Bridging the depleted MORB mantle and the continental crust using titanium isotopes
Affiliations | Corresponding Author | Cite as | Funding information
Published by the European Association of Geochemistry
under Creative Commons License CC BY-NC-ND 4.0
Keywords: titanium isotopes, depleted mantle, continental crust, komatiites, MORBs
- Share this article
Article views:3,967Cumulative count of HTML views and PDF downloads.
- Download Citation
- Rights & Permissions
top
Abstract
Figures and Tables
 Figure 1 Positive correlation between δ49Ti and (La/Sm)N values for the komatiite and MORB samples. |  Figure 2 The change of δ49Ti values with time for the komatiite and MORB samples. The average δ49Ti values for the depleted and primitive groups are indicated. Also shown are the average of 12 chondrite groups with 2 sd uncertainty from Deng et al. (2018) and the continental crust value inferred from shale data from Greber et al. (2017b). |  Figure 3 Modelling the Ti isotopic fractionations in the partial melts and residues from partial melting of the mantle or the crust. The grey area in (a) indicates the remaining Ti fraction in the residual mantle after ≈ 8-9 % partial melting (DTi ≈ 0.132; Prytulak and Elliott, 2007), i.e. process (i) illustrated in (b). The orange field in a shows the Ti fraction after 20-30 % partial melting of a basaltic crust to produce the felsic melts equivalent to tonalite-trondhjemite-granodiorite (TTGs) (DTi ≈ 3.2-3.3; Martin et al., 2014), i.e. process (ii) in b. The Δ49Tiresidue−source values of TTGs and corresponding residues were calculated using the Ti isotope data of TTG samples from Greber et al. (2017b) and assuming E-MORBs as their sources. |
| Figure 1 | Figure 2 | Figure 3 |
top
Introduction
The Earth’s mantle is chemically and isotopically heterogeneous (Allègre, 1982
Allègre, C.J. (1982) Chemical geodynamics. Tectonophysics 81, 109−132.
; Zindler and Hart, 1986Zindler, A., Hart, S. (1986) Chemical geodynamics. Annual Review of Earth and Planetary Sciences 14, 493−571.
). Multiple mantle end members, inferred from the radiogenic isotopic compositions (Pb-Pb, Sm-Nd and Rb-Sr) of oceanic basalts, have been defined to describe these heterogeneities (Zindler and Hart, 1986Zindler, A., Hart, S. (1986) Chemical geodynamics. Annual Review of Earth and Planetary Sciences 14, 493−571.
). In particular, the depleted MORB mantle (DMM), i.e. the inferred source of the modern normal-type mid-ocean ridge basalts (N-MORB), is characterised by a significant depletion in the incompatible trace elements (Workman and Hart, 2005Workman, R.K., Hart, S.R. (2005) Major and trace element composition of the depleted MORB mantle (DMM). Earth and Planetary Science Letters 231, 53−72.
), and is to a first order compositionally complementary to the continental crust (Hofmann, 1988Hofmann, A.W. (1988) Chemical differentiation of the Earth: the relationship between mantle, continental crust, and oceanic crust. Earth and Planetary Science Letters 90, 297−314.
). Thus, the DMM has been interpreted as a residual mantle that was homogenised after the extraction of a component similar to the present day continental crust (Hofmann, 1988Hofmann, A.W. (1988) Chemical differentiation of the Earth: the relationship between mantle, continental crust, and oceanic crust. Earth and Planetary Science Letters 90, 297−314.
; McCulloch and Bennett, 1994McCulloch, M.T., Bennett, V.C. (1994) Progressive growth of the Earth's continental crust and depleted mantle: geochemical constraints. Geochimica et Cosmochimica Acta 58, 4717−4738.
; Workman and Hart, 2005Workman, R.K., Hart, S.R. (2005) Major and trace element composition of the depleted MORB mantle (DMM). Earth and Planetary Science Letters 231, 53−72.
). However, since the continental crust is too silicic to have been derived from a one-step partial melting of upper mantle peridotites, the origin of its complementarity with the DMM is still unclear. Intracrustal magmatic differentiation and removal of the mafic/ultramafic complement are required to drive the bulk continental crust to andesitic compositions (Rudnick, 1995Rudnick, R.L. (1995) Making continental crust. Nature 378, 571−578.
; Walter, 2003Walter, M.J. (2003) Melt extraction and compositional variability in mantle lithosphere. In: Carlson, R.W. (Ed.) Treatise on Geochemistry−The Mantle and Core. Elsevier-Pergamon, Oxford, 363−394.
).In order to evaluate the relationship between DMM and the continental crust, quantification of the processes causing the depletions in the incompatible trace element budget of DMM must be understood. Titanium stable isotopes are well suited to discriminate between mantle and crustal melting because they do not appear to fractionate during partial melting of the mantle, whereas they tend to become enriched in the heavier isotopes during magmatic differentiation in the crust (Millet et al., 2016
Millet, M.A., Dauphas, N., Greber, N.D., Burton, K.W., Dale, C.W., Debret, B., Macpherson, C.G., Nowell, G.M., Williams, H.M. (2016) Titanium stable isotope investigation of magmatic processes on the Earth and Moon. Earth and Planetary Science Letters 449, 197−205.
; Greber et al., 2017aGreber, N.D., Dauphas, N., Puchtel, I.S., Hofmann, B.A., Arndt, N.T. (2017a) Titanium stable isotopic variations in chondrites, achondrites and lunar rocks. Geochimica et Cosmochimica Acta 213, 534−552.
,bGreber, N.D., Dauphas, N., Bekker, A., Ptáček, M.P., Bindeman, I.N., Hofmann, A. (2017b) Titanium isotopic evidence for felsic crust and plate tectonics 3.5 billion years ago. Science 357, 1271−1274.
).Based on the enrichments of the radiogenic Nd and Hf isotopes in N-MORB samples (Vervoort et al., 1996
Vervoort, J.D., Patchett, P.J., Gehrels, G.E., Nutman, A.P. (1996) Constraints on early Earth differentiation from hafnium and neodymium isotopes. Nature 379, 624−627.
; Vervoort and Blichert-Toft, 1999Vervoort, J.D., Blichert-Toft, J. (1999) Evolution of the depleted mantle: Hf isotope evidence from juvenile rocks through time. Geochimica et Cosmochimica Acta 63, 533−556.
), the DMM should acquire its high Sm/Nd and Lu/Hf ratios during the late Archean. Archean komatiites are produced by high degree partial melting (25-40 %) of the mantle, and erupt at temperatures >1750 K (Arndt, 2003Arndt, N. (2003) Komatiites, kimberlites, and boninites. Journal of Geophysical Research: Solid Earth 108, doi: 10.1029/2002JB002157.
