Isotopic and elemental partitioning of boron between hydrous fluid and ...

American Mineralogist, Volume 87, pages 769–774, 2002

LETTERS Isotopic and elemental partitioning of boron between hydrous fluid and silicate melt RICHARD L. HERVIG,1,2,* GORDON M. MOORE,3 LYNDA B. WILLIAMS,1,2 SIMON M. PEACOCK,2 JOHN R. HOLLOWAY,2,3 AND KURT ROGGENSACK2 1

Center for Solid State Science, Arizona State University, Tempe, Arizona 85287, U.S.A. Department of Geological Sciences, Arizona State University, Tempe, Arizona 85287, U.S.A. 3 Department of Chemistry and Biochemistry, Arizona State University, Tempe, Arizona 85287, U.S.A. 2

ABSTRACT The fractionation of B and its isotopes between aqueous fluid and silicate melt has been studied from 550 to 1100 ∞C and 100–500 MPa. Fluid-melt partition coefficients are 1 for rhyolite melt. This shows that B is not always strongly extracted from melts into hydrous fluids. Boron isotopic fractionation is large compared with the carbon and oxygen stable isotopic systems (especially at high T) and is most simply explained by differences in coordination (trigonal vs. tetrahedral) among coexisting phases. Combined with earlier measurements on illite-water (300– 350 ∞C), B isotopic fractionation defines a temperature-dependent trend from 300 to 1100 ∞C. Because of the large magnitude and apparent low sensitivity to bulk composition, B isotopic fractionation can be readily applied to studies of diagenesis, hydrothermal alteration of planetary bodies, subduction-zone processing and arc magma generation, and magma chamber evolution.

INTRODUCTION Fluids play important and varied roles in many geologic processes such as metamorphism, magma generation, and ore formation. Fluids may act as agents to catalyze reactions, and reduce, oxidize, or transport elements. The mobility of fluids, which makes them effective catalysts and metasomatic agents, also limits their direct study. Boron is an element that has potential as a tracer for such processes because it can be quite soluble in fluids. Noting the large range of B isotopic ratios in naturally occurring samples (~9% range in terrestrial rocks and waters; Palmer and Swihart 1996), we have been curious to discover if this isotopic system might shed light on fluid-solid interactions. Understanding the significance of this isotopic variation first requires quantitative measurements of B fractionation. This work presents results of experimental studies that measured B isotopic fractionation between silicate melts and H2O fluid and analyses of natural samples that shed light on B isotopic fractionation and the partitioning of B into CO2rich fluids.

SAMPLES STUDIED Experimental run products Experiments to measure B partitioning (fluid/melt) and isotopic fractionation were conducted using basaltic and rhyolitic melt and aqueous fluid. The basalt was a mid-ocean ridge sample from the Juan de Fuca ridge (TT-152; Table 1). An aliquot of this glassy sample was combined with NIST SRM 951 * E-mail: [email protected] 0003-004X/02/0506–769$05.00

769

boric acid to a level of ~2000 ppm B, melted, reground, and remelted. The rhyolite (Table 1) was a natural peraluminous sample from Macusani, Peru, containing ~2000 ppm B (Pichavant et al. 1987). Conditions for the experiments are given in Table 2. The basalt was studied at 950–1100 ∞C and between 110 and 170 MPa in a rapid-quench, internally heated apparatus (Holloway et al. 1992). Experiments on the rhyolite used a non-end-loaded piston cylinder at 500 MPa and 750–850 ∞C (http://depthsoftheearth.com/catalog.html). Both sets of experiments followed a similar protocol. A chip of “dry” glass was sealed in a capsule (either Au-Pd or Au, see Table 2) with a 1:1 ratio of H2O. The run was taken up to pressure and temperature for a period of 24–52 hours and quenched. The capsules were weighed, punctured, and re-weighed after drying at 110 ∞C to ensure that a fluid phase had been present. Fragments of the glassy run products were mounted in epoxy with fragments of the starting material and polished. Natural samples One pelite from the Bradshaw Mountains (central Arizona) was studied. The metamorphic rocks in the area contain andalusite, sillimanite, staurolite, garnet, biotite, and muscovite; this assemblage constrains the peak metamorphic temperature at ~550 ∞C (O’Hara 1980). Boron isotopic analyses of coexisting tourmaline and muscovite were made under the assumption that the analyses would represent the isotopic fractionation between these two phases at the peak metamorphic temperature, thus providing a “natural experiment.” Several melt inclusions from olivine crystals found in the Cerro Negro (Nicaragua) volcano were analyzed for their H2O and CO2 contents (Roggensack et al. 1997) by infrared spec-

