An iron isotopic perspective for the material flow from off-axis mantle plume to mid-ocean ridge system

doi:10.21203/rs.3.rs-3744530/v1

Download PDF

Article

An iron isotopic perspective for the material flow from off-axis mantle plume to mid-ocean ridge system

https://doi.org/10.21203/rs.3.rs-3744530/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The interaction between the mid-oceanic ridge (MOR) and the mantle plume is a crucial aspect for understanding Earth’s internal works. However, the compositions and styles of material flow from off-axis plumes to the MOR system still remains elusive. This study examined the iron (Fe) isotope composition of basalts from the South Mid-Atlantic Ridge (SMAR) for the first time. The results showed a consistent relationship between the Fe isotope compositions, trace element ratios, and traditional radiogenic isotopic compositions of lavas from the SMAR and the off-axis Saint Helena plume. This suggests that variations in the Fe isotope compositions of the SMAR mantle source may be influenced by the contribution of the Saint Helena plume. We propose that heavy-Fe pyroxenites are widely distributed within the lithosphere-asthenosphere boundary (LAB) layer in the region between the plume and ridge. These plume-related pyroxenites can transport plume signatures with heavy Fe isotopic compositions into the MOR system at the LAB level.

Earth and environmental sciences/Solid Earth sciences/Geochemistry

Earth and environmental sciences/Solid Earth sciences/Petrology

Earth and environmental sciences/Solid Earth sciences/Geodynamics

The debate between the “top-down” (plate model) and “bottom-up” (plume model) approaches to understanding the Earth’s engine has persistently continued for over fifty years^{1, 2}. Mid-ocean ridges (MOR) and hotspots are surface manifestations of two geodynamic systems, namely plate tectonics and mantle plumes, and they offer valuable insights into the workings of the Earth's interior³. To understand deep processes within the Earth, such as the boundaries between the crust and mantle, the core and mantle, and the lithosphere and asthenosphere, lavas found in certain tectonic locations can be studied alongside experimental mineralogy, petrology, and seismic tomography techniques⁴. Despite these efforts^{5, 6}, unraveling the material exchange and dynamic features of the lithosphere-asthenosphere boundary (LAB) continues to be a challenge.

The global MOR system is usually influenced by on/off-axis mantle plumes^{7, 8}, and materials from these plumes tend to migrate towards extensional and fracture environments in the MOR region^{9, 10}. When off-axis plumes interact with ridges, the material from the mantle plumes may transport along the bottom of the lithosphere or the top of the asthenosphere into the MOR system^{11, 12}. The heterogeneity in the lithology and composition of the MOR mantle sources in the vicinity of mantle plumes has been regarded as a result of the influence of mantle plumes^{13, 14}. However, the specific process through which off-axis mantle plumes affect the lithology and composition of the MOR mantle sources at the LAB layer level remains unclear¹⁵.

Previous studies have showed that melt deposits with a low degree of partial melting in the LAB layer can penetrate and crystallize into the bottom of the lithosphere, forming lithosphere-type pyroxenite under high pressure conditions^{16, 17}. These melt deposits have the potential to randomly enter the depleted peridotite mantle domain and undergo partial melting in the MOR system^{17, 18, 19}. Moreover, the distribution of the enriched mantle plume compositional signature along the MOR system can indicate the propagation and migration path of plume materials into the MOR mantle source^{20, 21}. However, it remains unclear whether pyroxenite originated from mantle plumes (referred to as plume-type pyroxenite), which serves as carrier materials with specific mantle plume componential signatures, can be directly transported from the off-axis plume into the MOR magma system. Recent studies have shown that Fe isotope variability of MORB (mid-ocean ridge basalts) and OIB (ocean island basalts) can be linked to source heterogeneity^{17, 22, 23, 24, 25}, suggesting that Fe isotopes, along with trace elements and radiogenic isotopes, can serve as effective tracers to identify lithological heterogeneity in the Earth’s mantle and provide geochemical evidence for the origin of pyroxenite components in the LAB layer.

The South Mid-Atlantic Ridge (SMAR) system, coupled with the presence of the Saint Helena mantle plume, provides an excellent research area for investigating the interaction between off-axis mantle plumes and MOR. The Saint Helena plume, as the earliest influence on the lithosphere underlying Gondwana continent in the South Atlantic, played an important role in the opening of the South Atlantic and the formation of the SMAR system^{26, 27}. Currently, the off-axis Saint Helena plume still exerts a significant influence on the geochemical and geophysical properties of the nearby SMAR system^{20, 21, 28, 29, 30}. This study focuses on the section of the SMAR system closest to the Saint Helena plume (Fig. 1A) to analyze the Fe isotope composition of the SMAR basalts (Fig. 1B) and explore the compositions and dynamics of the LAB layer located between the mantle plume and the ridge.

The Fe isotope signatures of the SMAR MORBs. The basaltic samples show a wide variation in δ⁵⁶Fe values, ranging from + 0.10 to + 0.22‰ with an average of + 0.13 ± 0.03‰ (2SD) (Fig. S1). This indicates significant variability in mantle Fe isotopes over a large scale (> 1400 km) on the ridge segment. The positive correlations observed between δ⁵⁶Fe values and K (R² = 0.69) and Nb (R² = 0.70) (Fig. S2) suggest that seafloor alteration does not influence these samples¹⁷. Although there is no correlation between δ⁵⁶Fe values and indices of magma differentiation (e.g., Mg#) (R² < 0.01) (Fig. S3), which suggests that fractional crystallization is not a major factor in explaining the observed δ⁵⁶Fe variations^{33, 34}, correction for the effect of fractional crystallization is still necessary. Hence, the Fe isotopes were corrected to a melt with Mg#=72, resulting in a decrease of only 0.01–0.05‰ from the measured values (within acceptable analytical error). The corrected δ⁵⁶Fe values showed a significant correlation with the measured values (R² = 0.91, N = 19) (Fig. S4), indicating that magma differentiation has a limited effect on the δ⁵⁶Fe of the SMAR MORBs^{19, 34}. Refer to the supplementary materials for more details.

