TY - JOUR

T1 - Lithium isotope fractionation between mica, quartz, amphibole, feldspars, and granitic melt

T2 - Experimental approach and implications for natural granitic systems

AU - Gao, Xu

AU - Michaud, Julie Anne Sophie

AU - Koch, Lennart

AU - Zhou, Zhenhua

AU - Zhang, Chao

AU - Horn, Ingo

AU - Almeev, Renat R.

AU - Weyer, Stefan

AU - Holtz, François

N1 - Publisher Copyright: © 2025 The Author(s)

PY - 2025/10/15

Y1 - 2025/10/15

N2 - Lithium isotope fractionation has been extensively used to investigate magmatic and hydrothermal processes over the past decade. Thus, knowledge of Li isotope fractionation factors between minerals and melts is essential for the interpretation of Li isotope data. However, Li isotope fractionation between granitic melts and common silicate minerals has not been directly determined experimentally. To address this issue and to investigate the effect of NaCl-bearing fluids on lithium isotopic fractionation, we conducted Cl-free and Cl-bearing experiments aimed at investigating the isotope fractionation factors between silicate minerals and hydrous melt at 575 − 600 °C and 200 MPa. The run products are composed of Li-mica, K-rich feldspar, Na-rich feldspar, ferroholmquistite (Fhlm; a Li-bearing amphibole), quartz, and melt. In Cl-free experiments, quartz is isotopically heaviest with Li isotope fractionation between quartz and melt ΔQz-melt = +7.0 ‰ (translating to an isotope fractionation factor α = 1.0070), followed by Li-mica with ΔLi-mica-melt = +3.1 ‰ (α = 1.0031), K-rich feldspar with ΔK-fsp-melt = +0.1 ‰ (α = 1.0001), ferroholmquistite with ΔFhlm-melt = − 1.9 ‰ (α = 0.9981) and Na-rich feldspar with ΔNa-fsp-melt = − 2.7 ‰ (α = 0.9973). Our experimental data indicate that Li-mica has a higher δ7Li value than granitic melt. This observation differs from previous findings, based on bond-energy estimations, according to which micas are expected to be isotopically lighter than the coexisting melt. This discrepancy may be attributed to the coordination environment in minerals, which can be distorted, influencing Li-O bonding energies. The Li isotope fractionation factors between mica, K-rich feldspar, Na-rich feldspar, ferroholmquistite, and melt in Cl-bearing experiments are very similar to those of Cl-free systems. This implies that the presence of NaCl-bearing fluids in a closed magmatic system has a limited effect on Li isotope fractionation during magmatic processes. The results from a multi-stage quantitative fractionation model suggest that granitic residual melts evolve to isotopically lighter δ7Li values during crystal fractionation due to the high αmica-melt and αquartz-melt values (>1). A high degree of crystal fractionation in Li-poor muscovite-bearing granitic systems could lead to a limited but still measurable Li isotope shift in residual melts (>1‰), whereas shifts up to 6 ‰ are observed in Li-rich systems. Lithium-rich mica is thus more effective in causing Li isotope fractionation as compared to muscovite and biotite. Our findings imply that large lithium isotopic fractionation observed in natural granitic systems could be caused by magmatic processes, even if water–rock interaction in an open system does not occur.

AB - Lithium isotope fractionation has been extensively used to investigate magmatic and hydrothermal processes over the past decade. Thus, knowledge of Li isotope fractionation factors between minerals and melts is essential for the interpretation of Li isotope data. However, Li isotope fractionation between granitic melts and common silicate minerals has not been directly determined experimentally. To address this issue and to investigate the effect of NaCl-bearing fluids on lithium isotopic fractionation, we conducted Cl-free and Cl-bearing experiments aimed at investigating the isotope fractionation factors between silicate minerals and hydrous melt at 575 − 600 °C and 200 MPa. The run products are composed of Li-mica, K-rich feldspar, Na-rich feldspar, ferroholmquistite (Fhlm; a Li-bearing amphibole), quartz, and melt. In Cl-free experiments, quartz is isotopically heaviest with Li isotope fractionation between quartz and melt ΔQz-melt = +7.0 ‰ (translating to an isotope fractionation factor α = 1.0070), followed by Li-mica with ΔLi-mica-melt = +3.1 ‰ (α = 1.0031), K-rich feldspar with ΔK-fsp-melt = +0.1 ‰ (α = 1.0001), ferroholmquistite with ΔFhlm-melt = − 1.9 ‰ (α = 0.9981) and Na-rich feldspar with ΔNa-fsp-melt = − 2.7 ‰ (α = 0.9973). Our experimental data indicate that Li-mica has a higher δ7Li value than granitic melt. This observation differs from previous findings, based on bond-energy estimations, according to which micas are expected to be isotopically lighter than the coexisting melt. This discrepancy may be attributed to the coordination environment in minerals, which can be distorted, influencing Li-O bonding energies. The Li isotope fractionation factors between mica, K-rich feldspar, Na-rich feldspar, ferroholmquistite, and melt in Cl-bearing experiments are very similar to those of Cl-free systems. This implies that the presence of NaCl-bearing fluids in a closed magmatic system has a limited effect on Li isotope fractionation during magmatic processes. The results from a multi-stage quantitative fractionation model suggest that granitic residual melts evolve to isotopically lighter δ7Li values during crystal fractionation due to the high αmica-melt and αquartz-melt values (>1). A high degree of crystal fractionation in Li-poor muscovite-bearing granitic systems could lead to a limited but still measurable Li isotope shift in residual melts (>1‰), whereas shifts up to 6 ‰ are observed in Li-rich systems. Lithium-rich mica is thus more effective in causing Li isotope fractionation as compared to muscovite and biotite. Our findings imply that large lithium isotopic fractionation observed in natural granitic systems could be caused by magmatic processes, even if water–rock interaction in an open system does not occur.

KW - Crystal fractionation

KW - In situ lithium isotope analysis

KW - Li isotope fractionation

KW - Mica

KW - Mineral/melt Li isotope fractionation coefficient

UR - http://www.scopus.com/inward/record.url?scp=105014295013&partnerID=8YFLogxK

U2 - 10.1016/j.gca.2025.08.028

DO - 10.1016/j.gca.2025.08.028

M3 - Article

AN - SCOPUS:105014295013

VL - 407

SP - 12

EP - 31

JO - Geochimica et cosmochimica acta

JF - Geochimica et cosmochimica acta

SN - 0016-7037

ER -