Oxygen isotope fractionation between calcite and tremolite: an ...

8 downloads 0 Views 751KB Size Report
Oct 5, 1993 - Abstract Oxygen isotope partitioning between calcite and tremolite was experimentally calibrated in the pres- ence of small amounts of a ...
Contrib Mineral Petrol (1994) 118:249-255

9 Springer-Verlag 1994

Y.-F. Zheng 9P. Metz 9 M. Satir

Oxygen isotope fractionation between calcite and tremolite: an experimental study

Received: 5 October 1993 / Accepted: 21 April 1994

Abstract Oxygen isotope partitioning between calcite and tremolite was experimentally calibrated in the presence of small amounts of a supercritical CO2-H20 fluid at temperatures from 520 to 680~ C and pressures from 3 to 10 kbar. The experiments were carried out within the stability field of the calcite-tremolite assemblage based on phase equilibrium relationships in the system CaO-MgO-SiOz-CO2-H20, so that decomposition of calcite and tremolite was avoided under the experimental conditions. Appropriate proportions of carbon dioxide to water were used to meet this requirement. Large weight ratios of mineral to fluid were employed in order to make the isotopic exchange between calcite and tremolite in the presence of a fluid close to that without fluid. The data processing method for isotopic exchange in a three-phase system has been applied to extrapolate partial equilibrium data to equilibrium values. The determined fractionation factors between calcite (Cc) and tremolite (Tr) are expressed as: 103 In ~Co-Tr= 3.80 x 106/T 2 - 1.67 By combining the present data with the experimental calibrations of Clayton et al. (1989) on the calcite-quartz system, we obtain the fractionation for the quartztremolite system: 10 3 In

~Qz_Tr= 4.18 x 106/T 2 - 1.67

Our experimental calibrations are in good agreement with the theoretical calculations of Hoffbauer et al.

Y.-F. Zheng 9P. Metz ([g~3). M. Satir Institute for Mineralogy, Petrology and Geochemistry, University of Tiibingen, Wilhelmstrage 56, D-72074 Tfibingen, Germany Y.-F. Zheng Department of Earth and Space Sciences, University of Science and Technology of China, Hefei, Anhui 230026, PR China Editorial responsibility: J. Hoefs

(1994) and the empirical estimates of Bottinga and Javoy (1975) based on isotopic data from naturall assemblages. At 700~ good agreement also exists between our experimental data and theoretical values calculated by Zheng (1993b). With decreasing temperature, however, an increasing difference between these data appears. Retrograde isotopic reequilibration by oxygen diffusion may be common for amphibole relative to diopside in metamorphic rocks. However, isotopic equilibrium in amphibole can be preserved in cases of rapid cooling.

Introduction Amphiboles are important hydroxyl-bearing minerals occurring in nature. They are formed in a wide range of igneous and metamorphic rocks. Oxygen isotope analyses of amphiboles were first carried out by Taylor and Epstein (1962), who showed that this mineral group occupies a specific position in the sequence of common rock-forming minerals with respect to the tendency to concentrate 180. A set of coexisting minerals commonly exhibits decreasing 180/160 ratios in the sequence: quartz, alkali feldspar, plagioclase, muscovite, amphibole, biotite, and magnetite. Earlier analyses of amphiboles were also reported by Garlick and Epstein (1967), Shieh and Taylor (1969) and Vogel and Garlick (1970). These authors indicate that oxygen isotope fractionations between quartz and amphibole in metamorphic rocks generally decrease with increasing metamorphic grade judged on petrological and geological grounds. Using data on naturally occurring oxygen isotope fractionation among coexisting minerals in igneous and metamorphic rocks, Bottinga and Javoy (1975) deduced empirical expressions for the temperature dependence of oxygen isotope partitioning in amphibole for a temperature range from 500 to 800~ C. Matthews et al. (1983) attempted experimentally to determine the oxygen isotope fractionation between tremolite and water at 600-700 ~ C by means of the three-isotope method but did not obtain consistent results. Neither empirical nor

250 experimental calibrations have presented satisfactory solutions of the equilibrium oxygen isotope fractionation in amphiboles. Direct exchange reactions with c a r b o n a t e under nominally dry conditions at high pressures have been applied by Clayton et al. (1989) and C h i b a et al. (1989) to calibrate the oxygen isotope fractionation a m o n g m a j o r a n h y d r o u s rock-forming minerals. A l t h o u g h having m a n y advantages over h y d r o t h e r m a l experiments, the a n h y d r o u s c a r b o n a t e - e x c h a n g e technique is not suitable for hydroxyl-bearing minerals such as a m phiboles which will d e c o m p o s e under a n h y d r o u s conditions when exchange experiments are not p e r f o r m e d within their P-T-X field of stability. In order to circumvent such a difficulty, a new experimental a p p r o a c h , which conducts oxygen isotope exchange between calcite and silicate in the presence of a supercritical CO2H 2 0 fluid within the stability field of the calcite-silicate assemblage, has been developed and successfully applied to the calcite-forsterite system (Zheng et al. 1994a). In this connection, a d a t a processing m e t h o d for isotopic exchange in a three-phase system (e.g. calcite-silicate-fluid) has been established to extrapolate partial equilibrium d a t a to equilibrium values. The present study is devoted to determining the equilibrium oxygen isotope fractionation between calcite and tremolite using the oxygen isotope exchange between the two minerals in the presence of a CO2-H20 fluid. The choice of the calcite-tremolite-CO2-H20 system is m a d e for several reasons: (1) phase equilibrium data for this system are available f r o m experimental determinations (Slaughter et al. 1975; P u h a n and Metz 1987) and t h e r m o d y n a m i c calculations (Gottschalk 1990); (2) tremolite Ca2Mgs[SisO22](OH)2 is p e r h a p s the least complicated mineral of the a m p h i b o l e g r o u p a n d thus has the a d v a n t a g e of a relatively simple chemistry; (3) the oxygen diffusion experiments of Farver and Giletti (1985) suggest that hydroxyl oxygens are not exchanged at a faster rate t h a n other oxygens in a m p h i bole, enabling application of the partial equilibrium technique; (4) oxygen isotope fractionation between calcite and tremolite has been calculated theoretically by Hoffbauer et al. (1994) and Zheng (1993b) using m o d i fied increment methods. Thus the present experimental data can be used to test the validity of the theoretical results.