Correlations between Fe isotopes and incompatible element ratios, as well as radiogenic Sr-Nd-Pb-Hf isotopes. Several processes, including seawater alteration, fractional crystallization, partial melting, and magma mixing, could potentially contribute to the observed variation in δ⁵⁶Fe in the MORBs^{33, 35, 36, 37, 38, 39}. The Fe isotope composition (δ⁵⁶Fe_corr) after fractional crystallization correction is positively correlated with the isotopic ratios of ²⁰⁶Pb/²⁰⁴Pb (R² = 0.45), ⁸⁷Sr/⁸⁶Sr (R² = 0.33), and incompatible element ratios of (La/Sm)_N (R² = 0.83), Sm/Yb (R² = 0.73), Nb/Zr (R² = 0.70), and K₂O/TiO₂ (R² = 0.83) (Fig. 2). On the contrary, Fe isotope composition is negatively correlated with the isotopic ratios of ¹⁴³Nd/¹⁴⁴Nd (R² = 0.60) and ¹⁷⁶Hf/¹⁷⁷Hf (R² = 0.58) (Fig. 2). This suggests that the δ⁵⁶Fe variation in the SMAR 12–25°S MORBs is primarily derived from mantle source beneath these SMAR segments^{17, 19}. Furthermore, an increase in Fe isotope composition indicates a greater contribution of magma derived from pyroxenite-induced melts from the Saint Helena mantle plume (Fig. 2). It is important to note that radiogenic isotope ratios are influenced by parent-daughter variations in the source of melts over time, rather than petrogenesis⁴⁰. Thus, the correlations between corrected Fe isotopes and radiogenic Sr-Nd-Pb-Hf isotopes reflect a mixing between two physically distinct components in the sub-ridge mantle source.

The variation in Fe isotopic composition observed in SMAR MORBs cannot solely be explained by fractional crystallization^{33, 37, 38, 39} or melting of peridotite mantle^{35, 36, 41, 42, 43}. Instead, the melting of mantle plume-type pyroxenite (Fig. 3A) and oceanic lithosphere-type pyroxenite (Fig. S5) is responsible for the range of Fe isotope variation^{17, 19}, indicating that the mantle source region in the study area could be influenced by mantle plume-related material components. The systematic correlation between Fe isotopes and incompatible element ratios, as well as radiogenic Sr-Nd-Pb-Hf isotopes, suggests that the enrichment materials of the Saint Helena plume contribute to the progressively heavier Fe isotope components in SMAR MORBs (Fig. 2). Furthermore, the spatial distribution of Fe isotopes along the ridge axis indicates that the ridge region affected by the Saint Helena plume has a relatively heavier Fe isotopic composition, with the closest ridge segment (SMAR 15.4°S) to the Saint Helena plume exhibiting the heaviest Fe isotopic composition (Fig. 4A) (Fig. S4). Therefore, there appears to be a correlation between the lithosphere-type pyroxenite and plume-type pyroxenite present in the MOR mantle domain and the interaction between the Saint Helena plume and the SMAR system. The origin of these pyroxenites with heavy Fe isotopic compositions in the SMAR mantle source can be explained by multiple models, which can provide insights into the material component and geodynamics of LAB layer between the off-axis mantle plume and ridge.

Model I: Fe-heavy pyroxenites are primarily formed through low-F melt metasomatism at the base of the lithosphere. The lithosphere in different regions and the low-degree of partial melting (low-F) melts derived from different mantle domains will exhibit distinct compositional features⁶. Consequently, based on the occurrences of pyroxene-rich rocks in the upper mantle inspired the “marble cake model”⁴⁶, it is highly unlikely to observe a systematic correlation among Fe-Sr-Nd-Pb-Hf isotopes and incompatible elements in lithosphere-type pyroxenites resulting from metasomatism between low-F melts and lithosphere within a such large area (1300 km×1300 km) in this study. Furthermore, even if these lithosphere-type pyroxenites happen to have similar geochemical, radiogenic isotopic, and Fe isotopic compositions, their distribution in the mantle is random. However, this study reveals that the spatial distribution of Fe isotopic composition in basaltic rocks remains consistent with the previously indicated characteristics of mantle plume influence on the SMAR system (Fig. 4A)²⁰. Therefore, the propagation and migration mechanism of these lithosphere-type pyroxenites in the asthenosphere should be controlled by the material flow of mantle plume. Thus, if Fe-heavy pyroxenites were primarily formed through low-F melt metasomatism, it would require the formation of melts with unified Fe-Sr-Nd-Pb-Hf isotope and incompatible element compositions through low-degree partial melting processes in the LAB layer. Simultaneously, these melts will interact with the lower part of the overlying lithosphere to form lithosphere-type pyroxenites with unified compositions, which then systematically move into the MOR magmatic system influenced by plume and ridge interaction (Fig. 4B).

Furthermore, it suggests the presence of a coherent asthenosphere flow that propagates from the mantle plume to the MOR region at the base of the lithosphere. It can also explain the formation of low-F melts with similar Fe-Sr-Nd-Pb-Hf isotope and incompatible element compositions, as these melts can be derived from the asthenosphere influenced by the same mantle plume. Thus, this model can be achieved based on the regularity of asthenosphere flow under plume-ridge interaction conditions. In addition, due to limitations such as the quantity of melts formed through low-degree melt metasomatism⁶, the propagation ability of melts in the solid phase⁴⁷, and consumption of melt components by interacting lithosphere⁴⁸, it will be difficult for low-F melts to propagate along the base of the lithosphere to distances exceeding hundreds of kilometers (the shortest distance between the SMAR and Saint Helena Island is about 750 km). Therefore, there should be extensive melt-rock interactions in the LAB layer located between mantle plumes and MORs, indicating widespread material exchange in the LAB layer (Fig. 4C).

