Journal of Petrology | Volume 39 | Number 3 | Pages 469-493 | 1998
© Oxford University Press 1998
Texture–Temperature–Geochemistry Relationships in the Upper Mantle as Revealed from Spinel Peridotite Xenoliths from Wangqing, NE China
1 Department of Geology, Royal Holloway University of London Egham Hill, Egham,Surrey TW20 0ex, UK
2 Guangzhou Institute of Geochemistry, Chinese Academy of Sciences 510640 Wushan, Guangzhou, People's Republic of China
3 Département Des Sciences De La Terre, Université De La Rochelle Avenue Marillac, F-17000 La Rochelle, France
4 Institute of Geology, State Seismological Bureau 100029 Beijing, People's Republic of China
Received November 19, 1996; Revised typescript accepted October 3, 1997
 |
ABSTRACT |
|---|
 TOP ABSTRACT IntroductionSample Location and PetrographyAnalytical TechniquesResultsTemperature estimationBulk-rock major and minor...REE in clinopyroxenes and...Sr and Nd isotopes...DiscussionConcluding RemarksReferences |
|---|
Spinel peridotite xenoliths from Wangqing, NE China, exhibit correlated variations in texture, temperature, geochemistry and radiogenic isotopes. Protogranular and transitional peridotites are less refractory than equigranular samples, which are predominantly harzburgites and clinopyroxene-poor lherzolites. Deformed harzburgites and lherzolites are variably enriched in light rare earth elements (LREE), and Ce/Yb–MgO variations indicate that these olivine-rich rocks have been preferentially infiltrated by metasomatic agents. Clinopyroxenes from the fertile protogranular and transitional lherzolites have high 143Nd/144 Nd (
Nd = 10–18) and low 87Sr/86Sr (0.7020–0.7036) consistent with long-term time-integrated depletions (>1.34 Gy). By contrast, clinopyroxenes in the majority of the equigranular samples have higher 87Sr/86Sr (>0.7035) and less radiogenic 143Nd/144Nd (Nd <4.5) values. Mineral chemistry shows intra-grain and inter-grain heterogeneity, and the degree of heterogeneity decreases from protogranular to equigranular peridotites. Thermometric calculation further reveals a cooling event from >1200°C to 835–930°C at which all samples last equilibrated. Integration of these data suggests that the Wangqing xenoliths may represent fragments of the lithosphere that has been isolated from the asthenosphere for a long time (>1.34 Gy). This aged lithosphere was locally altered by asthenosphere-derived fluids, which in turn enhanced grain boundary migration and recrystallization, giving rise to coupled textural and geochemical variations. During the thermal erosion of the lithosphere that took place since the late Mesozoic, the base of lithosphere represented by the precursor rocks of the Wangqing peridotites became rheologically similar to the thermal boundary layer such that it was able to intrude with the asthenosphere diapirically the rigid uppermost lithospheric mantle at a time just before the host basalt eruption. KEY WORDS: petrography; trace element; isotope compositions; peridotite xenoliths; Eastern China
|
Introduction |
|---|
|
TOPABSTRACTIntroductionSample Location and PetrographyAnalytical TechniquesResultsTemperature estimationBulk-rock major and minor...REE in clinopyroxenes and...Sr and Nd isotopes...DiscussionConcluding RemarksReferences |
|---|
Compositional heterogeneity of the upper mantle beneath eastern China is well known and many areas, beneath crust of diverse age, have a geochemical (and geophysical) similarity to the mantle found beneath oceanic basins (e.g. E & Zhao, 1987
; Song & Frey, 1989; Tatsumoto et al., 1992; Qi et al., 1995). Based on geological evidence, and thermobarometric and isotopic data, it has been proposed that the lithospheric mantle beneath eastern China, in particular the Sino-Korean craton, represents a mixture of old lithosphere and newly accreted asthenosphere, which replaced the Archaean lithospheric keel through thermal erosion over a period of 400 my (Menzies et al., 1993; Griffin et al., 1998). Despite these achievements, more data, especially trace element and isotopic composition of mantle-derived xenoliths, are ultimately needed to better constrain the evolution of lithospheric mantle underlying this region. In addition, some of these early studies were disadvantaged as they focused on a few samples from many, widely separated volcanoes (Fan & Hooper, 1989; Tatsumoto et al., 1992) and a systematic investigation of samples from a single volcano was not undertaken. With this strategy, we undertook a petrological and geochemical investigation on a suite of spinel peridotite xenoliths from Wangqing, NE China, whereby the basin development and associated magmatism are thought to be related either to lithosphere rifting (Liu et al., 1992) or to subduction of the Pacific plate (Tian et al., 1992). Major and trace element abundances and Sr–Nd isotope compositions as well as petrographic data on these xenoliths will be used (1) to investigate any relationship between equilibrium temperature, texture and geochemistry in the mantle; (2) to identify geochemically distinct mantle domains in this region and to assess whether the mantle lithosphere is affected by possible subduction processes; and finally (3) to propose a petrogenetic model within the context of the evolutionary history of eastern China. |
Sample Location and Petrography |
|---|
|
TOPABSTRACTIntroductionSample Location and PetrographyAnalytical TechniquesResultsTemperature estimationBulk-rock major and minor...REE in clinopyroxenes and...Sr and Nd isotopes...DiscussionConcluding RemarksReferences |
|---|
The Wangqing volcanic field (K–Ar age 6.3–2.1 Ma; Liu et al., 1992
) in Jilin province is a part of the Miocene Fushun-Mishan volcanic zone and covers an area of about 2 km * 3 km (Fig. 1). The basaltic volcanism in this region is largely believed to be associated with the activation and development of the north part of the Tan-Lu fault and its deviate the Fushun-Mishan fault (Liu et al., 1992), although a relationship between basin development, magmatism and subduction of the Pacific plate was also proposed (Tian et al., 1992). Geophysical data indicate that the lithosphere in this rift zone is relatively thin (60–80 km) as a result of upwelling of the asthenosphere (Chen et al., 1991). The Wangqing basaltic flows form three different platforms, with the upper and lower layers being rich in xenoliths. All rocks are basanites (mg-number 0.65) which have undergone little crystal–liquid fractionation (Fan & Hooper, 1991; Liu et al., 1994).
|
The xenoliths range in size from 5 to >30 cm in diameter with an average size of 15 cm. Most are fresh and exhibit clear contacts with the host basalt. All studied xenoliths are anhydrous peridotites and belong to the Group I xenoliths of Frey & Prinz, (1978)
. Using the textural classification of Mercier & Nicolas, (1975), we can distinguish three textural types; these are, in order of increasing degree of deformation: (1) coarse-grained protogranular, (2) transitional and (3) tabular equigranular.- Protogranular lherzolites are characterized by a coarse grain size of
4-5 mm (Fig. 2a), occasionally up to 1 cm. Olivine and orthopyroxene (opx) grains are equant and exhibit curvilinear grain boundaries. In some samples, vermicular spinels and surrounding opxs form spinel–pyroxene clusters. The rocks belonging to this textural category are exclusively lherzolites that contain >10% clinopyroxene (cpx) (Table 1).
- Transitional samples are intermediate between those peridotites with protogranular and porphyroclastic texture (Mercier & Nicolas, 1975). These samples differ from the protogranular ones by a slightly greater degree of recrystallization (Fig. 2b). Some samples have kinked olivines and triple junctions at 120° angle. Samples of this group are predominantly spinel lherzolites.
