Matthew F. Thirlwall

Arsenic in groundwater: Testing pollution mechanisms for sedimentary aquifers in Bangladesh

In the deltaic plain of the Ganges-Meghna-Brahmaputra Rivers, arsenic concentrations in groundwater commonly exceed regulatory limits (.50 m gL 21 ) because FeOOH is microbially reduced and releases its sorbed load of arsenic to groundwater. Neither pyrite oxidation nor competitive exchange with fertilizer phosphate contribute to arsenic pollution. The most intense reduction and so severest pollution is driven by microbial degradation of buried deposits of peat. Concentrations of ammonium up to 23 mg L 21 come from microbial fermentation of buried peat and organic waste in latrines. Concentrations of phosphorus of up to 5 mg L 21 come from the release of sorbed phosphorus when FeOOH is reductively dissolved and from degradation of peat and organic waste from latrines. Calcium and barium in groundwater come from dissolution of detrital (and possibly pedogenic) carbonate, while magnesium is supplied by both carbonate dissolution and weathering of mica. The 87 Sr/ 86 Sr values of dissolved strontium define a two-component mixing trend between monsoonal rainfall (0.711 6 0.001) and detrital carbonate (,0.735).

Petrogenetic evolution of late Cenozoic, post-collision volcanism in western Anatolia, Turkey

Following an Eocene continent-arc collision, the Western Anatolia region experienced a complete cycle of thickening and orogenic collapse. The early stage of collision-related volcanism, which was most evident during the Early Miocene (<21 Ma), produced a considerable volume of lavas and pyroclastic deposits of basaltic andesite to rhyolite composition. The volcanic activity continued into the Middle Miocene with a gradual change in eruptive style and magma composition. The Middle Miocene activity formed in relation to localised extensional basins and was dominated by lava flows and dykes of basalt to andesite composition. Both the Early and Middle Miocene rocks exhibit calc-alkaline and shoshonitic character. The Late Miocene volcanism (<11 Ma) was marked by alkali basalts and basanites erupted along the zones of localised extension. The Early–Middle Miocene volcanic rocks exhibit enrichment in large ion lithophile elements (LILE) and light rare earth elements (LREE) relative to the high field strength elements (HFSE) and have high 87Sr/86Sr (0.70757–0.70868) and low 143Nd/144Nd (0.51232–0.51246) ratios. Modelling of these characteristics indicates a mantle lithospheric source region carrying a subduction component inherited from a pre-collision subduction event. Perturbation of this subduction-metasomatised lithosphere by either delamination of the thermal boundary layer or slab detachment is the likely mechanism for the initiation of the post-collision magmatism. Petrographic characteristics and trace element systematics (e.g. phenocryst assemblages and relative depletion in MREE and heavy rare earth elements (HREE)) suggest that the Early–Middle Miocene magmas underwent hydrous crystallisation (dominated by plagioclase+pyroxene+pargasitic amphibole) in deep crustal magma chambers. Subsequent crystallisation in shallower magma chambers follows two different trends: (1) anhydrous (pyroxene+plagioclase-dominated); and (2) hydrous (edenitic amphibole+plagioclase+pyroxene dominated). AFC modelling shows that the Early–Middle Miocene magmas evolved through assimilation combined with fractional crystallisation, and that the effects of assimilation decreased gradually from the Early Miocene into the Middle Miocene. This may indicate a progressive crustal thinning related to the extensional tectonics that prevailed from the latest Early Miocene onwards. In contrast, the Late Miocene alkaline rocks are characterised by low 87Sr/86Sr (0.70311–0.70325) and high 143Nd/144Nd (0.51293–0.51298) ratios and have OIB-type like trace element patterns characterised by enrichment in LILE, HFSE, LREE and MREE, and a slight depletion in HREE, relative to average N-MORB. REE modelling indicates that these rocks formed by partial melting of a garnet-bearing lherzolite source. Trace element and isotope systematics are consistent with an origin by decompression melting of an enriched asthenospheric mantle source.