Model II: Fe-heavy pyroxenites are originated from mantle plumes. Previous studies have explained the correlation between Fe isotopes and incompatible element ratios as a result of ancient low-F melt and rock interaction in localized areas¹⁷. Recent research based on global data has indicated a weak correlation between Fe isotopes and incompatible element ratios in the mantle source of global MORB, suggesting that this correlation may also be a result of low-F melt and lithosphere interaction¹⁹. However, there are several unconfirmed assumptions in this regard. First, it is assumed that the metamorphic process acts as a “closed system” where the geochemical signatures, including Fe isotopes, of metasomatized veins/veinlets are not modified^{38, 49}. Second, the behavior of heavy iron isotopes relative to light iron isotopes in the melting process, under conditions of unknown mantle redox, resembling incompatible elements, has not been experimentally confirmed⁴³, and this model does not assume that heavy iron isotopes behave similarly to incompatible elements and are more easily incorporated into melts⁴². Furthermore, no correlation between δ⁵⁶Fe-Fe³⁺/Fe²⁺ and ancient low-F enrichment in MORB, which are believed to be enriched by low-degree melts, has been found¹⁹.

In addition, a low-F melt has high ratios of incompatible elements and can generate enriched pyroxene rocks/veinlets through interaction with depleted mantle peridotite⁵⁰. Such pyroxenites of this origin would also exhibit the composition of radiogenic isotopes found in low-F melts⁴⁸. However, existing research data shows significant differences in the radiogenic isotopes composition in incompatible element-enriched MORBs, and there is no consistent correlation between heavy Fe isotopes and enriched radiogenic isotope components¹⁹. Therefore, the observed unified correlation between Fe isotopes, incompatible elements, and radiogenic isotope composition in basaltic magmas within small regions may be the result of metasomatism between low-F melts and lithosphere. However, it could also be the result of contamination from mantle source with heavier Fe isotopes and enriched components related to mantle plume. On a global scale, there is a weak positive correlation between heavy Fe isotopes and incompatible elements in MORB, but no correlation with radiogenic isotopes (especially Pb isotopes)¹⁹. This cannot solely be attributed to low-F melts and lithosphere metasomatism but may also be influenced by different origins of enriched mantle plume components in different MOR regions.

The systematic variations in incompatible element ratios and radiogenic isotope signatures of MORB in this study area are influenced by the HIMU-type Saint Helena plume²⁰. Therefore, the correlation between Fe isotopes and incompatible element ratios with radiogenic isotope signatures indicates that the Fe isotope composition in SMAR MORB is also influenced by the Saint Helena mantle plume (Fig. 2). Furthermore, we modeled the mixing of HIMU plume-type pyroxenite with depleted mantle peridotite in the mantle source region (Fig. 3). The results show that the Fe isotope and radiogenic Sr-Nd-Hf isotope composition in the study area is consistent with a mixture of depleted peridotite mantle undergoing 5% (or 10%) partial melting and HIMU-type mantle undergoing 10% (or 25%) partial melting in varying proportions (Fig. 3B, C, D). This result indicates the presence of pyroxene rock components from the HIMU-type mantle plume in the SMAR mantle source region. In the melting region of the ridge, as the HIMU plume-type pyroxenites take place partial melting with surrounding depleted mantle peridotite, it leads to the formation of basaltic lavas with the HIMU-type radiogenic isotope and heavy Fe isotope compositions along the SMAR system. Finally, this model is consistent with the composition and spatial characteristics of Fe isotopes of the SMAR basalts (Fig. 4A), which further indicates the presence of a large amount of plume-type pyroxenites in the plume-affected asthenosphere flow between the mantle plume and the MOR. Moreover, under the control of asthenosphere flow affected by the interaction of the plume and ridge^{20, 21}, these plume-type pyroxenites, as carriers of the enriched mantle plume component signal, can be systematically transported to the sub-ridge melting zone.

Model III: Fe-heavy pyroxene in the SMAR mantle source is a combination of lithosphere- and plume-type pyroxenite. This model can explain the Fe isotope, incompatible element, and radiogenic isotope composition of the SMAR MORB. Low-F melt metasomatism-induced lithosphere-type pyroxenites also exhibit enrichment characteristics of mantle plumes with heavy Fe isotopes (Fig. 4B, C)^{17, 19}. Additionally, the flow of plume-affected asthenosphere transports both mantle plume-type and lithosphere-type pyroxenites, which are randomly distributed at the bottom of the lithosphere, into the melting zone of the MOR system (Fig. 4C)^{19, 51}. The plume-affected asthenosphere flow, with its higher temperature⁵², would promote extensive melting and metasomatism, causing more lithosphere-type pyroxenites to transition into the MOR region regularly (Fig. 4C). This model suggests that lithological heterogeneity is enhanced in the plume-affected MOR mantle source regions, as evidenced by the Fe isotope composition and its correlation with other elements and radiogenic isotopes.