- Tabular equigranular samples are the most abundant among the Wangqing xenolith population (>60%). They are principally lherzolite, harzburgite and dunite (Table 1). It is noted that the cpx contents in equigranular lherzolites (1–8%) are significantly lower than those in protogranular ones (Table 1). The grain size varies from 0.3 to 1 mm. Most crystals form a noticeable foliation. Grain boundaries with well-developed 120° triple junctions are common. Some samples contain rare relict unmixed opx porphyroclasts (Fig. 2c), indicating a genetic relation between porphyroclastic and equigranular rocks. Several samples (e.g. WQ91-5) are ‘banded’, comprising layers of neoblasts with different grain sizes (Fig. 2d), similar to the samples with zones of overgrown olivine observed by Cabanes & Mercier, (1988) in a xenolith suite from San Quintin (Baja California, Mexico).
|
|
WQ91-28 is a peculiar spinel harzburgite with a porphyroclastic texture. A large opx (>1.5 cm) with abundant exsolution lamellae of cpx and spinel provides evidence for significant subsolidus re-equilibration. Such a large opx is particularly suitable for an evaluation of the complex thermal history of the upper mantle (Witt & Seck, 1987
). |
Analytical Techniques |
|---|
|
TOPABSTRACTIntroductionSample Location and PetrographyAnalytical TechniquesResultsTemperature estimationBulk-rock major and minor...REE in clinopyroxenes and...Sr and Nd isotopes...DiscussionConcluding RemarksReferences |
|---|
The mineral major element chemistry data for 18 xenoliths were obtained in the Centre of Camparis at University of Paris 6, using a Cameca Camebax Microbeam. Operating conditions were the same as those reported by Xu et al., (1993)
. In situ analyses were performed with particular attention to the chemical heterogeneity in a single porphyroclast grain (core vs rim) and between porphyroclast and neoblast, especially for pyroxenes.
A subset of ten xenoliths was sawn into slabs and fresh central portions without basaltic veins were taken for analyses. Bulk abundances of major and minor elements (Cr, Ni, V, Zr, Y, Rb, Sr, Ga, Ba, Zn, Cu) were determined using the Philips PW1400 X-ray fluorescence spectrometer at Royal Holloway, University of London, on glass discs and pressed pellets. Pre-ignition data were used to determine the loss on ignition (LOI) before major element analyses. Negative LOI (as a result of oxidation of Fe2+) indicates that the absence of significant water or volatiles in the samples. Analytical uncertainties for the majority of the major elements were estimated to be <1% (between 0.2 and 0.6%). Bulk-rock trace element data [rare earth elements (REE), Sr, Y, Ba, U, Rb, Th, Pb and Zr] were obtained by inductively coupled plasma-mass spectrometry (ICP-MS) at CARE Imperial College (Silwood Park). Analytical procedures have been described elsewhere (Jarvis, 1990). Zr was determined by isotopic dilution following the technique of Thirlwall, (1982), and the abundances of Zr, Sr and Y determined by ICP-MS are virtually identical to those obtained by XRF.
Inclusion-free clinopyroxenes were hand-picked under a binocular microscope from concentrates produced by magnetic separation. The separates (100 mg) were leached with hot 6 M sub-boiled HCl for 30 min and then dissolved in distilled HF–HNO3 and 6 M HCl in Savillex screwtop Teflon beakers at 150°C for 4–10 days. Each sample was split into two aliquots: two-thirds was used for Sr and Nd isotope composition and one-third, which was mixed with a dilute REE spike, was used for REE determination (Thirlwall, 1982). Sr, Nd and REE were separated using a standard ion-exchange technique. A blank was processed with each set of five samples. Samples were loaded onto single Ta beads for Sr, single Re beads for Nd and triple Ta–Re–Ta beads for REE. The isotopic and REE analyses were performed on a VG354 five-collector mass spectrometer at the University of London Isotope Laboratory, using experimental procedures described previously by Thirlwall, (1991a, 1991b). The Sr and Nd blanks during the period of analyses are 0.5 ng and 0.3 ng, respectively. Analyses of standards during the period of analysis (1994–1995) are as follows: SRM987 gave 87Sr/86Sr = 0.710244 ± 18 (2 SD, n = 78) and low Aldrich gave 143Nd/144Nd = 0.512420 ± 9 (2 SD, n = 23). The isotopic ratios measured are normalized to 86Sr/88Sr = 0.1194 and 146Nd/144Nd = 0.7219.
|
Results |
|---|
|
TOPABSTRACTIntroductionSample Location and PetrographyAnalytical TechniquesResultsTemperature estimationBulk-rock major and minor...REE in clinopyroxenes and...Sr and Nd isotopes...DiscussionConcluding RemarksReferences |
|---|
Mineral chemistry
Representative electron microprobe analyses of four constituent phases of xenoliths of different textural types are listed in Table 2. A full list is available from the senior author upon request.
|
Olivine
The compositions of olivine correlate, to some degree, with texture (Table 2a). Olivines from protogranular lherzolites display a limited range in Mg/(Mg + Fe) (mg-number 0.893–0.898), which falls within the ‘primitive mantle’ range (Glücklich & Mercier, 1989
). In contrast, olivines from equigranular samples vary in mg-number (0.892–0.926), with most being between 0.90 and 0.926. In the same thin section, porphyroclast and neoblast olivines have nearly identical compositions. No significant zoning was found in any of the olivines.
Spinel
Spinels in protogranular samples have a Cr/(Cr + Al) (cr-number) ratio of 0.088–0.12, with mg-numbers of 0.808–0.837, outlining again their fertile character, whereas spinels in equigranular samples tend to be depleted with typical values of cr-number >0.3 (Table 2b).
|
Orthopyroxene
Orthopyroxenes in protogranular and transitional samples show variations in composition depending on their textural position (porphyroclast or neoblast) and distance to the grain boundary (core vs rim). Four different opx compositions can be distinguished, and are best exemplified by sample WQ91-28 (Table 2c) (compare Witt & Seck, 1987
). It is assumed that the unmixed opx comprises essentially cpx lamellae and that the later recrystallization process did not significantly affect the bulk chemical composition. The primary composition of the pre-exsolution opx can be obtained by applying scanning techniques with an electron microprobe (Mercier et al., 1984; Witt & Seck, 1987). Thus, the reconstructed composition of this opx is characterized by significantly high Ca, Cr and Al contents relative to those of the neoblasts and exsolved hosts, reflecting high temperature before exsolution (Witt & Seck, 1987). Both exsolved and exsolution-free opx porphyroclasts exhibit considerable chemical zoning within each grain, with higher Cr, Al and Ca contents in the cores than in the rims. Neoblasts of opx have a composition similar to that of the porphyroclast rims. Such chemical zoning patterns are illustrated in a correlation diagram of the Al content of opx and the Cr of coexisting spinel (Fig. 3a). The variations of Al content within individual grains and between grains are presented as vertical tie lines. The resulting eye-fitted curve is similar to that for the xenoliths from the Massif Central, France (Brown et al., 1980). Orthopyroxenes from equigranular samples are also compositionally zoned, but with a more limited compositional range compared with that of the protogranular and transitional samples (Fig. 3a). The exception is WQ91-28 because it contains an exceptionally large opx.
|
|
) xenoliths (except for WQ91-28) are smaller than those in protogranular (
) and transitional (
) peridotites.Clinopyroxene
Like opx, cpx in protogranular and transitional samples has a wider range in composition than that from tabular equigranular samples. The Al content in cpx correlates with the Cr/(Cr + Al) of the coexisting spinel (Fig. 3b). As observed by Brown et al., (1980)
for the xenoliths from the French Massif Central, Al in cpx decreases with increasing Cr/(Cr + Al) ratio of spinel more rapidly than Al in the coexisting opx. This is consistent with the experimental observation that cpx is consumed before other phases during partial melting (Mysen & Kushiro, 1977). However, the cpxs in three samples (WQ91-22, WQ91-28 and WQ91-77) are anomalously enriched in Al relative to the coexisting spinel composition and they deviate conspicuously from the general variation pattern (Fig. 3b). In summary, mineral chemistry in the Wangqing peridotites can be broadly correlated with textural variation. Although there is some overlap between protogranular and transitional peridotites, it is clear that these samples are less refractory than the equigranular ones. On the other hand, the degree of chemical equilibrium in the peridotites seems to increase from protogranular via transitional to equigranular samples.
