–954.
Ravich MI and Valyashko VM. (1965) Solubility of sodium fluoride at elevated temperatures. Russian Journal of Inorganic Chemistry 10:107–109.
Rutlin JL. (1998) Chemical reactions and mineral formation during sodium aluminium fluoride attack on aluminosilicate and anorthite based refractories. Dr-Ing. thesis (Norwegian University of Science and Technology, Trondheim)167.
Scaillet B and Macdonald R. (2004) Fluorite stability in silicic magmas. Contributions to Mineralogy and Petrology 147:319–329.[CrossRef][Web of Science]
Schaller T, Dingwell DB, Keppler H, Knoeller W, Merwin L, Sebald A. (1992) Fluorine in silicate glasses: a multinuclear nuclear magnetic resonance study. Geochimica et Cosmochimica Acta 56:701–707.[CrossRef][Web of Science]
Schmidt BC, Riemer T, Kohn SC, Behrens H, Dupree R. (2000) Different water solubility mechanisms in hydrous glasses along the Qz–Ab join: evidence from NMR spectroscopy. Geochimica et Cosmochimica Acta 64:513–526.[CrossRef][Web of Science]
Shand SJ. (1927) Eruptive Rocks: Their Genesis, Composition, Classification, and Their Relation to Ore-Deposits, with a Chapter on Meteorites(New York, Wiley) pp. 360.
Shannon RD. (1976) Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides. Acta Crystallographica A32:751–767.
Stalder R, Ulmer P, Thompson AB, Günther D. (2000) Experimental approach to constrain second critical end-points in fluid silicate systems: near-solidus fluids and melts in the system albite–H2O. American Mineralogist 85:68–77.
temprok M. (1991) Ongonite from Ongon Khairkhan, Mongolia. Mineralogy and Petrology 43:255–273.[CrossRef][Web of Science]
Stewart DB. (1978) Petrogenesis of lithium-rich pegmatites. American Mineralogist 63:970–980.[Abstract]
Stilling AB. (1998) Bulk composition of the Tanco pegmatite at Bernic Lake, Manitoba, Canada. M.Sc. thesis (University of Manitoba, Winnipeg)76.
Tagirov B and Schott J. (2001) Aluminum speciation in crustal fluids revisited. Geochimica et Cosmochimica Acta 65:3965–3992.[CrossRef][Web of Science]
Tagirov B, Schott J, Harrichourry J-C, Salvi S. (2002) Experimental study of aluminum speciation in fluoride-rich supercritical fluids. Geochimica et Cosmochimica Acta 66:2013–2024.[CrossRef][Web of Science]
Thomas R and Klemm W. (1997) Microthermometric study of silicate melt inclusions in Variscan granites from SE Germany: volatile contents and entrapment conditions. Journal of Petrology 38:1753–1765.[CrossRef][Web of Science]
Thomas R, Förster H-J, Rickers K, Webster JD. (2005) Formation of extremely F-rich hydrous melt fractions and hydrothermal fluids during differentiation of highly evolved tin–granite magmas: a melt/fluid-inclusion study. Contributions to Mineralogy and Petrology 148:582–601.[CrossRef][Web of Science]
Tuttle OF and Bowen NL. (1958) Origin of Granite in the Light of Experimental Studies in the System NaAlSi3O8–KAlSi3O8–SiO2–H2O(Geological Society of America, Memoir) 74: pp. 153.
Urusova MA and Ravich MI. (1966) Phase equilibria in the potassium fluoride–water system at elevated temperatures. Russian Journal of Inorganic Chemistry 11:353–357.
