Journal of Earth Science  2017, Vol. 28 Issue (4): 578-587 PDF     0
Typical Oxygen Isotope Profile of Altered Oceanic Crust Recorded in Continental Intraplate Basalts
Huan Chen1,2,3, Qun-Ke Xia1, Etienne Deloule4, Jannick Ingrin3
1. School of Earth Sciences, Zhejiang University, Hangzhou 310027, China;
2. School of Earth and Space Sciences, University of Science and Technology of China, Hefei 230026, China;
3. UMET, UMR CNRS 8207, Université de Lille1, 59655 Villeneuve d'Ascq, France;
4. CRPG, UMR 7358, CNRS Université de Lorraine, 54501 Vandoeuvre-les Nancy, France
Abstract: Recycled oceanic crust (ROC) has long been suggested to be a candidate introducing enriched geochemical signatures into the mantle source of intraplate basalts. The different parts of oceanic crust are characterized by variable oxygen isotope compositions (δ18O=3.7‰ to 13.6‰). To trace the signatures of ROC in the mantle source of intraplate basalts, we measured the δ18O values of clinopyroxene (cpx) phenocrysts in the Cenozoic basalts from the Shuangliao volcanic field, NE China using secondary ion mass spectrometer (SIMS). The δ18O values of the Shuangliao cpx phenocrysts in four basalts ranging from 4.10‰ to 6.73‰ (with average values 5.93‰±0.36‰, 5.95‰±0.30‰, 5.58‰±0.66‰, and 4.55‰± 0.38‰, respectively) apparently exceed those of normal mantle-derived cpx (5.6‰±0.2‰) and fall in the typical oxygen isotope range of altered oceanic crust. The δ18O values display the negative correlations with the Eu, Sr anomalies of whole rocks and erupted ages, demonstrating that (1) the ROC is the main enriched component in the mantle source of the Shuangliao basalts and (2) the contributions of ROC varied with time. The basalt with the lowest δ18O value is characterized by a significant K positive anomaly, highest H2O/Ce and Ba/Th ratios, suggesting that the mantle source of basalts with low δ18O can also include a water-rich sediment component that may be the trigger for partial melting. Considering the continuous subduction of the Pacific slab, the temporal heterogeneity of the source components is likely to be caused by the Pacific slab subduction.
Keywords: continental basalt    oxygen isotope    recycled oceanic crust    Pacific slab    eastern China
0 INTRODUCTION

The small-volume continental intraplate basalts are widely spread in each continent (e.g., eastern China, western Africa, western US, western/central Europe, southeastern Australia), and the genesis of these basalts are hotly debated (Farmer, 2014). Especially in eastern China, the continental basalts extend from the northernmost to the southernmost, along the continental margin (Lei et al., 2013), which form an important part of the volcano belt of the western circum-Pacific rim (Fig. 1). These basalts are characterized by typical ocean island basalt (OIB)-like trace element patterns and Sr-Nd-Pb isotopic compositions, which indicate the presence of enriched components in the mantle sources (e.g., Chen et al., 2007; Zhang et al., 2001; Jung and Hoernes, 2000; Marzoli et al., 2000; Zou et al., 2000; Rogers et al., 1995; Zhou and Armstrong, 1982). Recently, seismic tomography studies have revealed the presence of recycled oceanic slab in the Earth's mantle (e.g., Liu et al., 2017; Fichtner and Villaseñor, 2015; Wei et al., 2012; Huang and Zhao, 2006). Therefore, the recycled oceanic crust (ROC) has increasingly been suggested to be one of the major enriched components in their mantle sources (Chen et al., 2017, 2015a, b; Liu et al., 2015a, b; Wang et al., 2015; Xu et al., 2012; Kuritani et al., 2011).

 Download: larger image Figure 1. (a) Simplified tectonic divisions and the distribution of the Cenozoic intraplate basalts in eastern China as well as the location of the Shuangliao volcanic field (modified from Xu et al., 2012); (b) the distribution and sample locations (red stars) of the Shuangliao basalts. DTGL. Daxin'anling-Taihangshan Gravity Lineament.

Oxygen isotope exchanges between oceanic crust and seawater occurs during hydrothermal alteration (Muehlenbachs and Clayton, 1976). Because the temperature of hydrothermal alteration varied from the lower to the upper part of the oceanic crust, different layers of altered oceanic crust acquired distinct oxygen isotope compositions through hydrothermal exchange with seawater (Taylor, 1974). The general profile of oxygen isotope compositions of altered oceanic crust exhibits δ18O values higher than the normal mantle in the upper part and δ18O values lower than the normal mantle in the lower part (Fig. 2) (Gao et al., 2006; Eiler, 2001; Hoffman et al., 1986; Gregory and Taylor, 1981). Therefore, oxygen isotope is a powerful tool to trace ROC component in the mantle source of continental intraplate basalts (e.g., Chen et al., 2017; Liu et al., 2015a, b; Wang et al., 2015; Kokfelt et al., 2006; Eiler et al., 2000; Putlitz et al., 2000; Woodhead et al., 1993).

 Download: larger image Figure 2. Typical oxygen isotope profile of an altered sediment-covered oceanic crust. The blue curve shows the average δ18O values of the altered oceanic crust. The data for the oceanic crust are based on the Samail ophiolite (Taylor, 1974), the δ18O values of marine sediments are from Eiler (2001).

The Shuangliao volcanic field is located in the south part of NE China, which consists of eight volcanoes. Xu et al. (2012) and Chen et al. (2015b) conducted detailed geochemical studies including Ar-Ar erupted ages, major and trace elements, Sr-Nd-Pb isotopes and H2O content on these basalts. The Shuangliao basalts are characterized by high Fe2O3, HIMU (high µ)-type trace element patterns and significant correlations between Ba/Th, Ce/Pb and H2O/Ce ratios, which suggested that the recycled oceanic slab component may be involved in the mantle source and the contributions of these source components changed with time. Hence, this area is an appropriate place to trace oxygen isotopic signatures of ROC in continental basalts.