|
Temperature estimation |
|---|
|
TOPABSTRACTIntroductionSample Location and PetrographyAnalytical TechniquesResultsTemperature estimationBulk-rock major and minor...REE in clinopyroxenes and...Sr and Nd isotopes...DiscussionConcluding RemarksReferences |
|---|
Temperatures (Table 3) were estimated using the two-pyroxene and Ca-in-opx thermometers of Brey & Köhler, (1990)
, and the Cr–Al-opx thermometer of Witt-Eickschen & Seck, (1991). For a given fully equilibrated sample, these thermometers should yield similar estimates (Witt-Eickschen & Seck, 1991). Simultaneous application of these different calibrations therefore allows us to test the validity of such estimates. Given the ubiquitous chemical zoning in pyroxenes in the Wangqing peridotites, we have performed temperature calculations with porphyroclast–neoblast and core–rim distinctions for each sample. Figure 4a shows the histogram of temperatures calculated applying the Ca-in-opx thermometer. These data are broadly consistent with those estimated by the Cr–Al-opx thermometer (Table 3). The core compositions of opx porphyroclasts in transitional and protogranular rocks yield a temperature interval of 967–1063°C. In contrast, temperatures calculated from the compositions of porphyroclast rims and neoblasts for these two textural types are much lower (840–911°C). Such a temperature variation is best seen in WQ91-28, for which detailed core–rim traverse analyses on a large opx porphyroclast and neoblasts have been obtained (Fig. 4b).
|
|
Compared with this relatively large core–rim temperature range (1063–840°C) for the protogranular and transitional peridotites, the range for the equigranular samples is narrow (950–835°C). It is interesting to note that the temperature range (835–933°C) constrained by rim compositions of the equigranular samples is very similar to that for the neoblasts and porphyroclast rims in protogranular and transitional samples. Rare relict opx porphyroclasts in some equigranular samples yield temperatures as high as those obtained for porphyroclasts in protogranular and transitional lherzolites (Fig. 4a). An even higher temperature (1200°C) is estimated based on the reconstructed primary core composition before unmixing (for sample WQ91-28). All of these observations lead us to believe that the precursor rocks for equigranular samples underwent deformation and cooling similar to that observed in the protogranular and transitional peridotites. This happened before the peridotites developed their present texture and thermal state.
However, the two-pyroxene thermometer does not produce a bimodal distribution of temperature as revealed by both Ca-in-opx and Cr–Al-opx thermometers (Table 3; Fig. 4). For instance, the Ca-in-opx thermometer gives core-rim temperatures of 1063–866°C for WQ91-8, whereas the two-pyroxene thermometer yields a much smaller temperature interval (885–845°C; Table 3). The reason for these inconsistent estimates is that the cores of cpx and opx porphyroclasts do not always retain compositions corresponding to the same thermal state, whereas the application of two-pyroxene thermometers requires complete equilibrium between coexisting minerals. Cpx in the Wangqing xenoliths recrystallized to greater degrees than coexisting orthopyroxenes given the rare cpx porphyroclasts and smaller sizes of cpx relative to opx. It is possible that cpx may compositionally re-equilibrate to new temperatures, whereas opx cores still partly retain the composition corresponding to the initial temperature before cooling. This interpretation is supported by the similarity between two-pyroxene temperatures and the rim temperatures obtained by the Ca-in-opx thermometer (Table 3), because the former thermometer largely depends on cpx compositions (Brey & Köhler, 1990). These data emphasize that two-pyroxene thermometers are not applicable unless chemical equilibrium between corresponding minerals (core vs core, rim vs rim) is demonstrated.
|
Bulk-rock major and minor element compositions |
|---|
|
TOPABSTRACTIntroductionSample Location and PetrographyAnalytical TechniquesResultsTemperature estimationBulk-rock major and minor...REE in clinopyroxenes and...Sr and Nd isotopes...DiscussionConcluding RemarksReferences |
|---|
The Wangqing peridotites display a wide compositional range in major elements (Table 4). This overlaps with the range already defined for Chinese xenoliths (E & Zhao, 1987
; Song & Frey, 1989; Qi et al., 1995). In Fig. 5, major element compositions are plotted against MgO, which is a good indicator of the degree of depletion in mantle peridotites. The correlations between MgO and modal contents of cpx and opx indicate that bulk-rock chemistry is controlled by modal phase compositions. Among the samples analysed, WQ91-37 shows the lowest MgO and highest CaO and TiO 2 contents, similar to primitive mantle values (Jagoutz et al., 1979; Frey et al., 1985; Hart & Zindler, 1986; McDonough, 1990). CaO, Al2O3 , TiO2 and SiO2 are negatively correlated with MgO (Fig. 5), and define similar trends to those of spinel peridotites from other world-wide occurrences (e.g. Frey et al., 1985; McDonough, 1990). TiO2 content in WQ91-22 deviates from the common trend because of the presence of ilmenite in this sample. The poor correlation between MgO and Na 2 O is largely due to an anomalously high concentration of Na2 O in two protogranular lherzolites. The variation in bulk composition correlates with textural variation, with protogranular and transitional lherzolites being less depleted than deformed equigranular harzburgites or lherzolites.
|
|
Moderately incompatible elements [e.g. V, Sc, Ga, Cu, Zn, heavy REE (HREE)] are also negatively correlated with MgO (Fig. 6). Ni is positively correlated with MgO, consistent with its compatible character during melt-peridotite equilibrium. However, highly incompatible elements (e.g. Sr, Zr, Ba) do not show a linear correlation with MgO (Fig. 6). For example, all the protogranular and transitional samples, as well as one equigranular peridotite (CaO >2%), define a negative correlation between MgO and Sr, whereas a positive trend is observed for the rest of the equigranular samples (CaO <2%).
|
|
REE in clinopyroxenes and trace elements in whole rocks |
|---|
|
TOPABSTRACTIntroductionSample Location and PetrographyAnalytical TechniquesResultsTemperature estimationBulk-rock major and minor...REE in clinopyroxenes and...Sr and Nd isotopes...DiscussionConcluding RemarksReferences |
|---|
Bulk-rock trace element data and REE contents in cpx are given in Tables 4 and 5 and illustrated in Figs 7 and 8. For most samples, the chondrite-normalized REE patterns of cpx are essentially similar to those in bulk rocks, but at higher concentrations. This is consistent with the major role of cpx in controlling REE distribution in anhydrous peridotites (McDonough et al., 1992
). The cpx in WQ91-77 and WQ91-21 shows light REE (LREE) depletion patterns, whereas their bulk rocks display less pronounced depletions in LREE or relatively flat patterns. This indicates the presence of some intergranular LREE-enriched components. There are three distinct patterns, which are broadly related to different texture types and major element compositions (Fig. 7).
|
|
|
- LREE-depleted patterns [(La/Yb)n, cpx = 0.3–0.7] are found for two protogranular samples and one equigranular (WQ91-21) sample that is less depleted in basaltic elements (CaO >2%).
- Spoon-shaped REE patterns [(La/Sm)n, cpx = 1.1–1.3] are observed in two transitional samples (WQ91-6 and WQ91-11). They differ from the LREE-depleted patterns in that an upward inflection occurs for La to Ce. Similar REE distribution patterns have been observed in both oceanic and continental xenoliths (Menzies et al., 1985; Stosch et al., 1986; Downes & Dupuy, 1987; Kempton, 1987; Song & Frey, 1989). The REE contents in samples with this type of pattern and with LREE-depleted patterns are negatively correlated with MgO (Fig. 6), suggesting that the abundances of these elements were not significantly modified by late-stage metasomatism.
- LREE-enriched patterns [(La/Yb)n, cpx = 2.5–21) are uniquely found in equigranular harzburgites and cpx-poor lherzolites that are highly depleted in basaltic compositions (CaO <2%). These samples have low HREE abundances relative to those with LREE-depleted patterns, suggesting that these patterns are linked to the LREE enrichment in originally REE-depleted samples. Specifically, cpx in WQ91-5 exhibits a fractionated pattern for the LREE and middle REE (MREE) but an almost flat HREE pattern. This probably indicates that metasomatism has affected concentrations of the LREE and MREE, but not the HREE. However, this is not the case for WQ91-22, which shows strongly fractionated pattern from La to Lu (Fig. 7).
