- Email: [email protected]
A practical guide to clumped isotope geochemistry
Goldschmidt Conference Abstracts 2006
A practical guide to clumped isotope geochemistry J.M. EILER
A157
Low-calcium olivine crystals in subductionrelated magmas: Messengers from the mantle or the magma chamber? M.A. ELBURG1,2, V.S. KAMENETSKY2,3, R. ARCULUS4, R. THOMAS5
Geological and Planetary Sciences, Caltech, Pasadena, CA 91125, USA ([email protected]) 1
Clumped isotope geochemistry examines the spatial organization of rare isotopes within molecules and structural units of minerals. It principally focuses on the ordering, or ‘clumping’ of rare isotopes into bonds with or near each other rather than with isotopically normal atoms. Measurements of isotopic clumping encounter unique problems and limitations. Here, I discuss and illustrate the practical sides of this field, by way of a set of pithy aphorisms that should be followed when making such measurements by gas source isotope ratio mass spectrometry. (1) Cleanliness is next to godliness. Abundances of clumped isotopic species (e.g., 13C18O16O) are influenced by isobaric interferences from contaminants (e.g., 12C35Cl) making up as little as 1 ppb of the sample. This calls for aggressive sample purification procedures, often tailored to each sample type. (2) Water is root of all evil. Even trace amounts of water can cause ‘exchangeable’ compounds (e.g., CO2) to undergo isotopic re-ordering, even in the absence of measurable changes in bulk isotopic composition. This calls for strenuous efforts to dry samples, gas handling lines, etc. (3) Location, location, location. The information content of clumped isotope geochemistry stems from isotopic ordering in specific bonds. Therefore, analytes must be extracted from samples with negligible, or minimal and controlled, extents of isotopic ‘scrambling’. This generally rules out high-temperature processes and many heterogenous reactions. (4) You are only as good as your reference frame. Clumped isotope measurements must be standardized against materials of known composition and known state of isotopic ordering, and must be very closely matched in composition to unknowns to avoid corrections for instrumental non-linearity. This calls for specialized approaches to standardization. (5) The devil is in the details. The dynamic range of clumped isotope effects is less than 1& in most cases. Thus, this is a game that is only interesting when data have precision of hundredths of permil. (6) Size matters. Analytical precision is limited by exceptionally small relative intensities of ion beams for clumped isotopic species (ca. 10 5 · the major beam for most gases). Therefore, useful data generally require unusually large samples (ca. 10’s or 100’s of lmoles). (7) Time is on your side. For the same reasons one strives for large samples, it is important to maximize counting times—generally an hour or more of analysis per gas is a minimum.
doi:10.1016/j.gca.2006.06.1380
Department of Geology and Soil Science, Ghent University, 9000 Ghent, Belgium 2 MPI Chemistry, Geochemistry Department, Mainz, Germany ([email protected]) 3 CODES and School of Earth Sciences, University of Tasmania, Hobart, Tasmania 7001, Australia 4 Department of Earth and Marine Sciences, Australian National University, Canberra, ACT 0200, Australia 5 GeoForschungsZentrum Potsdam, Telegrafenberg B 120, Potsdam, D-14473, Germany Subduction-related magma’s are complex mixtures of silicate liquid(s) with one or more populations of crystals, which do not need to be related to the liquid in which they are found. Unraveling the origin of different components contributing to arc magmas is a prerequisite to understanding the processes and geochemical components that are involved in their origins. Our work has focused on olivine- and clinopyroxene-bearing picritic magmas from geographically diverse subduction-related areas (South Sulawesi, Valu Fa Ridge, Solomon Islands, Kamchatka). The magmas’ whole rock compositions are high in calcium and magnesium and undersaturated in silica. A significant part of the olivine population has low calcium and high nickel contents, which is at odds with the calcic whole rock composition. The low-Ca olivines contain inclusions of orthopyroxene, which is unstable in Si-undersaturated magma. Olivine crystals with low Ca and high Ni contents are often interpreted as mantle xenocrysts, but we propose that these crystals have formed in a silica-rich magma with low calcium contents. This interpretation is based on the shape and chemical zoning of the crystals, the composition of included minerals and the presence of melt inclusions. The co-existence within the subduction environment of silica-rich, calcium-poor magmas, documented by these low-Ca olivine crystals, and silica-undersaturated, high-calcium magma’s represented by the bulk of the sample, is striking. It is possible that both magma types represent localised dissolution-precipitation reactions in a shallow magma chamber, as recently proposed for high-calcium magmas by several authors. This, however, does not solve the question as to the origin and fate of the hypothetical, hot (>1250 C), ‘normal’ subduction-related magma that caused these localised reactions. doi:10.1016/j.gca.2006.06.1381