Here, we measured the δ18O values of the clinopyroxene (cpx) phenocrysts in Cenozoic basalts from the Shuangliao volcanic field using secondary ion mass spectrometry (SIMS). The typical oxygen isotope profile of altered oceanic crust in the Shuangliao basalts clearly confirms that the recycled oceanic slab was present in the mantle source. The mass balance calculation demonstrates that a water-rich sediment component may also be involved in the mantle source although the basalts display the lowest δ18O, in contrast with intuitively expected elevated δ18O. Combined with the erupted ages, the changing source components is likely to be caused by the ongoing Pacific slab subduction.

1 GEOLOGICAL BACKGROUND AND SAMPLES

Northeast China (NE China) lies in the Xing'an-Mongolia orogenic belt (XMOB), which belongs to the east part of the Paleozoic Central Asian orogenic belt (Fig. 1a). It is composed of several minor blocks (e.g., Erguna, Xing'an, Songliao, Jiamusi) amalgamated during subduction and collision among the Siberian Craton, the North China Craton (NCC) and the Pacific Plate (Li, 2006; Sengör and Natal'in, 1996; Sengör et al., 1993). The tectonic evolutions of NE China mainly include the closure of the Paleo-Asian Ocean, the amalgamation of several minor blocks and the subduction of the Pacific Plate since Late Mesozoic (Maruyama et al., 1997; Sengör and Natal'in, 1996; Sengör et al., 1993).

The Shuangliao volcanic field is located in the southeast of the Songliao Basin, which consists of eight volcanoes: Aobaoshan (ABS), Bobotushan (BBT), Bolishan (BLS), Shitoushan (STS), Dahalabashan (DHLB), Xiaohalabashan (XHLB), Datuerjishan (DTEJ) and Xiaotuerjishan (XTEJ) (Fig. 1b). Seismic tomography has shown the presence of a stagnant subducted Pacific slab in the mantle transition zone beneath the Shuangliao volcanic field (Liu et al., 2017; Wei et al., 2012; Huang and Zhao, 2006). The Shuangliao basalts vary from basanite, alkaline olivine basalt, and transitional basalts to dolerites (Chen et al., 2015b; Xu et al., 2012). Abundant peridotite xenoliths were carried out by the eruption of these volcanoes (Yu et al., 2009). The Ar-Ar dating results have shown that all volcanoes erupted between 51.0 and 41.6 Ma (Xu et al., 2012). Volcanic cones with high alkalinity erupted between 51 and 48.5 Ma, while transitional or subalkaline cones erupted between 43.0 and 41.6 Ma, indicating that the alkalinity of the Shuangliao volcanic rocks decreased with time (Xu et al., 2012).

Our samples were collected from BBT, BLS, DHLB and XTEJ (red stars in Fig. 1b). According to Chen et al. (2015b) and Xu et al. (2012), these rocks can be classified into three types: basanite (BBT and BLS), alkali olivine basalt (DHLB) and transitional basalt (XTEJ). The basanites from BBT and BLS contain 20% of olivine (ol) and cpx phenocrysts. Meanwhile, in alkali olivine basalts and transitional basalts, the phenocrysts are composed of olivine, clinopyroxene and plagioclase (pl). The mantle xenoliths are mainly spinel lherzolites with minor spinel harzburgites. The Ar-Ar plateau ages of BBT, BLS, DHLB and XTEJ basalt are 50.1±0.8, 49.7±0.2, 51.0±0.5, and 43.0±0.4 Ma, respectively (Xu et al., 2012).

2 ANALYTICAL METHODS 2.1 In-Situ Oxygen Isotope Measurement

In-situ oxygen isotope measurement of cpx phenocryst was carried on a Cameca IMS-1270 at the Centre de Recherches Pétrographiques et Géochimiques, Centre National de la Recherche Scientifique (CRPG-CNRS, Nancy), following the analytical procedures described in Liu et al. (2015b) and Chen et al. (2017). The sample coated with Au was compensated for the charging by an electron gun. The Cs+ primary ion beam was focused into 20 μm in diameter on the sample surface at ~5.3 nA and 10 kV. Secondary negative ions were extracted at 10 kV and the 17OH and 18O ions were distinguished by the 3000 mass resolution power. Two off-axis Faraday cups (L'2 and H1) were used to detect the 16O and 18O of the samples in multi-collection mode. The oxygen isotope values of the cpx phenocrysts were reported in δ18O relative to the reference standards (Vienna Mean Standard Ocean Water, VSMOW). The internal precision of single analysis was typically less than 0.1‰ (2σ).

2.2 Corrections of Instrumental Mass Fractionation

The correlation between instrumental mass fractionation (IMF) and the Mg# (=100×Mg2+/(Mg2++Fe2+)) of cpx can be used for the matrix effect during the measurement of oxygen isotopes by SIMS (Gurenko et al., 2001). Thus, the Nüshan cpx megacrysts (NSH2, NSH5, NSH8, NSH9, NSH10, and NSH14) from eastern China, with homogeneous oxygen isotope compositions and the similar chemical compositions as those of cpx phenocrysts in this study (Xia et al., 2004), were selected for the correction of the matrix effect. The IMF can be generally defined as the deviation between the measured and true values and can occur at the production, transport and detection of secondary ions. The transport and detection of secondary ions are primarily controlled by the state of the instrument, which can be monitored by periodically measuring the standard (NSH9). The main proportion of mass fractionation occurs at the production of secondary ions, which depends on type and composition of the mineral (commonly referred to as the "matrix effect"). For this portion, a group of standards (NSH2, NSH5, NSH8, NSH9, NSH10, NSH14), with chemical compositions bracketing the unknown samples was used to correct the "matrix effect" (Deegan et al., 2016; Hartley et al., 2012; Kita et al., 2010, 2009; Page et al., 2010; Valley and Kita, 2009; Eiler et al., 1997).

For the deviation caused by transport and detection processes, the fractionation factor (δ18O*) could be calculated from the measured value of NSH9 (δ18Oraw) and the "true" value (δ18Otrue), using the following equation (Kita et al., 2009)

 ${\delta ^{18}}{{\rm{O}}^*} = {\rm{bia}}{{\rm{s}}_{(1)}} = {\delta ^{18}}{{\rm{O}}^{{\rm{NSH}}9}}_{({\rm{raw}})}-{\delta ^{18}}{{\rm{O}}^{{\rm{NSH}}9}}_{({\rm{true}})}$ (1)

For sessions with a linearly time-dependent bias drift, a linear correlation between this bias and time is used to correct.