The primitive mantle-normalized trace element distributions in the bulk xenoliths from Wangqing are complex (Fig. 8). They are very different from those in their host basalts (Fig. 8a). The basalts have minor positive Sr and Zr anomalies in basalts, whereas the xenoliths typically have very low contents of Nb and Ba and show negative or no Zr anomalies. It therefore can be concluded that trace element compositions of the bulk peridotites are not significantly affected by host basalts. However, the highly incompatible elements, particularly those mobile in sub-surface environments (U, Rb) can be seriously affected by post-eruption processes (Ionov et al., 1995). Most samples in Fig. 8 show minor enrichment in La, which is widely believed to be immobile in surface processes. The progressive enrichment from Ce to U, Th and Rb in WQ91-11 and WQ91-6 thus suggests that relatively high abundances of U and Th in these two samples are not results of secondary alterations. This is consistent with the absence of serpentinization in these samples.
The LREE-depleted whole-rock samples show ubiquitous La inflections. This is not seen for Nb, which follows a coherent depletion pattern from Lu to Ba (Fig. 8a). The fertile samples show small or no Zr and Ti anomalies. In contrast, pronounced Zr–Ti depletions are seen in HREE-poor peridotites irrespective of LREE enrichment (Fig. 8). Two harzburgites (WQ91-5 and WQ91-13) present marked Y anomalies, but the cause for these Y anomalies is unknown.
|
Sr and Nd isotopes in clinopyroxenes |
|---|
|
TOPABSTRACTIntroductionSample Location and PetrographyAnalytical TechniquesResultsTemperature estimationBulk-rock major and minor...REE in clinopyroxenes and...Sr and Nd isotopes...DiscussionConcluding RemarksReferences |
|---|
Cpx separates display a wide isotopic variation, with 87Sr/86Sr varying over a range of 0.7020–0.7050 and 143Nd/144Nd over a range of 0.5127–0.5136 (
Nd = + 18.3 to + 2.8; Table 6). This range is typical of suites of spinel peridotite xenoliths from single localities (Menzies et al., 1985; Stosch & Lugmair, 1986; Stosch et al., 1986; McDonough & McCulloch, 1987; Roden et al., 1988; Ionov et al., 1992). Consistent with previous results (W. M. Fan, unpublished data, 1992), these values suggest that there is no genetic link between peridotite xenoliths and the host basalts, as the host basalts have a very different, more restricted isotopic composition (Fig. 9; Zhou & Zhu, 1992).
|
|
Sr and Nd isotopic ratios also correlate with textural variation and the refractory character of the rock (Fig. 10), compatible with the correlation between chemistry (major and trace element) and texture. On an Sr–Nd isotope diagram, cpxs from protogranular and transitional samples fall within or above the field of mid-ocean ridge basalts (MORB) with high
Nd >9.8 and low 87Sr/86Sr <0.7036. For example, the Nd values (14.8–18.3) of WQ91-1 and WQ91-77 are higher than those measured in MORB or abyssal peridotite samples and are similar to those proposed for depleted mid-ocean ridge mantle isotopic end-member (DMM) (Zindler & Hart, 1986). WQ91-11 has a similarly high 143Nd/144Nd ratio (0.5133) but a higher 87Sr/86Sr ratio compared with the mantle array. Similar isotopic relationships have been observed in independent analyses on xenoliths from the same locality (W. M. Fan, unpublished data, 1992). In contrast, four out of five equigranular and refractory samples (CaO <2%) display much higher 87Sr/86Sr (>0.7033) and lower 143Nd/144Nd (Nd <4.5). Compared with the protogranular and transitional lherzolites, these LREE-enriched samples show much less scatter in Sr–Nd isotopic composition (Fig. 9). However, the least depleted equigranular lherzolite (WQ91-21) has an isotopic composition similar to that of the protogranular and transitional samples. This matches the LREE-depleted pattern of this sample, suggestive of the original character of the equigranular samples.
|
|
Discussion |
|---|
|
TOPABSTRACTIntroductionSample Location and PetrographyAnalytical TechniquesResultsTemperature estimationBulk-rock major and minor...REE in clinopyroxenes and...Sr and Nd isotopes...DiscussionConcluding RemarksReferences |
|---|
Geochemical variations during depletion and enrichment processes
Nature and timing of depletion events
The Wangqing peridotites range from fertile lherzolites to cpx-poor lherzolites to harzburgites. Basaltic elements (CaO, Al2O3 , TiO2 , Na2 O) and moderately incompatible elements (Sc, Yb) are negatively correlated with MgO (Figs 5 and 6). Similarities between these trends and those defined by the xenoliths from world-wide localities indicate that the Wangqing peridotites represent the refractory residues left after extraction of basalts of variable degrees (e.g. Frey & Prinz, 1978
; Frey et al., 1985; Kempton, 1987; McDonough, 1990). Such a depletion event is also evident from normalized REE and trace element distribution patterns (Figs 7 and 8). For instance, the samples (e.g. WQ91-1, WQ91-37) show stronger depletions in highly incompatible elements (Ba, Nb, La) than in moderately incompatible elements (HREE), as expected for a melt extraction process. However, WQ91-11 and WQ91-21 have negative Zr and Ti anomalies relative to the adjacent REE of similar compatibility. These anomalies may be a pre-enrichment feature of the xenoliths, as metasomatism has only affected highly incompatible elements rather than moderately incompatible REE in these samples (Fig. 8a and b). Both WQ91-21 and WQ91-11 show the lowest HREE contents among LREE-depleted samples (Fig. 8a and b) suggesting that the pronounced Zr–(Ti-) depletion is restricted to cpx-poor peridotites, i.e. those which experienced highest degrees of partial melting. Jochum et al., (1989) reported no Zr–Ti anomalies in fertile spinel peridotite xenoliths. McDonough et al., (1992) suggested that positive Ti and Zr anomalies in opx can compensate for the negative anomalies in cpx, resulting in no high field strength element (HFSE) anomalies in bulk rocks. However, new ICP-MS analyses on xenoliths from Sikhote-Alin by Ionov et al., (1995) found no or only slight positive Zr anomaly in opx. As some Wangqing xenoliths represent mantle fragments that did not suffer any significant metasomatic alteration, they may retain important information about the age of depletion. In Fig. 11, these LREE-depleted peridotites do not define a simple isochron. Given the correlations shown in Fig. 5, the Wangqing samples may be originally derived from a geochemically uniform reservoir. In this sense, the non-isochron correlation may suggest that the Wangqing samples have experienced more than one depletion event. The calculated Nd model age for these LREE-depleted samples, using a single-stage model from a primitive mantle source, is between 1.36 and 2.5 Ga (Table 5). This indicates a series of depletion episodes in the mantle beneath Wangqing in the Proterozoic. However, it is equally possible that they were derived from a heterogeneous mantle that is compositionally zoned from primitive mantle to depleted mantle (DM). This latter interpretation is particularly attractive because the isotope data for protogranular and transitional peridotites (except WQ91-37) define a line passing through the present-day MORB. Linear regression of the data gives an age of 1.34 ± 0.05 Ga, which is very similar to Bulk-Earth model age (1.36 Ga) of the LREE-depleted equigranular WQ91-21. WQ91-37 and WQ91-77 define a line passing through the Bulk Earth corresponding to an age of 1.89 Ga. These ages can be related to the major orogenic events at this region. Given the large degrees of depletion experienced by the Wangqing xenoliths, we prefer the second interpretation.
|
The Sm–Nd ‘isochron’ line defined by protogranular and transitional peridotites can be alternatively interpreted as a result of mixing, given the trace element evidence for introduction of an LREE-rich component in some transitional peridotites (Fig. 6). However, the lack of linear correlation between 143Nd/144Nd and 1/Nd argues against this mixing model. Moreover, the cpx in transitional lherzolites has Sm/Nd ratios higher than that of the primitive mantle (0.325), suggesting that the enrichment has only affected La and Ce but probably not Sm and Nd. This is partly supported by lack of correlation between 143Nd/144Nd and 147Sm/144Nd for four equigranular samples which show variable enrichment of LREE (including Sm and Nd).
Nature of enrichment processes
The LREE in most equigranular samples are positively correlated with MgO (Fig. 6). Specifically, sample WQ91-5, with the highest (La/Yb)n (>6), has the lowest CaO (<1%) and the highest mg-number (>0.923). Moreover, most of these equigranular samples show enrichment of La and other highly incompatible elements (Rb, U, Th; Fig. 8). These observations are not consistent with an origin as partial melting residues, but require a two-stage model (Frey & Green, 1974) involving partial melting followed by metasomatic events.
