Reaction history of garnet‐sapphirine granulites and conditions of Archaean high‐pressure granulite‐facies metamorphism in the Central Limpopo Mobile Belt, Zimbabwe

Abstract

Sequential reaction textures in Archaean garnet-corundum-sapphirine granulites from the Central Zone of the Limpopo Belt document a progression from early, coarse-grained, high-pressure (P '9.5 kbar) granulite-facie assemblages (Ml) to late, low-pressure (P < 6 kbar) granulite-facies sub-assemblagesThe stable M1 assemblage was garnet (57% pyrope; Mg/(Mg + Fe) = 62) + sapphirine + corundum + gedrite+phlogopite+rutile. Late-M1 boron-free kornerupine grew at the expense of garnet and corundum, and coexisted with garnet, sapphirine and gedrite. Partial or complete breakdown of coarse garnet and komerupine during h42 resulted in the development of pseudomorphs and coronas consisting of fine-grained symplectic intergrowths of cordiente, gedrite and sapphirine (later, spinel).The majority of reaction textures can be explained in terms of a stable reaction sequence, and a model time-sequence of mineral facies can be constructed. When compared with a qualitative petrogenetic grid of (Fe, Mg)-discontinuous reactions in the FMASH multisystem sapphirine-garnet-corundum-spinelcordierite-gedrite-kornerupine, the facies-sequence indicates decompression at essentially constant T assuming constant u(H,O).Exhumation of M1 corundum inclusions during A42 breakdown of kornerupine resulted in production of metastable spinel by a disequilibrium reaction with gedrite. A second disequilibrium reaction of the spinel with cordierite produced sapphirine. The operation of such reaction while pressure was decreasing (the opposite dP from that implied by the texture if assumed to be the product of an equilibrium reaction) has serious implications for the use of reaction textures in the construction of P-T vectors.Garnet-biotite thermometry on garnet interiors and phlogopite inclusions in corundum yields temperatures of cu. 850" C for the M1 stage. A minimum late-M1 pressure of cu. 7 kbar is indicated by the former association of kornerupine and corundum. Relict M1 kyauites reported by other workers indicate a minumum early-Ml pressure of 9.5 kbar, implying metamorphism at depths of at least 33 km (probably ~3 8 k m ) . The high-pressure granulite-facies metamorphism was followed by an almost isothermal pressure decrease of >5 kbar, indicative of rapid uplift. The P-7' path is interpreted as the product of a single metamorphic cycle which probably took place in response to tectonic thickening of the crust. Such a process contrasts with the extensional origin recently proposed for isobaridly cooled granulite-facies terranes.

Conflicting P–T paths within the Central Zone of the Limpopo Belt: A consequence of different thermobarometric methods?

“…6A-D) are consistent with the modes, compositions and zonations in the observed mineral assemblages, which define a P-T path that traverses from peak conditions of 8-9 kbar and 840 +/À 10°C along a decompression cooling path to 4.5-5.5 kbar and $650°C. The peak P-T conditions from this study are in agreement with those determined by Watkeys et al (1983), Windley et al (1984), Droop (1989), Hisada and Miyano (1996) and Perchuk et al (2000). The retrograde path, characterized by decompression cooling, is in agreement with the paths defined by Zeh et al (2004), Van Reenen et al (2004), and Boshoff et al (2006) and the DC1 path of Perchuk et al (2008).…”

“…2). The peak metamorphic P-T conditions agree well with the peak P-T estimates obtained by Droop (1989) and Hisada and Miyano (1996). However, the retrograde path, defined by matrix assemblages, garnet resorption textures, mineral zoning and diffusion patterns (Zeh et al, 2004) requires a simultaneous P-T decrease, in agreement with the DC paths defined by Perchuk et al (2000), Van Reenen et al (2004), Hisada et al (2005) and Rigby et al (2008b) but in stark contrast to retrograde paths defined by ITD (e.g.…”

“…The retrograde path, characterized by decompression cooling, is in agreement with the paths defined by Zeh et al (2004), Van Reenen et al (2004), and Boshoff et al (2006) and the DC1 path of Perchuk et al (2008). The path is, however, inconsistent with isothermal decompression (Droop, 1989) followed by isobaric cooling (Hisada and Miyano, 1996). Retrograde paths characterized by ITD (Droop, 1989) have been used as evidence to support a 'pop-up' style of exhumation for the CZ (e.g.…”

“…However, the retrograde path, defined by matrix assemblages, garnet resorption textures, mineral zoning and diffusion patterns (Zeh et al, 2004) requires a simultaneous P-T decrease, in agreement with the DC paths defined by Perchuk et al (2000), Van Reenen et al (2004), Hisada et al (2005) and Rigby et al (2008b) but in stark contrast to retrograde paths defined by ITD (e.g. Droop, 1989;Hisada and Miyano, 1996) and subsequent IBC (Hisada and Miyano, 1996).…”

“…Whereas, the relative timing, style and extent of metamorphic activity in CZ has, over the years, come into question by numerous workers (e.g. Harris and Holland, 1984;Droop, 1989;Hisada and Miyano, 1996;Jaeckel et al, 1997;Holzer et al, 1998;Kröner et al, 1998Kröner et al, , 1999Smit et al, 2001;Van Reenen et al, 2004;Zeh et al, 2004;Zeh et al, 2005a,b;Hisada et al, 2005;Boshoff et al, 2006;Zeh et al, 2007;Rigby et al, 2008b;Gerdes and Zeh, 2009;Van Reenen et al, 2008;Perchuk et al 2008) who have defined different P-T paths. This has ultimately led to a great deal of controversy surrounding the tectonic evolution of the Limpopo Belt (Rigby et al, 2008a).…”