For the deviation caused by the mineral type and composition during production of secondary ions (matrix effect), the correlation between the relative IMF and the Mg# of a group of cpx standards was estimated (Fig. 3; Gurenko et al., 2001). A linear function correlation between IMF and cpx's Mg# for each session was obtained

 ${\rm{IMF}} = f({\rm{M}}{{\rm{g}}^\# }) = A \times {\rm{M}}{{\rm{g}}^\# } + B$ (2)

where A and B are the parameters in the linear regression. The bias caused by the matrix effect between the unknown samples and the NSH9 was corrected using the following equation

 ${\rm{bia}}{{\rm{s}}_{(2)}} = {\rm{IMF}}-{\rm{IM}}{{\rm{F}}_{({\rm{NSH}}9)}} = f\left( {{\rm{M}}{{\rm{g}}^\# }} \right)-{\rm{IM}}{{\rm{F}}_{({\rm{NSH}}9)}}$ (3)

Overall, the "true" value $(\delta {}^{18}{\rm{O}}_{{\rm{corrected}}}^{{\rm{sample}}})$ for unknown samples was estimated by removing the effect of the bias(1) and bias(2)

 $\begin{array}{l} \quad \delta {}^{18}{\rm{O}}_{{\rm{corrected}}}^{{\rm{sample}}}\\ = \alpha {}^{18}{\rm{O}}_{{\rm{raw}}}^{{\rm{sample}}} - {\rm{bia}}{{\rm{s}}_{(1)}} - {\rm{bia}}{{\rm{s}}_{(2)}}\\ = \alpha {}^{18}{\rm{O}}_{{\rm{raw}}}^{{\rm{sample}}} - (\delta {}^{18}{\rm{O}}_{{\rm{raw}}}^{{\rm{NSH9}}} - \delta {}^{18}{\rm{O}}_{{\rm{true}}}^{{\rm{NSH9}}}) - \\ \quad (f({\rm{M}}{{\rm{g}}^\# }) - {\rm{IM}}{{\rm{F}}_{({\rm{NSH}}9)}}) \end{array}$ (4)

The standards were analyzed every day before measuring the unknown samples. The correlation between Mg# and IMF of standards in each day (each session) was estimated. The analytical results of each session for the standards are listed in Fig. 3 and Table 1.

 Download: larger image Figure 3. The correlation between Mg# and IMF of clinopyroxene standards in each session
Table 1 The SIMS analysis results of standards for matrix effect correction
2.3 Precision and Accuracy

The error includes the internal precision of the instrument, the reproducibility or repeatability during measurement and the additional error caused by IMF corrections.

The internal precision of Cameca IMS 1270 at CRPG-Nancy is typically in the range of ±0.08‰-0.14‰ (in most cases < 0.1‰) (2SE). The reproducibility or repeatability during measurement can be calculated from a standard deviation of N measurements (Kita et al., 2009; Valley and Kita, 2009; Fitzsimons et al., 2000). During the measurement, several analyses are conducted (generally 3 times) on each cpx grain. The standard deviation of the mean (SE) is typically less than ±0.3‰. The additional error caused by IMF corrections usually contain three portions: 1) the uncertainty regarding the standards' oxygen isotopic composition (less than ±0.2‰, 2SD, Xia et al., 2004); 2) the uncertainty regarding the chemical composition of cpx (the error of Mg# is near 1 unit for EPMA analysis, Chen et al., 2015b); and 3) the uncertainty from "matrix effect" line calibrated by the linear least-square (Eq. 2). Based on Eq. 2, the uncertainty of oxygen isotopes introduced by the error of Mg# is less than ±0.1‰. The uncertainty caused by the least-square regression line (Eq. 2) can be calculated by the equation

 ${\rm{S}}{{\rm{E}}_{\delta \;{\rm{pred}}}} = {\rm{S}}{{\rm{E}}_{\delta \;{\rm{std}}}}\sqrt {\frac{1}{n} + \frac{{{{({\rm{M}}{{\rm{g}}^{\# *}} - \overline {{\rm{M}}{{\rm{g}}^\# }} )}^2}}}{{\sum\nolimits_{i = 1}^n {{{({\rm{Mg}}_i^\# - \overline {{\rm{M}}{{\rm{g}}^\# }} )}^2}} }}}$ (5)

where Mg#* is the Mg number of cpx, n is the number of standards involved in the regression line, ${{\rm{M}}{{\rm{g}}^\# }}$ is the average Mg# of all of the standards, Mg#i is the Mg# of the i-th standard, and the SEδ std is the standard deviation of the data points regarding the regression line, which is calculated as follows

 ${\rm{S}}{{\rm{E}}_{\delta \;{\rm{std}}}} = \sqrt {\sum\nolimits_{i = 1}^n {\frac{{{{({y_i} - A{x_i} - B)}^2}}}{{n - 2}}} }$ (6)

where A and B are same as in Eq. 2.

The calculated results show that the uncertainty caused by the least-square regression line is less than ±0.1‰ for sessions 1, 3 and 4 and less than ±0.2‰ for Session 2 (Fig. 4).

 Download: larger image Figure 4. The relationship between the errors caused by matrix effect regression line and the Mg# of clinopyroxene, calculated by Eqs. (5) and (6).

Overall, the total errors on δ18O are generally about ±0.5‰.

3 RESULTS

The chemical compositions of the cpx phenocrysts selected for SIMS analyses have been determined by Chen et al. (2015b). The BSE (backscattered electron) images show that most of the cpx phenocrysts are euhedral and homogeneous (Fig. 5), and only a few of them have an inherited core. These cpx phenocrysts exhibit chemical homogeneity within individual grains (Chen et al., 2015b). They are augitic to diopsidic cpx and have relatively lower Mg# (68.2-82.2) and Cr2O3 ( < 0.8 wt.%), higher TiO2 (0.7 wt.%-4 wt.%) compared to cpx in the peridotite xenoliths carried by the Shuangliao basalts (91.5-92.8 Mg#, 0.72 wt.%-1.42 wt.% Cr2O3, 0.16 wt.%-0.32 wt.% TiO2). Most of the studied cpx has TiO2 higher than 0.7 wt.% and the concentration of TiO2 show a negative correlation with the MgO (Fig. 6), which suggests that these cpx are igneous phenocrysts. The inherited core of some cpx has a relatively lower TiO2concentration ( < 0.7 wt.%), which is consistent with the chemical composition of the cpx from the lower continental crust in eastern China (Zhang and Zhang, 2007 and the references therein). Moreover, the highest Mg# values of the cpx phenocrysts are similar to those of the coexisting ol phenocrysts (~84.3), suggesting that these cpx phenocrysts were syn-crystallized with the ol phenocrysts.