Long-term reproducibility of multicollector Sr and Nd isotope ratio analysis☆

High-precision multidynamic analyses for Sr and Nd isotopic standards are reported for a 4-year period on the VG 354® mass spectrometer at Royal Holloway and Bedford New College (RHBNC). The data are assessed in 8 (Nd) or 11 (Sr) subsets separated by major machine maintenance. Within each subset, results for the Sr standard SRM987 show a consistent correlation between the mean triple-collector multidynamic 87Sr/86Sr result, normalised for fractionation by the power law, and the mean 86Sr/88Sr (gradient −0.020±0.007 2sd). This correlation is perfectly modelled by an exponential fractionation law calculated from exact masses. Both the La Jolla and laboratory Nd standards show consistent correlations in multidynamic data normalised by power law to 146Nd/144Nd=0.7219. The mean triple-collector 143Nd/144Nd vs. mean double-collector 142Nd/144Nd results show a positive correlation of gradient +0.176 ±0.090 (2sd), while triple-collector 145Nd/144Nd shows a negative correlation with 142Nd/144Nd, of gradient −0.046±0.040 (2sd). Fractionation-normalisation by exponential law improves the best data a little, but cannot explain the correlation, which may instead relate to minor degeneration of ion optical or collector performance. Exponential correction to the Sr data, and an empirical correction for Nd, result in mean external reproducibility (2sd) of ±0.000019 for 87Sr/86Sr and ±0.000008 for 143Nd/144Nd. Such reproducibility has important applications in Sr isotope stratigraphy, Sm-Nd geochronology and Nd isotope geochemistry. The best 142Nd/144Nd data only agree with previously published values if an exponential fractionation correction is applied. This requires the mean present-day chondrite 143Nd/144Nd to be 0.512646.

Strontium isotope profile of the early Toarcian (Jurassic) oceanic anoxic event, the duration of ammonite biozones, and belemnite palaeotemperatures

We profile 87Sr/86Sr, δ13C, δ18O, Sr/Ca, Mg/Ca, and Na/Ca in belemnites through Pliensbachian and Toarcian strata on the Yorkshire coast, UK, which include the early Jurassic oceanic anoxic event. The 87Sr/86Sr profile shows that the relative duration of ammonite subzones differ by a factor of up to 30: the Lower Jurassic exaratum subzone is 30 times longer than the clevelandicum subzone because the exaratum subzone in Yorkshire, which contains the anoxic event, is condensed by a factor of between 6.5 and 12.2 times, relative to adjacent strata. Using our 87Sr/86Sr profile, the resolution in correlation and dating attainable in the interval is between ±1.5 m and ±15 m of section, and better than 0.25 Myr. In parts of the sequence, this stratigraphic resolution equals that attainable with ammonites. A new age model is provided for late Pliensbachian and early Toarcian time that is based on the 87Sr/86Sr profile. Through the sequence, the Sr/Ca, Mg/Ca, Na/Ca and δ18O of belemnite carbonate covary, suggesting that elemental ratios may be useful for palaeotemperature measurement. Our δ13Cbelemnite data splits into three the previously reported positive isotope excursion (to +6.5‰) in the early Toarcian. We speculate that the excursion(s) resulted from addition to surface waters of isotopically heavy CO2 via ebullition of methanogenic CO2 from the sediment during early burial of organic rich (>10% TOC) sediments

Multicollector ICP-MS analysis of Pb isotopes using a 207pb-204pb double spike demonstrates up to 400 ppm/amu systematic errors in Tl-normalization

Pb isotope analyses using multicollector magnetic sector inductively coupled plasma mass spectrometers (ICP-MS) may potentially be corrected for mass bias, a major source of uncertainty, by assuming that the mass bias for Tl is similar to that of Pb. Around 7 ng of Pb has been analysed in this study using a Micromass IsoProbe MC-ICP-MS. With detailed attention to backgrounds arising from instrument memory and tailing from adjacent peaks, external precision of 55–180 ppm 2sd has been achieved over a year on Pb isotope ratios normalized to constant 208Pb/206Pb. This involves use of a wide range of SRM981 solutions and two different nebulizers. Mean ratios are very close to multidynamic TIMS data normalized in the same way, provided that anomalous TIMS behaviour of 207Pb [Chem. Geol. Isot. Geosci. Sect. 163 (2000) 299] is corrected to the least fractionated ratios measured. Tl-normalized Pb isotope ratios vary substantially between different SRM981 solutions and between nebulizers, with external precision over the year being up to 541 ppm (2sd) on 208Pb/204Pb. However, a 207pb-204pb double spike (DS) can accurately correct for mass bias on the IsoProbe, and external precision of at worst 245 ppm (2sd) on 208Pb/204Pb has been obtained over the year using a wide range of SRM981 solutions and two different nebulizers. Mean ratios on DS-corrected SRM981 and silicate samples are within error of DS-TIMS results, even using an independent calibration for the DS on the IsoProbe. A new 208Pb/206Pb value for SRM981 of 2.1677±2 is proposed, normalized to SRM982 208Pb/206Pb=1.00016. Use of SRM981 to optimize the 205Tl/203Tl normalizing ratio would give Tl-normalized sample ratios ∼400 ppm/amu higher than those calculated using the DS correction. It is shown that these problems with Tl normalization cannot result from isobaric interferences, but they appear to be due to unusual solution chemistry of Tl in the nebulizer.