). These characteristics mitigate chemical and isotopic fractionations during partial melting of their sources, and therefore are sensitive records of the spatio-thermal evolution of the composition of Earth's mantle over their eruptive history (Walter, 2003Walter, M.J. (2003) Melt extraction and compositional variability in mantle lithosphere. In: Carlson, R.W. (Ed.) Treatise on Geochemistry−The Mantle and Core. Elsevier-Pergamon, Oxford, 363−394.
; Puchtel et al., 2009Puchtel, I.S., Walker, R.J., Brandon, A.D., Nisbet, E.G. (2009) Pt-Re-Os and Sm-Nd isotope and HSE and REE systematics of the 2.7 Ga Belingwe and Abitibi komatiites. Geochimica et Cosmochimica Acta 73, 6367−6389.
; Sossi et al., 2016Sossi, P.A., Eggins, S.M., Nesbitt, R.W., Nebel, O., Hergt, J.M., Campbell, I.H., O'Neill, H.S.C., Van Kranendonk, M., Davies, D.R. (2016) Petrogenesis and geochemistry of Archean komatiites. Journal of Petrology 57, 147−184.
). In contrast with Archean komatiites, MORB samples are derived from lower degrees of partial melting (5-15 %) of the modern upper mantle (Asimow and Langmuir, 2003Asimow, P.D., Langmuir, C.H. (2003) The importance of water to oceanic mantle melting regimes. Nature 421, 815−820.
; Workman and Hart, 2005Workman, R.K., Hart, S.R. (2005) Major and trace element composition of the depleted MORB mantle (DMM). Earth and Planetary Science Letters 231, 53−72.
), and are subdivided based on their concentration in incompatible elements from light rare earth element depleted (N-MORB) to enriched (E-MORB) (Workman and Hart, 2005Workman, R.K., Hart, S.R. (2005) Major and trace element composition of the depleted MORB mantle (DMM). Earth and Planetary Science Letters 231, 53−72.
).Here we present a high precision Ti isotopic study of Archean komatiites, and N-MORB and E-MORB. The comparison of the Ti isotopic composition of MORBs with that of Archean komatiites is used to confirm isotopic effects by partial melting of the mantle, and further trace Ti isotopic variations in the mantle since the Archean.
top
Results
Ti Isotopic Variations in the Archean and Modern Mantle-Derived Rocks. Individual komatiites and MORBs have δ49Ti (the per mille deviation of the 49Ti/47Ti ratio relative to the OL-Ti standard) values between +0.053 ± 0.018 ‰ and −0.020 ± 0.016 ‰ (Tables S-1 and S-2). On average, N-MORB samples are isotopically lighter (δ49Ti = +0.001 ± 0.008 ‰; 2 se, n = 5) than E-MORB samples (δ49Ti = +0.035 ± 0.007 ‰; 2 se, n = 5). Similarly, among Archean komatiites, the late Archean (2.9-2.7 Ga) ones, which are depleted in incompatible trace elements, have a lower average δ49Ti value (+0.003 ± 0.013 ‰, 2 se, n = 5) than the middle Archean (3.5-3.3 Ga) ones (δ49Ti = +0.038 ± 0.018 ‰; 2 se, n = 4), which have primitive mantle-like trace element abundances (Fig. 1). The Ti isotopic composition of the middle Archean komatiites overlaps with the average composition of the 12 chondrite groups (δ49Ti = +0.070 ± 0.054 ‰, 2 sd, n = 12; Deng et al., 2018
Deng, Z., Moynier, F., van Zuilen, K., Sossi, P.A., Pringle, E.A., Chaussidon, M. (2018) Lack of resolvable titanium stable isotopic variations in bulk chondrites. Geochimica et Cosmochimica Acta 239, 409−419.
) but is lower than that of the continental crust as inferred from shale data (δ49Ti = +0.181 ± 0.015 ‰, 2 se, n = 78; Greber et al., 2017bGreber, N.D., Dauphas, N., Bekker, A., Ptáček, M.P., Bindeman, I.N., Hofmann, A. (2017b) Titanium isotopic evidence for felsic crust and plate tectonics 3.5 billion years ago. Science 357, 1271−1274.
) (Fig. 2). These observations imply the presence of either Ti isotopic heterogeneities in the mantle or Ti isotopic fractionations during partial melting and fractional crystallisation, or both.Figure 1 Positive correlation between δ49Ti and (La/Sm)N values for the komatiite and MORB samples.
Figure 2 The change of δ49Ti values with time for the komatiite and MORB samples. The average δ49Ti values for the depleted and primitive groups are indicated. Also shown are the average of 12 chondrite groups with 2 sd uncertainty from Deng et al. (2018)
Deng, Z., Moynier, F., van Zuilen, K., Sossi, P.A., Pringle, E.A., Chaussidon, M. (2018) Lack of resolvable titanium stable isotopic variations in bulk chondrites. Geochimica et Cosmochimica Acta 239, 409−419.
and the continental crust value inferred from shale data from Greber et al. (2017b)Greber, N.D., Dauphas, N., Bekker, A., Ptáček, M.P., Bindeman, I.N., Hofmann, A. (2017b) Titanium isotopic evidence for felsic crust and plate tectonics 3.5 billion years ago. Science 357, 1271−1274.
.top
Discussion
Ti Isotopic Compositions of the Mantle Sources of Archean Komatiites. Two arguments demonstrate that Archean komatiites have inherited the Ti isotopic compositions of their mantle sources. (i) Komatiites were produced by high degree partial melting (25-40 %) at mantle potential temperature >2000 K (Arndt, 2003
Arndt, N. (2003) Komatiites, kimberlites, and boninites. Journal of Geophysical Research: Solid Earth 108, doi: 10.1029/2002JB002157.
; Sossi et al., 2016Sossi, P.A., Eggins, S.M., Nesbitt, R.W., Nebel, O., Hergt, J.M., Campbell, I.H., O'Neill, H.S.C., Van Kranendonk, M., Davies, D.R. (2016) Petrogenesis and geochemistry of Archean komatiites. Journal of Petrology 57, 147−184.
), which minimises potential Ti isotopic fractionation between komatiitic liquids and their sources. (ii) The fractional crystallisation of olivine at low pressure does not produce significant Ti isotopic fractionation in the residual melts, because Ti is incompatible in olivine (Sossi and O'Neill, 2016Sossi, P.A., O’Neill, H.S.C. (2016) Liquidus temperatures of komatiites and the effect of cooling rate on element partitioning between olivine and komatiitic melt. Contributions to Mineralogy and Petrology 171, 49. doi: 10.1007/s00410-016-1260-x.
). Thus, the change of Ti isotopic composition of komatiites with time reflects a secular depletion in the heavier Ti isotopes of the mantle during the mid-late Archean (Fig. 2). The timing of this change in the Ti isotopic composition of the mantle matches the formation age of the DMM inferred from the Sm-Nd and Lu-Hf isotope compositions of juvenile crustal rocks of various ages (Vervoort et al., 1996Vervoort, J.D., Patchett, P.J., Gehrels, G.E., Nutman, A.P. (1996) Constraints on early Earth differentiation from hafnium and neodymium isotopes. Nature 379, 624−627.
