Anthony E. Williams-Jones

The role of hydrothermal processes in concentrating high-field strength elements in the Strange Lake peralkaline complex, northeastern Canada

Abstract The middle-Proterozoic peralkaline pluton at Strange Lake, Quebec/Labrador, comprises hypersolvus to subsolvus phases which are unusually enriched in Zr, Y, REEs, Nb, Be, and F, as exotic alkali and alkaline-earth silicate minerals. The highest concentrations of these elements are in subsolvus granite, which underwent intense low temperature (≤200°C) hydrothermal alteration involving hematization and the replacement of alkali high-field strength element (HFSE) minerals by calcic equivalents. This alteration is interpreted to have been caused by meteoric or formational waters. High temperature (≥ 350°C) alteration, attributed to orthomagmatic fluids, is evident in other parts of the subsolvus granite by the replacement of arfvedsonite by aegirine. Comparisons of the chemical compositions of fresh and altered rocks indicate that rocks subjected to high temperature alteration were chemically unaffected, except for depletion in Zr, Y, and HREEs. These elements were appreciably enriched in rocks that underwent low temperature alteration. Other elements affected by low temperature alteration include Ca and Mg, which were added and Na, which was removed. Available data on HFSE speciation in aqueous fluids and the chemistry of the pluton, suggest that the HFSEs were transported as fluoride complexes. If this was the case, the low temperature fluid could not have been responsible for HFSE transport, because the high concentration of Ca and low solubility of fluorite would have buffered F − activity to levels too low to permit significant complexation. We propose that HFSE mineralization and accompanying alteration were the result of mixing, in the apical parts of the pluton, of a F-rich, essentially Ca-free orthomagmatic fluid containing significant concentrations of HFSEs, with an externally derived meteoric-dominated fluid, enriched in Ca as a result of interaction with calc-silicate gneisses and gabbros. According to this interpretation, the latter fluid was responsible for the exchange of Ca for alkalis, mainly Na, in HFSE-rich minerals and, by sharply reducing F − activity in the mixed fluid through fluorite precipitation and/or increasing pH, destabilised the HFSE-fluoride complexes, causing deposition of HFSE-bearing minerals. An important implication of this study is that major HFSE enrichment may be restricted to those rare cases where F-rich, Ca-free, metal leaching environments and Ca-rich depositional environments are juxtaposed.

The role of hydrothermal processes in the granite-hosted Zr, Y, REE deposit at Strange Lake, Quebec/Labrador: Evidence from fluid inclusions

Abstract The Strange Lake Zr, Y, REE, Nb, and Be deposit is hosted by a small, high-level, Late-Proterozoic peralkaline granite stock that intruded into high-grade metamorphic gneisses on the Quebec-Labrador border. The stock is extensively altered. Early alteration is manifested by the replacement of arfvedsonite with aegirine. Later alteration involved Ca-Na exchange. Zr, Ti, Y, REEs, Nb, and Be are concentrated in Ca-bearing minerals that, together with quartz, commonly pseudomorph Na-bearing minerals. Fluid inclusions in pseudomorphs comprise several distinct types: high-salinity (13 to 24 wt% NaCl eq.), Ca-rich aqueous inclusions that homogenize to liquid between 135 and 195°C; mixed aqueousmethane inclusions; methane inclusions; and solid-bearing inclusions. Aqueous-methane inclusions represent heterogeneous entrapment of immiscible high-salinity aqueous liquid and methane. Bastnasite (tentatively identified by SEM analysis) occurs as a daughter mineral. Other daughter or trapped minerals include a Y, HREE-bearing mineral, possibly gagarinite, and hematite, galena, sphalerite, fluorite, pyrochlore, kutnahorite (?), and griceite (?). The first three inclusion types also occur in quartz in pegmatites and veins together with lower-temperature, lower-salinity, Na-dominated aqueous inclusions. The entrapment temperature inferred for the aqueous inclusions from microthermometry and the Na-K-Ca geothermometer range from 155 to 195°C for the higher-salinity inclusions and 100 to 165°C for the low-salinity inclusions. A model is proposed in which the intrusion of a peralkaline granite to high crustal levels initiated a ground/formational water-dominated hydrothermal system in adjacent gabbroic, calc-silicate, and graphitic gneisses. Reaction of the high-salinity, Ca-rich liquid with the graphitic gneisses led to the production of an immiscible methane gas. Subsequent interaction of this liquid with the granite led to extensive replacement of sodic minerals by calcium analogues at temperatures of less than 200°C. Some time after the onset of Ca metasomatism the high-salinity liquid mixed with a Ca-poor, low-salinity, low-temperature liquid that had leached F and rare metals from the granite. Yttrium and REE mineral deposition occurred as a result of the decreased ligand concentration that accompanied fluorite deposition during mixing of the Ca-rich and Ca-poor aqueous liquids.