Furthermore, there may be another possible model: after off-axis mantle plume material ascends, it undergoes metasomatic reactions with the overlying oceanic lithospheric mantle, resulting in the formation of plume-induced pyroxenite at the top of the mantle plume⁵³. Under conditions of ridge-plume interaction, this plume-induced pyroxenite, which inherits the enriched components of the mantle plume and heavy Fe isotope composition, is transported to the MOR magma system within the asthenosphere. In this model, extensive low-degree partial melting and metasomatic reactions (forming lithosphere-type pyroxenite) at the LAB layer and direct entry of plume-type pyroxenite into the MOR mantle source region may be not required for the formation of basalts with Fe isotope composition characteristics observed in this study area. Instead, the formation of basalts with these characteristics in the MOR region can be attributed to the widespread distribution of plume-induced pyroxenite in the asthenosphere located between plume and ridge. This model may also explain the lack of significant heavy Fe isotope characteristics in the basalts from Saint Helena Island⁴⁵. It suggests that the composition of these basalts is the result of partial melting of plume-induced pyroxenite, generated through metasomatic reactions between the mantle plume and lithosphere, along with partial melting of surrounding depleted peridotite, rather than direct and extensive melting of plume-type pyroxenite in the Saint Helena Island mantle source region.

Our research indicates that the heterogeneity of the mantle source in the SMAR may originate from different types of pyroxenite, including lithosphere-type, plume-type, plume-induced, or a combination of them. Regardless of their origin, our findings consistently show that the Fe isotope composition in SMAR basalts is affected by pyroxenite associated with the Saint Helena plume. This is the first time that the geochemical evidence demonstrates that plume-related pyroxenite acts as carriers, transmitting the characteristics of mantle plume composition, such as enriched geochemical and radiogenic isotopes, as well as heavy Fe isotopes, into the MOR magmatic system. This transmission of pyroxenite occurs through the movement of plume-influenced asthenosphere at the base of the lithosphere, revealing the material component and dynamics of the LAB layer located between off-axis mantle plumes and the MOR system.

Sample collection and preparation. The materials for this study were collected from the SMAR 13.2–24.2°S segments that are within about 750 km of the Saint Helena hotspot (Fig. 1). Fresh basaltic samples with glassy chilled margins were collected from 19 underwater stations (with depths between 1513 m and 3386 m) using an underwater remotely operated vehicle equipped with manipulator arms and cameras (TVG) for 17 stations, a remote operated vehicle (ROV) for one station (S031-ROV02), and a deep-sea optics investigation vehicle (SHX) for one station (L003-SHX01).

Samples were sawn into slabs, and the fresh central parts (> 250 g) were used for bulk-rock analyses. The central parts were first crushed into decimeter-sized chunks with a hydraulic press and then into centimeter-sized chips with a stainless-steel pestle and mortar. After the altered and secondary vein materials and olivine and plagioclase phenocrysts were handpicked and removed under a binocular microscope, the remaining chips were powdered in an aluminum ceramic mill. Before digestion for isotope measurements, the chips were leached for 1 h in a 1:1 mixture of H₂O₂ and 4N HNO₃ at room temperature, and rinsed with Milli-Q water.

Whole-rock major and trace element analyses. The bulk-rock major element oxides compositions were determined using inductive coupled plasma-optical emission spectrometry (ICP-OES), and trace elements including high field strength elements (HFSE) and rare earth elements (REE), were analyzed using an X Series II inductively coupled plasma-mass spectrometer (ICP-MS) at the Key Laboratory of Marine Geology and Metallogeny at the Ministry of Natural and Resources in China (KLMGM-MNR). The BHVO-2 standards were used to monitor drift in mass response during mass spectrometric measurements. Data precisions is ± 0.2%−2% for major elements present in concentrations > 1 wt% and ~ ± 2%−5% for minor elements present in concentrations < 1.0 wt%. Loss on ignition (LOI) was also measured at 1050 ℃ after samples were dried at a temperature of 100 ℃ at KLMGM-MNR (Table S1).

Whole-rock Sr-Nd-Pb-Hf isotope analyses. About 200 mg of rock powder for each sample was dissolved with double distilled HNO₃ + HCl + HF in a high-pressure jacket equipped Teflon beaker at 200 ℃ for 24 h. The samples were then dried and re-dissolved with 2 ml 3N HNO₃ for 2 h. The final sample solution was first loaded onto Sr-spec resin columns to separate Sr and Pb, where the eluted sample solution was collected and then loaded onto AG 50W-X8 resin columns to separate out the REE. The eluted sample solution was then collected from the AG 50W-X8 resin columns was collected and then loaded onto Ln-spec resin columns to collect Hf. The separated REE solution was dried and re-dissolved with 0.25 N HCl before being loaded onto Ln-spec resin columns to collect Nd. This streamlined methodology is a modified version of the method used by Sun et al. (2020)¹⁷, and we ran these analyses at the Institute of Oceanology, Chinese Academy of Sciences (IOCAS). The Sr and Nd isotopes were measured on a high-resolution multi-collector inductively coupled plasma-mass spectrometer (MC-ICP-MS) (Nu Instruments Ltd., Wrexham, North Wales, UK) at KLMGM-MNR. During the period of analysis, the NBS987 standard yielded an average value of ⁸⁷Sr/⁸⁶Sr = 0.710264 ± 13 (2σ, n = 30), the Jndi-1 standard gave an average value of ¹⁴³Nd/¹⁴⁴Nd = 0.512116 ± 7 (2σ, n = 30). Total procedural blanks were < 200 pg for Sr and < 50 pg for Nd. Details of analytical results are presented in Table S1. The Pb and Hf isotopes were analyzed by using a Nu Plasma MC-ICP-MS in the IOCAS. International standards of Alfa Hf and NBS-981 were used as bracketing standards every five samples to monitor the instrument drift during the analysis of Hf and Pb isotopes, respectively. The measured ¹⁷⁶Hf/¹⁷⁷Hf isotope ratios were normalized for instrumental mass fraction using the exponential law with ¹⁷⁹Hf/¹⁷⁷Hf = 0.7325. The Alfa Hf standard (n = 12) for Hf isotope yielded an average value of ¹⁷⁶Hf/¹⁷⁷Hf = 0.282218 ± 0.000013 (2σ). Pb isotope ratios were normalized for instrumental mass fraction relative to NBS/SRM 997 ²⁰³Tl/²⁰⁵Tl = 0.41891. Analyses of the NBS981 standard (n = 12) for Pb isotope yielded average values of ²⁰⁶Pb/²⁰⁴Pb = 16.930 ± 0.004 (2σ), ²⁰⁷Pb/²⁰⁴Pb = 15.481 ± 0.005 (2σ), and ²⁰⁸Pb/²⁰⁴Pb = 36.676 ± 0.018 (2σ) (Table S1).

