TY - JOUR
AU - Breaks, Frederick, W
AB - Abstract We constrain the oxygen fugacity (⁠ fO2 ⁠) of strongly peraluminous granites [SPGs; i.e. granites (sensu lato) generated through the partial melting of sediments] across the Archean–Proterozoic boundary, which coincides roughly with the Great Oxygenation Event (GOE), to understand whether secular changes in atmospheric O2 levels may be imprinted on the fO2 of igneous rocks. SPGs were chosen to maximize the potential effects of sediments in their sources on the fO2 of the magmas. We studied 28 Archean (2685–2547 Ma) and 31 Meso- to Paleoproterozoic (1885–1420 Ma) geographically distributed samples from North America, spanning two cratons (Superior and Wyoming) and both orogenic and anorogenic Proterozoic provinces (Trans-Hudson Orogen, Wopmay Orogen, and SW USA). We present an analysis of both new and previously published whole-rock major and trace element data and mineral major element chemistry from the samples. All the studied samples are peraluminous high-silica plutonic rocks (all contain >67 wt % SiO2, and 92% are true granites with >69 wt % SiO2), and biotite + muscovite ± garnet ± tourmaline ± sillimanite are the primary aluminous minerals in all samples. Whole-rock major element and trace element abundances of all samples are consistent with derivation by partial melting of aluminous sediments. To constrain the fO2 of crystallization of the SPGs, we developed an alphaMELTS-based method that takes advantage of the sensitivity of biotite FeT/(FeT + Mg) ratios to fO2 ⁠. This method is able to reproduce experimental and empirical data where biotite compositions and whole-rock compositions, pressures and temperatures of crystallization and fO2 are known. For the SPGs in this study, alphaMELTS modeling indicates that 68% of Proterozoic samples crystallized at an fO2 between NNO –1 and NNO +1·1 (where NNO is nickel–nickel oxide buffer), whereas the remaining Proterozoic samples (32%) and most of the Archean samples (75%) crystallized at ≤NNO –2. The simplest explanation of these results is that the Proterozoic SPGs were derived from metasedimentary source rocks that on average had more oxidized bulk redox states relative to their Archean counterparts. The bulk redox state of the metasedimentary source rocks of SPGs of all ages is defined by the relative abundances of oxidized (e.g. Fe3+ and S6+) and reduced (e.g. organic matter) material. The crystallization of both Archean and Proterozoic samples at fO2 values of ≤NNO –2 is consistent with them having their fO2 buffered by graphite (formed from organic carbon) in their source regions. However, the dominantly low fO2 (≤NNO –2) values of the Archean SPGs plausibly reflects the presence of organic material and relatively reduced metasedimentary rocks in their source region prior to the GOE. In contrast, the elevated fO2 values of the majority of the Proterozoic SPGs may reflect enhanced sulfate contents or increased Fe3+/FeT in sediments after the GOE, which, in terms of the bulk redox state of their metasedimentary source region, would have offset the reducing nature of organic matter present there. INTRODUCTION Interaction between magmas and materials that once resided at or near the surface of the Earth contributes to the diversity of igneous rocks (Bowen, 1928; Taylor, 1980; DePaolo, 1981a). Geochemical signatures of supracrustal rocks are incorporated into magmas in primarily two ways. First, near-surface sedimentary and igneous rocks are recycled into the mantle through subduction. Subsequent mass transfer from the down-going plate to the overlying mantle wedge can transfer distinctive geochemical features to the source regions of arc magmas (e.g. White & Dupré, 1986; Pearce & Peate, 1995; Thirlwall et al., 1996). Second, sedimentary rocks are buried and metamorphosed and then either incorporated into magmas through assimilation or partially melted to produce granites (e.g. White & Chappell, 1977; Clemens & Vielzeuf, 1987; Ague & Brimhall, 1988a; Holtz & Johannes, 1991; Patiño Douce & Johnston, 1991). Quantifying the extent to which incorporation of sediments has influenced the geochemistry and petrology of igneous rocks and the mechanisms by which the sedimentary signature enters magmas is a long-standing problem in petrology. Both assimilation of sediments and their presence in the mantle or crustal source regions of magmas have been identified through the examination