The disproportionation of gold(I) chloride complexes at 25 to 200°C

Abstract The disproportionation of aqueous Au(I) chloride complexes at elevated temperature has been investigated experimentally using the solubility method. At 300°C, the dominant gold species in aqueous HCl solutions is AuCl2−. Upon cooling, this aurous complex partially decomposes according to the following reaction: 3 AuC 1 2 − = 2 Au ( s ) + AuC 1 4 − + 2 C 1 − (A1) Log KA1 values were obtained at 100°C (4.42 ± 0.22), 150°C (2.86 ± 0.12), and 200°C (1.45 ± 0.19). The results are in excellent agreement with the earlier potentiometric study of Nikolaeva et al. (1972) at 25–80°C. The combined data were used to obtain the following polynomial: log KA1 = −13.55 + 8593/T−700610/T2 (T =Kelvin, valid from 25 to 200°C). The rate of reaction (A1) at 25°C was investigated by monitoring the production of AuCl4− after quench using a UV spectrophotometer. The rates were very slow for the first 5–10 min, but then rapidly increased to values that remained approximately constant with further reaction progress. The measured reaction rates fell in the range 2.1 · 10−8 to 3.7 · 10−6 moles AuCl4− · kg H2O−1 · minute−1. In general, faster rates were obtained for samples with high initial AuCl2− concentrations. Addition of gold foils caused an abrupt increase in rate, indicating that the reaction is catalyzed by the native metal. Gold crystals formed during the disproportionation reaction at 25°C show a variety of morphologies, including examples with anomalous fivefold symmetry. Our results indicate that the stability of AuCl2− relative to AuCl4− increases quickly with temperature. At 25°C, AuCl4− is unlikely to be of geochemical importance, with the possible exception of oxidized, acidic solutions that are also rich in chloride ion. In contrast, AuCl2− may be the dominant form of dissolved gold in brines with near-neutral pH (e.g., seawater), as well as hydrothermal fluids that are both saline and oxidized. Cooling or dilution of solutions saturated with AuCl2− could result in deposition of Au via a disproportionation reaction, as in our experiments.

The aqueous geochemistry of Zr and the solubility of some Zr-bearing minerals

Abstract Literature data on the thermodynamics of complexation of Zr with inorganic species, at 25°C, have been critically reviewed. The preponderance of published complexation constants deal with F− and OH− ions. Stability constants for the complexation reactions are relatively independent of ionic strength and thus recomended values for each ligand type are averages of the most reliable data. Complexation constants under elevated conditions (T ⪕ 250°C andPv = PH2O) have been predicted for various Zr complexes (F−, Cl−, SO42− and OH−) using Helgesons electrostatic approach. Predominance diagrams (calculated for simple systems with these constants) suggest that, over a wide range of pH conditions, Zr(OH)4(aq) will dominate the aqueous geochemistry of Zr except under very high activities of competing ligands (e.g., F−, SO42−). The solubilities of vlasovite [Na2ZrSi4O11] and weloganite [Sr3Na2Zr(CO3)6·3H2O have been measured in KCI solutions (0.5–1.0 M) at 50°C. Weloganite dissolution is complicated by the predictable precipitation of strontianite (SrCO3) whereas vlasovite dissolves incongruently. Solubility products for the dissolution of welonganite and vlasovite are determined to be −28.96±0.14 and −20.40±1.18, respectively. Concentrations of Zr up to 10−3 m were present in the experimental solutions; the presence of large amounts of Zr in aqueous solutions support the possibility of extensive remobilization of Zr during hydrothermal mineralization.

The aqueous geochemistry of the rare earth elements and yttrium: VI. Stability of neodymium chloride complexes from 25 to 300°C