770

HERVIG ET AL.: BORON ISOTOPES

TABLE 1. Composition of starting materials wt% P 2 O5 SiO2 TiO2 Al2O3 FeO MnO MgO CaO Na2O K2O F B (ppm) * Pan et al. (1992). † Pichavant et al. (1987).

TT-152* 0.22 50.8 1.84 13.7 12.4 0.22 6.67 11.5 2.68 0.15 n.a. 2300

RESULTS Macusanite† 0.54 72.5 0.03 16.0 0.62 0.06 0.02 0.19 4.12 3.68 1.33 1926

trometry, and show a degassing trend dominated by the loss of CO2. We analyzed these melt inclusions for their B contents to examine the effect of CO2 degassing on the fluid/melt partitioning behavior of B.

The results of the experiments are given in Table 2. Fluid/ melt partition coefficients for B are compared with other literature values on Figure 1. For the rhyolitic composition at 500 MPa, we found partition coefficients slightly above unity, in the range of earlier experiments on the same rhyolite at 200 MPa and experiments on haplogranitic composition at 100 MPa (London et al. 1988; Pichavant 1981). Partition coefficients for the basaltic composition are all ~100 ∞C (Bassett 1976), whereas trace B is tetrahedrally coordinated in illite and muscovite. Boron-rich silicate melts (e.g., wt% levels) contain mostly trigonal B (Morgan et al. 1990; Yun and Bray 1978). The coordination of trace B in silicate melts awaits spectroscopic studies (e.g., Sen et al. 1994), but the approximate colinearity of the silicate melt-hydrous fluid with muscovitetourmaline and illite-water on Figure 3 indicates to us that trace B occupies tetrahedral sites in silicate melts. In support of this conclusion, it has been observed that decreasing the B in melts synthesized in the Na2O-B2O3-SiO2 system results in increasing amounts of tetrahedrally coordinated B at the expense of trigonal B (Yun and Bray 1978). We note that the B-O bond length is 8% larger for IVB compared to IIIB species (Hawthorne et al. 1996). The resulting differences in bond strengths and

differences in the electronic environments between the two coordination states may contribute to the large isotopic fractionation. In addition, because the B-O bond is strong (i.e., stronger than Si-O bonds; Navrotsky 1996), B-O interactions may overwhelm the effect of next-nearest neighbors, thus leading to small effects from changing bulk chemistry. Comparison with other studies The hypothesis that coordination differences are the major reason for the large B isotopic fractionation at high temperature can be partially tested by examining published B isotopic analyses of coexisting phases for which coordination is known. These analyses are shown in Figure 3 as open squares. At the highest temperature, analyses were made of trace B in basaltic melt and coexisting silicate phenocrysts (Benton et al. 1999). We suggest that B coordination in the phenocrysts and glass are identical. The B content of the phenocrysts is very small, and so the analyses may be strongly influenced by contamination from glass, but, taken at face value, the analyses show no difference in d11B between the crystals and the glass. Spivack et al. (1990) showed that B isotopic compositions were nearly identical between liquid water and water vapor, two phases expected to have the same B coordination. Peacock and Hervig (1999) analyzed coexisting phengitic muscovite and actinolite. They found no difference in B isotopic ratio between these phases expected to contain B in tetrahedral coordination. Leeman et al. (1992) analyzed brines and water vapor, finding small differences in the B isotopic composition of the two phases (both containing trigonal B). At low temperatures, analyses of sassolite and water at 25 ∞C (both containing B in trigonal coordination) showed very small B isotopic fractionation, whereas analyses of borax (which contains B in both trigonal and tetrahedral coordination) and coexisting water (Oi et al. 1991) showed significant fractionation. Oi et al. (1991) used their analyses of borate minerals to calculate the isotopic fractionation of B between tetrahedral and trigonal coordination at 25 ∞C as –39 ± 20‰. This value is close to the extrapolation of our high temperature data to 25 ∞C (1000lna ~ –36). Experiments synthesizing tourmaline at hydrothermal conditions do not support coordination-dependent isotopic fractionation of B. Palmer et al. (1992) measured B isotopic contents of synthetic tourmaline and coexisting fluid from 350–750 ∞C and observed a temperature-dependent fractionation that was also pressure-dependent. Results of their experiments at 200 MPa are shown on Figure 3. Measurements at lower pressures showed larger fractionation. We would have expected negligible isotopic fractionation between these phases with common (trigonal) B coordination at any pressure. Palmer et al. (1992) noted that the pressure-dependence of the B isotopic fractionation might relate to kinetic effects during synthesis. Low-temperature studies at low B concentrations are discussed separately here. Kakihana et al. (1977) used theoretical determinations of reduced partition function ratios to calculate B isotopic fractionation between trigonal and tetrahedral sites. Their result is shown as the thin dashed line on Figure 3, and indicates smaller isotopic fractionation than we have measured at higher temperatures, but showed good agreement with their ion-exchange studies. Hemming et al. (1995) measured sig-