Veksler IV. (2004) Liquid immiscibility and its role at the magmatic–hydrothermal transition: a summary of experimental studies. Chemical Geology 210:7–31.[CrossRef][Web of Science]
Veksler IV, Dorfman AM, Kamenetsky M, Dulski P, Dingwell DB. (2005) Partitioning of lanthanides and Y between immiscible silicate and fluoride melts, fluorite and cryolite and the origin of the lanthanide tetrad effect in igneous rocks. Geochimica et Cosmochimica Acta 69:2847–2860.[CrossRef][Web of Science]
Webster JD. (1990) Partitioning of F between H2O and CO2 fluids and topaz rhyolite melt. Implications for mineralizing magmatic–hydrothermal fluids in F-rich granitic systems. Contributions to Mineralogy and Petrology 104:424–438.[CrossRef][Web of Science]
Webster JD and Rebbert CR. (1998) Experimental investigation of H2O and Cl– solubilities in F-enriched silicate liquids; implications for volatile saturation of topaz rhyolite magmas. Contributions to Mineralogy and Petrology 132:198–207.[CrossRef][Web of Science]
Webster JD, Holloway JR, Hervig RL. (1987) Phase equilibria of a Be, U and F-enriched vitrophyre from Spor Mountain, Utah. Geochimica et Cosmochimica Acta 51:389–402.[CrossRef][Web of Science]
Webster JD, Thomas R, Rhede D, Förster H-J, Seltmann R. (1997) Melt inclusions in quartz from an evolved peraluminous pegmatite: geochemical evidence for strong tin enrichment in fluorine-rich and phosphorus-rich residual liquids. Geochimica et Cosmochimica Acta 61:2589–2604.[CrossRef][Web of Science]
Webster JD, Thomas R, Förster H-J, Seltmann R, Tappen C. (2004) Geochemical evolution of halogen-enriched granite magmas and mineralizing fluids of the Zinnwald tin–tungsten mining district, Erzgebirge, Germany. Mineralium Deposita 39:452–472.[Web of Science]
Weidner JR and Martin RF. (1987) Phase equilibria of a fluorine-rich leucogranite from the St. Austell pluton, Cornwall. Geochimica et Cosmochimica Acta 51:1591–1597.[CrossRef][Web of Science]
Williams TJ, Candela PA, Piccoli PM. (1997) Hydrogen–alkali exchange between silicate melts and two-phase aqueous mixtures: an experimental investigation. Contributions to Mineralogy and Petrology 128:114–126.[CrossRef][Web of Science]
Williamson BJ, Stanley CJ, Wilkinson JJ. (1997) Implications from inclusions in topaz for greisenisation and mineralisation in the hensbarrow topaz granite, Cornwall, England. Contributions to Mineralogy and Petrology 127:119–128.[CrossRef][Web of Science]
Williamson BJ, Wilkinson JJ, Luckham PF, Stanley CJ. (2002) Formation of coagulated colloidal silica in high-temperature mineralizing fluids. Mineralogical Magazine 64:547–553.
Wyllie PJ. (1979) Magmas and volatile components. American Mineralogist 64:469–500.[Abstract]
Wyllie PJ and Ryabchikov ID. (2000) Volatile components, magmas, and critical fluids in upwelling mantle. Journal of Petrology 41:1195–1206.
Wyllie PJ and Tuttle OF. (1961) Experimental investigation of silicate systems containing two volatile components. Part II. The effects of NH3 and HF, in addition to H2O on the melting temperatures of albite and granite. American Journal of Science 259:128–143.[Web of Science]
Xiong X-L, Rao B, Chen F-R, Zhu J-C, Zhao Z-H. (2002) Crystallization and melting experiments of a fluorine-rich leucogranite from the Xianghualing Pluton, South China, at 150 MPa and H2O-saturated conditions. Journal of the Asian Earth Sciences 21:175–188.
Zeng Q and Stebbins JF. (2000) Fluoride sites in aluminosilicate glasses: high-resolution 19F NMR results. American Mineralogist 85:863–867.
Zhu J-C and Liu W. Maurice YT. (1990) Topazite–ongonite relationships and its bearing on vertical zonation in rare-metal granites: evidence from Xianghualing district, Hunan Province, China. Proceedings of the Eighth Quadrennial IAGOD SymposiumStuttgartSchweitzerbart pp. 303–313.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||