For WQ91-11 and WQ91-6, a fluid rich in LREE (in particular La), Rb, Th and U, but low in Nb and Ba has to be invoked. Three of four LREE-enriched equigranular peridotites have negative Ti–Zr anomalies (Fig. 8). Similar to the observations of Ionov et al., (1995) on Sikhote-Alin xenoliths, Zr and Ti do not always behave coherently with Nb. For example, there are both Nb and Zr–Ti anomalies in WQ91-20, whereas WQ91-5 and WQ91-13 are very low in Ti and Zr but show no anomalies for Nb. WQ91-22 shows no Zr-Ti anomalies but has a high Ba content contrasting with low Ba abundances in the other three equigranular samples (Fig. 8c). The diversity of these enriched patterns may reflect multiple metasomatic events involving different metasomatic agents. For WQ91-5, WQ91-13 and WQ91-20, the metasomatic agent is rich in LREE and other highly incompatible elements but does not carry significant Ba, Ti and Zr. Taking into account the experimental results of Schneider & Eggler, (1986) and the ICP-MS analyses of rare carbonate-bearing peridotites (Ionov et al., 1993), we argue that this metasomatic agent may be a fluid or carbonate melt rather than a silicate melt. This inference is consistent with the fact that the LREE-enriched peridotites show no evidence of reheating, which would be expected in case of interaction with silicate melts, in particular with a mechanism of percolation-controlled chromatographic metasomatism as indicated by spoon-shaped patterns in cpxs from some transitional peridotites (Navon & Stolper, 1987). By contrast, the metasomatic agent responsible for LREE enrichment in WQ91-22 may be compositionally similar to host basalts, although its exact nature is difficult to identify.
It should be emphasized that pronounced LREE enrichment is restricted to the cpx-poor samples (CaO <2%), as also observed for the peridotite xenoliths from world-wide occurrences (Frey & Prinz, 1978; Menzies et al., 1985; Stosch et al., 1986; Downes & Dupuy, 1987; Kempton, 1987; Chen et al., 1989; McDonough & Frey, 1989; Song & Frey, 1989; Qi et al., 1995). Moreover, the observation that (Ce/Yb)n, cpx increases with increasing MgO (Fig. 10) requires that the largest amounts of melt penetrated into mostly depleted peridotites. This means that a highly LREE-rich fluid preferentially reacts with refractory and deformed samples. The potential for this selective metasomatism is supported by the experimental evidence (Toramaru & Fujii, 1986) that connectivity of a melt phase through grain-boundaries may be favoured in olivine-rich peridotites relative to adjacent olivine-poor peridotites and pyroxenites, and by the theoretical consideration (Vasseur et al., 1991) that the fluid transport and metasomatic processes at mantle level are proportional to the deformation, and increase with decreasing grain-size.
Metasomatic enrichment is also reflected by isotopic ratios, as the LREE-enriched samples are characterized by relatively high 87Sr/86Sr and low 143Nd/144Nd ratios (Figs 9 and 10). However, the majority of the LREE-enriched cpxs still have depleted Sr-Nd isotopic signatures (Fig. 9). The lack of correlation between Sm/Nd and 143Nd/144Nd and the small variation of Nd of the equigranular peridotites are similar to those observed for hydrous peridotites from West Eifel (Stosch & Lugmair, 1986). It is possible that the LREE-enrichment events occurred recently such that insufficient time has elapsed for Sm–Nd system to evolve to form the positive correlation similar to that observed for LREE-depleted samples. Three mantle components may be invoked to explain the isotopic variations displayed by the Wangqing peridotites:
- Depleted MORB-source Mantle (DMM), which is represented by the LREE-depleted protogranular and some transitional xenoliths (WQ91-1, WQ91-77). These samples might represent the mantle composition before metasomatic enrichment.
- An enriched component with relatively low 87Sr/86Sr and low Nd that is EM1-like (Basu et al., 1991; Tatsumoto et al., 1992). This enriched component may be related to the pyroxenites at Wangqing or their parental melts, which are isotopically similar to BSE (W. M. Fan, unpublished data, 1992). The mixing of DMM and EM1 can explain the isotopic spectrum displayed by most of the peridotites (Fig. 9).
- A second enriched component characterized by relatively high 87Sr/86Sr similar to EM2 is required for the samples (i.e. WQ91-11, WQ91-20; Fig. 9) that plot elsewhere. Such an EM2 component has been suggested by Tatsumoto et al., (1992) based on the Sr–Nd–Pb isotopic systematics of xenoliths and basalts from several localities in eastern China.
Texture–temperature relationship: a diapir model
Based on studies of peridotite xenoliths from young volcanic regions of the Massif Central, Brown et al., (1980) indicated that protogranular xenoliths commonly have higher temperatures than equigranular xenoliths, with the porphyroclastic group having transitional characteristics. A similar texture-temperature relationship has also been observed in peridotite xenoliths from the Rhenish Massif, Germany (Witt-Eickschen & Seck, 1991). These observations imply a partly stratified upper mantle, with deformed and depleted peridotites overlying undeformed and fertile ones. This idea finds supports from ocean ridge melting models (McKenzie, 1985). However, the data presented in this study (Table 3, Fig. 4) suggest that all textural groups including both undeformed and deformed xenoliths last equilibrated at similar temperatures (835–930°C), implying that they were situated at similar depths at the time of basaltic eruption.
All the Wangqing xenoliths display similar chemical zoning patterns in pyroxenes with decreasing Cr, Al and Ca contents from core to rim. According to Witt & Seck, (1987), this may be the result of a cooling event. Application of the Ca-in-opx thermometer of Brey & Köhler, (1990) to protogranular and transitional samples yields porphyroclast core-rim temperatures of 1063–840°C (Table 3). Similar high temperatures are obtained for rare relict opx in equigranular peridotites, suggesting that all the Wangqing peridotites have experienced a similar cooling event from >1100°C to 835–930°C. The difference is that the re-equilibration was almost complete for the equigranular peridotites, probably because of assistance of deformation, but not for the protogranular ones.
Diapiric intrusion (Nicolas et al., 1987; Witt & Seck, 1987) would be a reasonable geodynamic model to account for this cooling. It involves an uplift of a hotter lower part of the mantle into a relatively colder uppermost mantle and subsequent thermal re-equilibration between these two thermally different units. The remaining chemical zonations in pyroxenes in turn suggest a relatively short thermal relaxation period undergone by the Wangqing xenoliths before they were transported to the surface, because diffusion would smooth out such heterogeneities on time scales of less than 100 ky at mantle temperature (Witt & Seck, 1987). The decreasing degree of within-grain heterogeneity from protogranular to equigranular samples (Fig. 3a and b) would thus reflect the distinct sites occupied by different textural peridotites within a diapir. It is likely that the undeformed protogranular samples would have been situated in the interior of the diapiric body. Their grain cores commonly retain information on the temperature before diapiric upwelling and only the rims of crystals have compositionally readjusted to low temperatures. On the other hand, the equigranular samples would have been localized within the flanks of the diapiric body or alternatively they may represent the uppermost mantle itself. In the first case, the deformation is associated with the diapiric emplacement. In the second case, the deformation existed before diapiric intrusion, and these deformed rocks were preferentially enriched in incompatible elements during metasomatic events that happened some time earlier than the age of emplacement. It is difficult to choose between these two alternatives, given the presence of both LREE-enriched and LREE-depleted equigranular xenoliths. We thus suggest that they came from both diapiric flanks and adjacent pre-existing lithospheric mantle.