nificant B isotopic fractionation between calcite and water and aragonite and water at 25 ∞C. The coordination of B in the carbonates was measured by NMR (Sen et al. 1994), which showed trigonal B in calcite and tetrahedral B in aragonite. Despite the difference in coordination between the two phases, the fractionation of B between the two different carbonates and water was indistinguishable (~16‰ lighter than the water). The interpretation presented by Hemming et al. (1995) was that tetrahedrally coordinated B adsorbed on the carbonate surface was directly incorporated into the growing crystal. They noted that the isotopic composition of both carbonate minerals mimicked the isotopic composition of the tetrahedral B in solution (using the model of Kakihana et al. 1977) and concluded that only tetrahedral B interacted with the surface of the growing calcite. The studies of Kakihana et al. (1977) and Hemming et al. (1995) show that coordination-dependent fractionation of B isotopes does occur, but with significantly smaller fractionations than the extrapolation of our high-temperature data (approximately –20‰ at 25 ∞C compared to our estimate near –36‰). Since the initial work by Kakihana et al. (1977), other resins have been tested in ion-exchange columns and have shown different fractionation. For example, Oi et al. (1997) found isotopic fractionation between trigonal and tetrahedral species to vary from 18 to 22‰ when different resins were used. They concluded that smaller fractionation results from the adsorption of some trigonal B species as well as the dominant tetrahedral species, with different resins adsorbing more or less of the trigonal B. All the resins studied so far are hydrocarbon based, and the suspected form of adsorption is a B(OH)2(OC)2 cluster where the two C atoms are components of the resin surface and the B is tetrahedrally coordinated to O (e.g., Oi et al. 1997). When studying adsorption of B onto clay minerals at 25 ∞C, Palmer et al. (1987) found that the silicate surfaces adsorbed B up to 32‰ lighter than the B dissolved in the coexisting water, much larger than the adsorption on carbonates or hydrocarbon-based resins, but similar to the extrapolation of our high-temperature measurements on silicates. Newer theoretical studies of the reduced partition function ratios for B now suggest that the isotopic fractionation between trigonal and tetrahedral coordination is approximately –26‰ at 25 ∞C (Oi 2000), but is very sensitive to the species in solution— reaching values of –36‰ for some polyborate units. These different observations alert us to the potential role of speciation and surface-dependent adsorption processes in controlling B isotopic fractionation on different minerals at low temperatures. Estimating thermodynamic values The B isotope-exchange reaction between phase 1 (all trigonal B) and phase 2 (all tetrahedral B) can be written as: III 11

( B1) + IV(10B2) = III(10B1) + IV(11B2)

(1)

with a distribution coefficient, or isotopic fractionation factor a = IV(11B/10B)2/III(11B/10B)1 .

