Staff and Postgraduate Students
Publication details for Prof Jon DavidsonKnesel, K.M & Davidson, J.P (2002). Insights into Collisional Magmatism from Isotopic Fingerprints of Melting Reactions. Science 296(5576): 2206-2208.
- Publication type: Journal papers: academic
- ISSN/ISBN: 0036-8075, 1095-9203
- DOI: 10.1126/science.1070622
- View online: Online version
- Durham research online: DRO record
Author(s) from Durham
Piston-cylinder experiments in the granite system demonstrate that a wide variety of isotopically distinct melts can arise from progressive melting of a single source. The relation between the isotopic composition of Sr and the stoichiometry, of the observed melting reactions suggests that isotopic signatures of anatectic magmas can be used to infer melting reactions in natural systems. Our results also indicate that distinct episodes of dehydration and fluid-fluxed melting of a single, metapelitic source region may have contributed to the bimodal geochemistry of crustally derived leucogranites of the Himalayan orogen
1. T. Hammouda, M. Pichavant, M. Chaussidon, Earth
Planet. Sci. Lett. 144, 109 (1996).
2. K. M. Knesel, J. P. Davidson, Geology 24, 243 (1996).
3. S. Tommasini, G. R. Davies, Earth Planet. Sci. Lett.
148, 273 (1997).
4. M. B. Baker, E. M. Stolper, Geochim. Cosmochim. Acta
58, 2811 (1994).
5. J. C. Ayers, J. B. Brenan, B. E. Watson, D. A. Wark,,
W. G. Minarik, Am. Mineral. 77, 1080 (1992).
6. Experiments were carried out in 2.54-cm, NaCl-graphite
furnace assemblies. At or above 900íC, graphite furnaces
were isolated from the salt pressure-transmitting medium
by a Pyrex sleeve. Temperature was measured using
Pt-PtRh thermocouples. Uncertainties in temperature
measurements were not determined. However, doublethermocouple
experiments (5) indicate that the high
thermal conductivity of Ni capsules of the same dimensions
used here reduces the thermal gradient to ;5íC
over a sample length of 8 mm at 10 kbar and 1000íC. As
a conservative measure, uncertainties in temperature are
taken as 610íC.
7. Glass was dissolved in a warm 50:50 Hf-H2O mixture in
sealed Te߯n beakers for 4 hours. The mixture was spun
with a centrifuge, and the acid containing the dissolved
glass was removed by pipette and dried in preparation
for cation-exchange chemistry. Leaching tests indicate
that the diamond powder contributes less than 0.15% to
the total Sr analyzed. The isotopic compositions of the
experimental glasses were measured on a VG sector
mass spectrometer following procedures outlined in earlier
work (25). Repeated analysis of the Sr standard SRM
987 over the study period yielded a value of 0.7102276
25 (2s SD, n 5 22 analyses).
8. A. E. Patinouce, H. Harris, J. Petrol. 39, 689 (1998).
9. N. Petford, J. Geophys. Res. 100, 15735 (1995).
10. E. W. Sawyer, J. Petrol. 32, 701 (1991).
11. N. Harris, M. Ayres, J. Geol. Soc. London Spec. Publ.
138 (1998), p. 171.
12. L. Barbero, C. Villaseca, G. Rogers, P. E. Brown, J.
Geophys. Res. 100, 15745 (1995).
13. M. T. George, J. M. Bartlett, Tectonophysics 260, 167
14. T. M. Harrison et al. J. Petrol. 40, 3 (1998).
15. N. Peford, K. Gallagher, Earth Planet. Sci. Lett. 193,
16. C. France-Lanford, P. Lefort, J. Trans. R. Soc. Edinburgh
Earth Sci. 79, 183 (1988).
17. S. Guillot, P. Le Fort, Lithos 35, 221 (1995).
18. N. Harris, M. Ayres, J. Massey, J. Geophys. Res. 100,
19. P. Le Fort, J. Geophys. Res. 86, 10545 (1981).
20. E. H. Rutter, D. H. K. Neumann, J. Geophys. Res. 100,
21. R. S. DՌemos, M. Brown, R. A. Strachan, J. Geol. Soc.
London 149, 487 (1992).
22. M. Ayres, N. Harris, D. Vance, Mineral. Mag. 61, 29
23. T. Rushmer, J. Trans. R. Soc. Edinburgh Earth Sci. 87,
24. M. Brown, Y. A. Averkin, E. L. McLellan, J. Geophys.
Res. 100, 15655 (1995).
25. K. M. Knesel, J. P. Davidson, Contrib. Mineral. Petrol.
136, 285 (1999).
26. J. Hertogen, R. Gijbels, Geochim. Cosmochim. Acta
40, 313 (1976).
27. C.-H. Chen, D. J. DePaolo, C.-Y. Lan, Earth Planet. Sci.
Lett. 143, 125 (1996).
28. The source rock used in the melting models is a kynaitegrade,
two-mica schist [sample PAN-3 of (11)], which
was studied experimentally (8) to constrain phase relations
during melting of likely sources of Himalayan
leucogranites. The schist (113 ppm Sr; 87Sr/86Sr23Ma,
19Ma 5 0.7606, 0.7609) consists of 29% muscovite (128
ppm Sr; 87Sr/86Sr23Ma, 19Ma 5 0.7722, 0.7726), 11% plagioclase
(An28; 321 ppm Sr; 87Sr/86Sr23Ma, 19Ma 5 0.7376,
0.7377), 38% quartz, 13% biotite, 6% garnet, and 4%
kyanite1staurolite1tourmaline. 87Sr/86Sr ratios were
calculated at 23 and 19 million years ago (Ma) for
muscovite-dehydration and ߵid-ߵxed melting, respectively,
assuming bulk-rock homogenization at 500 Ma as
reported in (11).
29. The concentration of Sr in anatectic melt was modeled
by the following expression for nonmodal, fractional
D S1 2
P 2 1D
where Co is the concentration in the initial solid, CL is
the concentration in the liquid, Do is the bulk solidliquid
distribution coefޣient weighted according to
the modal assemblage in the source, F is the degree
of melting, and P is the bulk distribution coefޣient
weighted according to the proportion of phases entering
the melt. The effect of product mineral growth
on melt chemistry during incongruent breakdown of
muscovite was accounted for by replacement of P in
the equation above with Q (26)
P 2 paOb
ti k i
1 2 pa ~1 2 tl!
where pi is the mass contribution of phase i to the
liquid, ti is the mass fraction of the incongruent phase
contributed to the liquid and product mineral phases.
Sr-partition coefޣients are given in (18).
30. We thank C. Manning and R. Newton for guidance in
the piston-cylinder laboratory at the University of
California at Los Angeles and the journal reviewers
for their helpful comments. Supported by the NSF.