; Vervoort and Blichert-Toft, 1999Vervoort, J.D., Blichert-Toft, J. (1999) Evolution of the depleted mantle: Hf isotope evidence from juvenile rocks through time. Geochimica et Cosmochimica Acta 63, 533−556.
). In addition, late Archean komatiites have a trace element inventory similar to that of the DMM (Workman and Hart, 2005Workman, R.K., Hart, S.R. (2005) Major and trace element composition of the depleted MORB mantle (DMM). Earth and Planetary Science Letters 231, 53−72.
; Sossi et al., 2016Sossi, P.A., Eggins, S.M., Nesbitt, R.W., Nebel, O., Hergt, J.M., Campbell, I.H., O'Neill, H.S.C., Van Kranendonk, M., Davies, D.R. (2016) Petrogenesis and geochemistry of Archean komatiites. Journal of Petrology 57, 147−184.
), e.g., (La/Sm)N values of 0.37-0.78 where the subscript 'N' denotes normalisation to the primitive mantle (McDonough and Sun 1995McDonough, W.F., Sun, S.S. (1995) The composition of the Earth. Chemical Geology 120, 223−253.
). Thus, the late Archean komatiites have likely sampled a mantle reservoir that has experienced melt extraction, potentially causing its Ti isotopic composition to trend towards that of the contemporary DMM.Lack of Resolvable Ti Isotopic Fractionation during the Genesis of MORB Melts. Differentiated magmas tend to be enriched in the heavier Ti isotopes as a result of the preferential incorporation of the lighter Ti isotopes into the VI-fold sites of Fe-Ti oxides, relative to the lower-coordinated ones (more IV- and V-fold) of the silicate melts (Millet et al., 2016
Millet, M.A., Dauphas, N., Greber, N.D., Burton, K.W., Dale, C.W., Debret, B., Macpherson, C.G., Nowell, G.M., Williams, H.M. (2016) Titanium stable isotope investigation of magmatic processes on the Earth and Moon. Earth and Planetary Science Letters 449, 197−205.
). Following a similar rationale, Ti isotopes may be also fractionated during the generation of MORB melts or their subsequent differentiation. However, the magnitude of this isotopic fractionation cannot be assessed precisely from previous data or from theoretical considerations: additional experimental or theoretical work would be required to quantify these effects precisely. In the following, we estimate the magnitude of this fractionation from the present δ49Ti data from the MORB and komatiite samples.Assuming a Rayleigh distillation process for the extraction of MORB melts, which is consistent with near-fractional melting (Asimow et al., 1997
Asimow, P.D., Hirschmann, M.M., Stolper, E.M. (1997) An analysis of variations in isentropic melt productivity. Philosophical Transactions of the Royal Society of London A 355, 255−281.
), the δ49Ti difference of the melting residues or the partial melts relative to their sources, hereafter referred respectively as Δ49Tiresidue−source and Δ49Timelt−source, should follow:Eq. 1
and
Eq. 2
where f is the remaining Ti fraction in the melting residues and α49/47melt-crystal is the melt−crystal Ti isotopic fractionation factor defined as:
Eq. 3
The remaining Ti fraction in the melting residues can be simply modelled by a fractional melting process:
Eq. 4
where F is the melt fraction and DTi is the partition coefficient of Ti. The values of F and DTi are estimated based on experimental data. To define the value of 103 × ln(α49/47melt-crystal), the Ti isotopic ratios of the sources are compared to that of the partial melts or of the melting residues, i.e. the values of Δ49Timelt−source or of Δ49Tiresidue−source need to be determined.
During partial melting of mantle peridotites, Ti is a moderately incompatible element (DTi ≈ 0.132 for spinel peridotite; Prytulak and Elliott, 2007
Prytulak, J., Elliott, T. (2007) TiO2 enrichment in ocean island basalts. Earth and Planetary Science Letters 263, 388−403.
). The genesis of MORB melts (F = 5-15 %) should thus extract ≈ 33-71 % of the Ti from the source regions (Equation 4; Asimow and Langmuir, 2003Asimow, P.D., Langmuir, C.H. (2003) The importance of water to oceanic mantle melting regimes. Nature 421, 815−820.
; Workman and Hart, 2005Workman, R.K., Hart, S.R. (2005) Major and trace element composition of the depleted MORB mantle (DMM). Earth and Planetary Science Letters 231, 53−72.
). Using f = 0.5 (i.e. a mean F ≈ 8.7 %) in Equation 2, the value of Δ49Timelt-source for MORBs is:Eq. 5
Taking the Ti isotopic compositions of the late Archean komatiites as representative of the DMM, Δ49Timelt−source is equal to Δ49TiN-MORB−late Archean komatiite = −0.002 ± 0.015 ‰ (2 se). This corresponds to 103 × ln(α49/47melt-crystal) = −0.003 ± 0.022 ‰ (2 se) (Fig. 1).
The lack of significant Ti isotopic fractionation between crystal and melt during mantle melting raises a question about the origin of the Ti isotopic difference between the depleted and primitive mantle reservoirs (Δ49Tidepleted−primitive = −0.034 ± 0.011 ‰; Fig. 2). Based on its average TiO2 content (≈ 0.119 wt. %), the DMM would be at a maximum 50 % depleted in Ti relative to the primitive mantle (≈ 0.217 wt. %) (McDonough and Sun 1995
McDonough, W.F., Sun, S.S. (1995) The composition of the Earth. Chemical Geology 120, 223−253.
; Workman and Hart, 2005Workman, R.K., Hart, S.R. (2005) Major and trace element composition of the depleted MORB mantle (DMM). Earth and Planetary Science Letters 231, 53−72.
). To produce a Δ49Tidepleted−primitive = −0.034 ± 0.011 ‰ would require a 103 × ln(α49/47melt-crystal) of +0.049 ± 0.016 ‰ using f = 0.5 in Equation 1, which is too high compared to the limited 103 × ln(α49/47melt-crystal) value during partial melting of the mantle peridotites (Fig. 3). Thus, another mechanism is required to explain the Ti isotopic difference between the depleted and primitive mantle reservoirs.Figure 3 Modelling the Ti isotopic fractionations in the partial melts and residues from partial melting of the mantle or the crust. The grey area in (a) indicates the remaining Ti fraction in the residual mantle after ≈ 8-9 % partial melting (DTi ≈ 0.132; Prytulak and Elliott, 2007
Prytulak, J., Elliott, T. (2007) TiO2 enrichment in ocean island basalts. Earth and Planetary Science Letters 263, 388−403.
), i.e. process (i) illustrated in (b). The orange field in a shows the Ti fraction after 20-30 % partial melting of a basaltic crust to produce the felsic melts equivalent to tonalite-trondhjemite-granodiorite (TTGs) (DTi ≈ 3.2-3.3; Martin et al., 2014Martin, H., Moyen, J.F., Guitreau, M., Blichert-Toft, J., Le Pennec, J.L. (2014) Why Archean TTG cannot be generated by MORB melting in subduction zones. Lithos 198, 1−13.
