Manfred J. van Bergen

Geochemical trends across an arc-continent collision zone : Magma sources and slab-wedge transfer processes below the Pantar Strait volcanoes, Indonesia

Abstract Four volcanoes in the Pantar Strait, the westernmost part of the extinct sector of the east Sunda arc, show remarkable across-arc variation in elemental abundances (K2O: 1.2 to 4.3%), trace element ratios (Pb/Ce: 0.4 to 0.18; Ce/Yb: 20 to 55) and isotope ratios (143Nd/144Nd: 0.51263 to 0.51245; 87Sr/86Sr: 0.7053 to 0.7068; 206Pb/204Pb: 19.29 to 19.15). Pb isotopes are decoupled from Sr and Nd isotopes, with the frontal volcanoes showing the higher Nd and Pb and lower Sr isotopic ratios. The isotopic and trace element ratios of the volcanic samples are best explained by modification of a MORB-type source (with Indian Ocean island basalt–type Pb isotopic characteristics) by a fluid and a partial melt of subducted continental material (SCM). The frontal volcano contains the highest proportion of the fluid component, with a small contribution of partial melt. The source of the rear-arc volcano is strongly influenced by a partial melt of SCM that had undergone a previous dehydration event, by which it lost most of its fluid-mobile elements such as Pb. The SCM partial melt was in equilibrium with both rutile and garnet, whereas mantle melting took place in the presence of residual mica. The relatively large across-arc increase in incompatible elements can be explained by a combination of increasing addition of SCM partial melt, changing mantle wedge fertility and smaller degrees of partial melting toward the rear of the arc. Comparison with a more westerly across-arc transect shows that the relatively low 143Nd/144Nd ratios of the frontal volcano, and the decoupling of Pb from Sr and Nd isotopes are unique to the Pantar Strait volcanoes. This is likely to reflect magma generation in a collisional environment, where the leading edge of the Australian continent, rather than subducted sediment, contributes to the magma source.

Dual role of seawater and hydrothermal fluids in Early Archean chert formation: Evidence from silicon isotopes

The great variety and abundance of chert deposits in Archean terrains constitute one of the most unusual features that mark Earth9s early geological history. Proposed explanations for their origin largely relying on field observations, trace element patterns, or oxygen isotope signatures have not yielded an encompassing model. Here we document silicon isotope systematics in cherts from 3.5–3.0 Ga units in the Pilbara Craton (Western Australia) as evidence of their formation by several distinct processes in Early Archean near-surface environments. Our δ 30 Si results, in combination with geochemical and mineralogical signatures and field relations, point to three end-member sources of silica derivation. One chert type is inferred to have originated through massive transformation of precursor material by silica added from sea water. At least 2‰ differences in δ 30 Si between the two other types, produced by direct chemical precipitation on the seafloor or in conduits, discriminate seawater from hydrothermal fluid as a source of silica. A virtually continuous Si isotope trend in cherts from this group is consistent with interaction between silica-carrying fluids at submarine vent systems.

Determination of sulfur isotope ratios and concentrations in water samples using ICP-MS incorporating hexapole ion optics

Sulfur isotope ratios are difficult to determine by quadrupole ICP-MS due to interfering O 2 + and NO + molecular ions of high signal intensity at isotopes 32 S and 34 S. Rf-only hexapole devices have recently been introduced into ICP-MS instrumentation to facilitate ion transfer from interface to analyser. By introducing a mixture of ‘reactive’ gases into the hexapole, a series of ion-molecule reactions can be induced to reduce or remove interfering polyatomic species. The effects of various gas mixtures (He, H 2 and Xe) on the transfer of sulfur ions through the hexapole and the breakdown of interfering O 2 + and NO + molecular ions at m/z=32 and m/z=34 were investigated. A rapid charge transfer reaction between O 2 + and Xe gives at least a factor of 10 improvement in the S + /O 2 + ratio. A further reduction in O 2 + is achieved by the addition of H 2 . δ 34 S variations were investigated in crater-lake waters and waters obtained from springs and rivers on the flanks of volcanoes in Java, Indonesia. Under optimum conditions (S=10-50 mg l –1 ), the 34 S/ 32 S measurement precision for standards and samples was <0.3% RSD. Mass bias errors were corrected by using a concentration-matched in-house standard of average North Atlantic sea-water (δ 34 S=20.5). Results compare favorably against published data measured by standard gas source mass spectrometric techniques. The proposed technique is potentially useful as a survey tool due to the large δ 34 S variation (±20) encountered in nature and the accuracy and reproducibility of the technique (±3-5).

