Andri Stefánsson

Dissolution of primary minerals of basalt in natural waters: I. Calculation of mineral solubilities from 0°C to 350°C

Abstract The solubilities of forsterite, fayalite, enstatite, ferrosilite, hedenbergite, diopside, anorthite, high-albite, magnetite, hematite, ulvospinel, ilmenite, F-apatite and OH-apatite and olivine, plagioclase, orthopyroxene, clinopyroxene and Fe–Ti oxide solid solutions of fixed composition were calculated in the temperature range 0–350°C at saturated water vapour pressure. The thermodynamic database used for end-member minerals was that of Robie and Hemingway [Robie, R.A., Hemingway, B.S., 1995. Thermodynamic properties of minerals and related substances at 298.15 K and 1 bar (105 Pascals) pressures and at higher temperatures. U.S. Geol. Surv. Bull. 2131, 461 pp.] except for plagioclases [Arnorsson, S., Stefansson, A., 1999. Assessment of feldspar solubility constants in water in the range of 0° to 350°C at vapor saturation pressures. Am. J. Sci. 299, 173–209.] and that of Shock and Helgeson [Shock, E.L., Helgeson, H.C., 1988. Calculation of the thermodynamic and transport properties of aqueous species at high pressures and temperatures: correlation algorithms for ionic species and equation-of-state predictions to 5 kb and 1000°C. Geochim. Cosmochim. Acta 53, 2009–2036.] and Shock et al. [Shock, E.L., Oelkers, E.H., Johnson, J.W., Sverjensky, D.A., Helgeson, H.C., 1992. Calculation of the thermodynamic properties of aqueous species at high pressures and temperatures: effective electrostatic radii, dissociation constants, and standard partial molal properties to 1000°C and 5 kbar. J. Chem. Soc., Faraday Trans. 88, 803–826.] for most aqueous species. For aqueous Fe(OH) 4 − and Al(OH) 4 − , the thermodynamic properties reported by Diakonov et al. and Pokrovskii and Helgeson [Pokrovskii, V.A., Helgeson, H.C., 1995. Thermodynamic properties of aqueous species and the solubilities of minerals at high pressures and temperatures: the system Al 2 O 3 –H 2 O–NaCl. Am. J. Sci. 295, 1255–1342], respectively, were used. In the present study, the standard partial molal properties and the HKF equation-of-state parameters for aqueous H 4 SiO 4 0 were revised to better describe the recent experimental results at low temperatures. For H 4 SiO 4 0 , the new parameters are also consistent with quartz solubility experiments up to 900°C and 5 kbar. Further, the HKF equation-of-state parameters for aqueous Ti(OH) 4 0 were estimated from rutile solubility [Ziemniak, S.E., Jones, M.E., Combs, K.E.S., 1993. Solubility behaviour of titanium (IV) oxide in alkaline media at elevated temperatures. J. Sol. Chem. 22, 601–623.], which enables the calculations of the solubility of Ti-bearing minerals at elevated temperatures and pressures. Much higher solubilities were found for the silicate minerals below 100°C than previously reported, which is related to higher quartz solubility at low temperature and correspondingly new data on the thermodynamic properties of H 4 SiO 4 0 . The present results are particularly important for the stabilities of primary basaltic minerals of natural composition under weathering conditions. They are also of importance for the study of equilibrium/dis-equilibrium conditions in active geothermal systems.

Feldspar saturation state in natural waters

The saturation state of feldspar minerals in natural waters ranging from 0°C to over 300°C was studied. Waters above 200°C have closely approached equilibrium with microcline and low-albite. This is consistent with the occurrence of these minerals as hydrothermal minerals in active geothermal systems with temperature in excess of some 200°C. The Na+/K+ activity ratio of geothermal waters with temperature as low as 50°C closely approaches that predicted from thermodynamic data for the reaction low-albite + K+ = microcline + Na+, suggesting that Na and K ion activities in these geothermal waters are controlled by simultaneous equilibrium with these two feldspars. Geothermal waters are undersaturated with primary disordered plagioclases and alkali-feldspars of compositions typically found in volcanic rocks. Accordingly these feldspars tend to dissolve in such waters simultaneously with precipitation of ordered alkali-feldspars. Surface- and non-thermal groundwaters are usually undersaturated with respect to igneous and metamorphic feldspars regardless of composition and Al-Si ordering and tend, therefore, to dissolve under weathering conditions.