A practical guide to clumped isotope geochemistry J.M. EILER
A157
Low-calcium olivine crystals in subductionrelated magmas: Messengers from the mantle or the magma chamber? M.A. ELBURG1,2, V.S. KAMENETSKY2,3, R. ARCULUS4, R. THOMAS5
Geological and Planetary Sciences, Caltech, Pasadena, CA 91125, USA ([email protected]) 1
Clumped isotope geochemistry examines the spatial organization of rare isotopes within molecules and structural units of minerals. It principally focuses on the ordering, or ‘clumping’ of rare isotopes into bonds with or near each other rather than with isotopically normal atoms. Measurements of isotopic clumping encounter unique problems and limitations. Here, I discuss and illustrate the practical sides of this field, by way of a set of pithy aphorisms that should be followed when making such measurements by gas source isotope ratio mass spectrometry. (1) Cleanliness is next to godliness. Abundances of clumped isotopic species (e.g., 13C18O16O) are influenced by isobaric interferences from contaminants (e.g., 12C35Cl) making up as little as 1 ppb of the sample. This calls for aggressive sample purification procedures, often tailored to each sample type. (2) Water is root of all evil. Even trace amounts of water can cause ‘exchangeable’ compounds (e.g., CO2) to undergo isotopic re-ordering, even in the absence of measurable changes in bulk isotopic composition. This calls for strenuous efforts to dry samples, gas handling lines, etc. (3) Location, location, location. The information content of clumped isotope geochemistry stems from isotopic ordering in specific bonds. Therefore, analytes must be extracted from samples with negligible, or minimal and controlled, extents of isotopic ‘scrambling’. This generally rules out high-temperature processes and many heterogenous reactions. (4) You are only as good as your reference frame. Clumped isotope measurements must be standardized against materials of known composition and known state of isotopic ordering, and must be very closely matched in composition to unknowns to avoid corrections for instrumental non-linearity. This calls for specialized approaches to standardization. (5) The devil is in the details. The dynamic range of clumped isotope effects is less than 1& in most cases. Thus, this is a game that is only interesting when data have precision of hundredths of permil. (6) Size matters. Analytical precision is limited by exceptionally small relative intensities of ion beams for clumped isotopic species (ca. 10 5 · the major beam for most gases). Therefore, useful data generally require unusually large samples (ca. 10’s or 100’s of lmoles). (7) Time is on your side. For the same reasons one strives for large samples, it is important to maximize counting times—generally an hour or more of analysis per gas is a minimum.
doi:10.1016/j.gca.2006.06.1380
Department of Geology and Soil Science, Ghent University, 9000 Ghent, Belgium 2 MPI Chemistry, Geochemistry Department, Mainz, Germany ([email protected]) 3 CODES and School of Earth Sciences, University of Tasmania, Hobart, Tasmania 7001, Australia 4 Department of Earth and Marine Sciences, Australian National University, Canberra, ACT 0200, Australia 5 GeoForschungsZentrum Potsdam, Telegrafenberg B 120, Potsdam, D-14473, Germany Subduction-related magma’s are complex mixtures of silicate liquid(s) with one or more populations of crystals, which do not need to be related to the liquid in which they are found. Unraveling the origin of different components contributing to arc magmas is a prerequisite to understanding the processes and geochemical components that are involved in their origins. Our work has focused on olivine- and clinopyroxene-bearing picritic magmas from geographically diverse subduction-related areas (South Sulawesi, Valu Fa Ridge, Solomon Islands, Kamchatka). The magmas’ whole rock compositions are high in calcium and magnesium and undersaturated in silica. A significant part of the olivine population has low calcium and high nickel contents, which is at odds with the calcic whole rock composition. The low-Ca olivines contain inclusions of orthopyroxene, which is unstable in Si-undersaturated magma. Olivine crystals with low Ca and high Ni contents are often interpreted as mantle xenocrysts, but we propose that these crystals have formed in a silica-rich magma with low calcium contents. This interpretation is based on the shape and chemical zoning of the crystals, the composition of included minerals and the presence of melt inclusions. The co-existence within the subduction environment of silica-rich, calcium-poor magmas, documented by these low-Ca olivine crystals, and silica-undersaturated, high-calcium magma’s represented by the bulk of the sample, is striking. It is possible that both magma types represent localised dissolution-precipitation reactions in a shallow magma chamber, as recently proposed for high-calcium magmas by several authors. This, however, does not solve the question as to the origin and fate of the hypothetical, hot (>1250 C), ‘normal’ subduction-related magma that caused these localised reactions. doi:10.1016/j.gca.2006.06.1381