Green clinopyroxenes and associated phases in a potassium-rich lava from the Leucite Hills, Wyoming

Green, salitic pyroxenes occur as megacrysts and as cores in diopsidic pyroxene phenocrysts and microphenocrysts in a wyomingite lava from Hatcher Mesa, Leucite Hills, Wyoming. Al-rich phlogopite (16–21% Al2O3), apatite, Fe-Ti-oxide, Mg-rich olivine (Fo93) and orthopyroxene (En61) also occur as megacrysts or as inclusions in diopside phenocrysts. All of these phases are found in ultramafic xenoliths in the host lava, and petrographic and chemical evidence is presented that the megacrysts originate by the disaggregation of the xenoliths. It is concluded that the latter are accidental fragments of the wall rocks traversed by the wyomingite magma and it is suggested that the clinopyroxene-rich xenoliths, from which the green pyroxenes are derived, formed in the upper mantle as a result of local metasomatism or by crystallization from magmas of unknown composition during an earlier igneous event. The precise role of the clinopyroxene-rich xenoliths (which also contain apatite, Fe-Ti-oxide and amphibole) in the genesis of the Leucite Hills magmas cannot be elucidated on the basis of the available data, but it is unlikely that they represent the source material from which these magmas are derived.

COMPLEX ZONING OF CLINOPYROXENES IN THE LAVAS OF VULSINI, LATIUM, ITALY: EVIDENCE FOR MAGMA MIXING

Microprobe analyses of pyroxene phenocrysts occurring in two tephritic leucitites, two leucite phonolites and one trachyte from Vulsini are reported. Three compositionally distinct types of pyroxene occur in the tephritic leucitites: (a) salite, forming resorbed cores in pyroxene phenocrysts; (b) diopside, forming euhedral-anhedral cores in pyroxene phenocrysts: and (c) pyroxene of intermediate composition to (a) and (b), which occurs as mantels around the phenocryst cores, as separate phenocrysts and as microphenocrysts. The pyroxenes in the leucite phonolites and in the trachyte do not show discontinuous zoning such as that shown by pyroxenes in the tephritic leucitites. The compositional characteristics of the latter are most easily explained by the mixing of two magmas, both tephritic leucitite in composition, one of which was relatively evolved (Fe-rich) and carried salite phenocrysts, the other relatively primitive (Mg-rich) which carried diopside phenocrysts. A review of petrographic and geochemical evidence indicates that magma mixing may have been an important process, in addition to fractional crystallization, at Vulsini and at other central Italian volcanic centres. The implications for theories about the origin of potassium-rich magmas in western Italy are briefly discussed.

Determination of silicon isotope ratios in silicate materials by high-resolution MC-ICP-MS using a sodium hydroxide sample digestion method

Silicon isotope ratios (28Si, 29Si and 30Si) can be measured with high precision by multi-collector inductively coupled plasma mass spectrometers (MC-ICP-MS). However, the problematic extraction of silicon from geological materials has been a major disadvantage in previous silicon isotope studies with conventional gas source mass spectrometry, whereas available silicon isotope results obtained by MC-ICP-MS techniques have been mainly restricted to waters and high purity silica. We show here that high yields of silicon (>97%) can be achieved from samples ranging from pure silica to basaltic compositions (45–52 wt.% SiO2) via a three-step digestion and purification procedure. Silicon isotope measurements, performed with a Finnigan Neptune MC-ICP-MS used in medium-resolution mode (resolving power: 2500), indicate that polyatomic interferences can be resolved and that both δ29Si and δ30Si can be determined with high accuracy and precision on interference-free peak plateaux in the mass spectrum. Instrumental blanks (20–65 mV) were reduced to acceptable values with a Cetac Aridus desolvating device fitted with a sapphire injector in the torch. Sensitivity in medium-resolution mode is in the range of ∼6 V per μg g−1 for 28Si. δ29Si and δ30Si have been determined for silicon isotope standards IRMM-018 (δ30Si = −1.75‰), IRMM-018-76 (δ30Si = −1.42‰), Diatomite (δ30Si = 1.34‰) and Big Batch (δ30Si = −10.52‰), for USGS standards BHVO-2 (δ30Si = −0.09‰) and AGV-2 (δ30Si = −0.01‰), and for Aldrich pure silica powder (δ30Si = −0.32‰). Precision on δ30Si is 0.18–0.41‰ (2 s.d.). Our combined procedure for sample preparation followed by high-resolution MC-ICP-MS analysis facilitates straightforward and safe measurement of silicon isotope ratios in silicate materials.

Minette inclusions in the rhyodacitic lavas of Mt-Amiata (Central Italy): mineralogical and chemical evidence of mixing between Tuscan and Roman type magmas