), i.e. process (ii) in b. The Δ49Tiresidue−source values of TTGs and corresponding residues were calculated using the Ti isotope data of TTG samples from Greber et al. (2017b)Greber, N.D., Dauphas, N., Bekker, A., Ptáček, M.P., Bindeman, I.N., Hofmann, A. (2017b) Titanium isotopic evidence for felsic crust and plate tectonics 3.5 billion years ago. Science 357, 1271−1274.
and assuming E-MORBs as their sources.Recycling of Melting Residues from the Generation of the Continental Crust into the Late Archean Mantle. In contrast with the lack of melt−crystal Ti isotopic fractionation during partial melting of the peridotitic mantle, magma differentiation in the crust produces significant Ti isotopic variations (Fig. 2; Millet et al., 2016
Millet, M.A., Dauphas, N., Greber, N.D., Burton, K.W., Dale, C.W., Debret, B., Macpherson, C.G., Nowell, G.M., Williams, H.M. (2016) Titanium stable isotope investigation of magmatic processes on the Earth and Moon. Earth and Planetary Science Letters 449, 197−205.
; Greber et al., 2017aGreber, N.D., Dauphas, N., Puchtel, I.S., Hofmann, B.A., Arndt, N.T. (2017a) Titanium stable isotopic variations in chondrites, achondrites and lunar rocks. Geochimica et Cosmochimica Acta 213, 534−552.
,bGreber, N.D., Dauphas, N., Bekker, A., Ptáček, M.P., Bindeman, I.N., Hofmann, A. (2017b) Titanium isotopic evidence for felsic crust and plate tectonics 3.5 billion years ago. Science 357, 1271−1274.
). Significant volumes of felsic plutonic rocks, i.e. TTGs, have been generated in the late Archean (Martin, 1986Martin, H. (1986) Effect of steeper Archean geothermal gradient on geochemistry of subduction-zone magmas. Geology 14, 753−756.
). Contemporaneously, the δ49Ti value of the mantle is observed to have changed, as indicated by komatiities. These TTGs likely originate from partial melting of protoliths of basaltic compositions (Martin et al., 2014Martin, H., Moyen, J.F., Guitreau, M., Blichert-Toft, J., Le Pennec, J.L. (2014) Why Archean TTG cannot be generated by MORB melting in subduction zones. Lithos 198, 1−13.
). Since Ti-rich minerals are stable under crustal temperatures and pressures, Ti acts as a compatible element during melt-crystal fractionation in the crust, e.g., DTi ≈ 3.2-3.3 (Martin et al., 2014Martin, H., Moyen, J.F., Guitreau, M., Blichert-Toft, J., Le Pennec, J.L. (2014) Why Archean TTG cannot be generated by MORB melting in subduction zones. Lithos 198, 1−13.
). The generation of TTG-like melts would then extract only ≈ 10 % of Ti from the sources, using F = 0.20-0.30 and DTi = 3.2-3.3 in Equation 4. The 2.98 Ga TTGs from Kaapvaal Craton have been shown to be significantly heavier in Ti isotopes than mantle-derived rocks (δ49Ti = +0.361 ± 0.088 ‰, 2 se, n = 10; Greber et al., 2017bGreber, N.D., Dauphas, N., Bekker, A., Ptáček, M.P., Bindeman, I.N., Hofmann, A. (2017b) Titanium isotopic evidence for felsic crust and plate tectonics 3.5 billion years ago. Science 357, 1271−1274.
). Assuming a source with an E-MORB-like isotopic composition, these TTGs indicate a Δ49Timelt−source ≈ +0.326 ± 0.088 ‰, corresponding to a Δ49Tiresidue−source ≈ −0.036 ± 0.010 ‰ and a 103 × ln(α49/47melt-crystal) ≈ +0.344 ± 0.093 ‰ when using f = 0.9 in Equations 1 and 2 (Fig. 3).The much larger Ti isotopic fractionation produced during partial melting of crust, compared to partial melting of peridotites, likely arises from the facts that: (i) melt-crystal segregation in the crust occurs at lower temperatures (< 1273 K) than that of the normal decompression melting of the mantle (≥ 1673 K) (Martin et al., 2014
Martin, H., Moyen, J.F., Guitreau, M., Blichert-Toft, J., Le Pennec, J.L. (2014) Why Archean TTG cannot be generated by MORB melting in subduction zones. Lithos 198, 1−13.
), and (ii) Ti is more IV- and V-fold coordinated in silicic melts than in mafic melts (Farges et al., 1996Farges, F., Brown Jr, G.E., Rehr, J.J. (1996) Coordination chemistry of Ti (IV) in silicate glasses and melts: I. XAFS study of titanium coordination in oxide model compounds. Geochimica et Cosmochimica Acta 60, 3023−3038.
). These thermal and structural differences significantly increase 103 × ln(α49/47melt-crystal) during the genesis of TTG-like melts. As a consequence, the residues (cumulates or restites) would be enriched in the lighter Ti isotopes with a Δ49Tiresidue−source ≈ −0.036 ± 0.010 ‰. Thus, the lighter Ti isotopic compositions of the DMM relative to the primitive mantle reservoirs could be due to mantle recycling of the residues from the generation of the late Archean felsic continental crust. To produce a δ49Ti difference of 0.034 ‰ between E-MORBs and N-MORBs, an addition of ≥ 3 % shale-like sediments (TiO2 ≈ 0.64 wt. % and δ49Ti ≈ +0.18 ‰; Greber et al., 2017bGreber, N.D., Dauphas, N., Bekker, A., Ptáček, M.P., Bindeman, I.N., Hofmann, A. (2017b) Titanium isotopic evidence for felsic crust and plate tectonics 3.5 billion years ago. Science 357, 1271−1274.
) into the DMM (TiO2 ≈ 0.12 wt. % and δ49Ti ≈ + 0.00 ‰) would be needed, which is unlikely since E-MORBs have only slightly higher 87Sr/86Sr and lower 143Nd/144Nd ratios than N-MORBs (Workman and Hart, 2005Workman, R.K., Hart, S.R. (2005) Major and trace element composition of the depleted MORB mantle (DMM). Earth and Planetary Science Letters 231, 53−72.
). Therefore, we consider that E-MORBs are derived from a less depleted mantle reservoir that has not been subjected to crustal residue recycling rather than the one contaminated by recycled sediments.Since the DMM (Δ49Tidepleted−primitive = −0.034 ± 0.011 ‰) seems to inherit the Ti isotopic composition of the residues from the formation of the late Archean TTGs (Δ49Tiresidue−source ≈ −0.036 ± 0.010 ‰; Fig. 3a), a significant proportion of Ti (94 ± 40 %) in the DMM (i.e. fTi-DMM) must have been processed by magma differentiation in the crust, if following the relation of
. This would require that, during middle to late Archean, basaltic crust having the Ti isotopic composition of the mantle was (i) continuously extracted, and (ii) re-melted to produce a felsic continental crust, with the residues from this process being mixed into the upper mantle (Fig. 3b). The delivery of melting residues from the crust into the mantle can be achieved by either slab subduction (Martin, 1986Martin, H. (1986) Effect of steeper Archean geothermal gradient on geochemistry of subduction-zone magmas. Geology 14, 753−756.