Dissolution of primary minerals in natural waters: II. Mineral saturation state

Abstract The saturation state of olivine, plagioclase, clinopyroxene, orthopyroxene, Fe–Ti oxides and apatite of variable composition has been assessed in natural waters in Iceland, with temperature ranging from 0°C to 300°C and in situ pH from below 5 to above 10. Cold waters are undersaturated with respect to olivine, orthopyroxene, clinopyroxene and plagioclases indicating that all these minerals tend to dissolve under weathering conditions. With increasing pH and temperature, the waters approach saturation with pyroxene, olivine, and plagioclases. Also, the degree of undersaturation of olivine and orthopyroxene decreases with increasing Fe content of the minerals and Fe-rich olivine and orthopyroxene are stable between 50°C and 150°C, whereas Mg-rich ones tend to dissolve. Natural waters in Iceland are saturated with respect to pure albite when above 50°C. They are, on the other hand, undersaturated with Ca-rich plagioclase up to 250°C where the waters reach saturation. Pure magnetite and hematite are stable at all temperatures. With increasing titanium content, the minerals become unstable, and the waters are undersaturated with respect to pure ulvospinel at temperatures up to 300°C. Pure ilmenite is, however, close to saturation under weathering conditions but undersaturated above 200°C. F-apatite is close to saturation at all temperatures. On the other hand, undersaturation with respect to OH-apatite is observed at all temperatures. The weathering susceptibilities of primary minerals of basalt in Icelandic waters in increasing order are Mg-olivine>Fe-olivine, Ti-rich magnetite>Ca-plagioclase, Mg-orthopyroxene>Fe-orthopyroxene, clinopyroxene>Na-plagioclase, F-apatite>Ti-rich ilmenite≫Ti-poor magnetite, Ti-poor hematite.

Stability of chloridogold(I) complexes in aqueous solutions from 300 to 600°C and from 500 to 1800 bar

Abstract The solubility of gold has been measured in the system H2O+H2+HCl+NaCl+NaOH at temperatures from 300 to 600°C and pressures from 500 to 1800 bar in order to determine the stability and stoichiometry of chloride complexes of gold(I) in hydrothermal solutions. The experiments were carried out in a flow-through autoclave system. This approach permitted the independent determination of the concentrations of all critical aqueous components in solution for the determination of the stability and stoichiometry of gold(I) complexes. The solubilities (i.e. total dissolved gold) were in the range 9.9 × 10−9 to 3.26 × 10−5 mol kg−1 (0.002–6.42 mg kg−1) in solutions of total dissolved chloride between 0.150 and 1.720 mol kg−1, total dissolved sodium between 0.000 and 0.975 mol kg−1 and total dissolved hydrogen between 4.34 × 10−6 and 7.87 × 10−4 mol kg−1. A nonlinear least squares treatment of the data demonstrates that the solubility of gold in chloride solutions is accurately described by the reactions, Au(s) + 2Cl − + H + = AuCl 2 − + 0.5 H 2 ( g ) K s ,020 Au(s) + H 2 O = AuOH(aq) + 0.5 H 2 ( g ) K s ,001 where AuCl2− predominates in acidic chloride solutions and AuOH(aq) in neutral to alkaline chloride and chloride-free solutions. The solubility constant, logKs,020, increases with increasing temperature and decreases with increasing pressure from a minimum of −5.43 (±0.29) at 300°C and 500 bar to a maximum of −0.15 (±0.16) at 600°C and 1000 bar, with the pressure effects becoming more important with increasing temperature. The equilibrium solubility constant for AuOH(aq) has been previously determined by Stefansson and Seward (2003) . The solubility of gold at pH >5 was found to be independent of chloride concentration up to 1 mol kg−1 and identical to the solubility of gold with respect to AuOH(aq). The stability of AuClOH− was estimated to be 3 to 6 orders of magnitude less stable than AuOH(aq) and AuCl2− in hydrothermal solutions. Hence, gold(I) chloride complexes play an important role in transporting gold in aqueous acidic chloride solutions above 400°C.