Abstract Major-element, trace-element and mineralogical data are reported for a series of mafic inclusions occurring in the rhyodacitic lavas of Mt. Amiata (southern Tuscany, Italy), a volcanic complex belonging to the Tuscan Magmatic Province. The inclusions form, together with some subordinate mafic latitic lavas and the host rock rhyodacites, a continuous series, which varies from olivine and minor nepheline normative to quartz, hypersthene and corundum normative rock-types over a range of 48–67% SiO 2 . All of the rocks are characterized by high contents of potassium (5–6% K 2 O) and many trace elements, particularly the incompatible elements. Following straight-line correlation trends, Al, Fe, Mg, Ca, Ti, P, Sr and Ba decrease with increasing SiO 2 , whereas Na and Li increase, and K, Rb, Zr, La, Ce, Nb and Y remain virtually constant. The primary mineral assemblage of the inclusions consists of diopsidic pyroxene, forsteritic olivine and mica, and is representative for minettes. Xenocrysts of plagioclase, sanidine, orthopyroxene clinopyroxene, biotite, ilmenite and olivine occur in increasing amounts in the more siliceous samples. The composition of these xenocrysts is identical to that of the phenocryst assemblage in the rhyodacites. The latitic lavas are rich in xenocrysts and differ from the minettes mainly by the absence of primary mica. The data strongly suggest a magma mixing origin for the minettes, latites and possibly part of the rhyodacites. Injection of mafic magma into a siliceous magma chamber has resulted in the formation of hybrid inclusions and a latitic magma that could only reach the surface after eruption of the voluminous rhyodacitic lavas. The rhyodacites which erupted early and are slightly richer in SiO 2 than most of the other rhyodacites are considered to represent the acid mixing component, whereas the mafic end-member has close affinities to the potassic alkaline lavas of the adjacent Roman Province. The inclusions have the modal composition of a minette and this implies that volatiles were important constituents of the hybrid magmas. This may be attributed to the primary character of the Roman-type end-member involved, or explained by a process of selective enrichment operating during mixing. Although upper mantle derived Roman magmas have interacted with Tuscan magmas of crustal anatectic origin at Mt. Amiata and probably other volcanic centres in central Italy, the specific chemistry of lavas in both provinces cannot be explained by simple mixing of these magmas in shallow chambers.

Sulfur-induced offsets in MC-ICP-MS silicon-isotope measurements

Sample preparation methods for MC-ICP-MS silicon-isotope measurements often involve a cation-exchange purification step. A previous study has argued that this would suffice for geological materials, as the occasional enrichment of anionic species would not compromise silicon-isotope analysis. Here we report significant offsets in MC-ICP-MS silicon-isotope measurements induced by the presence of sulfur. We show that offsets in δ30Si become significant above SO4/Si ratios (wt.) of 0.02, reaching up to ca. +1.4‰ at SO4/Si ratios above ∼0.2. This finding is particularly relevant for studies of sulfur-rich waters and silicified rocks where alteration was accompanied by sulfur-enrichments. We propose an additional purification step to remove sulfur from solid sample material.

Complex interaction of aluminous metasedimentary xenoliths and siliceous magma; an example from Mt. Amiata (Central Italy)

AbstractAluminous, silica-deficient metasedimentary xenoliths in siliceous lavas of Mt. Amiata have preserved composite zoning-patterns indicative for complex processes of magma-rock interaction. Petrographic observations and small-scale mineralogical and chemical differences between up to five distinct zones (including the core and envelope of lava) provide evidence that:1.Partial melt formed in and extracted from the xenoliths was more mafic than the host magma and had a Mg/Fe ratio higher than that of the restite, at least during the peak of thermometamorphism.2.Liquid-state interdiffusion occurred at the interface between partial melt and the enclosing magma.3.Certain mineral phases in the restite (notably, hercynitic spinel) became unstable in the presence of a siliceous liquid. Quartz-poor muscovite-biotite schists are considered to be the most likely parent rocks and it is believed that release of volatiles from decomposing micas played a significant role in the high-temperature metamorphic evolution of the xenoliths and their interaction with the magma.The conditions favourable for assimilation were enhanced by injection of mafic magma into the magma chamber. Although this concurrent operation of magma-mixing precludes a quantitative estimate of contamination from the wall-rocks (which was probably of minor importance) the present example indicates that dry acid magma may potentially become more mafic by interaction with partially melted hydrous rocks.

Common trace-element characteristics of crustal- and mantle-derived K-rich magmas at Mt. Amiata (central Italy)☆

Rare-earth and other trace-element data are reported for K-rich mafic inclusions and siliceous lavas from Mt. Amiata (southern Tuscany, Italy), which form an almost complete mixing series as a result of magmachamber replenishment. Despite a wide range of SiO2 contents (48–67 wt.%), most of the rocks are equally enriched in light REE (200–300×) and heavy REE (10–15×) relative to chondritic values, whereas intermediate REE are slightly less abundant in the lavas compared with the inclusions. Eu shows negative anomalies (EuEu★ = 0.49–0.70) that are most pronounced in the lavas. The compatible elements Cr, V, Sc and Co are found to behave like Sr, Ba and most major elements, as they show a distinct inverse correlation with SiO2, while Cs follows Na and Li, and markedly increases with SiO2. However, K and a number of other, predominantly incompatible elements, viz. Rb, Zr, La, Ce, Yb, Lu, Nb, Y, Hf, Th and U, display surprisingly little or no variation, and all of the rocks have unusually high concentrations of many of these elements. Based on arguments that mixing in the magma chamber was not accompanied by significant selective enrichment of certain elements, these chemical characteristics are attributed to the end-members which can be identified as a silica-saturated Tuscan-type magma and a primitive silica-undersaturated Roman-type magma of the high-K series. It is tentatively suggested that the different sources in the lower crust and upper mantle from which the end-member magmas are thought to originate, were metasomatized by the same fluids prior to melting.