Whole-rock Fe isotope analyses. Iron isotopic analyses were performed at the Metal Stable Isotope Laboratory of the University of Science and Technology of China (USTC), China, following the procedure described in previous studies^{54, 55, 56}. A brief description is given here.

After cleaned by an ultra-sonic bath with ethanol for 10 minutes for three times, whole-rock sample powders were dissolved in 7 ml capped Savillex beakers using double-distilled concentrated HF + HNO₃, HCl + HNO₃ and HCl successively. After full dissolution and evaporation of acid, the final chlorides were dissolved in 1 ml 6 N HCl to provide a stock bulk solution ready for chromatographic purification of Fe. Iron was purified using 1 ml pre-cleaned Bio-Rad anion resin (AG1-X8) conditioned with 6 N HCl. Matrix elements were removed using 6 N HCl, and washes were collected to assess Fe loss for yield check. Iron was eluted by 0.4 N HCl followed by 8 N HNO₃ and H₂O. The column chemistry was performed twice to efficiently purify Fe from matrix elements and obtain a pure Fe solution. The Fe yields through column chemistry are nearly 100% and the total procedural blank (from sample dissolution to mass spectrometry) is 4 ng, 7.4 ng, and 23.8 ng for Fe, respectively, negligible compared to ≥ 50 µg Fe loaded on the resin.

Iron isotope ratios were determined using a Thermo-Fisher Neptune Plus multi-collector inductively coupled mass spectrometry (MC-ICPMS) at the USTC. A sample-standard bracketing method was applied to correct instrumental mass bias and time drifts. Iron solutions were diluted to 2 ppm in 2% HNO₃ and bracketed with IRMM04 at the same concentration. ⁵⁴Fe, ⁵⁶Fe, and ⁵⁷Fe (along with ⁵³Cr for correcting any ⁵⁴Cr interference) were analyzed in static mode on L1, C, and H1 Faraday cups, respectively. Measurements were performed in high resolution (M/DM of ~ 8000) using Ni(H) + Ni (Jet) cone assembly with the ⁵⁶Fe sensitivity of ~ 6 V/ppm. Iron isotope ratios are reported in standard δ-notation in per mil relative to Fe standard IRMM04: δ^XFe = [(^XFe/⁵⁴Fe)_sample/(^XFe/⁵⁴Fe)_IRMM04 − 1] ×1000‰, where X = 56 or 57. In this study, the δ⁵⁶Fe of three USGS reference standards are 0.092 ± 0.035‰ (2SD, n = 3) for AVG-2, 0.054 ± 0.062‰ (2SD, n = 3) for BCR-2, and 0.122 ± 0.013‰ (2SD, n = 3) for G-2 (Table S2), consistent with those of previous studies^{42, 54, 56, 57, 58, 59} (Table S2). The plot of δ⁵⁷Fe vs. δ⁵⁶Fe for whole rocks, mineral separates, and standards defines a linear trend with a slope of ~ 0.674 (Fig. S1), consistent with the theoretical kinetic and equilibrium fractionation values (0.672 and 0.678, respectively)⁶⁰. All these evidence assures the accuracy and precision of our Fe isotopic data. These procedures for sample dissolution, column chemistry, and instrumental analysis have been previously reported in detail⁵⁶.

Calculation of Fe isotope fractionation during fractional crystallization. Fe isotope fractionation during fractional crystallization was calculated by correcting the effects of magma fractional crystallization back to the expected primary melts with Mg# = 72, following the methods of Humphreys and Niu (2009)⁶¹. The fractional crystallization model was explored using Petrolog3⁶², starting with the calculated primary melts with Mg# = 72. Olivine, plagioclase, and clinopyroxene were set to crystallize at 0.1 GPa pressure and QFM oxygen fugacity buffer. The modeling results show that olivine and plagioclase are the major liquid phases in the melts. Because plagioclase is a Fe-poor mineral, we corrected the Fe isotope compositions for the effect of fractional crystallization of olivine, following Sossi et al. (2016)³⁹ and Chen et al. (2019)³³. For the correction, we used a starting Fe³⁺/∑Fe = 0.15 and the isotope fractionation factor between olivine and melt ∆⁵⁶Fe_Ol−Melt = 2853 × (< F > _Ol − <F > _Melt)/T² after Dauphas et al. (2014)⁴³, where the melt temperature (T) is in Kelvin calculated by T (℃) = 1000 + 20 × MgO (wt%)⁶³ and < F > is the average force constant of Fe − O bonds (< F > _Ol = 197 N/m, <F > _Melt = Fe³⁺/∑Fe × <F > _Fe³⁺ _{in melt} (351 N/m) + (1 − Fe³⁺/∑Fe × <F > _Fe²⁺ _{in melt} (199 N/m)). The melts were incrementally adjusted with the addition of olivine until melt Mg# = 72. The calculated ∆⁵⁶Fe_Ol−Melt values were multiplied by following He et al. (2019)⁴⁴. The modeling results and the corrected Fe isotope compositions are given in Supplementary Table S3.