Gas pressures and redox reactions in geothermal fluids in Iceland

Abstract The gas and redox chemistry of 100–300 °C geothermal fluids in Iceland has been studied as a function of fluid temperature and fluid composition. The partial pressures of CO 2 in dilute ( m Cl m Cl>500 ppm) geothermal fluids above 200 °C are controlled by the mineral buffer clinozoisite+prehnite+calcite+quartz. Two buffers are considered to control the H 2 S and H 2 partial pressures above 200 °C depending on fluid salinity, epidote+prehnite+pyrite+pyrrhotite for dilute fluids and pyrite+prehnite+quartz+magnetite+anhydrite+clinozoisite+quartz for saline fluids. Below 200 °C, the partial pressures of CO 2 , H 2 S and H 2 also seem to be buffered but other minerals must be involved. Zeolites are expected to replace prehnite and epidote. Redox potential calculated on the assumption of equilibrium for the H + /H 2 redox couple decreases in dilute geothermal fluids with increasing temperature from about −0.5 V at 100 °C to −0.8 V at 300 °C, whereas saline geothermal fluids at 250 °C display a redox potential of about −0.45 V. A systematic discrepancy between redox couples of about 0.05–0.09 V is observed in the redox potential for the dilute geothermal fluids, whereas redox potentials agree within 0.02–0.04 V for saline geothermal waters. The discrepancies in the calculated redox potential for dilute geothermal fluids are thought to be due to a general lack of equilibrium between CH 4 , CO 2 and H 2 and between H 2 S, SO 4 and H 2 . It is, accordingly, concluded that an overall equilibrium among redox species has not been reached for dilute geothermal fluids whereas it appears to be more closely approached for the saline geothermal fluids. The latter conclusion is based on limited database and should be treated with care. Since the various redox components are not in an overall equilibrium in geothermal fluids in Iceland these fluids cannot be characterised by a unique hydrogen fugacity, oxygen fugacity or redox potential at a given temperature and pressure.

The hydrolysis of gold(I) in aqueous solutions to 600°c and 1500 bar

Abstract The solubility of gold has been measured in aqueous solutions at temperatures between 300 and 600°C and pressures from 500 to 1500 bar to determine the stability and stoichiometry of the hydroxy complexes of gold(I) in hydrothermal solutions. The experiments were carried out using a flow-through autoclave system. The solubilities, measured as total dissolved gold, were in the range 1.2 × 10 −8 to 2.0 × 10 −6 mol kg −1 (0.002 to 0.40 mg kg −1 ), in solutions of total dissolved sodium between 0.0 and 0.5 mol kg −1 , and total dissolved hydrogen between 4.0 × 10 −6 and 4.0 × 10 −4 mol kg −1 . At constant hydrogen molality, the solubility of gold increases with increasing temperature and decreases with increasing pressure. The solubilities were found to be independent of pH but increased with decreasing hydrogen molality at constant temperature and pressure. Consequently, gold dissolves in aqueous solutions of acidic to alkaline pH according to the reactionAu(s)+H 2 O(l)=AuOH(aq)+0.5H 2 (g) K s,1 The solubility constant, log K s,1 , increases with increasing temperature from a minimum of −8.76 (±0.18) at 300°C and 500 bar to a maximum of −7.50 (±0.11) at 500°C and 1500 bar and decreases to −7.61 (±0.08) at 600°C and 1500 bar. From the equilibrium solubility constant and the redox potential of gold, the formation constant to form AuOH(aq) was calculated. At 25°C the complex formation is characterised by an exothermic enthalpy and a positive entropy. With increasing temperature and decreasing pressure, the formation reaction becomes endothermic and is accompanied by a large positive entropy, indicating a greater electrostatic interaction between Au + and OH − .