 Download: larger image Figure 5. The BSE images of the cpx phenocrysts in the Shuangliao basalts. Abbreviation cpx. clinopyroxene.
 Download: larger image Figure 6. TiO2 versus MgO of the clinopyroxenes in the Shuangliao basalts. The dashed area represent the composition region of the clinopyroxenes from the lower continental crust in eastern China (from Zhang and Zhang (2007) and references therein).

The oxygen isotope compositions of the Shuangliao cpx phenocrysts are reported in Table 2 and plotted in Fig. 7. The δ18O values vary widely, ranging from 4.10‰ to 6.73‰, and the average δ18O value of the cpx phenocrysts in BBT, BLS, DHLB and XTEJ is 5.93‰±0.36‰, 5.95‰±0.30‰, 5.58‰±0.66‰, and 4.55‰±0.38‰, exceeding the value of cpx in typical mid-ocean ridge basalt (MORB) and mantle peridotites (5.4‰-5.8‰) (Eiler et al., 1997; Mattey et al., 1994). Interestingly, this variation is time-dependent. From 51 to 43 Ma, the δ18O of the Shuangliao basalts varied from the values higher than those of the normal mantle to the values lower than those of the normal mantle, which displayed an oxygen isotope profile similar to that of the altered oceanic crust (Gao et al., 2006; Eiler, 2001; Hoffman et al., 1986; Gregory and Taylor, 1981).

Table 2 The oxygen isotope composition of the cpx phenocrysts in the Shuangliao basalts
 Download: larger image Figure 7. Oxygen isotope compositions of the clinopyroxenes phenocrysts in the Shuangliao basalts. The range of δ18O values of the clinopyroxenes phenocrysts in N-MORB, EM1, EM2 and HIMU are calculated from the δ18O values of the olivine phenocrysts (Eiler et al., 1997), assuming an equilibrium fractionation of 0.4‰ (Mattey et al., 1994). The erupted ages of the Shuangliao basalts are from Xu et al. (2012).
4 DISCUSSION 4.1 Heterogeneity of Oxygen Isotope Compositions

The δ18O values of the cpx phenocrysts in the Shuangliao basalts have two primary features: (1) the average δ18O values vary widely, from 4.55‰ to 5.95‰, falling in the typical oxygen isotope profile of altered oceanic crust (Gao et al., 2006; Eiler, 2001; Hoffman et al., 1986; Gregory and Taylor, 1981); and (2) the δ18O values within individual samples also display considerable variations from 1.04‰ to 1.75‰.

During the magma formation and evolution, the oxygen isotope compositions of magma can be affected by many factors, including partial melting of the mantle source, fractional crystallization, crustal contamination (or assimilation-fractional crystallization [AFC] process), devolatilization and water-melt interaction during ascent to the surface and alteration on the surface.

With MgO between 8 wt.% and 3 wt.%, the fractionation of oxygen isotopes caused by partial melting, fractional crystallization and/or devolatilization will be less than 0.1‰ for δ18O of the basaltic melt (Eiler, 2001). Therefore, partial melting, fractional crystallization and devolatilization are unlikely to be the primary cause of δ18O variations in our samples. In addition, the previous studies have shown that crustal contamination for the Shuangliao basalts is limited (Chen et al., 2015b; Xu et al., 2012), thus excluding the effect of crustal contamination on the observed oxygen isotopic variations. Theoretically, an AFC process would significantly affect the oxygen isotopic compositions of melts if they underwent during ascent (Taylor, 1974). According to the equation in Taylor (1974), to increase the δ18O value, high degree of fractional crystallization and crustal contamination is required. However, the sample (BBT2) with the highest δ18O value is characterized by highest Mg# (66.9), highest Ni (525.6 ppm) and Cr (373 ppm) contents, highest Nb/U (38.4) and Ce/Pb (24.9) ratios (Chen et al., 2015b), which is not consistent with this AFC process. Although the high-temperature water-melt interaction during crystallization (Wang and Eiler, 2008) can affect the oxygen isotope composition of the magma (decreasing the δ18O values of the sample), the water content of XTEJ4 basalt (with the lowest δ18O value) is lower than those of other basalts in the Shuangliao volcanic field (Chen et al., 2015b) and argues against such an interaction as the cause for the low δ18O values. Additionally, the selected cpx phenocrysts are all fresh without crack and inclusion (Chen et al., 2015b), which can eliminate the effect of the alteration on the surface.

Overall, the typical oxygen isotope profile of altered oceanic crust recorded by the cpx phenocrysts in the Shuangliao basalts represents the heterogeneity of the mantle source, which means both components with high and low δ18O values were involved in the mantle source. On the other hand, the variation in oxygen isotope compositions within each sample is most likely caused by magma mixing in a complex magma plumbing system shortly before the eruption. In fact, the δ18O features of the cpx phenocrysts within the single sample of the Shuangliao basalts are consistent with the results of other basalts from the Chaihe-aershan and Taihang volcanoes in eastern China (Chen et al., 2017; Liu et al., 2015b), Cannary Island (Gurenko et al., 2011) and Yellowstone (Wotzlaw et al., 2015).

4.2 Recycled Oceanic Crust in the Mantle Source

Detailed geochemical studies have shown that the Shuangliao basalts are characterized by high Fe2O3, HIMU-like trace element pattern and Sr-Nd isotope composition, which indicate that the recycled oceanic slab may have been involved in the mantle source (Xu et al., 2012).