Such a model encounters some difficulties with respect to isotopic compositions. A recently accreted asthenosphere to the lithosphere would imply an Sr–Nd isotopic composition typical of MORB source for the diapir body (e.g. Stosch et al., 1986). However, two samples of the Wangqing xenoliths have Nd values (14.8, 18.3) significantly higher than that of MORB. Such high Nd values are commonly interpreted as a result of isolation from the convecting mantle on the time scale of a few billion years (e.g. Ionov et al., 1992; Table 6) and thus argue against an asthenospheric origin for all the Wangqing xenoliths. In fact, the highest temperature recorded in an opx porphyroclast core is 1200°C, which is slightly lower than the potential temperature of normal asthenosphere (1280°C, McKenzie & Bickle, 1988). As discussed above, all the Wangqing peridotites may have experienced a similar cooling event. We thus infer that the precursor rocks of the Wangqing xenoliths (at least of some protogranular and transitional lherzolites) may have been located at the base of the lithosphere. These considerations, combined with the accumulating evidence for extensive thermal erosion of the lithosphere (i.e. loss of >120 km of Archaean lithospheric keel) in eastern China during the last 400 my (Menzies et al., 1993; Griffin et al., 1998), lead us to propose the following petrogenetic model. Most Wangqing xenoliths may represent fragments of the lithosphere that have been isolated from the asthenosphere for a long time (>1.34 Gy). This has resulted in high time-integrated Nd values. With continuing upwelling of the asthenosphere, this old lithosphere was thermally eroded and thinned. During this process, the base of the lithosphere could become rheologically similar to the thermal boundary layer, allowing it to intrude with the asthenosphere diapirically the more rigid uppermost lithosphere mantle at a time just before the host basalt eruptions. On the other hand, its time-integrated Sr–Nd isotopic signature has been preserved. This old lithosphere has repeatedly been infiltrated by small-degree melt fractions or fluids that were produced by decompression melting before or during asthenosphere upwelling. This process resulted in the enrichment of incompatible elements in peridotites. This enrichment was preferentially confined to olivine-rich rocks and deformation zones. As a result of feedback processes, the presence of a melt or fluid film at grain boundaries enhanced grain boundary migration and recrystallization, resulting in coupled textural and geochemical variations.
Texture–geochemistry relationship: a comparison with other xenoliths
At Wangqing, the fertile protogranular peridotites are largely immune to metasomatism, whereas the depleted tabular equigranular ones were affected to various degrees and enriched in incompatible elements. Similar texture–chemistry correlation has been reported for peridotite xenoliths from many localities in the world (Downes & Dupuy, 1987; Downes, 1990; Downes et al., 1992; Bedini et al., 1997). Downes, (1990) has attributed such a relation to metasomatism linked to deformation. This raises the question of whether metasomatism and deformation take place simultaneously. Timing of these events is difficult to constrain. Xu et al., (1996) have conducted a detailed investigation of trace element and O-isotope compositions on sheared (mylonitic, grain size <0.1 mm) and granular peridotites (including protogranular to equigranular, grain size >1 mm) from Miocene alkali basalts in Yitong, which is situated 200 km west of Wangqing (Fig. 1). They found that the sheared samples display LREE-depleted patterns, whereas some granular samples are strongly enriched in LREE. Similarly, one equigranular sample (WQ91-21) from Wangqing is depleted in LREE in contrast to the LREE-enriched patterns exhibited by the majority of equigranular peridotites. These observations are not consistent with preferential LREE enrichment in deformed peridotites. This paradox can be reconciled if the presence of fluid is not necessary for plastic deformation and the associated LREE enrichment in deformed samples is a consequence of preferential circulation of fluids along deformed zones (Xu et al., 1996). The mylonitic texture in some Yitong xenoliths has been attributed to the existence of a major lithospheric shear zone (Xu et al., 1993), which is not the case for Wangqing. Diffusion data and recrystallization mechanism all suggest that this shearing is a recent event. The low equilibration temperature (750–800°C) further indicates a location of these sheared peridotites in the uppermost part of the mantle just beneath the Moho discontinuity (Xu et al., 1993). It is therefore possible that the metasomatic melts or fluids coming from depth may not have reached that level. Compared with these shallow sheared xenoliths, the Wangqing xenoliths represent fragments of a deeper lithospheric mantle which has been thermally and chemically affected by upwelling asthenosphere and its derived melts or fluids. Similar enrichment is indeed observed in high-temperature (900–1000°C) granular peridotites from Yitong.
As mentioned above, the LREE enrichment preferentially occurs in CaO-poor peridotites at Wangqing. The LREE-depleted sample WQ91-21 is not different from the other LREE-enriched equigranular samples in terms of texture and equilibrium temperature, but it has relatively high CaO content (2.1%, Table 4). Combined with high CaO contents (>2%) in LREE-depleted protogranular and transitional peridotites, this would imply that chemical composition plays a more important role than texture in preferential LREE-enrichment processes. This can partly account for the LREE depletion in highly sheared peridotites from Yitong because they are dominantly fertile lherzolites (Xu et al., 1993).
Provenance of mantle domains beneath Wangqing
The development of the Mesozoic–Cenozoic rift system and associated magmatism in eastern China was believed to be related to the subduction of the Kula Plate in Jurassic–Cretaceous time and to the later subduction of the Pacific plate (e.g. Tian et al., 1992). Given the relatively large distance between the active island arc of Japan and NE China (>1000 km), it was proposed that the earlier phase involved a shallowly dipping Benioff zone to allow the subducted slab to reach the mantle underlying NE China. In this tectonic model, it would be expected that the thermal and chemical structure of the lithospheric mantle beneath N and NE China may have been affected by subduction-related processes. WQ91-11 plots above the mantle array because of the relatively high 87Sr/86Sr at given 143Nd/144Nd. Similar isotopic signatures are also observed on mantle xenoliths from other localities and are believed to be of primary mantle origin associated with ancient subduction processes (Stosch & Lugmair, 1986; Menzies & Hawkesworth, 1987). However, this interpretation is not consistent with very low Sr content in this sample. Moreover, the Wangqing xenoliths (except WQ91-5) have relatively constant Sr/Nd ratios of 13–20 (averaged value 15.6 ± 2.5), irrespective of LREE enrichment, and no hydrous phases such as amphibole and phlogopite. These contrast with the expected signatures (high Sr/Nd ratios of 30–35 and presence of hydrous phases) of the lithosphere developed in a subduction-related environment (McDonough & McCulloch, 1987; Maury et al., 1992). WQ91-5 has a positive Sr anomaly resulting in a very high Sr/Nd ratio (34), within the range of subduction-related volcanics. Nevertheless, other characteristics typical of subduction-related metasomatism (e.g. Nb and Ta depletion, hydrous phases) are absent in this sample, and metasomatism involving carbonate fluids and melts can account for the high Sr content. The Wangqing data are therefore consistent with a rift-type continental mantle. No conclusive evidence for subduction-related processes in the composition of the upper mantle beneath the regions northeast of Wangqing has been found by Ionov et al., (1995), on the basis of a detailed trace element study of peridotite xenoliths from Sikhote-Alin.
|
Concluding Remarks |
|---|
|
TOPABSTRACTIntroductionSample Location and PetrographyAnalytical TechniquesResultsTemperature estimationBulk-rock major and minor...REE in clinopyroxenes and...Sr and Nd isotopes...DiscussionConcluding RemarksReferences |
|---|
Integration of petrological, textural and geochemical data for a peridotite xenolith suite from Wangqing, NE China, indicates a complex evolution history of the upper mantle at this region.
- The regular correlations between major element contents, LREE-depleted patterns and high Nd values suggest that the Wangqing xenoliths represent residues of multiple melting episodes within a heterogeneous mantle ranging from primitive mantle to DM. The Sm–Nd isotope composition indicates major depletion events at 1.34 –1.89 Ga, which can be related to known tectonic–thermal events in the region.
- The relatively high contents of Sr, Th, Nb and LREE in some equigranular peridotites suggest that the Wangqing mantle has locally been modified during metasomatic events that occurred before the eruption of host basalts. The metasomatic fluids may be released by decompression melting of the upwelling asthenosphere. There is no conclusive evidence for a role of subduction-related processes in these enrichment events. The association of LREE enrichment in deformed and depleted harzburgites and cpx-poor lherzolites indicates a selective metasomatic mechanism that is consistent with the existing experimental results. Modal composition may play a more important role than texture in preferential LREE-enrichment processes.