Modeling melting-induced mixing of two mantle component. The model of melting-induced mixing of two mantle components describes the scenario where varying proportions of pyroxenite, which is of low-degree melt metasomatic origin, are randomly or non-uniformly dispersed in the depleted mantle matrix. Melting-induced mixing⁵¹ can be understood as either melting two different lithologies to varying extents and then mixing the resulting melts, or melting a composite mantle containing a pre-mechanically well-mixed enriched pyroxenitic lithology with depleted peridotitic lithology in varying proportions. First, the incompatible element composition was modeled by incongruent melting of a spinel lherzolite and a garnet pyroxenite to different degrees (1–35%) to generate different melts⁴⁸. The incongruent melting relationship for garnet pyroxenite follows: 0.9 cpx + 0.1 garnet = 1.000 Melt, with initial modes of 0.7 cpx and 0.3 garnet²². Relevant partition coefficients were taken from Gale et al. (2011)⁶⁴. The incompatible element compositions of the peridotite were from Workman and Hart (2005)⁶⁵, and the averaged compositions of pyroxenite were from Guo et al. (2023)¹⁹. Second, the Fe isotope components were further modeled. The δ⁵⁶Fe composition of the melts was assumed to be a mixture of the peridotite-derived melts (Table S4 and S5) and lithosphere-type (Table S6 and S7) and plume-type pyroxenite-derived melts (Table S8 and S9). The δ⁵⁶Fe composition of the melts was calculated using non-modal batch melting mass balance equations, mainly following the methods of Williams and Bizimis (2014)²².

$${{\delta }}^{56}{\text{F}\text{e}}_{\text{m}\text{e}\text{l}\text{t}} = {{\delta }}^{56}{\text{F}\text{e}}_{\text{s}\text{o}\text{u}\text{r}\text{c}\text{e}}- \left(1-\text{F}\times \frac{{\text{F}\text{e}\text{O}}_{\text{m}\text{e}\text{l}\text{t}}}{{\text{F}\text{e}\text{O}}_{\text{s}\text{o}\text{u}\text{r}\text{c}\text{e}}}\right)\times {\varDelta }^{56}{\text{F}\text{e}}_{\text{r}\text{e}\text{s}\text{i}\text{d}\text{u}\text{e}-\text{m}\text{e}\text{l}\text{t}}$$

where F is the degree of partial melting. FeO_melt is here treated approximately as trace elements. The FeO and δ⁵⁶Fe compositions of the melts derived from different partial melting degrees (1–35%) of the lherzolite and pyroxenite components were calculated separately (Table S10). The initial FeO and δ⁵⁶Fe compositions of the lherzolite and pyroxenite are listed in Table S4, S6, and S8. Using these results, we are then able to calculate the final δ⁵⁶Fe composition of the mixed melt:

$${{\delta }}^{56}{\text{F}\text{e}}_{\text{m}\text{i}\text{x}} = \frac{\text{p}\ast {\text{F}\text{e}\text{O}}_{\text{L}}\ast {{\delta }}^{56}{\text{F}\text{e}}_{\text{L}} +\left(1-\text{p}\right)\ast {\text{F}\text{e}\text{O}}_{\text{P}}\ast {{\delta }}^{56}{\text{F}\text{e}}_{\text{P}}}{\text{p}\ast {\text{F}\text{e}\text{O}}_{\text{L}}+\left(1-\text{p}\right)\ast {\text{F}\text{e}\text{O}}_{\text{P}}}$$

Where p is the lherzolite-derived melts fraction in the mixed melts, δ⁵⁶Fe_L and FeO_L represent lherzolite-derived melts, and δ⁵⁶Fe_P and FeO_P represents pyroxenite-derived melts.

Finally, we conducted mixing calculations of Sr-Nd-Hf isotopes and δ⁵⁶Fe values for the melt components originating from HIMU-type pyroxenite and DMM end-member with different degrees of partial melting. The initial FeO and δ⁵⁶Fe compositions of the HIMU-type pyroxenite and DMM are listed in Table S11. The Sr-Nd-Hf isotopic data and element components are cited from Zhang et al. (2021)²⁴. The results of the mixing calculations are presented in Table S12.

Competing interests

The authors declare no competing interests.

Author contributions

H.T.Z. and Q.S.Y. conceived this research and wrote the manuscript; X.F.S. and C.S.L. designed this project; F.Y.S. completed the calculation of Fe isotope fractionation during fractional crystallization; F.H., W.Q.Y. and R.R.W. completed the laboratory analyses; All authors contributed extensively to the interpretation of the data, discussions and preparation of the manuscript.

Acknowledgements

We would like to thank the scientists and crew onboard the R/V DAYANGYIHAO for their help with observations during the Chinese Ocean Survey Expedition nos. 22 and 26. We are grateful to Dr. P.Y. Guo for helpful discussion. This work was supported by National Natural Science Foundation of China (grant number 42276078); Basic Scientific Fund for National Public Research Institutes of China (grant number GY0222Q04); China Ocean Mineral Resources R & D Association Project (grant numbers DY135-S2-2, DY135-S2-2-01); Project of Laoshan Laboratory (grant number LSKJ202204103); Taishan Scholarship from Shandong Province.