of trace elements (DePaolo, 1981a; White & Dupré, 1986), stable and radiogenic isotopes (O’Neil et al., 1977; Taylor, 1980), and, in young arc volcanic rocks, the presence of cosmogenic 10Be (Brown et al., 1982; Morris et al., 1990). In extreme cases, such as strongly peraluminous granites (SPGs), which are derived wholly or largely through the partial melting of Al-rich clastic sedimentary rocks, the mineralogy, petrology, and geochemistry of the magmas are strongly controlled by the character of the sedimentary source material (White & Chappell, 1977; Sylvester, 1998). In addition to major element, trace element, and isotope geochemistry, it is plausible that the oxidation states of magmas are affected by incorporation of rocks that have interacted with near-surface environments. This is because of large variations in the availability of oxygen in near-surface and crustal environments, ranging from sediments such as arkoses that reflect the oxygen content of the atmosphere (which has been oxygenated since the Proterozoic and highly oxygenated for at least the past several hundred million years) to organic-rich muds that form under reducing deep-water conditions. Assimilation and/or partial melting of both highly oxidizing and reducing materials have been suggested as influences on the oxygen fugacity (⁠ fO2 ) of magmas. For example, subduction zone basalts have higher fO2 values than mid-ocean ridge basalts (Carmichael, 1991; Kelley & Cottrell, 2009), generally believed to be due to incorporation of oxidized material from the subducted slab into the mantle wedge from which the arc magmas are primarily derived (Parkinson & Arculus, 1999; Kelley & Cottrell, 2012; Brounce et al., 2014). In contrast, Phanerozoic SPGs and granites strongly contaminated by supracrustal sediments are generally significantly more reduced than metaluminous granites of arc batholiths (Ishihara, 1977, 2004; Ague & Brimhall, 1988b; Whalen & Chappell, 1988); their relatively low fO2 values have been ascribed to the presence of reduced organic carbon in their metasedimentary source regions (Wyborn et al., 1981; Ague & Brimhall, 1988a; Whalen & Chappell, 1988). The partial pressure of oxygen in the atmosphere, and therefore mechanisms of weathering/alteration and redox conditions during sediment deposition and diagenesis, has varied over Earth’s history (Rasmussen & Buick, 1999; Holland, 2006; Guo et al., 2009; Sverjensky & Lee, 2010; Lyons et al., 2014). It might therefore be expected that this would be expressed as secular variations in the fO2 values of igneous rocks that have sediments as source components or contaminants. The largest recognized increase in the partial pressure of atmospheric oxygen occurred at c. 2·4 Ga during what is known as the ‘Great Oxidation Event’ (GOE). The GOE left clear imprints on the sedimentary rock record, including the appearance of fluvial and nearshore red beds and retention of iron in paleosols (Holland, 1984), the disappearance of readily oxidized detrital minerals such as pyrite and uraninite from clastic sedimentary rocks (Grandstaff, 1980; Rasmussen & Buick, 1999; Johnson et al., 2016), and the appearance of evaporative sulfate mineral deposits (Chandler, 1988). Thus, SPGs derived from sedimentary rocks deposited on either side of the GOE might have significantly different fO2 values, mirroring abrupt changes in the redox states of Fe and S in their sources. However, it is also possible, despite variations in atmospheric pO2, that the bulk redox state of sedimentary source material depends strongly on the details of its depositional environment and the presence or absence of organic carbon. Indeed, total organic carbon contents in Archean and Proterozoic/Phanerozoic shales display similar distributions (Dimroth & Kimberley, 1976; Holland, 1984; Lyons et al., 2014). As a result, it is possible that the organic C contents of sediments would overwhelm the potential effects of variation in atmospheric pO2 on the oxidation states of the sources of SPGs and that they would not record such variations. In this study, we develop constraints on the fO2 values during crystallization of Archean and Proterozoic SPGs to determine whether changes in atmospheric pO2 levels across the GOE are recorded in granites from the continental crust. To isolate the effects of the sedimentary contribution to igneous rocks and to provide the most direct link between igneous rocks