Paleoceanographic changes of the Late Pliensbachian–Early Toarcian interval: a possible link to the genesis of an Oceanic Anoxic Event

Abstract Secular records of the elemental and isotopic composition of belemnite calcite were studied in Pliensbachian and Toarcian sections from the Yorkshire coast, UK, and Southern Germany, to investigate oceanographic change during an interval prior to and including the Toarcian Oceanic Anoxic Event (OAE). Records from Southern Germany are correlated to the UK stratigraphy using strontium isotope stratigraphy. The geochemical trends measured from belemnite calcite are consistent between the two sections, and are interpreted in terms of temperature and salinity of the northwest European epi-continental sea. The data suggest that a dramatic environmental change coincided with the Toarcian OAE. Belemnite Mg/Ca, Sr/Ca, and Na/Ca ratios increase by a factor of between 1.7 and 2 coincident with a 3‰ negative shift in δ18O from the mid-tenuicostatum zone until the lower falciferum zone of the UK ammonite biostratigraphy (a period of ∼0.6–0.7 Myr). Taken at face value, the Mg/Ca and δ18O data argue for an abrupt warming of 6–7°C and substantial freshening during this interval. Global warming accompanied by an accelerated hydrological cycle and increased runoff is proposed to explain these changes. Prior to these events, data from lower in the Yorkshire section suggest a possible cooling accompanied by a shift to more saline waters during the period from the upper Pliensbachian margaritatus zone to the Toarcian lower tenuicostatum zone. This earlier event may also have been important in causing density stratification in the northwest European epi-continental sea.

An assessment of mass discrimination in MC-ICPMS using Nd isotopes

The stability of mass discrimination in multiple-collector magnetic-sector inductively coupled mass spectrometry (MC-ICPMS)—with time, between elements and between samples—makes it potentially much simpler to deal with than in TIMS. However, while the stability of mass bias across limited areas of the mass spectrum is critical to the derivation of precise isotope ratios, the fundamentals of mass bias behaviour of MC-ICPMS instruments are still incompletely characterised. In this paper, we present Nd isotope data for standards and samples with the aim of using the well-known Nd isotopic system to obtain systematic information on the nature of the mass bias in MC-ICPMS. An extensive Nd dataset was obtained on a Micromass IsoProbe and more limited data on two different Nu Instrument machines. The IsoProbe data were obtained over 18 months between March 2000 and September 2001. The standard approach of using 146Nd/144Nd to normalise other Nd isotope ratios leads to both inaccurate (by around 100 ppm in the case of the exponential-law normalised 143Nd/144Nd ratio) and relatively imprecise (2 rsd=45 ppm for 143Nd/144Nd) results. On the IsoProbe, this is due to the fact that the magnitude of the exponential mass bias itself varies, albeit by a small amount, over limited mass ranges such as that for Nd. The result is that the inaccuracy is much greater for isotope ratios that have an average mass further away from that of the normalising ratio—for example, >500 ppm for 150Nd/144Nd vs. <30 ppm for 145Nd/144Nd. Both accuracy and precision increase dramatically if a normalising ratio is used that is close in mass to the ratio to be normalised. 143Nd/144Nd (average mass=143.5) normalisation with 145Nd/142Nd (average mass=143.5) yields a value identical to TIMS. An alternative approach is to use post-normalisation linear correlations between isotope ratios to do a secondary mass bias correction. Such an approach with the 143Nd/144Nd ratio using 142Nd/144Nd yields a value identical to the TIMS value and a long-term reproducibility of 14–20 ppm. This compares with a reproducibility of 45 ppm using simple normalisation to 146Nd/144Nd. We have tested both these approaches on standards and samples with 143Nd/144Nd up to 30 e units different from our in-house standard and identical results to TIMS are obtained. Post-normalisation correlations between isotope ratios obtained on the Nu Instruments MC-ICPMS are qualitatively very similar to those obtained on the IsoProbe and suggest a common cause. This, despite the very different physical characteristics of the various instruments. Furthermore, it also appears that qualitatively very similar effects, though at much smaller magnitude, are observed in TIMS. The data suggest that the quasi-empirical exponential law does not perfectly correct for mass discrimination on any mass spectrometer. This inadequacy becomes important, for precise isotope ratio analysis, when dealing with the large mass discriminations inherent in MC-ICPMS.