Abstract The stability and stoichiometry of Nd (III) chloride complexes have been experimentally determined in the temperature range 40 to 300°C, P = Psat. The solubility of AgCl (s) was measured in solutions of fixed HC1 + NaC1 concentration (0.01 to 5.0 m) and varying ΣNd/ΣCl molar ratio (0.0 to 0.5), following the method of Gammons (1995). The results of over 250 individual solubility experiments were regressed to obtain the following smoothed values for the first and second cumulative association constants for the Nd(III) chloride complexes: Nd3+ + Cl− = NdC12+ (β1); Nd3+ + 2Cl− = NdCl2+ (β2):25°C50°C100°C150°C200°C250°C300°log β10.060.210.661.312.173.224.48±.50±.30±.20±.20±.15±.30±.50log β2——0.131.082.524.456.87±.50±.30±.15±.50c]±.50These are the first experimentally determined equilibrium constants for chloride complexes of any rare earth element (REE) at elevated temperature. At 25°C, neodymium exists mainly as Nd3+ in the absence of high concentrations of Cl− and other ligands (F−, CO3−, SO4−). However, complexation with chloride is greatly enhanced by increase in temperature, such that NdC12+, NdC12+, and possibly NdCl30 become the dominant species for NaCl HCl H2O brines at 300°C. The experimental data indicate a higher degree of complexation than predicted from earlier theoretical studies (Wood, 1990b; Haas et al., 1995), particularly in the case of log β2. Calculations of monazite solubility in seafloor hydrothermal systems (Wood and Williams-Jones, 1994) are re-evaluated in light of our new experimental data. Chloride complexes are shown to dominate the aqueous Nd socciation at 300°C, and lead to solubilities that are (1) much higher than previously estimated and (2) much closer to the maximum concentrations that have been reported from active black smokers. However, the large fluxes of REEs in altered rock beneath ancient massive sulfide deposits are still difficult to explain assuming that modern seafloor hydrothermal systems are direct analogs to ore-forming processes. Significant differences in fluid chemistry (e.g., lower pH, higher Cl− or F− concentrations) and/or duration and intensity of hydrothermal activity (higher water/rock ratio) are required to explain the REE systematics in ancient volcanogenic massive sulfide deposits.

An experimental study of the stability of copper chloride complexes in water vapor at elevated temperatures and pressures

Abstract The solubility of copper chloride in liquid-undersaturated HCl-bearing water vapor was investigated experimentally at temperatures of 280 to 320°C and pressures up to 103 bars. Results of these experiments show that the solubility of copper in the vapor phase is significant and increases with increasing fH2O, but is retrograde with respect to temperature. This solubility is attributed to the formation of hydrated copper-chloride gas species, interpreted to have a copper-chlorine ratio of 1:1 (e.g., CuCl, Cu3Cl3, etc.) and a hydration number varying from 7.6 at 320°C, to 6.0 at 300°C, and 6.1 at 280°C. Complex formation is proposed to have occurred through the reaction: A1 3 CuCl solid +nH 2 O gas ⇋ Cu 3 Cl 3 ·(H 2 O) n gas Log K values determined for this reaction are −21.46 ± 0.05 at 280°C (n = 7.6), −19.03 ± 0.10 at 300°C (n = 6.0), and −19.45 ± 0.12 at 320°C (n = 6.1), if it is assumed that the vapor species is the trimer, Cu3Cl3(H2O)6–8. Calculations based on the above data indicate that at 300°C and HCl fluxes encountered in passively degassing volcanic systems, the vapor phase could transport copper in concentrations as high as 280 ppm. Theoretically, this vapor could form an economic copper deposit (e.g., 50 million tonnes of 0.5% Cu) in as little as ∼20,500 yr.

Fischer-Tropsch synthesis of hydrocarbons during sub-solidus alteration of the Strange Lake peralkaline granite, Quebec/Labrador, Canada

Abstract The composition of the carbonic phase(s) of fluid inclusions in pegmatite quartz from the Strange Lake peralkaline complex has been analysed by gas chromatography using online extraction of inclusion contents and a PoraPLOT® Q capillary column. The measured gas species are, in order of abundance, CH 4 , H 2 , C 2 H 6 , CO 2 , N 2 , CA, n -C 4 H 10 , n -C 5 H 12 , C 2 H 2 , i -C 4 H 10 and C 2 H 4 Minor amounts of i -C 5 H 12 , n -C 6 H 14 , i -C 6 H 14 , and neo-C 6 H 14 were also detected (but not quantified) in some samples. A suite of quartz samples from Ca-metasomatised pegmatites contains fluid inclusions with a similar distribution of hydrocarbons but much higher proportions of C0 2 . The carbonic fluid coexisted immiscibly with a brine ( Salvi and Williams-Jones, 1992 ), which on the basis of field and petrographic evidence, was interpreted to have originated from the magma. However, thermodynamic calculations indicate that the above gas species, specifically the hydrocarbons, could not have coexisted at equilibrium in the proportions measured, at any geologically reasonable conditions either prior to or post entrapment. We propose, instead, that the gas compositions measured in the Strange Lake inclusions, and in inclusions from other alkalic complexes, resulted from the production of H 2 during the alteration of arfvedsonite to aegirine, and the subsequent reaction of this H 2 with orthomagmatic C0 2 and CO to form hydrocarbons in a magnetite-catalysed Fischer-Tropsch synthesis. Locally, influx of an oxidised calcic brine, derived externally from the pluton, altered the original composition of the fluid by converting hydrocarbons to C0 2 .