Experimental methods The experimental procedures have been described by Zheng et al. (1994a). In practice a powdered mixture of calcite and tremolite with a small amount of a CO2-H20 fluid was held at the desired temperature and pressure for a given time. It is important to perform all experiments strictly within the T-P-Xco2 stability field of the calcite-tremolite assemblage. Upon completion of an exchange experiment, the product mineral phases were extracted and separately analysed for their isotopic compositions. The weights of calcite and tremolite used in each experiment were chosen to contain the same number of oxygen atoms, and were about either 10 mg or 15 mg each. The grain size of the

Table 1 Isotopic composition of starting materials

Material

51SOsMow %0

~13CpDB%0

Calcite I (Merck) Calcite II (Merck) Calcite pA (Merck) Calcite NBS-18 (carbonatite) Tremolite (Nord Talgje) CO2 (AgaC204) H20

18.65

-22.37

17.93

- 20.09

15.41

- 7.29

7.84

-

5.13

11.52 25.16

- 19.93

- 12.03

starting materials was mostly in the 5-10 gm range. The weight proportion of CO2 to H20 was chosen to maintain the stability of the calcite-tremolite assemblage at the experimental temperatures and pressures. Phase diagrams calculated thermodynamically by Gottschalk (1990) were used for this purpose. Silver oxalate (Ag2C204) was used to produce pure CO 2 by thermal decomposition (Holloway and Reese 1974). Charges consisted of 20 or 30 mg solids and 0.806 or 2.63 mg fluid loaded together into a capsule made from gold tubing. Isotopic equilibrium was approached from opposite directions by an appropriate choice of the oxygen isotope composition of the starting materials. Four different calcites were used to exchange oxygen isotopes with a common tremolite (from Nord Talgje, Norway; Miiller and Strauss 1984). The isotopic compositions of the starting materials are given in Table 1. The experiments were carried out either in a piston-cylinder apparatus at 10_+0.5 kbar (1.0_+0.05 GPa) using a NaC1 pressure cell, or in conventional hydrothermal autoclaves at 5 or 3_+0.05 kbar (500 or 300___5 MPa) where CO2 was used as pressure medium. The temperatures of the experiments ranged from 520 to 680~ C. Uncertainties in the temperature measurement were about __5~ C in the hydrothermal autoclaves and about _+10~ C in the piston-cylinder apparatus. SEM and X-ray diffraction measurements on representative run products showed that no other mineral phase than calcite plus tremolite was present. However, dissolution and recrystallization have taken place significantly, which may be the principal mechanism of the isotopic exchange between calcite and tremolite. After the runs the capsules were pierced and dried at 110~ C before performing the isotopic analyses. The solid phases were reacted with 100% phosphoric acid for one day at 25~ C in the standard procedure for isotopic analyses of carbonates (McCrea 1950). The residual tremolite was washed repeatedly with distilled water to remove acid and then dried at 110~ C before analysis. Oxygen was extracted from the tremolites by reaction with BrF 5 (Clayton and Mayeda 1963). All the oxygen and carbon isotopic ratios were measured on a MAT 252 mass spectrometer and are reported in ~180 relative to SMOW (standard mean ocean water) and in ~13C relative to PDB (Peedee belemnite). Analytical uncertainties are about +0.2%0 for both ~180 and 513C measurements.

Results and discussion The analytical results for all exchange experiments are presented in Table 2. Since complete isotopic equilibriu m was not obtained in individual experiments, an ext r a p o l a t i o n procedure is required to infer the equilibrium values. For this purpose, the d a t a processing

251 Table 2 Experimental data for calcite-tremolite oxygen isotope exchange in the presence of a CO2-H20 fluid, k J8 orco va values are obtained by calculating the 8z80~n~o from mass balance before and after the isotopic exchange in the total system; y and x are defined by Eqs. (3) and (4) in the text. M / F represents the weight T

P

t

(o C)

(kbar)

(days)

M/F

680

10

3

Xco 2

818 Offi uid

Run no.

0.36

TC '15 TC '16 TC 25 TC '1 TC '5 TC '6 TC '49 TC '50 TC '51 TC 35 TC