Anderson DL (2001) Top-down tectonics? Science 293:2016–2018
Meibom A, Anderson D (2003) The statistical upper mantle assemblage. Earth Planet Sci Lett 217:123–139
Helffrich GR, Wood BJ (2001) The Earth's mantle. Nature 412:501–507
French SW, Romanowicz BA (2014) Whole-mantle radially anisotropic shear velocity structure from spectral-element waveform tomography. Geophys J Int 199:1303–1327
Hoink T, Jellinek A, Lenardic A (2011) Viscous coupling at the lithosphere-asthenosphere boundary. Geochem Geophys Geosyst 12:Q0AK02
Niu YL, Green DH (2018) The petrological control on the lithosphere-asthenosphere boundary (LAB) beneath ocean basins. Earth-Sci Rev 185:301–307
Ribe NM (1996) The dynamics of plume-ridge interaction: 2. off-ridge plumes. J Geophys Res 101(B7):16195
Ito G, Lin J, Graham D (2003) Observational and theoretical studies of the dynamics of mantle plume–mid-ocean ridge interaction. Rev Geophys 41(4):1017
Ito G, Bianco T (2014) Patterns in Galápagos magmatism arising from the upper mantle dynamics of plume-ridge interaction, in The Galapagos: A Natural Laboratory for the Earth Sciences, 245–261
Mittal T, Richards MA (2017) Plume-ridge interaction via melt channelization at Galápagos and other near-ridge hotspot provinces. Geochem Geophys Geosyst 18:1711–1738
Montagner JP, Ritsema J (2001) Interaction between ridges and plumes. Science
Murton BJ, Taylor RN, Thirlwall MF (2002) Plume–ridge interaction: a geochemical perspective from the Reykjanes Ridge. J Petrol 43:1987–2012
Ito G, Keken PE (2007) V. 7.09–hot spots and melting anomalies. Treatise on Geophysics 17(6):371–435
Kelley KA, Kingsley R, Schilling JG (2013) Composition of plume-influenced Mid‐ocean ridge lavas and glasses from the Mid‐Atlantic ridge, East Pacific rise, Galápagos spreading center, and Gulf of Aden. Geochem Geophys Geosyst 14(1):223–242
French SW, Romanowicz BA (2015) Broad plumes rooted at the base of the Earth’s mantle beneath major hotspots. Nature 525:95–99
Pilet S, Baker MB, Stolper EM (2008) Metasomatized lithosphere and the origin of alkaline lavas. Science 320:916–919
Sun P et al (2020) Large iron isotope variation in the eastern Pacific mantle as a consequence of ancient low-degree melt metasomatism. Geochim Cosmochim Acta 286:269–288
Halliday AN et al (1995) Incompatible trace elements in OIB and MORB and source enrichment in the sub-oceanic mantle. Earth Planet Sci Lett 133:379–395
Guo PY et al (2023) Low-degree melt metasomatic origin of heavy Fe isotope enrichment in the MORB mantle. Earth Planet Sci Lett 601:117892
Zhang HT et al (2021) Tracing material contributions from Saint Helena plume to the South Mid-Atlantic ridge system. Earth Planet Sci Lett 572:117130
Zhang HT et al (2023) The involvement of deep plume–related materials in the South Atlantic Ocean asthenosphere as indicated by isotopic independent component analysis of basalts. Int J Earth Sci 112:1293–1309
Williams HM, Bizimis M (2014) Iron isotope tracing of mantle heterogeneity within the source regions of oceanic basalts. Earth Planet Sci Lett 404:396–407
Konter JG et al (2016) Unusual δ⁵⁶ Fe values in Samoan rejuvenated lavas generated in the mantle. Earth Planet Sci Lett 450:221–232
Nebel O et al (2019) Reconciling petrological and isotopic mixing mechanisms in the Pitcairn mantle plume using stable Fe isotopes. Earth Planet Sci Lett 521:60–67
Soderman CR, Shorttle O, Matthews S, Williams HM (2022) Global trends in novel stable isotopes in basalts: Theory and observations. Geochim Cosmochim Acta 318:388–414
Wilson M (1992) Magmatism and continental rifting during the opening of the South Atlantic Ocean: A consequence of lower cretaceous super-plume activity? Geo Soc London Special Pub 68(1):241–255
Matos RMD, Krueger A, Norton I, Casey K (2021) The fundamental role of the Borborema and Benin–Nigeria provinces of NE Brazil and NW Africa during the development of the South Atlantic Cretaceous Rift system. Mar Petrol Geol 127:104872
Schilling JG (1985) Upper mantle heterogeneities and dynamics. Nature 314(6006):62–67
Schilling JG (1991) Fluxes and excess temperatures of mantle plumes inferred from their interaction with migrating mid-ocean ridges. Nature 352(6334):397–403
Zhang HT et al (2020) Petrology and geochemistry of South Mid-Atlantic Ridge (19°S) lava flows: implications for magmatic processes and possible plume-ridge interactions. Geosci Front 11:1953–1973
O’Conner JM, le Roex AP (1992) South Atlantic hot spot-plume systems: 1. Distribution of volcanism in time and space. Earth Planet Sci Lett 113:343–364
Amante C, Eakins BW (2009) ETOPO1 1 arc-minute global relief model: Procedures, data sources and analysis. Natl Geophys. Data Center, Boulder
Chen S et al (2019) Iron isotope fractionation during mid-ocean ridge basalt (MORB) evolution: evidence from lavas on the East Pacific Rise at 10°30′N and its implications. Geochim Cosmochim Acta 267:227–239
Chen Y, Niu Y, Duan M, Gong H, Guo P (2021) Fractional crystallization causes the iron isotope contrast between mid-ocean ridge basalts and abyssal peridotites. Commun Earth Environ 2:65
Williams HM et al (2004) Iron isotope fractionation and the oxygen fugacity of the mantle. Science 304:1656–1659