Generally, the lower gabbroic oceanic crust is enriched in plagioclase (pl) that is characterized by the positive Eu and Sr anomalies (Drake and Weill, 1975; Ching-oh et al., 1974). Thus, the lower gabbroic oceanic crust should have positive Eu and Sr anomalies (e.g., Bach et al., 2001). Meanwhile, the upper oceanic crust is composed by MORB, which has no Eu and Sr anomalies (e.g., Kelley et al., 2003). In Figs. 8a and 8b, negative correlations can be observed between the δ18O values of the cpx phenocrysts and Eu anomalies (Eu/Eu*), Sr anomalies (Sr/Sr*) of whole rocks for the Shuangliao basalts, suggesting two components in the mantle source: Component Ⅰ with low δ18O value and positive Eu and Sr anomalies; Component Ⅱ with high δ18O value and weak Eu and Sr anomalies. Because the Eu and Sr anomalies are typically associated with pl (Drake and Weill, 1975; Ching-oh et al., 1974) and the pl is rare in the Shuangliao basalts (Chen et al., 2015b; Xu et al., 2012), the pl contribution with the low δ18O values should have been present in the mantle source of the basalts. This is consistent with the low δ18O characteristics of the high temperature hydrothermally altered lower gabbroic oceanic crust (Gao et al., 2006; Bach et al., 2001; Hoffman et al., 1986; Gregory and Taylor, 1981). In addition, the basalts with high δ18O values have no Eu anomaly and relatively weak Sr anomaly which are similar to the features of the upper oceanic crust undergone the low temperature hydrothermal alteration (Gao et al., 2006; Kelley et al., 2003; Hoffman et al., 1986; Gregory and Taylor, 1981). Therefore, the negative correlations between the δ18O values of the cpx phenocrysts and the Eu and Sr anomalies are exactly correspond to the whole section of the oceanic crust, providing a clear evidence that a whole recycled oceanic crust was involved in the mantle source of these continental intraplate basalts and the proportions of these components varied with time (in the mantle source of earlier basalts (~51 Ma), the recycled upper oceanic crust component dominated, whereas the lower oceanic crust component was involved in during the younger eruptive cycle (~43 Ma)).

 Download: larger image Figure 8. Comparison of the range of δ18O values in the clinopyroxenes phenocrysts and the trace element ratios (Eu/Eu*, Sr/Sr* and H2O/Ce) of the Shuangliao basalts. The Eu/Eu*, Sr/Sr* and H2O/Ceare calculated from Chen et al. (2015b). The black lines are mix lines calculated between depleted mantle (DMM) and upper oceanic crust (UOC), lower oceanic crust (LOC), marine sediments (GLOSS). The δ18O values of DMM, GLOSS, UOC and LOC are from Eiler (2001). The H2O/Ce ratios of DMM, GLOSS, UOC and LOC are from Dixon et al. (2002) and Plank and Langmuir (1998). The ratio of Ba/Th for DMM, GLOSS, UOC and LOC is calculated from Bach et al. (2001), Kelley et al. (2003), Plank and Langmuir (1998) and Workman and Hart (2005).

It is worth noting that the H2O/Ce ratios of the Shuangliao basalts (158-737) display a positive correlation with (Ba/Th)n (n means primitive mantle normalization) and a negative correlation with Ce/Pb (Chen et al., 2015b), and the XTEJ4 basalt with the lowest δ18O compositions (4.10‰-5.32‰) is characterized by a significant K positive anomaly, high H2O/Ce (566) and Ba/Th (105) ratios. All these suggest the presence of another component with high (Ba/Th)n and low Ce/Pb in the mantle source. The marine sediments characterized by high (Ba/Th)n and low Ce/Pb can serve as such component (Dixon et al., 2002; Plank and Langmuir, 1998), but they also have high δ18O values (Eiler, 2001; Gregory and Taylor, 1981). Here, we conducted a mass balance calculation to determine whether the involvement of a sediment component can match the low δ18O in the Shuangliao mantle source. Basing on the equation

 $\begin{array}{l} \frac{{c_{{\rm{melt}}}^{\rm{i}}}}{{c_{melt}^j}} = (aC_{{\rm{DMM}}}^{\rm{i}} + bC_{{\rm{LOC}}}^{\rm{i}} + cC_{{\rm{GLOSS}}}^{\rm{i}})/\\ \quad \quad (aC_{{\rm{DMM}}}^{\rm{j}} + bC_{{\rm{LOC}}}^{\rm{j}} + cC_{{\rm{GLOSS}}}^{\rm{j}}) \end{array}$ (7)

where i and j represent the element; ${c_{{\rm{melt}}}^{\rm{i}}}$, $C_{{\rm{DMM}}}^{\rm{i}}$, $C_{{\rm{LOC}}}^{\rm{i}}$ and $C_{{\rm{GLOSS}}}^{\rm{i}}$ are the concentration of element i in melt, DMM (depleted mantle), LOC (lower oceanic crust) and GLOSS (marine sediments), a, b and c are the proportions of the components (a+b+c=1). The δ18O value, H2O/Ce and Ba/Th ratios were used for the mass-balance calculation. The results show that the XTEJ4 basalts with the lowest δ18O value and highest H2O/Ce ratio can be mixed by 72.5% DMM, 26% LOC and 1.5% GLOSS (the δ18O value, H2O/Ce and Ba/Th ratios of DMM, LOC and GLOSS are from Eiler (2001), Workman and Hart (2005), Dixon et al. (2002), Bach et al. (2001) and Plank and Langmuir (1998)), which means that the mantle source of the basalts with the low δ18O compositions can contain a water-rich sediment component. Furthermore, we calculated the mixing lines between DMM and UOC (upper oceanic crust), LOC, GLOSS. All points are within the areas of the mix lines and the relative contributions of these recycled oceanic components in the mantle source varied significantly within a limited time (Figs. 8c and 8d). In combination with the ages of the Shuangliao basalts, the contribution of the recycled sediments in the mantle source seems to increase from earlier volcanic events (~51 Ma) to the later ones (~43 Ma). Considering the water-rich sediment is generally most fusible, its addition to the mantle might be a trigger of the melting.