and changing surface conditions, we examine Archean and Paleo- to Mesoproterozoic SPGs that are formed by the partial melting of metasedimentary rocks. To place constraints on the fO2 values of the SPGs, we use empirical relationships based on previously published experimental data, data from volcanic and plutonic rocks, and alphaMELTS (https://magmasource.caltech.edu/alphamelts/1/alphamelts_manual.pdf) thermodynamic modeling to understand variations in FeT/(FeT + Mg) ratios in biotite (where FeT is total Fe, i.e. Fe3+ + Fe2+) versus those of their host rocks—which are sensitive to the fO2 ⁠, T, and aH2O during crystallization. Finally, we discuss our results in light of changes in the sedimentary redox budgets for C, Fe, and S across the GOE. DEFINITIONS, SAMPLES, REGIONAL GEOLOGY, AND PREVIOUS STUDIES Definition of strongly peraluminous granites Peraluminous granites are those that include corundum in their CIPW norms (Shand, 1927). This translates into excess Al2O3 in their bulk composition that cannot be accommodated in feldspar by coordinating with Ca, Na, and K. Here we define strongly peraluminous granites as granites (sensu lato) with an aluminum saturation index (ASI) or molar Al/(Ca – 1·78P + Na + K) > 1·1 (note reduction of Ca by the amount necessary to combine with P2O5 to make apatite). In terms of mineralogy, strongly peraluminous rocks contain an aluminous mineral such as muscovite, garnet, tourmaline, cordierite, kyanite, sillimanite, or andalusite. For all samples considered in this study, there is at least one of these aluminous phases present in the rock. Strongly peraluminous granites are often considered synonymous with ‘S-type’ granites, or those derived through the partial melting of (meta-)sedimentary rocks. In their original definition of ‘S-type’ granites, Chappell & White (1974) used a variety of criteria to distinguish sediment-derived granitoids of the Lachlan fold belt, Australia, in addition to the ASI. These included other geochemical features, such as whole-rock 87Sr/86Sr0 > 0·708 and δ18O ≥10‰, suggesting derivation from weathered sedimentary material. However, since the initial studies of the Lachlan fold belt, other studies have demonstrated that many ‘S-type’ granites from the Lachlan fold belt have a significant mantle contribution and thus are not pure melts of sediments (Gray, 1984; Collins, 1996; Healy et al., 2004). Therefore, rather than Chappell & White’s (1974) definition of ‘S-type’ granites, we have used the geochemical classification of strongly peraluminous granites [see Frost et al. (2001) and Frost & Frost (2008) for more thorough discussion of issues with the ‘alphabetic’ classification for granitoids] along with a careful review of available data, so as to include only localities with sufficient information to suggest the probable derivation from metasedimentary source regions. The data we take into account in addition to geochemistry include geological setting, field relationships, mineralogy, and oxygen isotope data (where available). Samples We studied North American SPGs from the Archean Superior and Wyoming cratons, from the Paleoproterozoic Trans-Hudson and Wopmay orogens, and from the Mesoproterozoic anorogenic magmatic provinces of the southwestern USA (see Fig. 1). Detailed information on sample localities is given in Table 1, including crystallization ages, source material characteristics and age, petrogenetic interpretations, and supporting references. We summarize the regional geology of the studied suites in the main text below and give more detailed descriptions for each locality in the Supplementary Data Appendix, including field photographs in Fig. A2 (supplementary data are available for downloading at http://www.petrology.oxfordjournals.org). The data used in this paper are a combination of newly collected data and a compilation of previously published data. Data sources for each sample are given in Table 2. Table 1 Summary of regional geology of strongly peraluminous granite localities Locality Craton/orogenic belt Age (Ma) Lithologies Source rocks (lithology and age) Petrogenesis/tectonic interpretation References Ghost Lake batholith (ON, Canada) Superior Province (boundary of Wabigoon and Winnipeg River subprovinces) ∼2685, U–Pb mnz bt + crd granites to evolved ms + grt + tur leucogranites Zealand–Minnitaki–Warclub metasedimentary rocks: turbiditic wackes and mudstones. Depositional age 2733–2706 