Williams HM et al (2005) Systematic iron isotope variations in mantle rocks and minerals: the effects of partial melting and oxygen fugacity. Earth Planet Sci Lett 235:435–452
Teng FZ, Dauphas N, Helz RT (2008) Iron isotope fractionation during magmatic differentiation in Kilauea Iki lava lake. Science 320(5883):1620–1622
Schuessler JA, Schoenberg R, Sigmarsson O (2009) Iron and lithium isotope systematics of the Hekla volcano, Iceland — evidence for Fe isotope fractionation during magma differentiation. Chem Geol 258:78–91
Sossi PA, Nebel O, Foden J (2016) Iron isotope systematics in planetary reservoirs. Earth Planet Sci Lett 452:295–308
Zindler A, Hart SR (1986) Chemical geodynamics. Ann Rev Earth Planet Sci 14:493–571
Weyer S, Ionov DA (2007) Partial melting and melt percolation in the mantle: the message from Fe isotopes. Earth Planet Sci Lett 259:119–133
Dauphas N et al (2009) D. Iron isotopes may reveal the redox conditions of mantle melting from Archean to Present. Earth Planet Sci Lett 288:255–267
Dauphas N et al (2014) Magma redox and structural controls on iron isotope variations in Earth’s mantle and crust. Earth Planet Sci Lett 398:127–140
He Y et al (2019) A nephelinitic component with unusual δ⁵⁶Fe in Cenozoic basalts from eastern China and its implications for deep oxygen cycle. Earth Planet Sci Lett 512:175–183
Zhao J et al (2022) The effect of Fe–Ti oxide separation on iron isotopic fractionation during basalt differentiation. Contrib Mineral Petr 177:101
Allègre CJ, Turcotte DL (1986) Implications of a two-component marble-cake mantle. Nature 323(6084):123–127
Lambart S, Baker MB, Stolper EM (2016) The role of pyroxenite in basalt genesis: Melt-PX, a melting parameterization for mantle pyroxenites between 0.9 and 5 GPa. J Geophys Res 121(8):5708–5735
Niu YL (1997) Mantle melting and melt extraction processes beneath ocean ridges: Evidence from abyssal peridotites. J Petrol 38:1047–1074
Schoenberg R et al (2009) Fe isotope systematics of coexisting amphibole and pyroxene in the alkaline igneous rock suite of the Ilímaussaq Complex, South Greenland. Chem Geol 258:65–77
Niu YL, Hékinian R (2004) Ridge suction drives plume-ridge interactions. In: Hékinian R, Stoffers P (eds) Oceanic Hotspots. Springer-Verlag, New York, pp 285–307. (Chap. 9)
Niu YL, Regelous M, Wendt JI, Batiza R, O’Hara MJ (2002) Geochemistry of near-EPR seamounts: Importance of source vs. process and the origin of enriched mantle component. Earth Planet Sci Lett 199:327–345
Dalton CA, Langmuir CH, Gale A (2014) Geophysical and geochemical evidence for deep temperature variations beneath mid-ocean ridges. Science 344:80–83
Liu CZ et al (2022) Archean cratonic mantle recycled at a mid-ocean ridge. Sci Adv 8:eabn6749
Huang F, Zhang Z, Lundstrom CC, Zhi X (2011) Iron and magnesium isotopic compositions of peridotite xenoliths from Eastern China. Geochim Cosmochim Acta 75:3318–3334
An Y, Huang JX, Griffin WL, Liu C, Huang F (2017) Isotopic composition of Mg and Fe in garnet peridotites from the Kaapvaal and Siberian cratons. Geochim Cosmochim Acta 200:167–185
Huang J, Guo S, Jin QZ, Huang F (2020) Iron and magnesium isotopic compositions of subduction-zone fluids and implications for arc volcanism. Geochim Cosmochim Acta 278:376–391
Craddock PR, Warren JM, Dauphas N (2013) Abyssal peridotites reveal the near-chondritic Fe isotopic composition of the Earth. Earth Planet Sci Lett 365:63–76
Teng FZ, Dauphas N, Huang S, Marty B (2013) Iron isotopic systematics of oceanic basalts. Geochim Cosmochim Acta 107:12–26
Liu SA, Li D, Li S, Teng FZ, Ke S, He Y, Lu Y (2014) High-precision copper and iron isotope analysis of igneous rock standards by MC-ICP-MS. J Anal At Spectrom 29:122–133
Young ED, Galy A, Nagahara H (2002) Kinetic and equilibrium mass-dependent isotope fractionation laws in nature and their geochemical and cosmochemical significance. Geochim Cosmochim Acta 66:1095–1104
Humphreys ER, Niu YL (2009) On the composition of ocean island basalts (OIB): The effects of lithospheric thickness variation and mantle metasomatism. Lithos 112:118–136
Danyushevsky LV, Plechov P (2011) Petrolog3: integrated software for modeling crystallization processes. Geochem Geophys Geosyst 12:1–32
Nisbet EG et al (1993) Constraining the potential temperature of the Archaean mantle: a review of the evidence from komatiites. Lithos 30(3):291–307
Gale A et al (2011) Enriched basalts at segment centers: the lucky strike (37°17′N) and menezgwen (37°50′N) segments of the mid-Atlantic ridge. Geochem Geophy Geosy 12(6):Q06016
Workman RK, Hart SR (2005) Major and trace element composition of the depleted MORB mantle (DMM). Earth Planet Sci Lett 231:53–72

There is NO Competing Interest.

Supplementarytables.xlsx
Supplementary tables
Supplementarymaterial.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

An iron isotopic perspective for the material flow from off-axis mantle plume to mid-ocean ridge system

Status:

Version 1

Abstract

Figures

Introduction

Results

Discussion

Methods

Declarations

Competing interests

Author contributions

Acknowledgements

References

Additional Declarations

Supplementary Files

Status:

Version 1