The δ18O as well as the trace element ratios (e.g., Ba/Th) of the Shuangliao basalts significantly varied with the eruption ages (Figs. 7 and 8), indicating that the contributions of different recycled oceanic components in the mantle source changed within the time interval of the Shuangliao volcanoes' formation. The Pacific slab is constantly subducting under eastern China for at least the last 80 Ma (Maruyama et al., 1997) and continuously transports recycled materials to the deep mantle. Therefore, the temporal heterogeneity of the source components shown by the Shuangliao basalts is likely to be caused by the ongoing Pacific slab subduction. In other words, the Pacific slab is likely the source of the enriched components in the continental basalts from eastern China (Chen et al., 2017).

5 CONCLUSIONS

(1) The δ18O values of the cpx phenocrysts in the Shuangliao Cenozoic basalts vary widely (from 4.10‰ to 6.73‰), exceeding the value of cpx in typical MORB and mantle peridotites and falling in the range of oxygen isotope compositions of an altered oceanic crust, which provides a clear evidence that a recycled oceanic slab was involved in the mantle source of these continental basalts.

(2) The relative contribution of the enriched source components in the mantle source has changed with time. More water-rich sediment component is observed in the mantle source of the youngest Shuangliao basalts, which might be the trigger for mantle melting.

(3) Given the Pacific slab is constantly subducting under eastern Asia and is expected to continuously transport recycled materials into the deep mantle, such ongoing subduction is a reasonable explanation for the variations of recycled components in the mantle source over time.

ACKNOWLEDGMENTS

This research was supported by the National Natural Science Foundation of China (Nos. 41225005 and 41173047). We thank Andrey Gurenko and Nordine Bouden for their help in SIMS analysis. The constructive comments and suggestions from two anonymous reviewers are greatly appreciated. The final publication is available at Springer via http://dx.doi.org/10.1007/s12583-017-0798-5.