(2)

The Gibbs free energy (DG) of this exchange reaction can be related to the reaction enthalpy (DH) and entropy (DS) and

773

the fractionation factor as follows. DG @ DH – TDS = –RTlna

(3)

where R is the universal gas constant and the approximation arises from ignoring second-order effects. Rearranging gives: lna = –DH/RT + DS/R.

(4)

From the slope and intercept of the line on Figure 2, we determine DH (~100 J/mol) and DS (~50 mJ/mol-deg), respectively, for B isotopic exchange. At 25 ∞C, DG is ~90 J/mol. The DG and DH are somewhat larger than observed for other stable isotopic exchange reactions (DG298 ~80 J/mol, DH ~40 J/mol; O’Neil 1986; Stolper and Epstein 1991). In addition, temperature-dependent heat capacity changes (DCP) for stable isotopic exchange reactions may shift a 1/T dependence of a at low T to a 1/T2 dependence as temperatures reach a few hundred degrees (O’Neil 1986). In contrast, B isotopic fractionation does not show any obvious deviation from a linear relation with 1/T (up to 1100 ∞C) and does not exhibit a change in sign in fractionation as found for O isotopes in some phases (Valley 1986). Limitations of boron isotopic fractionation studies Limitations in using this isotopic system include (1) low levels of B in many natural phases; (2) uncertainty in B coordination in some phases; and (3) kinetics of isotopic exchange. The first item relates to analytical sensitivity and control of B contamination (Chaussidon et al. 1997; Rose et al. 2001) whereas the second and third items are more difficult to address. We can assume that trace B dissolved in H2O fluids (8, conditions where tetrahedral B is present (Palmer and Swihart 1996) and that trace B substitutes for tetrahedral Al or Si in silicates. If B contents are sufficiently high in melts (e.g., several wt%), then B may enter trigonal coordination (Morgan et al. 1990; Yun and Bray 1978), but we argue above that trace B in most natural silicate melts will be tetrahedral. In borate and borosilicate minerals, B is completely or partly trigonally coordinated, but in most rock-forming phases in nature, trace B substitutes for silica; as a result, the assumption of tetrahedral B is appropriate. The third item may be quite important, but the kinetics of B reaction/diffusion are incompletely characterized. The studies by Baker (1992) and Chakraborty et al. (1993) on melts (dacitic to haplogranitic composition) show that trace B diffuses at about the same rate as Si. If this result is also true for crystals, then B isotopes may be difficult to exchange after crystallization because of slow diffusion.

SUMMARY The partitioning of B between hydrous fluid and melt varies with melt composition. Boron is preferentially lost to hydrous fluids from rhyolitic melts at low pressure and is compatible in basaltic melts compared with hydrous or CO2rich fluids. In contrast, the partitioning of B isotopes between hydrous fluids and melt appears independent of melt composition. The range of B isotopic ratios observed in nature may be

774


explained in large part by coordination-dependent fractionation of trace B between tetrahedral and trigonal sites in coexisting phases. This isotopic fractionation remains large at high temperature, is apparently linear in 1/T, and is not strongly dependent on bulk chemistry.

ACKNOWLEDGMENTS The authors appreciate helpful comments by Julie Morris, Bill Leeman, and Ed Grew, and the technical support provided by Al Higgs to keep the SIMS operating smoothly. This material is based upon work supported by the National Science Foundation under Grant No. 9706263 and 9973040 and by the U.S. Dept. of Energy (DE FG04 97ER14414).