- The texture–temperature–geochemistry relationships observed in the Wangqing peridotites indicate that this old lithosphere and underlying asthenosphere diapirically intruded the shallow mantle. The undeformed protogranular samples would have been situated at the interior of this diapiric body and the equigranular samples within the flanks of diapir, although some equigranular samples may alternatively represent the uppermost mantle itself. The short time interval between diapiric intrusion and eruption of the host basalt does not allow the diffusion to smooth out the ubiquitous within-grain chemical zoning produced during diapiric intrusion (>1200°C to 850°C). This upwelling process may be related to the extensive thermal erosion of lithosphere in eastern China, which has taken place since the late Mesozoic (Menzies et al., 1993; Griffin et al., 1998).
|
|
Acknowledgements |
|---|
Comments by Drs D. A. Ionov, H. Downes, P. Kempton and Q. Qi on an earlier version, and patient editorial handling by Dr. P. Kempton are greatly appreciated. Thanks also go to J. V. Ross, L. B. Shi, X. O. Zhang and Q. F. Yang for their help in the field, to Dr K. Jarvis for the access to the NERC-funded inductively coupled plasma-mass spectrometer at Imperial College at Silwood site, and to M. Fialin, G. Ingram, M. Thirlwall, N. Holloway and J. Willis for technical assistance at every step of this work. The samples studied in this paper were collected within a collaborative scheme of INSU-CNRS (France)–NSREC (Canada)–SSB/NNSFC (People' Republic of China). Financial support was from the Royal Society (London), the Industrial Association (Department of Geology, RHUL), INSU-CNRS (France) and NSFC (China). Y. G. X. acknowledges support from the Chinese Academy of Science and National Education Committee during the revision of this paper.
* Corresponding author. Present address: Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, 510640 Wushan, Guangzhou, People's Republic of China.
|
References |
|---|
|
TOPABSTRACTIntroductionSample Location and PetrographyAnalytical TechniquesResultsTemperature estimationBulk-rock major and minor...REE in clinopyroxenes and...Sr and Nd isotopes...DiscussionConcluding RemarksReferences |
|---|
Anders E. &Grevesse, N. Abundances of the elements: meteoritic and solar. Geochimica et Cosmochimica Acta (1989) 53:197–214.[Web of Science]
Basu A. R., Wang J. W., Huang W. K., Xie G. H., Tatsumoto M. Major element, REE and Pb, Nd and Sr isotopic geochemistry of Cenozoic volcanic rocks of eastern China: implications for origin from suboceanic-type mantle reservoirs. Earth and Planetary Science Letters (1991) 105:149–169.[Web of Science]
Bedini M., Bodinier J.-L., Dautria J. M., Morten L. Evolution of LILE-enriched small melt fractions in the lithospheric mantle: a case study from the East Africa Rift. Earth and Planetary Science Letters (1997) 153:67–83.[Web of Science]
Brey G. P., Köhler T. Geothermobarometry in four-phase lherzolites II. New thermobarometers, and practical assessment of existing thermobarometers. Journal of Petrology (1990) 31:1353–1378.
Brown G. M., Pinsent R. H., Coisy P. The petrology of spinel-peridotites from the Massif Central, France. American Journal of Science (1980) 280-A:471–498.[Web of Science]
Cabanes N. &Mercier, J. C. C. Insight into the upper mantle beneath an active extensional zone: the spinel-peridotite xenoliths from San Quintin (Baja California, Mexico). Contributions to Mineralogy and Petrology (1988) 100:374–382.[Web of Science]
Chen C. Y., Frey F. A., Song Y. Evolution of the upper mantle beneath southeast Australia: geochemical evidence from peridotite xenoliths in Mount Leura basanite. Earth and Planetary Science Letters (1989) 93:195–209.[Web of Science]
Chen G. Y., Song Z. H., An C. Q., Cheng L. H., Zhuang Z., Fu Z. W., Lu Z. L., Hu J. F. Three dimensional crust and upper mantle structure of the North China region (in Chinese). Acta Geophysica Sinica (1991) 34:172–181.
Downes H. Shear zones in the upper mantle-relation between geochemical enrichment and deformation in mantle peridotites. Geology (1990) 18:374–377.
Downes H., Dupuy C. Textural, isotopic and REE variations in spinel peridotite xenoliths, Massif Central, France. Earth and Planetary Science Letters (1987) 82:121–135.[Web of Science]
Downes H., Embey-Isztin A., Thirlwall M. F. Petrology and geochemistry of spinel peridotite xenoliths from the western Pannonian Basin (Hungary): evidence for an association between enrichment and texture in the upper mantle. Contributions to Mineralogy and Petrology (1992) 109:340–354.[Web of Science]
E M. L., Zhao D. S. Cenozoic Basalts and their Deep-seated Inclusions, Eastern China. (1987) Beijing: Scientific Press. 493. (in Chinese).
Fan Q., Hooper P. R. Mineral chemistry of ultramafic xenoliths of eastern China: implications for the upper mantle composition and paleogeotherms. Journal of Petrology (1989) 28:1117–1158.
Fan Q., Hooper P. R. The Cenozoic basaltic rocks of Eastern China: petrology and chemical composition. Journal of Petrology (1991) 32:765–810.
Frey F. A., Green D. H. The mineralogy, geochemistry and origin of lherzolite inclusions in Victorian basanites. Geochimica et Cosmochimica Acta (1974) 38:1023–1059.[Web of Science]
Frey F. A., Prinz M. Ultramafic inclusions from San Carlos, Arizona: petrologic and geochemical data bearing on their petrogenesis. Earth and Planetary Science Letters (1978) 38:129–176.[Web of Science]
Frey F. A., Suen C. J., Stockman H. W. The Ronda high temperature peridotite: geochemistry and petrogenesis. Geochimica et Cosmochimica Acta (1985) 49:2469–2491.[Web of Science]
Glücklich M.-E., Mercier J.-C. C. The Basin-and-Range lithospheric mantle: evidence for homogeneity at the regional scale. TERRA Abstracts (1989) 1:318.
Griffin W. L., Zhang A. D., O'Reilly S. Y., Ryan C. G. Phanerozoic evolution of the lithosphere beneath the Sino-Korean craton. In: Mantle Dynamics and Plate Interactions in East Asia. Geophysical Monograph, American Geophysical Union—Flower M. F., Chung S. L., Lo C. H., Lee T. Y., eds. (1998) (in press).
Hart S. R., Zindler A. In search of a bulk Earth composition. Chemical Geology (1986) 57:247–267.[Web of Science]
Ionov D. A., Kramm U., Stosch H.-G. Evolution of the upper mantle beneath the southern Baikal rift zone: an Sr-Nd isotope study of xenoliths from the Bartoy volcanoes. Contributions to Mineralogy and Petrology (1992) 111:235–247.[Web of Science]
Ionov D. A., Dupuy C., O'Reilly S. Y., Kopylova M. G., Genshaft Y. S. Carbonated peridotite xenoliths from Spitsbergen: implications for trace element signature of mantle carbonate metasomatism. Earth and Planetary Science Letter (1993) 119:283–297.
Ionov D. A., Prikhod'ko V. S., O'Reilly S. Y. Peridotite xenoliths in alkali basalts from the Sikhote-Alin, southeastern Siberia, Russia: trace element signatures of mantle beneath a convergent continental margin. Chemical Geology (1995) 111:235–247.
Jagoutz E., Palme H., Baddenhausen H., Blum K., Cendales M., Drebus G., Spettel B., Lorenz V., Wänke H. The abundances of major, minor and trace elements in the earth' mantle as derived from primitive ultramafic nodules. In: Proceedings of the 10th Lunar and Planetary Science Conference, Geochimica et Cosmochimica Acta, Supplement II (1979) 1031–2050.
Jarvis K. E. A critical evaluation of two sample preparation techniques for low-level determination of some geologically incompatible elements by inductively coupled plasma-mass spectrometry. Chemical Geology (1990) 83:83–103.
Jochum K. P., McDonough W. F., Palme H., Spettle B. Compositional constraints on the continental lithospheric mantle from trace elements in spinel peridotite xenoliths. Nature (1989) 340:548–550.
Kempton P. D. Mineralogic and geochemical evidence for differing styles of metasomatism in spinel lherzolite xenoliths: enriched mantle source regions of basalts. In: Mantle Metasomatism—Menzies M. A., Hawkesworth C. J., eds. (1987) New York: Academic Press. 45–89.