) or lower crustal recycling (Rudnick, 1995Rudnick, R.L. (1995) Making continental crust. Nature 378, 571−578.
). If the late Archean depleted mantle contained 20 % by mass of the bulk silicate Earth (BSE) (McCulloch and Bennett, 1994McCulloch, M.T., Bennett, V.C. (1994) Progressive growth of the Earth's continental crust and depleted mantle: geochemical constraints. Geochimica et Cosmochimica Acta 58, 4717−4738.
), the Ti isotope variations in the mantle would be produced by the extraction of a felsic continental crust constituting ≈ 0.38-0.56 % mass of the BSE during 3.5-2.7 Ga. Future Ti isotopic studies of komatiites and basalts covering the full age range of 3.5-2.7 Ga may be useful to test whether the continental extraction was continuous or abrupt. In the Archean, the hot convective mantle could homogenise the recycled melting residues into the ambient mantle. However, this mechanism should stop operating after the Archean, since the DMM was formed in the late Archean and has been stable in the upper mantle till the present day (Fig. 2). As the felsic continental crust continued to grow after the Archean (Dhuime et al., 2015Dhuime, B., Wuestefeld, A., Hawkesworth, C.J. (2015) Emergence of modern continental crust about 3 billion years ago. Nature Geoscience 8, 552−555.
), additional Ti isotope heterogeneities may exist in the mantle due to the recycling of ultramafic or mafic residues from the crust, which may be later sampled by intraplate magmas. Overall, the Earth's change across the Archean-Proterozoic boundary may be due to the decrease in the thermal gradient of the mantle (Korenaga, 2013Korenaga, J. (2013) Initiation and evolution of plate tectonics on Earth: theories and observations. Annual Review of Earth and Planetary Sciences 41, 117−151.
), likely signifying the prevalence of plate tectonics (Dhuime et al., 2015Dhuime, B., Wuestefeld, A., Hawkesworth, C.J. (2015) Emergence of modern continental crust about 3 billion years ago. Nature Geoscience 8, 552−555.
; Tang et al., 2016Tang, M., Chen K., Rudnick, R.L. (2016) Archean upper crust transition from mafic to felsic marks the onset of plate tectonics. Science 351, 372−375.
).top
Acknowledgements
We thank John Creech for introducing IsoSpike, Manuel Moreira for providing the MORB samples, and Marc-Alban Millet for sharing OL-Ti standard. Kirsten van Zuilen is thanked for her help in the lab. FM acknowledges funding from the ERC under the H2020 framework programme/ERC grant agreement #637503 (Pristine). FM and MC thank the financial support of the UnivEarthS Labex programme at Sorbonne Paris Cité (ANR-10-LABX-0023 and ANR-11-IDEX-0005-02). Parts of this work were supported by IPGP multidisciplinary programme PARI, and by Region Île-de-France SESAME Grant no. 12015908.
Editor: Cin-Ty Lee
top
References
Allègre, C.J. (1982) Chemical geodynamics. Tectonophysics 81, 109−132.
Show in context The Earth’s mantle is chemically and isotopically heterogeneous (Allègre, 1982; Zindler and Hart, 1986).
View in article
Arndt, N. (2003) Komatiites, kimberlites, and boninites. Journal of Geophysical Research: Solid Earth 108, doi: 10.1029/2002JB002157.
Show in contextArchean komatiites are produced by high degree partial melting (25-40 %) of the mantle, and erupt at temperatures >1750 K (Arndt, 2003).
View in article
(i) Komatiites were produced by high degree partial melting (25-40 %) at mantle potential temperature >2000 K (Arndt, 2003; Sossi et al., 2016), which minimises potential Ti isotopic fractionation between komatiitic liquids and their sources.
View in article
Asimow, P.D., Langmuir, C.H. (2003) The importance of water to oceanic mantle melting regimes. Nature 421, 815−820.
Show in context In contrast with Archean komatiites, MORB samples are derived from lower degrees of partial melting (5-15 %) of the modern upper mantle (Asimow and Langmuir, 2003; Workman and Hart, 2005), and are subdivided based on their concentration in incompatible elements from light rare earth element depleted (N-MORB) to enriched (E-MORB) (Workman and Hart, 2005).
View in article
The genesis of MORB melts (F = 5-15 %) should thus extract ≈ 33-71 % of the Ti from the source regions (Equation 4; Asimow and Langmuir, 2003; Workman and Hart, 2005).
View in article
Asimow, P.D., Hirschmann, M.M., Stolper, E.M. (1997) An analysis of variations in isentropic melt productivity. Philosophical Transactions of the Royal Society of London A 355, 255−281.
Show in context Assuming a Rayleigh distillation process for the extraction of MORB melts, which is consistent with near-fractional melting (Asimow et al., 1997), the δ49Ti difference of the melting residues or the partial melts relative to their sources, hereafter referred respectively as Δ49Tiresidue−source and Δ49Timelt−source, should follow:
View in article
Deng, Z., Moynier, F., van Zuilen, K., Sossi, P.A., Pringle, E.A., Chaussidon, M. (2018) Lack of resolvable titanium stable isotopic variations in bulk chondrites. Geochimica et Cosmochimica Acta 239, 409−419.
Show in context The Ti isotopic composition of the middle Archean komatiites overlaps with the average composition of the 12 chondrite groups (δ49Ti = +0.070 ± 0.054 ‰, 2 sd, n = 12; Deng et al., 2018) but is lower than that of the continental crust as inferred from shale data (δ49Ti = +0.181 ± 0.015 ‰, 2 se, n = 78; Greber et al., 2017b) (Fig. 2).
View in article
Figure 2 [...] Also shown are the average of 12 chondrite groups with 2 sd uncertainty from Deng et al. (2018) and the continental crust value inferred from shale data from Greber et al. (2017b).
View in article
Dhuime, B., Wuestefeld, A., Hawkesworth, C.J. (2015) Emergence of modern continental crust about 3 billion years ago. Nature Geoscience 8, 552−555.
Show in contextAs the felsic continental crust continued to grow after the Archean (Dhuime et al., 2015), additional Ti isotope heterogeneities may exist in the mantle due to the recycling of ultramafic or mafic residues from the crust, which may be later sampled by intraplate magmas.
View in article
Overall, the Earth's change across the Archean-Proterozoic boundary may be due to the decrease in the thermal gradient of the mantle (Korenaga, 2013), likely signifying the prevalence of plate tectonics (Dhuime et al., 2015; Tang et al., 2016).