References
 Bach W., Alt J. C., Niu Y. L., et al., 2001. The Geochemical Consequences of Late-Stage Low-Grade Alteration of Lower Ocean Crust at the SW Indian Ridge: Results from ODP Hole 735B (Leg 176). Geochimica et Cosmochimica Acta, 65(19): 3267–3287. DOI:10.1016/s0016-7037(01)00677-9 Chen H., Xia Q.-K., Ingrin J., 2015a. Water Content of the Xiaogulihe Ultrapotassic Volcanic Rocks, NE China: Implications for the Source of the Potassium-Rich Component. Science Bulletin, 60(16): 1468–1470. DOI:10.13039/501100001809 Chen H., Xia Q.-K., Ingrin J., et al., 2015b. Changing Recycled Oceanic Components in the Mantle Source of the Shuangliao Cenozoic Basalts, NE China: New Constraints from Water Content. Tectonophysics, 650: 113–123. DOI:10.13039/501100001809 Chen H., Xia Q.-K., Ingrin J., et al., 2017. Heterogeneous Source Components of Intraplate Basalts from NE China Induced by the Ongoing Pacific Slab Subduction. Earth and Planetary Science Letters, 459: 208–220. DOI:10.13039/501100001809 Chen Y., Zhang Y. X., Graham D., et al., 2007. Geochemistry of Cenozoic Basalts and Mantle Xenoliths in Northeast China. Lithos, 96(1/2): 108–126. DOI:10.1016/j.lithos.2006.09.015 Ching-oh S., Williams R. J., Shine-soon S., 1974. Distribution Coefficients of Eu and Sr for Plagioclase-Liquid and Clinopyroxene-Liquid Equilibria in Oceanic Ridge Basalt: An Experimental Study. Geochimica et Cosmochimica Acta, 38(9): 1415–1433. DOI:10.1016/0016-7037(74)90096-9 Deegan F. M., Whitehouse M. J., Troll V. R., et al., 2016. Pyroxene Standards for SIMS Oxygen Isotope Analysis and Their Application to Merapi Volcano, Sunda Arc, Indonesia. Chemical Geology, 447: 1–10. DOI:10.13039/501100004359 Dixon J. E., Leist L., Langmuir C., et al., 2002. Recycled Dehydrated Lithosphere Observed in Plume-Influenced Mid-Ocean-Ridge Basalt. Nature, 420(6914): 385–389. DOI:10.1038/nature01215 Drake M. J., Weill D. F., 1975. Partition of Sr, Ba, Ca, Y, Eu2+, Eu3+, and Other REE between Plagioclase Feldspar and Magmatic Liquid: An Experimental Study. Geochimica et Cosmochimica Acta, 39(5): 689–712. DOI:10.1016/0016-7037(75)90011-3 Eiler J. M., 2001. Oxygen Isotope Variations of Basaltic Lavas and Upper Mantle Rocks. Reviews in Mineralogy and Geochemistry, 43(1): 319–364. DOI:10.2138/gsrmg.43.1.319 Eiler J. M., Farley K. A., Valley J. W., et al., 1997. Oxygen Isotope Variations in Ocean Island Basalt Phenocrysts. Geochimica et Cosmochimica Acta, 61(11): 2281–2293. DOI:10.1016/s0016-7037(97)00075-6 Eiler J. M., Schiano P., Kitchen N., et al., 2000. Oxygen-Isotope Evidence for Recycled Crust in the Sources of Mid-Ocean-Ridge Basalts. Nature, 403(6769): 530–534. DOI:10.1038/35000553 Farmer, G. L. , 2014. Continental Basaltic Rocks. In: Holland, H. , Turekian, K. , eds. , Treatise on Geochemistry, Second Edition. Elsevier, Amsterdam. 75-110 Fichtner A., Villaseñor A., 2015. Crust and Upper Mantle of the Western Mediterranean-Constraints from Full-Waveform Inversion. Earth and Planetary Science Letters, 428: 52–62. DOI:10.1016/j.epsl.2015.07.038 Fitzsimons I. C. W., Harte B., Clark R. M., 2000. SIMS Stable Isotope Measurement: Counting Statistics and Analytical Precision. Mineralogical Magazine, 64(1): 59–83. DOI:10.1180/002646100549139 Gao Y. J., Hoefs J., Przybilla R., et al., 2006. A Complete Oxygen Isotope Profile through the Lower Oceanic Crust, ODP Hole 735B. Chemical Geology, 233(3/4): 217–234. DOI:10.1016/j.chemgeo.2006.03.005 Gregory R. T., Taylor H. P. Jr., 1981. An Oxygen Isotope Profile in a Section of Cretaceous Oceanic Crust, Samail Ophiolite, Oman: Evidence for δ18O Buffering of the Oceans by Deep ( > 5 km) Seawater-Hydrothermal Circulation at Mid-Ocean Ridges. Journal of Geophysical Research: Solid Earth, 86(B4): 2737–2755. DOI:10.1029/jb086ib04p02737 Gurenko A. A., Bindeman I. N., Chaussidon M., 2011. Oxygen Isotope Heterogeneity of the Mantle beneath the Canary Islands: Insights from Olivine Phenocrysts. Contributions to Mineralogy and Petrology, 162(2): 349–363. DOI:10.1007/s00410-010-0600-5 Gurenko A. A., Chaussidon M., Schmincke H. U., 2001. Magma Ascent and Contamination beneath one Intraplate Volcano: Evidence from S and O Isotopes in Glass Inclusions and Their Host Clinopyroxenes from Miocene Basaltic Hyaloclastites Southwest of Gran Canaria (Canary Islands). Geochimica et Cosmochimica Acta, 65(23): 4359–4374. DOI:10.1016/s0016-7037(01)00737-2 Hartley M. E., Thordarson T., Taylor C., et al., 2012. Evaluation of the Effects of Composition on Instrumental Mass Fractionation during SIMS Oxygen Isotope Analyses of Glasses. Chemical Geology, 334: 312–323. DOI:10.1016/j.chemgeo.2012.10.027 Hoffman S. E., Wilson M., Stakes D. S., 1986. Inferred Oxygen Isotope Profile of Archaean Oceanic Crust, Onverwacht Group, South Africa. Nature, 321(6065): 55–58. DOI:10.1038/321055a0 Huang J. L., Zhao D. P., 2006. High-Resolution Mantle Tomography of China and Surrounding Regions. Journal of Geophysical Research, 111(B9): B09305. DOI:10.1029/2005jb004066 Jung S., Hoernes S., 2000. The Major-and Trace-Element and Isotope (Sr, Nd, O) Geochemistry of Cenozoic Alkaline Rift-Type Volcanic Rocks from the Rhö n Area (Central Germany): Petrology, Mantle Source Characteristics and Implications for Asthenosphere-Lithosphere Interactions. Journal of Volcanology and Geothermal Research, 99(1/2/3/4): 27–53. DOI:10.1016/s0377-0273(00)00156-6 Kelley K. A., Plank T., Ludden J., et al., 2003. Composition of Altered Oceanic Crust at ODP Sites 801 and 1149. Geochemistry, Geophysics, Geosystems, 4(6). DOI:10.1029/2002gc000435 Kita N. T., Nagahara H., Tachibana S., et al., 2010. High Precision SIMS Oxygen Three Isotope Study of Chondrules in LL3 Chondrites: Role of Ambient Gas during Chondrule Formation. Geochimica et Cosmochimica Acta, 74(22): 6610–6635. DOI:10.1016/j.gca.2010.08.011 Kita N. T., Ushikubo T., Fu B., et al., 2009. High Precision SIMS Oxygen Isotope Analysis and the Effect of Sample Topography. Chemical Geology, 264(1/2/3/4): 43–57. DOI:10.1016/j.chemgeo.2009.02.012 Kokfelt T. F., Hoernle K., Hauff F., et al., 2006. Combined Trace Element and Pb-Nd-Sr-O Isotope Evidence for Recycled Oceanic Crust (Upper and Lower) in the Iceland Mantle Plume. Journal of Petrology, 47(9): 1705–1749. DOI:10.1093/petrology/egl025 Kuritani T., Ohtani E., Kimura J. I., 2011. Intensive Hydration of the Mantle Transition Zone beneath China Caused by Ancient Slab Stagnation. Nature Geoscience, 4(10): 713–716. DOI:10.1038/ngeo1250 Lei J. S., Xie F. R., Fan Q. C., et al., 2013. Seismic Imaging of the Deep Structure under the Chinese Volcanoes: An Overview. Physics of the Earth and Planetary Interiors, 224: 104–123. DOI:10.1016/j.pepi.2013.08.008 Li J. Y., 2006. Permian Geodynamic Setting of Northeast China and Adjacent Regions: Closure of the Paleo-Asian Ocean and Subduction of the Paleo-Pacific Plate. Journal of Asian Earth Sciences, 26(3/4): 207–224. DOI:10.1016/j.jseaes.2005.09.001 Liu J., Xia Q.