Liu R. X., Chen W. J., Sun J. Z., Li D. M. The K-Ar age and tectonic environment of Cenozoic volcanic rock in China. In: The Age and Geochemistry of Cenozoic Volcanic Rock in China—Liu R. X., ed. (1992) Beijing: Seismologic Press. 1–43. (in Chinese).
Liu C. Q., Masuda A., Xie G. H. Major- and trace-element compositions of Cenozoic basalts in eastern China: petrogenesis and mantle source. Chemical Geology (1994) 114:19–42.[Web of Science]
Maury R. C., Defant M., Joron J.-L. Metasomatism of the sub-arc mantle inferred from trace elements in Philippine xenoliths. Nature (1992) 360:661–663.
McDonough W. F. Constraints on the composition of the continental lithospheric mantle. Earth and Planetary Science Letters (1990) 101:1–18.[Web of Science]
McDonough W. F., Frey F. A. Rare earth elements in upper mantle rocks. In: Geochemistry and Mineralogy of Rare Earth Elements. Mineralogical Society of America, Reviews in Mineralogy—Lipin B., McKay G., eds. (1989) 21:99–145.
McDonough W. F., McCulloch M. T. The southeast Australian lithospheric mantle: isotopic and geochemical constraints on its growth and evolution. Earth and Planetary Science Letters (1987) 86:327–340.[Web of Science]
McDonough W. F., Stosch H.-S., Ware N. G. Distribution of titanium and the rare earth elements between peridotitic minerals. Contributions to Mineralogy and Petrology (1992) 110:321–328.[Web of Science]
McKenzie D. The extraction of magma from the crust and mantle. Earth and Planetary Science Letters (1985) 74:81–91.[Web of Science]
McKenzie D. P., Bickle M. J. The volume and composition of melt generated by extension of the lithosphere. Journal of Petrology (1988) 29:625–679.
Menzies M. A., Hawkesworth C. J. Upper mantle processes and composition. In: Mantle Xenoliths—Nixon P. H., ed. (1987) NewYork: John Wiley. 725–738.
Menzies M. A., Kempton P. D., Dungan M. Interaction of continental lithosphere and asthenospheric melts below the Geronimo Volcanic Field, Arizona, USA. Journal of Petrology (1985) 26:663–693.
Menzies M. A., Fan W. M., Zhang M. Palaeozoic and Cenozoic lithoprobes and the loss of >120 km of Archaean lithosphere, Sino-Korean craton, China. In: Magmatic Processes and Plate Tectonics. Geological Society, London, Special Publication—Prichard H. M., Alabaster T., Harris N. B. W., Neary C. R., eds. (1993) 76:71–78.
Mercier J.-C. C., Nicolas A. Textures and fabrics of upper mantle peridotites as illustrated by xenoliths from basalts. Journal of Petrology (1975) 16:454–487.
Mercier J.-C. C., Benoit V., Girardeau J. Equilibrium state of diopside-bearing harzburgites from ophiolites: geobarometric and geodynamic implications. Contributions to Mineralogy and Petrology (1984) 85:391–403.[Web of Science]
Mysen B. O., Kushiro I. Compositional variations of coexisting phases with degree of melting of peridotite in the upper mantle. American Mineralogist (1977) 62:843–865.[Abstract]
Navon O., Stolper E. Geochemical consequence of melt percolation: the upper mantle as a chromatographic column. Journal of Geology (1987) 95:285–307.[Web of Science]
Nicolas A., Lucazeau F., Bayer R. Peridotite xenoliths in Massif Central basalts: textural and geophysical evidence for asthenospheric diapirism. In: Mantle Xenoliths—Nixon P. H., ed. (1987) New York: John Wiley. 563–574.
Qi Q., Taylor L. A., Zhou X. Petrology and geochemistry of mantle peridotite xenoliths from SE China. Journal of Petrology (1995) 36:55–79.
Roden M. F., Irving A. J., Murthy V. R. Isotopic and trace element composition of the upper mantle beneath a young continental rift: results from Kilbourne Hole, New Mexico. Geochimica et Cosmochimica Acta (1988) 52:461–473.[Web of Science]
Schneider M. E., Eggler D. H. Fluids in equilibrium with peridotite minerals: implications for mantle metasomatism. Geochimica et Cosmochimica Acta (1986) 50:711–724.[Web of Science]
Song Y., Frey F. A. Geochemistry of peridotite xenoliths in basalt from Hannuoba, eastern China: implications for subcontinental mantle heterogeneity. Geochimica et Cosmochimica Acta (1989) 53:97–113.[Web of Science]
Stosch H. G., Lugmair G. W. Trace elements and Sr and Nd isotope geochemistry of peridotite xenoliths from the Eifel (W. Germany) and their bearing on the evolution of the subcontinental lithosphere? Earth and Planetary Science Letters (1986) 80:281–298.[Web of Science]
Stosch H. G., Lugmair G. W., Kovalenko V. I. Spinel peridotite xenoliths from the Tariat Depression, Mongolia. II: Geochemistry and Nd and Sr isotopic composition and their implication for the evolution of the sub-continental lithosphere. Geochimica et Cosmochimica Acta (1986) 50:2601–2614.[Web of Science]
Sun S.-S., McDonough W. F. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In: Magmatism in the Ocean Basins. Geological Society, London, Special Publication—Saunders A. D., Norry M. J., eds. (1989) 42:313–345.
Tatsumoto M., Basu A. R., Huang W. K., Wang J. W., Xie G. H. Sr, Nd, and Pb isotopes of ultramafic xenoliths in volcanic rocks of Eastern China: enriched components EMI and EMII in subcontinental lithosphere. Earth and Planetary Science Letters (1992) 113:107–128.[Web of Science]
Thirlwall M. F. A triple filament method for rapid and precise analysis of rare-earth elements by isotope dilution. Chemical Geology (1982) 35:155–166.[Web of Science]
Thirlwall M. F. Long-term reproducibility of multicollector Sr and Nd isotope ratio analysis. Chemical Geology (Isotope Geoscience Section) (1991a) 94:85–104.
Thirlwall M. F. High precision multicollector analysis of low levels of Nd as oxide. Chemical Geology (Isotope Geoscience Section) (1991b) 94:13–22.
Tian Z. Y., Han P., Xu K. D. The Mesozoic-Cenozoic eastern China rift system. Tectonophysics (1992) 208:341–363.[Web of Science]
Toramaru A., Fujii N. Connectivity of melt phase in a partially molten peridotite. Journal of Geophysical Research (1986) 91:9239–9252.
Vasseur G., Verniers J., Bodinier J.-L. Modelling of trace element transfer between mantle melt and heterogranular peridotite matrix. Journal of Petrology, Special Lherzolite Issue (1991) 41–54.
Witt G. E., Seck H. A. Temperature history of sheared mantle xenoliths from the west Eifel, West Germany: evidence for mantle diapirism beneath the Rhenish Massif. Journal of Petrology (1987) 28:475–493.
Witt-Eickschen G. E., Seck H. A. Solubility of Ca and Al in orthopyroxene from spinel peridotite: an improved version of an empirical geothermometer. Contributions to Mineralogy and Petrology (1991) 106:431–439.[Web of Science]
Xu Y. G., Ross J. V., Mercier J.-C. C. The upper mantle beneath the continental rift of Tanlu, Eastern China: evidence for the intralithospheric shear zones. Tectonophysics (1993) 225:337–360.[Web of Science]
Xu Y. G., Menzies M. A., Mattey D. P., Lowry D., Harte B., Hinton R. W. The nature of the lithospheric mantle near the Tancheng-Lujiang Fault, China: an integration of texture, chemistry and O isotopes. Chemical Geology (1996) 134:67–82.[Web of Science]
Zhou X., Zhu B. Q. Study on the isotopic system and mantle chemical zonation of Cenozoic basalt in eastern China. In: The Age and Geochemistry of Cenozoic Volcanic Rock in China—Liu R. X., ed. (1992) Beijing: Seismologic Press. 366–392. (in Chinese.
Zindler A., Hart S. R. Chemical geodynamics. Annual Review of Earth and Planetary Sciences (1986) 14:493–571.[Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||