View in article
Farges, F., Brown Jr, G.E., Rehr, J.J. (1996) Coordination chemistry of Ti (IV) in silicate glasses and melts: I. XAFS study of titanium coordination in oxide model compounds. Geochimica et Cosmochimica Acta 60, 3023−3038.
Show in contextThe much larger Ti isotopic fractionation produced during partial melting of crust, compared to partial melting of peridotites, likely arises from the facts that: (i) melt-crystal segregation in the crust occurs at lower temperatures (< 1273 K) than that of the normal decompression melting of the mantle (≥ 1673 K) (Martin et al., 2014), and (ii) Ti is more IV- and V-fold coordinated in silicic melts than in mafic melts (Farges et al., 1996).
View in article
Greber, N.D., Dauphas, N., Puchtel, I.S., Hofmann, B.A., Arndt, N.T. (2017a) Titanium stable isotopic variations in chondrites, achondrites and lunar rocks. Geochimica et Cosmochimica Acta 213, 534−552.
Show in context Titanium stable isotopes are well suited to discriminate between mantle and crustal melting because they do not appear to fractionate during partial melting of the mantle, whereas they tend to become enriched in the heavier isotopes during magmatic differentiation in the crust (Millet et al., 2016; Greber et al., 2017a,b).
View in article
In contrast with the lack of melt−crystal Ti isotopic fractionation during partial melting of the peridotitic mantle, magma differentiation in the crust produces significant Ti isotopic variations (Fig. 2; Millet et al., 2016; Greber et al., 2017a,b).
View in article
Greber, N.D., Dauphas, N., Bekker, A., Ptáček, M.P., Bindeman, I.N., Hofmann, A. (2017b) Titanium isotopic evidence for felsic crust and plate tectonics 3.5 billion years ago. Science 357, 1271−1274.
Show in context Titanium stable isotopes are well suited to discriminate between mantle and crustal melting because they do not appear to fractionate during partial melting of the mantle, whereas they tend to become enriched in the heavier isotopes during magmatic differentiation in the crust (Millet et al., 2016; Greber et al., 2017a,b).
View in article
The Ti isotopic composition of the middle Archean komatiites overlaps with the average composition of the 12 chondrite groups (δ49Ti = +0.070 ± 0.054 ‰, 2 sd, n = 12; Deng et al., 2018) but is lower than that of the continental crust as inferred from shale data (δ49Ti = +0.181 ± 0.015 ‰, 2 se, n = 78; Greber et al., 2017b) (Fig. 2).
View in article
Figure 2 [...] Also shown are the average of 12 chondrite groups with 2 sd uncertainty from Deng et al. (2018) and the continental crust value inferred from shale data from Greber et al. (2017b).
View in article
Figure 3 [...] The Δ49Tiresidue−source values of TTGs and corresponding residues were calculated using the Ti isotope data of TTG samples from Greber et al. (2017b) and assuming E-MORBs as their sources.
View in article
In contrast with the lack of melt−crystal Ti isotopic fractionation during partial melting of the peridotitic mantle, magma differentiation in the crust produces significant Ti isotopic variations (Fig. 2; Millet et al., 2016; Greber et al., 2017a,b).
View in article
The 2.98 Ga TTGs from Kaapvaal Craton have been shown to be significantly heavier in Ti isotopes than mantle-derived rocks (δ49Ti = +0.361 ± 0.088 ‰, 2 se, n = 10; Greber et al., 2017b).
View in article
To produce a δ49Ti difference of 0.034 ‰ between E-MORBs and N-MORBs, an addition of ≥ 3 % shale-like sediments (TiO2 ≈ 0.64 wt. % and δ49Ti ≈ +0.18 ‰; Greber et al., 2017b) into the DMM (TiO2 ≈ 0.12 wt. % and δ49Ti ≈ + 0.00 ‰) would be needed, which is unlikely since E-MORBs have only slightly higher 87Sr/86Sr and lower 143Nd/144Nd ratios than N-MORBs (Workman and Hart, 2005).
View in article
Hofmann, A.W. (1988) Chemical differentiation of the Earth: the relationship between mantle, continental crust, and oceanic crust. Earth and Planetary Science Letters 90, 297−314.
Show in context In particular, the depleted MORB mantle (DMM), i.e. the inferred source of the modern normal-type mid-ocean ridge basalts (N-MORB), is characterised by a significant depletion in the incompatible trace elements (Workman and Hart, 2005), and is to a first order compositionally complementary to the continental crust (Hofmann, 1988).
View in article
Thus, the DMM has been interpreted as a residual mantle that was homogenised after the extraction of a component similar to the present day continental crust (Hofmann, 1988; McCulloch and Bennett, 1994; Workman and Hart, 2005).
View in article
Korenaga, J. (2013) Initiation and evolution of plate tectonics on Earth: theories and observations. Annual Review of Earth and Planetary Sciences 41, 117−151.
Show in context Overall, the Earth's change across the Archean-Proterozoic boundary may be due to the decrease in the thermal gradient of the mantle (Korenaga, 2013), likely signifying the prevalence of plate tectonics (Dhuime et al., 2015; Tang et al., 2016).
View in article
Martin, H. (1986) Effect of steeper Archean geothermal gradient on geochemistry of subduction-zone magmas. Geology 14, 753−756.
Show in contextSignificant volumes of felsic plutonic rocks, i.e. TTGs, have been generated in the late Archean (Martin, 1986).
View in article
The delivery of melting residues from the crust into the mantle can be achieved by either slab subduction (Martin, 1986) or lower crustal recycling (Rudnick, 1995).
View in article
Martin, H., Moyen, J.F., Guitreau, M., Blichert-Toft, J., Le Pennec, J.L. (2014) Why Archean TTG cannot be generated by MORB melting in subduction zones. Lithos 198, 1−13.
Show in context Figure 3 [...] The orange field in a shows the Ti fraction after 20-30 % partial melting of a basaltic crust to produce the felsic melts equivalent to tonalite-trondhjemite-granodiorite (TTGs) (DTi ≈ 3.2-3.3; Martin et al., 2014), i.e. process (ii) in b.
View in article
These TTGs likely originate from partial melting of protoliths of basaltic compositions (Martin et al., 2014).
View in article
Since Ti-rich minerals are stable under crustal temperatures and pressures, Ti acts as a compatible element during melt-crystal fractionation in the crust, e.g., DTi ≈ 3.2-3.3 (Martin et al., 2014).
View in article
The much larger Ti isotopic fractionation produced during partial melting of crust, compared to partial melting of peridotites, likely arises from the facts that: (i) melt-crystal segregation in the crust occurs at lower temperatures (< 1273 K) than that of the normal decompression melting of the mantle (≥ 1673 K) (Martin et al., 2014), and (ii) Ti is more IV- and V-fold coordinated in silicic melts than in mafic melts (Farges et al., 1996).
View in article
McCulloch, M.T., Bennett, V.C. (1994) Progressive growth of the Earth's continental crust and depleted mantle: geochemical constraints. Geochimica et Cosmochimica Acta 58, 4717−4738.
