Transcript Foley Penrose

Trace elements and the formation of early crust
(1) Trace elements: what can we learn from them?
(2) TTG do not have the “subduction signature” (whatever that is)
(3) Mineral assemblages, not tectonic settings, control trace elements
(4) Igneous and metamorphic trace element partitioning patterns differ
A few collected, but not necessarily connected, thoughts:
(5) Eclogite xenoliths: what were they?
(6) What happens at subduction if the ocean crust is not basalt?
(7) Ultradepleted early Archaean uppermost mantle?
Subduction setting for Archaean TTG gneisses
Tarney (1976)
Cont crust TE
Low Ti, Nb, Ta is well known to be typical of subduction zone volcanics….
But we have to look closer
TTG are melts of garnet amphibolite
Garnet
amphibolite
Zr/Sm
Foley, Tiepolo & Vannucci (2002) Nature 417, 837-840
Explanation of Nb/Ta in rutile
Rutile must have DNb / DTa approx 0.6-0.7
Results in melts with high Nb/Ta
Optimal ionic
radius 58 pm
D Ru/Lq
Calcic amphiboles
Optimal ionic radius 62 - 66 pm
Highest where low Mg# and high DTi
Ta Nb
Ionic radius
Nb = 64 pm
Ta = 0.625 pm
Mantle eclogite xenoliths:
the best samples of Archaean oceanic
crust?
Mantle eclogite xenoliths
D. E. Jacob
Macquarie Island Ophiolite
Samail Ophiolite
Roberts Victor (South Africa, ca. 2.7Ga)
Bellsbank (South Africa)
Blaauwbosch (South Africa)
Newlands (South Africa)
Frank Smith (South Africa)
Sloan (USA)
Orapa (Botswana)
Grib (Arkhangelsk)
Ekati (Canada)
Mir (Yakutia)
Udachnaya (Yakutia, ca. 2.7 Ga)
Koidu (Sierra Leone, ca. 3.4 Ga)
Kuruman (South Africa)
Rietfontein (South Africa)
1
2
3
4
Oxygen isotopes
5
6
7
8
d18O (‰)
9
10
11
12
13
14
D.E.Jacob
Metamorphic trace element partitioning in eclogites
sample/chondrite
100
10
Garn et
Reconstructed Bulk
1
Cpx
0.1
0.01
La
Ce
Pr
Nd Sm
Eu Gd Tb Dy
Ho
Er Tm Yb
Lu
Eclogites melted as garnet amphibolites, then transformed to eclogites in the subsolidus
(high Nb/Ta rutiles must be metamorphic)
D.E. Jacob (2004)
Eclogites and continental crustal growth
Growth curves and eclogites
Continental crustal growth and survival
Growth curves and eclogites
Crustal
recycling
early
formation
Melting of basalt (MORB)
Melting curves and stabilities of
minerals well defined by experiments
Basaltic crust can melt as garnet
amphibolite on Geotherm 2
Early Archaean (pre-3600 Ma?)
If much hotter conditions:
Widespread melting of amphibolite in
early Archean should have produced
enormous quantities of silicic crust
Where is it?
If Geotherm 1 applied, then trace
elements of melts would not resemble
TTG
Melting of picriticultramafic crust
Picritic to ultramafic parts of the
crust will not form amphibolites during
metamorphism, but varieties of
pyroxenite instead.
If pyroxenites melt, they form
nephelinitic to basaltic melts, not
intermediate to silicic melts like TTG:
Production of silicic crust is stalled
Foley, Buhre & Jacob (2003): Nature 421, 249-252
Formation
of TTG
Supposing
subduction:
Foley, Buhre & Jacob (2003) Nature 421, 249-252
Near-surface origin of cratonic peridotites as depleted
uppermost mantle
1. Eclogites of near-surface origin entrained
2. Subcalcic garnets in garnet harzburgites require Cr-rich harzburgite - formed at MOR
3. Mounting textural evidence for re-enrichment
4. Trace elements show harzburgitic precursors
Philosophy for the Archaean
“When a man tells you he knows the exact truth about anything, you are
safe in inferring that he is an inexact man.”
Bertrand Russell (1931)
“Where nobody knows anything, there is no point in changing your mind.”
Bertrand Russell (1952)