Resolution of the effects of crustal assimilation, sediment subduction, and fluid transport in island arc magmas: Pb Sr Nd O isotope geochemistry of Grenada, Lesser Antilles

Abstract New lead, strontium, neodymium, and oxygen isotope and chemical data are reported for forty-three samples from the picrite to low-CaO andesite differentiation series (the M-series) of Grenada, Lesser Antilles island arc. Revised isotopic data for the low-Mg, high-Ca C-series basalts show tight correlations with MgO that confirm published interpretations of C-series evolution by coupled assimilation-fractional crystallisation (AFC). δ18O determined by laser fluorination on phenocryst augite increases from mantle values (+5.6) to the highest values observed in Grenada (+6.2) over the small observed range in 143Nd/144Nd, consistent with AFC. The most magnesian C-series basalts can be generated by about 16% olivine fractionation from picrites similar in major elements to M-series picrites, but with greater large ion lithophile element (LILE) enrichment and more MORB-like isotopic ratios. The strong enrichment in LILE relative to rare earth elements (REE), and in REEs relative to Nb and Zr, implies fluid transport of these elements from subducted altered ocean crust. This is consistent with constant 87Sr/86Sr around 0.7045, but lower 143Nd/144Nd and higher Δ207Pband Th/U than MORB require a small (ca. 0.2%) additional subducted sediment contribution. Some LILE (Pb, Rb) are not strongly enriched and Ce/Pb ratios are close to those of MORB. This may be due to release of a substantial part of the slab Pb at shallower depth than the zone of magma generation. Grenada high-silica andesites (>58% SiO2) lie on extensions of C-series AFC correlations, reflecting assimilation with amphibole-dominated fractionation from C-series parents. AFC does not lead to strongly radiogenic Pb in the andesites despite an inferred contaminant With 206Pb/204Pb > 20. This is because Pb is much more incompatible than Nd due to apatite and hornblende fractionation, resulting in progressively slower 206Pb/204Pb increase as magmatic Pb contents rise. M-series picrites and basic andesites display evidence for three-component mixing on Sr Nd isotope, Nd Pb isotope, and Pb/Nd ratio— 143Nd/144Nd diagrams. One endmember has isotopic compositions similar to the uncontaminated C-series lavas, although it has less LILE enrichment from subduction fluid. Two groups of M-series picrites are recognized with lower and higher La/Y; these have respectively higher (ca. 0.51290) and lower (ca. 0.51280) 143Nd/144Nd, but similar ranges in 87Sr/86Srand206Pb/204Pb. The sample with the most extreme low 143Nd/144Nd (0.51256) has the lowest 206Pb/204Pb in the M-series: all aspects of the chemistry and isotope systematics of this sample are consistent with generation from a mantle source with a relatively high contribution (ca. 2%) from subducted local seafloor sediment (206Pb/204Pb ≈ 19.2). A progressively greater contribution from this sediment can successfully explain the neodymium isotopic shift from the C-series through the low-La/Y M-series picrites to the high-La/Y M-series picrites. The extent of fluid modification of the mantle source, as monitored by LILE enrichment relative to LREE, broadly decreases as the subducted sediment component increases. Basic andesites of the M-series also display low- and high-La/Y groups, and can be generated from the picrites by amphibole-dominated fractional crystallisation. The highest 206Pb/204Pb,87Sr/86Sr, and Pb/Nd in Grenada are found in low-La/Y basic andesites and suggest that fractionation of low-La/Y picrite was accompanied by some 6% high-level assimilation of crust with high Sr/Ndand206Pb/204Pb > 20. This hypothesis is strongly supported by the presence of 18O-enriched quartz xenocrysts of crustal origin in two of the three basic andesites. Smaller amounts (2–5%) of the same crustal component can satisfactorily model the variation of Pb/Nd,206Pb/204Pb,and87Sr/86Sr at constant 143Nd/144Nd in the picrites, an interpretation supported by correlations between isotope ratios and MgO in the high-La/Y picrites. AFC relationships in the C-series can be modelled using a similar crustal assimilant, but it must have much lower Pb/Nd ratios. The C-series assimilant cannot be M-series magma products, since M-series Pb/Nd ratios are too high, and M-series magmas themselves only achieve high 206Pb/204Pb through crustal assimilation. Lead isotope compositions of Grenada magmas, even picrites, are largely controlled by high-level crustal assimilation, mostly through AFC processes. Since these magmas include some of the most magnesian lavas found in arcs worldwide, we recommend caution in interpreting the lead isotope chemistry of arc magmas elsewhere to be the result of mantle and subduction processes.