Major element chemistry of surface- and ground waters in basaltic terrain, N-Iceland.: I. primary mineral saturation

This contribution describes primary basalt mineral saturation in surface- and up to 90°C ground waters in a tholeiite flood basalt region in northern Iceland. It is based on data on 253 water samples and the mineralogical composition of the associated basalts. Surface waters are significantly under-saturated with plagioclase and olivine of the compositions occurring in the study area, saturation index (SI) values ranging from −1 to −10 and −5 to −20, respectively. With few exceptions these waters are also significantly under-saturated with pigeonite and augite of all compositions (SI = −1 to −7) and with ilmenite (SI = −0.5 to −6). The surface waters are generally over-saturated with respect to the titano-magnetite of the compositions occurring in the basalts of the study area, the range in SI being from −2 to +10. For crystalline OH-apatite, SI values in surface waters range from strong under-saturation (−10) to strong over-saturation (+5) but for crystalline F-apatite they lie in the range 0 to 15. Systematic under-saturation is, on the other hand, observed for “amorphous apatite,” i.e. an apatite of the kind Clark (1955) prepared by mixing Ca(OH)2 and H3PO4 solutions. Like surface waters, ground waters are under-saturated with plagioclase and olivine, its degree increasing with increasing Ca content of the plagioclase and increasing Fe content of the olivine, the SI values being −2 to −7 and 0 to −4 for the Ca-richest and Ca-poorest plagioclase, respectively, and about −3 to −18 and 0 to −15 for forsterite and fayalite, respectively. Ground waters are generally close to saturation with pigeonite and augite of all compositions. However, some non-thermal ground waters in highland areas are strongly under-saturated. Above 25°C the ground waters are ilmenite under-saturated but generally over-saturated at lower temperatures. These waters are titano-magnetite over-saturated at temperatures below 70°C, the SI values decreasing with increasing temperature from about 6 to 8 at 10°C to 0 at 70°C. The ground waters are highly over-saturated with both crystalline OH- and F-apatite, or by approximately 10 to 15 SI units but close to saturation with “amorphous apatite” containing about equal amounts of F and OH. The results presented here for the pyroxenes carry an unknown error because available thermodynamic data do not permit but a simple solid solution model for the calculation of their solubility. Published values on the dissociation constants for ferrous iron hydroxide complexes are very variable and those for ferric iron are limited. This casts an error of an unknown magnitude on the calculated SI values for all iron bearing minerals. This error may not be large for waters with a pH of less than 9 but it is apparently high for waters with a higher pH. Improved experimental data on the stability of ferrous and ferric hydrolysis constants are needed to improve the accuracy by which Fe-mineral saturation can be calculated in natural waters.

Experimental determination of the stability and stoichiometry of sulphide complexes of silver(I) in hydrothermal solutions to 400°C