REFERENCES CITED Baker, D.R. (1992) Tracer diffusion of network formers and multicomponent diffusion in dacitic and rhyolitic melts. Geochimica et Cosmochimica Acta, 56, 617– 631. Bassett, R.L. (1976) The geochemistry of boron in thermal waters. Stanford University, Stanford. Benton, L.D., Tomascak, P.B., and Helz, R.T. (1999) A study of boron isotopes in Kilauea Iki lava lake, Hawaii. Eos, Transactions, AGU (abstract), 80, F1197. Chakraborty, S., Dingwell, D.B., and Chaussidon, M. (1993) Chemical diffusivity of boron in melts of haplogranitic composition. Geochimica et Cosmochimica Acta, 57, 1741–1751. Chaussidon, M., Robert, F., Mangin, D., Hanon, P., and Rose, E. (1997) Analytical procedures for the measurement of boron isotope compositions by ion microprobe in meteorites and mantle rocks. Geostandards Newsletter, 21, 7–17. Hawthorne, F.C., Burns, P.C., and Grice, J.D. (1996) The crystal chemistry of boron. In E.S. Grew, and L.M. Anovitz, Eds. Boron: mineralogy, petrology, and geochemistry, p. 41–115. Mineralogical Society of America, Washington, D.C. Hemming, N.G., Reeder, R.J., and Hanson, G.N. (1995) Mineral-fluid partitioning and isotopic fractionation of boron in synthetic calcium carbonate. Geochimica et Cosmochimica Acta, 59, 371–379. Holloway, J.R., Dixon, J.E., and Pawley, A.R. (1992) An internally-heated rapidquench, high-pressure vessel. American Mineralogist, 77, 643–646. Kakihana, H., Kotaka, M., Satoh, S., Nomura, M., and Okamoto, M. (1977) Fundamental studies on the ion exchange separation of boron isotopes. Bulletin Chemical Society Japan, 50, 158–163. Leeman, W.P., Vocke, R.D., and McKibben, M.A. (1992) Boron isotopic fractionation between coexisting vapor and liquid in natural geothermal systems. In Y.K. Kharaka, and A.S. Maest, Eds., 7th International symposium on waterrock interaction proceedings, 7, p. 1007–1010. Balkema, Rotterdam. London, D., Hervig, R.L., and Morgan, G.B., VI. (1988) Melt-vapor solubility and elemental partitioning in peraluminous granite-pegmatite systems: Experimental results with Macusani glass at 200 MPa. Contributions to Mineralogy and Petrology, 99, 360–373. Matsuhisa, Y., Goldsmith, J.R., and Clayton, R.N. (1979) Oxygen isotopic fractionation in the system quartz-albite-anorthite-water. Geochimica et Cosmochimica Acta, 43, 1131–1140. Morgan, G.B., VI, London, D., and Kirkpatrick, R.J. (1990) Reconnaissance spectroscopic study of hydrous sodium aluminum borosilicate glasses. Geological Society of America Abstracts, 22, A167. Navrotsky, A. (1996) Thermochemistry of borosilicate melts and glasses—from pyrex to pegmatites. In E.S. Grew, and L.M. Anovitz, Eds., Boron: Mineralogy, Petrology, and Geochemistry, 33, p. 165–180. Reviews in Mineralogy, Mineralogical Society of America, Washington, D.C. Nies, N.P. and Campbell, G.W. (1964) Inorganic boron-oxygen chemistry: Boron, metallo-boron compounds, and boranes. 232 p. Interscience, New York. O’Hara, P.F. (1980) Metamorphic and structural geology of the Northern Bradshaw Mountains, Yavapai County, Arizona. Geology, p. 146. Arizona State University, Tempe. O’Neil, J.R. (1986) Theoretical and experimental aspects of isotopic fractionations. In J.W. Valley, J.H.P. Tayor, and J.R. O’Neil, Eds., Stable isotopes in high temperature geological processes, 16, p. 1–40. Reviews in Mineralogy, Mineralogical Society of America, Washington, D.C.

