37. OXYGEN- AND HYDROGEN-ISOTOPE COMPOSITION OF SOME BASALTS FROM DEEP SEA DRILLING PROJECT HOLE 504B, COSTA RICA RIFT, LEGS 69 AND 70

Whole-rock basalt samples from the upper half of Deep Sea Drilling Project Hole 504B have oxygen-isotope compositions typical of mid-ocean-ridge basalts which have experienced a moderate degree of low-temperature alteration by sea water. By contrast, δ 1 8 θ values in the lower half of the hole correspond to basalts which have experienced almost no detectable oxygen-isotope alteration. These observations suggest that the overall water/rock ratio was lower in the lower half of the drilled crust. A correlation between δ 1 8 θ values and Sr/Sr ratios suggests that the water/rock ratio, rather than temperature variation, was the main factor determining basalt δ 1 8 θ values. Hydrogen-isotope data appear to be consistent with a low water/rock ratio in the lower part of the crust. INTRODUCTION In order to provide further data on the extent of alteration of the 5.9-m.y.-old basalt crust drilled at Hole 504B (1°14'N, 83°44'W, near the Costa Rica Rift), the oxygen-isotope composition of whole-rock samples has been determined. Sr-isotope data (Barrett, this volume) indicate that the basalts in the upper part of the hole have been isotopically altered by basalt-sea-water interaction, whereas basalts in the lower part of the hole have maintained essentially primary Sr-isotope ratios. It is therefore of interest to establish whether δ 1 8 θ values reveal the same pattern of isotopic alteration. δD values were also measured to determine if variations with depth exist. Mid-ocean-ridge basalts which have not experienced sea-water alteration have a very restricted range in δ 1 8 θ values: 5.8±0.3% (Taylor, 1968; Muehlenbachs and Clayton, 1972; Muehlenbachs, 1980). However, basalts affected by low-temperature sea-water alteration display increased δ 1 8 θ values. This results mainly from the formation of secondary smectitic clays, which have δ 1 8 θ values notably higher than that of fresh basalt. Javoy and Fouillac (1980) reported δ 1 8 θ values of about 27% for vein smectites from Legs 51 to 53. A basalt containing 10% smectite of this isotopic composition would have an initial whole-rock value of 5.8% increased to 7.6%. The extent of oxygen-isotope alteration depends in part on the amount of oxygen made available to the basalts by circulating sea water, i.e., the water/rock ratio. A second factor is the temperature at which alteration takes place. With increasing temperatures, O-isotopic fractionations between silicates and water decrease, so that—all other factors being equal—whole-rock O enrichments are lessened. Eventually, at temperatures of Cann, J. R., Langseth, M. G., Honnorez, J., Von Herzen, R. P., White, S. M., et al., Init. Repts. DSDP, 69: Washington (U.S. Govt. Printing Office). 2 Present address: Dept, of Geology, University of Toronto, Toronto, Ontario M5S 1A1, Canada. about a few hundred degrees, rock-water fractionations decrease to the point where the δ 1 8 θ value of an altered basalt is essentially the same as that of fresh basalt. Basalts from ophiolite complexes metamorphosed in the zeolite to lower greenschist facies commonly have δ 1 8 θ values of 7 to 13% (Spooner et al., 1974; Heaton and Sheppard, 1977; McCulloch et al., 1980). However, for upper greenschist to amphibolite facies metabasic rocks, these authors report δ 1 8 θ values ranging from about 6%0 down to as low as 3%0. Such low values are due primarily to the much-reduced whole-rock-water fractionation factors at high temperatures. In the case of Hole 504B basalts, the presence of smectitic but absence of chloritic secondary minerals suggests that the temperature range over which alteration by sea water occurred probably was limited, and temperatures relatively low. At present, the temperature at the base of the hole (563 m sub-basement) is about 120°C. Over most of the crustal history, temperatures were likely lower, not higher, because of entrainment of cold sea water through basement outcrops now covered by sediment (R. Von Herzen, pers. comm., 1981). Therefore, temperature may not have been an important parameter in determining whole-rock δ 1 8 θ variations. Instead, as discussed below, the water/rock ratio appears to have been the controlling factor. EXPERIMENTAL The sample powders used for isotopic analysis represent splits of whole-rock samples which have also been analyzed (in a different laboratory) for strontium and lead isotopes (Barrett, this volume). Samples are vein-free, and do not include local breccia zones or the volumetrically insignificant glassy flow margins. Oxygen-isotope compositions were measured with a 60°, 15-cm single focusing mass spectrometer. Oxygen was extracted from powders by the bromine pentafluoride procedure (Clayton and Mayeda, 1963). Oxygen-isotope compositions are reported in common δ-notation relative to SMOW (Standard Mean Ocean Water). Routine reproducibility of δ 1 8 θ values of replicate analyses is typically 0.15%. Hydrogen was extracted from the samples, following overnight heating at 150°C, using the method described by Godfrey (1962). Typical reproducibility of δD values is 2%.