Abstract The solubility of silver sulphide (acanthite/argentite) has been measured in aqueous sulphide solutions between 25 and 400°C at saturated water vapour pressure and 500 bar to determine the stability and stoichiometry of sulphide complexes of silver(I) in hydrothermal solutions. The experiments were carried out in a flow-through autoclave, connected to a high-performance liquid chromatographic pump, titanium sampling loop, and a back-pressure regulator on line. Samples for silver determination were collected via the titanium sampling loop at experimental temperatures and pressures. The solubilities, measured as total dissolved silver, were in the range 1.0 × 10−7 to 1.30 × 10−4 mol kg−1 (0.01 to 14.0 ppm), in solutions of total reduced sulphur between 0.007 and 0.176 mol kg−1 and pHT,p of 3.7 to 12.7. A nonlinear least squares treatment of the data demonstrates that the solubility of silver sulphide in aqueous sulphide solutions of acidic to alkaline pH is accurately described by the reactions0.5Ag2S(s) + 0.5H2S(aq) = AgHS(aq) Ks,1110.5Ag2S(s) + 0.5H2S(aq) + HS− = Ag(HS)2− Ks,122Ag2S(s) + 2HS− = Ag2S(HS)22− Ks,232where AgHS(aq) is the dominant species in acidic solutions, Ag(HS)2− under neutral pH conditions and Ag2S(HS)22− in alkaline solutions. With increasing temperature the stability field of Ag(HS)2− increases and shifts to more alkaline pH in accordance with the change in the first ionisation constant of H2S(aq). Consequently, Ag2S(HS)22− is not an important species above 200°C. The solubility constant for the first reaction is independent of temperature to 300°C, with values in the range logKs,111 = −5.79 (±0.07) to −5.59 (±0.09), and decreases to −5.92 (±0.16) at 400°C. The solubility constant for the second reaction increases almost linearly with inverse temperature from logKs,122 = −3.97 (±0.04) at 25°C to −1.89 (±0.03) at 400°C. The solubility constant for the third reaction increases with temperature from logKs,232 = −4.78 (±0.04) at 25°C to −4.57 (±0.18) at 200°C. All solubility constants were found to be independent of pressure within experimental uncertainties. The interaction between Ag+ and HS− at 25°C and 1 bar to form AgHS(aq) has appreciable covalent character, as reflected in the exothermic enthalpy and small entropy of formation. With increasing temperature, the stepwise formation reactions become progressively more endothermic and are accompanied by large positive entropies, indicating greater electrostatic interaction. The aqueous speciation of silver is very sensitive to fluid composition and temperature. Below 100°C silver(I) sulphide complexes predominate in reduced sulphide solutions, whereas Ag+ and AgClOH− are the dominant species in oxidised waters. In high-temperature hydrothermal solutions of seawater salinity, chloride complexes of silver(I) are most important, whereas in dilute hydrothermal fluids of meteoric origin typically found in active geothermal systems, sulphide complexes predominate. Adiabatic boiling of dilute and saline geothermal waters leads to precipitation of silver sulphide and removal of silver from solution. Conductive cooling has insignificant effects on silver mobility in dilute fluids, whereas it leads to quantitative loss of silver for geothermal fluids of seawater salinity.

An oligarchic microbial assemblage in the anoxic bottom waters of a volcanic subglacial lake

In 2006, we sampled the anoxic bottom waters of a volcanic lake beneath the Vatnajökull ice cap (Iceland). The sample contained 5 × 105 cells per ml, and whole-cell fluorescent in situ hybridization (FISH) and PCR with domain-specific probes showed these to be essentially all bacteria, with no detectable archaea. Pyrosequencing of the V6 hypervariable region of the 16S ribosomal RNA gene, Sanger sequencing of a clone library and FISH-based enumeration of four major phylotypes revealed that the assemblage was dominated by a few groups of putative chemotrophic bacteria whose closest cultivated relatives use sulfide, sulfur or hydrogen as electron donors, and oxygen, sulfate or CO2 as electron acceptors. Hundreds of other phylotypes are present at lower abundance in our V6 tag libraries and a rarefaction analysis indicates that sampling did not reach saturation, but FISH data limit the remaining biome to <10–20% of all cells. The composition of this oligarchy can be understood in the context of the chemical disequilibrium created by the mixing of sulfidic lake water and oxygenated glacial meltwater.

Numerical modelling of CO2-water-basalt interaction

Abstract The effects of CO2 on water-basaltic glass interaction have been simulated at 25ºC. The calculations indicate that addition of CO2 (2–30 bar) to water significantly changes the reaction path. Initially, the pH is buffered between 4 and 6 by CO2 ionization, with dissolution of basaltic glass and the formation of secondary minerals with SiO2, Mg-Fe carbonates and dolomite predominating. Upon the dissolution of additional basaltic glass and mineral fixation of CO2, the pH increases to >8 and (Ca)-Fe-Mg smectites, SiO2, Ca-Na zeolites and calcite become the dominant secondary minerals forming. The overall reaction path depends on the initial water composition, reactive surface area, and the composition of the phyllosilicates and carbonates forming. The key factors are the mobility of Mg2+, Fe2+ and Ca2+ and the competing reactions for these solutes among secondary minerals.