Show in context Thus, the DMM has been interpreted as a residual mantle that was homogenised after the extraction of a component similar to the present day continental crust (Hofmann, 1988; McCulloch and Bennett, 1994; Workman and Hart, 2005).
View in article
If the late Archean depleted mantle contained 20 % by mass of the bulk silicate Earth (BSE) (McCulloch and Bennett, 1994), the Ti isotope variations in the mantle would be produced by the extraction of a felsic continental crust constituting ≈ 0.38-0.56 % mass of the BSE during 3.5-2.7 Ga.
View in article
McDonough, W.F., Sun, S.S. (1995) The composition of the Earth. Chemical Geology 120, 223−253.
Show in context In addition, late Archean komatiites have a trace element inventory similar to that of the DMM (Workman and Hart, 2005; Sossi et al., 2016), e.g., (La/Sm)N values of 0.37-0.78 where the subscript 'N' denotes normalisation to the primitive mantle (McDonough and Sun 1995).
View in article
Based on its average TiO2 content (≈ 0.119 wt. %), the DMM would be at a maximum 50 % depleted in Ti relative to the primitive mantle (≈ 0.217 wt. %) (McDonough and Sun, 1995; Workman and Hart, 2005).
View in article
Millet, M.A., Dauphas, N., Greber, N.D., Burton, K.W., Dale, C.W., Debret, B., Macpherson, C.G., Nowell, G.M., Williams, H.M. (2016) Titanium stable isotope investigation of magmatic processes on the Earth and Moon. Earth and Planetary Science Letters 449, 197−205.
Show in contextTitanium stable isotopes are well suited to discriminate between mantle and crustal melting because they do not appear to fractionate during partial melting of the mantle, whereas they tend to become enriched in the heavier isotopes during magmatic differentiation in the crust (Millet et al., 2016; Greber et al., 2017a,b).
View in article
Differentiated magmas tend to be enriched in the heavier Ti isotopes as a result of the preferential incorporation of the lighter Ti isotopes into the VI-fold sites of Fe-Ti oxides, relative to the lower-coordinated ones (more IV- and V-fold) of the silicate melts (Millet et al., 2016)
View in article
In contrast with the lack of melt−crystal Ti isotopic fractionation during partial melting of the peridotitic mantle, magma differentiation in the crust produces significant Ti isotopic variations (Fig. 2; Millet et al., 2016; Greber et al., 2017a,b).
View in article
Prytulak, J., Elliott, T. (2007) TiO2 enrichment in ocean island basalts. Earth and Planetary Science Letters 263, 388−403.
Show in contextDuring partial melting of mantle peridotites, Ti is a moderately incompatible element (DTi ≈ 0.132 for spinel peridotite; Prytulak and Elliott, 2007).
View in article
Figure 3 [...] The grey area in (a) indicates the remaining Ti fraction in the residual mantle after ≈ 8-9 % partial melting (DTi ≈ 0.132; Prytulak and Elliott, 2007), i.e. process (i) illustrated in (b).
View in article
Puchtel, I.S., Walker, R.J., Brandon, A.D., Nisbet, E.G. (2009) Pt-Re-Os and Sm-Nd isotope and HSE and REE systematics of the 2.7 Ga Belingwe and Abitibi komatiites. Geochimica et Cosmochimica Acta 73, 6367−6389.
Show in context These characteristics mitigate chemical and isotopic fractionations during partial melting of their sources, and therefore are sensitive records of the spatio-thermal evolution of the composition of Earth's mantle over their eruptive history (Walter, 2003; Puchtel et al., 2009; Sossi et al., 2016).
View in article
Rudnick, R.L. (1995) Making continental crust. Nature 378, 571−578.
Show in context Intracrustal magmatic differentiation and removal of the mafic/ultramafic complement are required to drive the bulk continental crust to andesitic compositions (Rudnick, 1995; Walter, 2003).
View in article
The delivery of melting residues from the crust into the mantle can be achieved by either slab subduction (Martin, 1986) or lower crustal recycling (Rudnick, 1995).
View in article
Sossi, P.A., O’Neill, H.S.C. (2016) Liquidus temperatures of komatiites and the effect of cooling rate on element partitioning between olivine and komatiitic melt. Contributions to Mineralogy and Petrology 171, 49. doi: 10.1007/s00410-016-1260-x.
Show in context (ii) The fractional crystallisation of olivine at low pressure does not produce significant Ti isotopic fractionation in the residual melts, because Ti is incompatible in olivine (Sossi and O'Neill, 2016).
View in article
Sossi, P.A., Eggins, S.M., Nesbitt, R.W., Nebel, O., Hergt, J.M., Campbell, I.H., O'Neill, H.S.C., Van Kranendonk, M., Davies, D.R. (2016) Petrogenesis and geochemistry of Archean komatiites. Journal of Petrology 57, 147−184.
Show in context These characteristics mitigate chemical and isotopic fractionations during partial melting of their sources, and therefore are sensitive records of the spatio-thermal evolution of the composition of Earth's mantle over their eruptive history (Walter, 2003; Puchtel et al., 2009; Sossi et al., 2016).
View in article
(i) Komatiites were produced by high degree partial melting (25-40 %) at mantle potential temperature >2000 K (Arndt, 2003; Sossi et al., 2016), which minimises potential Ti isotopic fractionation between komatiitic liquids and their sources.
View in article
In addition, late Archean komatiites have a trace element inventory similar to that of the DMM (Workman and Hart, 2005; Sossi et al., 2016), e.g., (La/Sm)N values of 0.37-0.78 where the subscript 'N' denotes normalisation to the primitive mantle (McDonough and Sun 1995).
View in article
Tang, M., Chen K., Rudnick, R.L. (2016) Archean upper crust transition from mafic to felsic marks the onset of plate tectonics. Science 351, 372−375.
Show in context Overall, the Earth's change across the Archean-Proterozoic boundary may be due to the decrease in the thermal gradient of the mantle (Korenaga, 2013), likely signifying the prevalence of plate tectonics (Dhuime et al., 2015; Tang et al., 2016).
View in article
Vervoort, J.D., Blichert-Toft, J. (1999) Evolution of the depleted mantle: Hf isotope evidence from juvenile rocks through time. Geochimica et Cosmochimica Acta 63, 533−556.
Show in context Based on the enrichments of the radiogenic Nd and Hf isotopes in N-MORB samples (Vervoort et al., 1996; Vervoort and Blichert-Toft, 1999), the DMM should acquire its high Sm/Nd and Lu/Hf ratios during the late Archean.
View in article
The timing of this change in the Ti isotopic composition of the mantle matches the formation age of the DMM inferred from the Sm-Nd and Lu-Hf isotope compositions of juvenile crustal rocks of various ages (Vervoort et al., 1996; Vervoort and Blichert-Toft, 1999).