Hydrothermal transport and deposition of the rare earth elements by fluorine-bearing aqueous liquids

New technologies, particularly those designed to address environmental concerns, have created a great demand for the rare earth elements (REE), and focused considerable attention on the processes by which they are concentrated to economically exploitable levels in the Earth’s crust. There is widespread agreement that hydrothermal fluids played an important role in the formation of the world’s largest economic REE deposit, i.e. Bayan Obo, China. Until recently, many researchers have assumed that hydrothermal transport of the REE in fluorine-bearing ore-forming systems occurs mainly due to the formation of REE-fluoride complexes. Consequently, hydrothermal models for REE concentration have commonly involved depositional mechanisms based on saturation of the fluid with REE minerals due to destabilization of REE-fluoride complexes. Here, we demonstrate that these complexes are insignificant in REE transport, and that the above models are therefore flawed. The strong association of H+ and F− as HF° and low solubility of REE-F solids greatly limit transport of the REE as fluoride complexes. However, this limitation does not apply to REE-chloride complexes. Because of this, the high concentration of Cl− in the ore fluids, and the relatively high stability of REE-chloride complexes, the latter can transport appreciable concentrations of REE at low pH. The limitation also does not apply to sulphate complexes and in some fluids, the concentration of sulphate may be sufficient to transport significant concentrations of REE as sulphate complexes, particularly at weakly acidic pH. This article proposes new models for hydrothermal REE deposition based on the transport of the REE as chloride and sulphate complexes.

The aqueous geochemistry of the rare-earth elements and yttrium 4. Monazite solubility and REE mobility in exhalative massive sulfide-depositing environments

Abstract Although there is considerable evidence that rare-earth elements (REE), and particularly the light REE (LREE), have been mobile in the alteration zones below many massive sulfide deposits, there is some disagreement over the cause of this mobility. Most researchers have argued or implied that the REE were mobilized by analogues of modern seafloor hydrothermal vent fluids. However, the few measurements that have been made of the REE concentrations of these modern fluids suggest that their ability to mobilize REE is negligible. In order to shed further light on this problem we have carried out calculations of monazite solubility (monazite is thought to be the principal host for the LREE) in a model vent fluid at temperatures from 200° to 300°C. These calculations show that at 300°C the model fluid will contain 0.08 ppb Ce, 0.07 ppb La and 0.03 ppb Nd. At 200°C the concentrations increase to 0.54 ppb, 0.37 ppb and 0.15 ppb, respectively. A decrease in pH of one unit at 300°C also increases REE solubility significantly (7.4 ppb Ce, 5.7 ppb La and 1.1 ppb Nd). These solubilities are similar to the concentrations of REE measured in seafloor hydrothermal vent fluids. However, they are considerably lower than required to account for the scale of REE mobility in alteration zones associated with massive sulfide deposits. If the mass ratio of typical hydrothermal vent fluid to rock reached 1000, concentrations of REE in the altered rock the predicted to be in the range

The stability of Au-chloride complexes in water vapor at elevated temperatures and pressures

Abstract The solubility of gold in liquid-undersaturated HCl-bearing water vapor was investigated experimentally at temperatures of 300 to 360°C and pressures up to 144 bars. Results of these experiments show that the solubility of gold in the vapor phase is significant and increases with increasing fHCl and fH2O. This behavior of gold is attributed to formation of hydrated gold-chloride gas species, interpreted to have a gold-chlorine ratio of 1:1 and a hydration number varying from 5 at 300°C to 3 at 360°C. These complexes are proposed to have formed through the following reaction: (A1) Au solid + m · HCl gas + n · H 2 O gas = AuCl m ·( H 2 O ) n gas + m 2 · H 2 gas which was determined to have log K values of −17.28 ± 0.36 at 300°C, −18.73 ± 0.66 at 340°C, and −18.74 ± 0.43 at 360°C. Gold solubility in the vapor was retrograde, i.e., it decreased with increasing temperature, possibly as a result of the inferred decrease in hydration number. Calculations based on our data indicate that at 300°C and fO2-pH conditions, encountered in high sulfidation epithermal systems, the vapor phase can transport up to 6.6 ppb gold, which would be sufficient to form an economic deposit (e.g., Nansatsu, Japan; 36 tonnes) in ∼ 30,000 yr.