-K., Deloule E., et al., 2015a. Recycled Oceanic Crust and Marine Sediment in the Source of Alkali Basalts in Shandong, Eastern China: Evidence from Magma Water Content and Oxygen Isotopes. Journal of Geophysical Research: Solid Earth, 120(12): 8281–8303. DOI:10.13039/501100001809 Liu J., Xia Q.-K., Deloule E., et al., 2015b. Water Content and Oxygen Isotopic Composition of Alkali Basalts from the Taihang Mountains, China: Recycled Oceanic Components in the Mantle Source. Journal of Petrology, 56(4): 681–702. DOI:10.1093/petrology/egv013 Liu X., Zhao D. P., Li S. Z., et al., 2017. Age of the Subducting Pacific Slab beneath East Asia and Its Geodynamic Implications. Earth and Planetary Science Letters, 464: 166–174. DOI:10.13039/501100001691 Maruyama S., Isozaki Y., Kimura G., et al., 1997. Paleogeographic Maps of the Japanese Islands: Plate Tectonic Synthesis from 750 Ma to the Present. The Island Arc, 6(1): 121–142. DOI:10.1111/j.1440-1738.1997.tb00043.x Marzoli A., Piccirillo E. M., Renne P. R., et al., 2000. The Cameroon Volcanic Line Revisited: Petrogenesis of Continental Basaltic Magmas from Lithospheric and Asthenospheric Mantle Sources. Journal of Petrology, 41(1): 87–109. DOI:10.1093/petrology/41.1.87 Mattey D., Lowry D., Macpherson C., 1994. Oxygen Isotope Composition of Mantle Peridotite. Earth and Planetary Science Letters, 128(3/4): 231–241. DOI:10.1016/0012-821x(94)90147-3 Muehlenbachs K., Clayton R. N., 1976. Oxygen Isotope Composition of the Oceanic Crust and Its Bearing on Seawater. Journal of Geophysical Research, 81(23): 4365–4369. DOI:10.1029/jb081i023p04365 Page F. Z., Kita N. T., Valley J. W., 2010. Ion Microprobe Analysis of Oxygen Isotopes in Garnets of Complex Chemistry. Chemical Geology, 270(1/2/3/4): 9–19. DOI:10.1016/j.chemgeo.2009.11.001 Plank T., Langmuir C. H., 1998. The Chemical Composition of Subducting Sediment and Its Consequences for the Crust and Mantle. Chemical Geology, 145(3/4): 325–394. DOI:10.1016/s0009-2541(97)00150-2 Putlitz B., Matthews A., Valley J. W., 2000. Oxygen and Hydrogen Isotope Study of High-Pressure Metagabbros and Metabasalts (Cyclades, Greece): Implications for the Subduction of Oceanic Crust. Contributions to Mineralogy and Petrology, 138(2): 114–126. DOI:10.1007/s004100050012 Rogers N. W., Hawkesworth C. J., Ormerod D. S., 1995. Late Cenozoic Basaltic Magmatism in the Western Great Basin, California and Nevada. Journal of Geophysical Research: Solid Earth, 100(B6): 10287–10301. DOI:10.1029/94jb02738 Sengör, A. M. C. , Natal'in, B. A. , 1996. Paleotectonics of Asia: Fragments of a Synthesis. In: Yin, A. , Harrison, M. , eds. , Tectonic Evolution of Asia. Cambridge University Press, Cambridge. 486-640 Sengör A. M. C., Natal'in B. A., Burtman V. S., 1993. Evolution of the Altaid Tectonic Collage and Palaeozoic Crustal Growth in Eurasia. Nature, 364(6435): 299–307. DOI:10.1038/364299a0 Taylor H. P., 1974. The Application of Oxygen and Hydrogen Isotope Studies to Problems of Hydrothermal Alteration and Ore Deposition. Economic Geology, 69(6): 843–883. DOI:10.2113/gsecongeo.69.6.843 Valley, J. W. , Kita, N. T. , 2009. In Situ Oxygen Isotope Geochemistry by Ion Microprobe. Mineralogical Association of Canada Short Course. Mineralogical Association of Canada, Toronto. 19-63 Wang X. C., Wilde S. A., Li Q. L., et al., 2015. Continental Flood Basalts Derived from the Hydrous Mantle Transition Zone. Nature Communications, 6: 7700. DOI:10.1038/ncomms8700 Wang Z., Eiler J. M., 2008. Insights into the Origin of Low-δ18O Basaltic Magmas in Hawaii Revealed from in situ Measurements of Oxygen Isotope Compositions of Olivines. Earth and Planetary Science Letters, 269(3/4): 377–387. DOI:10.1016/j.epsl.2008.02.018 Wei W., Xu J. D., Zhao D. P., et al., 2012. East Asia Mantle Tomography: New Insight into Plate Subduction and Intraplate Volcanism. Journal of Asian Earth Sciences, 60: 88–103. DOI:10.1016/j.jseaes.2012.08.001 Woodhead J. D., Greenwood P., Harmon R. S., et al., 1993. Oxygen Isotope Evidence for Recycled Crust in the Source of EM-Type Ocean Island Basalts. Nature, 362(6423): 809–813. DOI:10.1038/362809a0 Workman R. K., Hart S. R., 2005. Major and Trace Element Composition of the Depleted MORB Mantle (DMM). Earth and Planetary Science Letters, 231(1/2): 53–72. DOI:10.1016/j.epsl.2004.12.005 Wotzlaw J. F., Bindeman I. N., Stern R. A., et al., 2015. Rapid Heterogeneous Assembly of Multiple Magma Reservoirs Prior to Yellowstone Supereruptions. Scientific Reports, 5(1). DOI:10.1038/srep14026 Xia Q.-K., Dallai L., Deloule E., 2004. Oxygen and Hydrogen Isotope Heterogeneity of Clinopyroxene Megacrysts from Nushan Volcano, SE China. Chemical Geology, 209(1/2): 137–151. DOI:10.1016/j.chemgeo.2004.04.028 Xu Y. G., Zhang H. H., Qiu H. N., et al., 2012. Oceanic Crust Components in Continental Basalts from Shuangliao, Northeast China: Derived from the Mantle Transition Zone?. Chemical Geology, 328: 168–184. DOI:10.1016/j.chemgeo.2012.01.027 Yu S. Y., Xu Y. G., Huang X. L., et al., 2009. Hf-Nd Isotopic Decoupling in Continental Mantle Lithosphere beneath Northeast China: Effects of Pervasive Mantle Metasomatism. Journal of Asian Earth Sciences, 35(6): 554–570. DOI:10.1016/j.jseaes.2009.04.005 Zhang J., Zhang H. F., 2007. Compositional Features and P-T Conditions of Granulite Xenoliths from Late Cretaceous Mafic Dike, Qingdao Region. Acta Petrologica Sinica, 23(5): 1133–1140. Zhang M., Stephenson P. J., O'Reilly S. Y., et al., 2001. Petrogenesis and Geodynamic Implications of Late Cenozoic Basalts in North Queensland, Australia: Trace-Element and Sr-Nd-Pb Isotope Evidence. Journal of Petrology, 42(4): 685–719. DOI:10.1093/petrology/42.4.685 Zhou X., Armstrong R. L., 1982. Cenozoic Volcanic Rocks of Eastern China-Secular and Geographic Trends in Chemistry and Strontium Isotopic Composition. Earth and Planetary Science Letters, 58(3): 301–329. DOI:10.1016/0012-821x(82)90083-8 Zou H. B., Zindler A., Xu X. S., et al., 2000. Major, Trace Element, and Nd, Sr and Pb Isotope Studies of Cenozoic Basalts in SE China: Mantle Sources, Regional Variations, and Tectonic Significance. Chemical Geology, 171(1/2): 33–47. DOI:10.1016/s0009-2541(00)00243-6