Mantle components in Iceland and adjacent ridges investigated using double-spike Pb isotope ratios

High precision Sr-Nd isotope ratios together with Pb isotope ratios corrected for mass fractionation using a double spike are reported for an extensive suite of late Quaternary to Recent lavas of Iceland, the Kolbeinsey and Reykjanes Ridges, and a small number of basalts from further south on the Mid-Atlantic Ridge. Compared with global MORB, the Icelandic region is distinguished by having low 207Pb/204Pb for any given 206Pb/204Pb, expressed by negative [Delta]207Pb (-0.8 to -3.5) in all but four Icelandic samples. Most samples also have elevated 208Pb/204Pb (strongly positive [Delta]208Pb), which combined with their negative [Delta]207Pb is very unusual in MORB worldwide. The negative [Delta]207Pb is interpreted as a consequence of evolution in high-[mu] mantle sources for the last few hundred Ma. The region of negative [Delta]207Pb appears to correspond with the region of elevated 3He/4He, suggesting that both lithophile and volatile elements in melts from the whole region between 56 and 70[deg]N are dominantly sourced in a plume that has incorporated recycled Palaeozoic ocean crust and unradiogenic He, probably from the deep mantle. At least four mantle components are recognized on Iceland, two with an enriched character, one depleted and one that shows some isotopic affinities to EM1 but is only sampled by highly incompatible-element-depleted lavas in this study. Within restricted areas of Iceland, these components contribute to local intermediate enriched and depleted components that display near binary mixing systematics. The major depleted Icelandic component is clearly distinct in Pb isotopes from worldwide MORB, but resembles the depleted mantle source supplying the bulk of the melt to the Kolbeinsey and southern Reykjanes Ridges. However, an additional depleted mantle source is tapped by the northern Reykjanes Ridge, which with very negative [Delta]207Pb and less positive [Delta]208Pb is distinct from all Icelandic compositions. These components must mostly mix at mantle depths because a uniform mixture of three Icelandic components is advected southward along the Reykjanes Ridge.Despite strong covariation with isotope ratios, incompatible trace element ratios of Icelandic magmas cannot be representative of old mantle sources. The observed parent-daughter ratios in depleted and enriched Icelandic lavas would yield homogeneous Sr, Nd, Hf and 206Pb isotope signatures ~170 Ma ago if present in their sources. The heterogeneity in 207Pb/204Pb is not however significantly reduced at 170 Ma, and the negative present day [Delta]207Pb cannot be supported by the low [mu] observed in depleted lavas from Iceland or the adjacent ridges. Since [mu] is higher in melts than in their sources, it follows that all the depleted sources must be residues from <170 Ma partial melting events. These are thought to have strongly affected most incompatible trace element ratios.

A triple-filament method for rapid and precise analysis of rare-earth elements by isotope dilution

Abstract A method of mass spectrometric isotope dilution analysis for the poly-isotopic rare-earth elements (REE) is described, involving use of a triple-filament assembly. The method eliminates the need for chemical separation of the REE into groups, yet provides reproducibility of Sm/Nd to ∼ 0.1% and of Eu/Eu★ to ∼ 1%. All poly-isotopic REE except Lu may be analysed to high precision within 3 hr., and the method is easily fully automated on computer-controlled mass spectrometers.