HR: 10:50h
AN: V51K-03 INVITED [PDF]
TI: Stable Chlorine Isotopes in Ocean Crust Processes
AU: * Bach, W
EM: wbach@whoi.edu
AF: WHOI, Dept of Marine Chemistry and Geochemistry, Woods Hole, MA 02543 United States
AU: Layne, G
EM: glayne@whoi.edu
AF: WHOI, Dept of Geology and Geophysics, Woods Hole, MA 02543 United States
AU: Kent, A
EM: adam.kent@geo.orst.edu
AF: Oregon State University, Dept. of Geosciences, Corvallis, OR 97330 United States
AB:
The study of natural variations of Cl isotopic composition in ocean crustal rocks has large potential to further our
understanding of geochemical cycling of volatiles and elements soluble in saline aqueous solutions. Studies of oceanic basalt
suites to date confirm that Cl abundances are highly sensitive to the addition of saline components - either from addition
of subduction-related volatile fluxes in back-arc basins and volcanic arcs or via interaction between magmas and Cl-rich
seawater-derived components during melting, magma storage and transport. Recent data suggest that $\delta$$^{37}$Cl is much
more variable in the marine environment than originally thought, with strongly negative $\delta$$^{37}$Cl values (down to
-7.5 $\permil$) in marine pore waters and positive values (up to +7 $\permil$) in hydrothermal fluids from oceanic spreading
centers. Moreover, mantle-derived magmatic rocks reveal large variations in $\delta$$^{37}$Cl (-3 to +11 $\permil$),
reflecting mantle heterogeneity as well as assimilation of exogenic Cl by crystallizing magmas. The large isotopic variation
in low-Cl basalts has been explained by isotopic heterogeneities of the mantle, with very light $\delta$$^{37}$Cl values in
rocks from the southwest Chile Ridge that have island arc geochemical affinities and heavy $\delta$$^{37}$Cl values in
Reykjanes Ridge samples (Stewart, 2000, PhD Thesis, Duke University). The inference is that a slab-flux carries a negative
$\delta$$^{37}$Cl signature while recycled ocean crust in mantle plumes carries a strongly positive $\delta$$^{37}$Cl
signature, although this is not well constrained at present. Preferential release of isotopically light Cl from the
dewatering sediments is suggested by pore water data from the Barbados and Nankai accretionary prisms with $\delta$$^{37}$Cl
values down to -7.5 $\permil$ (Ransom et al. 1995, Geology, 23, 715). Volcanic fumaroles also appear to have negative
$\delta$$^{37}$Cl values. If this is the case then residual Cl in the subducting slab should become isotopically heavier as
$^{35}$Cl is preferentially released in the shallow subduction zone. The depleted MORB mantle is believed to have a
$\delta$$^{37}$Cl between 4 and 7 $\permil$, similar to C1-chondrite (Magenheim et al., 1995, EPSL, 131, 427). MORB with high
Cl and Cl/K tend to have $\delta$$^{37}$Cl close to 0 $\permil$, which has been explained by contamination of basaltic
magmas with seawater-derived Cl. However, the most evolved ferrobasalts and andesites from oceanic spreading ridges have
negative $\delta$$^{37}$Cl values, down to -1.7 $\permil$ (Magenheim, unpublished data). Together with data for oceanic
gabbros, the $\delta$$^{37}$Cl-[Cl] data for these highly evolved rocks form a trend that could be explained by an AFC-like
process, although the fact that the trend extends to negative $\delta$$^{37}$Cl values cannot be reconciled with batch mixing
of magma and salt or brine. Rather, it indicates that $^{35}$Cl is preferentially incorporated into the magma and may be
related to diffusive exchange between Cl in brine pools above the melt lens of an axial magma system. A more comprehensive
global dataset as well as spot analyses of Cl isotope ratios by IMP-SIMS (e.g., of melt inclusions) and the combination of
$\delta$$^{37}$Cl with other stable isotope systems (B, Li, O, H) are required before these tentative models for global
chlorine cycling and crustal assimilation at spreading ridges can be rigorously evaluated.
DE: 0330 Geochemical cycles
DE: 1040 Isotopic composition/chemistry
DE: 7220 Oceanic crust
DE: 8135 Hydrothermal systems (8424)
SC: Volcanology, Geochemistry, Petrology [V]
MN: 2003 Fall Meeting