View in article
Vervoort, J.D., Patchett, P.J., Gehrels, G.E., Nutman, A.P. (1996) Constraints on early Earth differentiation from hafnium and neodymium isotopes. Nature 379, 624−627.
Show in context Based on the enrichments of the radiogenic Nd and Hf isotopes in N-MORB samples (Vervoort et al., 1996; Vervoort and Blichert-Toft, 1999), the DMM should acquire its high Sm/Nd and Lu/Hf ratios during the late Archean.
View in article
The timing of this change in the Ti isotopic composition of the mantle matches the formation age of the DMM inferred from the Sm-Nd and Lu-Hf isotope compositions of juvenile crustal rocks of various ages (Vervoort et al., 1996; Vervoort and Blichert-Toft, 1999).
View in article
Walter, M.J. (2003) Melt extraction and compositional variability in mantle lithosphere. In: Carlson, R.W. (Ed.) Treatise on Geochemistry−The Mantle and Core. Elsevier-Pergamon, Oxford, 363−394.
Show in context Intracrustal magmatic differentiation and removal of the mafic/ultramafic complement are required to drive the bulk continental crust to andesitic compositions (Rudnick, 1995; Walter, 2003).
View in article
These characteristics mitigate chemical and isotopic fractionations during partial melting of their sources, and therefore are sensitive records of the spatio-thermal evolution of the composition of Earth's mantle over their eruptive history (Walter, 2003; Puchtel et al., 2009; Sossi et al., 2016).
View in article
Workman, R.K., Hart, S.R. (2005) Major and trace element composition of the depleted MORB mantle (DMM). Earth and Planetary Science Letters 231, 53−72.
Show in context In particular, the depleted MORB mantle (DMM), i.e. the inferred source of the modern normal-type mid-ocean ridge basalts (N-MORB), is characterised by a significant depletion in the incompatible trace elements (Workman and Hart, 2005), and is to a first order compositionally complementary to the continental crust (Hofmann, 1988).
View in article
Thus, the DMM has been interpreted as a residual mantle that was homogenised after the extraction of a component similar to the present day continental crust (Hofmann, 1988; McCulloch and Bennett, 1994; Workman and Hart, 2005).
View in article
In contrast with Archean komatiites, MORB samples are derived from lower degrees of partial melting (5-15 %) of the modern upper mantle (Asimow and Langmuir, 2003; Workman and Hart, 2005), and are subdivided based on their concentration in incompatible elements from light rare earth element depleted (N-MORB) to enriched (E-MORB) (Workman and Hart, 2005).
View in article
In addition, late Archean komatiites have a trace element inventory similar to that of the DMM (Workman and Hart, 2005; Sossi et al., 2016), e.g., (La/Sm)N values of 0.37-0.78 where the subscript 'N' denotes normalisation to the primitive mantle (McDonough and Sun 1995).
View in article
The genesis of MORB melts (F = 5-15 %) should thus extract ≈ 33-71 % of the Ti from the source regions (Equation 4; Asimow and Langmuir, 2003; Workman and Hart, 2005).
View in article
Based on its average TiO2 content (≈ 0.119 wt. %), the DMM would be at a maximum 50 % depleted in Ti relative to the primitive mantle (≈ 0.217 wt. %) (McDonough and Sun, 1995; Workman and Hart, 2005).
View in article
To produce a δ49Ti difference of 0.034 ‰ between E-MORBs and N-MORBs, an addition of ≥ 3 % shale-like sediments (TiO2 ≈ 0.64 wt. % and δ49Ti ≈ +0.18 ‰; Greber et al., 2017b) into the DMM (TiO2 ≈ 0.12 wt. % and δ49Ti ≈ + 0.00 ‰) would be needed, which is unlikely since E-MORBs have only slightly higher 87Sr/86Sr and lower 143Nd/144Nd ratios than N-MORBs (Workman and Hart, 2005).
View in article
Zindler, A., Hart, S. (1986) Chemical geodynamics. Annual Review of Earth and Planetary Sciences 14, 493−571.
Show in contextThe Earth’s mantle is chemically and isotopically heterogeneous (Allègre, 1982; Zindler and Hart, 1986).
View in article
Multiple mantle end members, inferred from the radiogenic isotopic compositions (Pb-Pb, Sm-Nd and Rb-Sr) of oceanic basalts, have been defined to describe these heterogeneities (Zindler and Hart, 1986).
View in article
top
Supplementary Information
The Supplementary Information includes:
- Materials
- Methods
- Supplementary Text
- Tables S-1 and S-2
- Figures S-1 and S-2
- Supplementary Information References
Download the Supplementary Information (PDF).
Figures and Tables
Figure 1 Positive correlation between δ49Ti and (La/Sm)N values for the komatiite and MORB samples.
Figure 2 The change of δ49Ti values with time for the komatiite and MORB samples. The average δ49Ti values for the depleted and primitive groups are indicated. Also shown are the average of 12 chondrite groups with 2 sd uncertainty from Deng et al. (2018)
Deng, Z., Moynier, F., van Zuilen, K., Sossi, P.A., Pringle, E.A., Chaussidon, M. (2018) Lack of resolvable titanium stable isotopic variations in bulk chondrites. Geochimica et Cosmochimica Acta 239, 409−419.
and the continental crust value inferred from shale data from Greber et al. (2017b)Greber, N.D., Dauphas, N., Bekker, A., Ptáček, M.P., Bindeman, I.N., Hofmann, A. (2017b) Titanium isotopic evidence for felsic crust and plate tectonics 3.5 billion years ago. Science 357, 1271−1274.
.
Figure 3 Modelling the Ti isotopic fractionations in the partial melts and residues from partial melting of the mantle or the crust. The grey area in (a) indicates the remaining Ti fraction in the residual mantle after ≈ 8-9 % partial melting (DTi ≈ 0.132; Prytulak and Elliott, 2007
Prytulak, J., Elliott, T. (2007) TiO2 enrichment in ocean island basalts. Earth and Planetary Science Letters 263, 388−403.
), i.e. process (i) illustrated in (b). The orange field in a shows the Ti fraction after 20-30 % partial melting of a basaltic crust to produce the felsic melts equivalent to tonalite-trondhjemite-granodiorite (TTGs) (DTi ≈ 3.2-3.3; Martin et al., 2014Martin, H., Moyen, J.F., Guitreau, M., Blichert-Toft, J., Le Pennec, J.L. (2014) Why Archean TTG cannot be generated by MORB melting in subduction zones. Lithos 198, 1−13.
), i.e. process (ii) in b. The Δ49Tiresidue−source values of TTGs and corresponding residues were calculated using the Ti isotope data of TTG samples from Greber et al. (2017b)Greber, N.D., Dauphas, N., Bekker, A., Ptáček, M.P., Bindeman, I.N., Hofmann, A. (2017b) Titanium isotopic evidence for felsic crust and plate tectonics 3.5 billion years ago. Science 357, 1271−1274.
and assuming E-MORBs as their sources.