Oi, T. (2000) Ab initio molecular orbital calculations of reduced partition function ratios of polyboric acids and polyborate anions. Zeitschrift Naturforschung A Physical Sciences, 55, 623–628. Oi, T., Kato, J., Ossaka, T., and Kakihana, H. (1991) Boron isotopic fractionation accompanying boron mineral formation from aqueous boric acid-sodium hydroxide solutions at 25 ∞C. Geochemical Journal, 25, 377–385. Oi, T., Shimazaki, H., Ishii, R., and Hosoe, M. (1997) Boron isotope fractionation in liquid chromatography with boron-specific resins as column packing material. Separation Science and Technology, 32, 1821–1834. Palmer, M.P., London, D., Morgan, G.B. VI, and Babb, H. (1992) Experimental determination of fractionation of 11B/10B between tourmaline and aqueous vapor: a temperature- and pressure-dependent isotopic system. Chemical Geology, 101, 123–129. Palmer, M.R. and Swihart, G.H. (1996) Boron isotope geochemistry: an overview. In E.S. Grew, and L.M. Anovitz, Eds., Boron mineralogy, petrology and geochemistry, 33, p. 709–744. Reviews in Mineralogy, Mineralogical Society of America, Washington, D.C. Palmer, M.R., Spivack, A.G., and Edmond, A.M. (1987) Temperature and pH controls over isotopic fractionation during adsorption of boron on marine clay. Geochimica et Cosmochimica Acta, 51, 2319–2323. Pan, V., Holloway, J.R., and Hervig, R.L. (1992) The pressure and temperature dependence of carbon dioxide solubility in tholeiitic basalt melts. Geochimica et Cosmochimica Acta, 55, 1587–1595. Peacock, S.M. and Hervig, R.L. (1999) Boron isotopic composition of subductionzone metamorphic rocks. Chemical Geology, 160, 281–290. Pichavant, M. (1981) An experimental study of the effect of boron on a water saturated haplogranite at 1 kbar vapour pressure. Contributions to Mineralogy and Petrology, 76, 430–439. Pichavant, M., Valencia Herrera, J., Boulmier, S., Briqueu, L., Joron, J.-L., Juteau, M., Marin, L., Michard, A., Sheppard, S.M.F., Treuil, M., and Vernet, M. (1987) The macusani glasses, SE Peru: evidence of chemical fractionation in peraluminous magmas. In B.O. Mysen, Ed., Magmatic Processes: Physicochemical Principles., 1, p. 359–373. Geochemical Society, University Park, Pennsylvania. Roggensack, K., Hervig, R.L., McKnight, S.B., and Williams, S.N. (1997) Explosive basaltic volcanism from Cerro Negro Volcano: Influence of volatiles on eruptive style. Science, 277, 1639–1642. Rose, E.F., Shimizu, N., Layne, G.D., and Grove, T.L. (2001) Melt production beneath Mt. Shasta from boron data in primitive melt inclusions. Science, 293, 281–283. Sen, S., Stebbins, J.F., Hemming, N.G., and Ghosh, B. (1994) Coordination environments of B impurities in calcite and aragonite polymorphs: A 11B MAS NMR study. American Mineralogist, 79, 819–825. Spivack, A.L., Berndt, M.E., and Seyfried, W.E., Jr. (1990) Boron isotope fractionation during supercritical phase separation. Geochimica et Cosmochimica Acta, 54, 2337–2339. Stolper, E. and Epstein, S. (1991) An experimental study of oxygen isotope partitioning between silica glass and CO2 vapor. In H.P. Taylor, Jr., J.R. O’Neil, and I.R. Kaplan, Eds., Stable Isotope Geochemistry: A Tribute to Samuel Epstein., Special Publication No. 3. The Geochemical Society, New York. Valley, J. (1986) Stable isotope geochemistry of metamorphic rocks. In J.W. Valley, J.H.P. Tayor, and J.R. O’Neil, Eds., Stable isotopes in high temperature geological processes, 16, p. 445–489. Reviews in Mineralogy, Mineralogical Society of America, Washington, D.C. Webster, J.D., Holloway, J.R., and Hervig, R.L. (1989) Partitioning of lithophile trace elements between H2O and H2O + CO2 fluids and topaz rhyolite melt. Economic Geology, 84, 116–134. Williams, L.B., Hervig, R.L., Holloway, J.R., and Hutcheon, I. (2001) Boron isotope geochemistry during diagenesis, Part I. Experimental determination of fractionation during illitization of smectite. Geochimica et Cosmochimica Acta, 65, 1769–1782. Yun, Y.H. and Bray, P.J. (1978) Nuclear magnetic resonance studies of the glasses in the system Na2O-B2O3-SiO2. Journal of Non-Crystalline Solids, 27, 363–380.

MANUSCRIPT RECEIVED AUGUST 7, 2001 MANUSCRIPT ACCEPTED JANUARY 7, 2002 MANUSCRIPT HANDLED BY ROBERT F. DYMEK