Ma Syncollisional, melting of sediments during collision of an arc (Wabigoon subprovince) and an older (∼3 Ga) microcontinent (Winnipeg River subprovince) Davis et al. (1988, 1990); Blackburn et al. (1991); Breaks & Janes (1991); Breaks & Moore (1992); Breaks et al. (2005) Sturgeon Lake granite (ON, Canada) Superior Province (Quetico subprovince) 2671 ± 2, U–Pb mnz bt + ms leucogranite, local grt, tur, cord, and brl Quetico metasedimentary rocks: greywacke protoliths deposited as forearc turbidites of the Wabigoon arc with mixed volcanic–plutonic source region. Depositional age ∼2690–2700 Ma Partial melting of Quetico metagreywackes resulting from collision of two arcs (Wabigoon and Wawa subprovinces) and burial of sediments Southwick & Sims (1979); Percival et al. (1985); Day & Weiblen (1986); Percival & Sullivan (1988); Sawyer & Barnes (1988); Percival & Williams (1989); Southwick (1991) Shannon Lake granite (MN, USA) Superior Province (Wawa subprovince) 2674 ± 5, U–Pb zrn bt + ms granite, local pegmatites with accessory grt, tur, and brl Not clearly identified. Probably local metasediments (greywackes, argillite, conglomerates) of the Wawa subprovince Post-deformational intrusion Boerboom & Zartman (1993) Preissac, Lacorne, Lamotte, Moly Hill plutons (QC, Canada) Superior Province (Pontiac subprovince and Lacorne block of Abitibi subprovince) Lamotte: 2647 ± 2, U–Pb mnz/ttn; Preissac: 2681–2660, U–Pb mnz/ttn ms + grt monzogranites Pontiac metasediments: mature turbidite wackes and pelites, interlayered with minor felsic, mafic, and ultramafic volcanic rocks. Depositional age >2685 Ma Syncollisional, remelting of mature metasedimentary rocks during continent–arc collision involving thrusting of Pontiac subprovince under Abitibi greenstone belt Dawson (1966); Feng & Kerrich (1992a, 1992b); Feng et al. (1993); Mulja et al. (1995a, 1995b); Ducharme et al. (1997) Mt Owen batholith, (WY, USA) Wyoming craton (Beartooth–Bighorn magmatic zone) 2547 ± 3, U–Pb zrn bt + ms leucogranite with local grt, pegmatitic to fine-grained zones Specific source rock not identified, but hydrous, peraluminous composition suggests formation through partial melting of pelitic or psammitic metasedimentary rocks Partial melting of quartzofeldspathic crustal source Reed & Zartman, (1973); Zartman & Reed (1998); Frost et al. (2006, 2018) Harney Peak granite (SD, USA) Trans-Hudson orogen 1715 ± 3, U–Pb mnz bt + ms ± grt granite, ms + tur ± grt granite. Associated pegmatite sills and satellite intrusions Metapelites and metagreywackes analogous to host rocks. Depositional age ∼2100–1880 Ma (for host metasedimentary rocks at surface) Partial melting of Paleoproterozoic (and possibly Archean) metasedimentary rocks during culmination of Trans-Hudson orogeny Redden et al., (1982; 1990); Duke et al. (1992); Nabelek et al., (1992a, 1992b; 1999, 2001) Hepburn intrusive suite (NWT, Canada) Wopmay orogen 1885, U–Pb zrn, upper intercept concordia bt + ms ± grt ± sil monzogranites and syenogranites Coronation Supergroup metapelites: a mixture of continental shelf, turbiditic, and deep-water sediments associated with the closure of a volcanic–sedimentary basin. Depositional age 1900 Ma Mantle-derived basalt generated in a back-arc setting, which subsequently assimilated and/or partially melted significant amounts of the sedimentary basin host rocks Lalonde (1986); Lalonde (1989); Lalonde & Bernard (1993) Silver Plume and St Vrain granites (CO, USA) SW US anorogenic Mesoproterozoic magmatism 1420–1450, Rb–Sr whole-rock and mineral isochron bt + ms ± sil granites Specific source rock not identified, but peraluminous composition and Sm–Nd isotope ratios suggest a Paleoproterozoic peraluminous quartzofeldspathic (probably dominantly metasedimentary) crustal source Anorogenic crustal melting, heat source not clearly identified Peterman et al. (1968); DePaolo (1981b); Anderson & Thomas (1985) Ak-Chin, Ruin, Oracle, Sierra Estrella granites (AZ, USA) SW US anorogenic Mesoproterozoic magmatism Ruin and Oracle granites: 1440 ± 20, U–Pb zrn bt + ms ± grt ± tur granites Specific source rock not identified, but peraluminous composition and Sm–Nd and Rb–Sr isotope ratios suggest a Paleoproterozoic peraluminous quartzofeldspathic (probably dominantly metasedimentary) crustal source Anorogenic crustal melting; heat source not clearly identified Silver et al. (1980); Farmer & DePaolo (1984); Nelson & DePaolo (1985); Anderson & Bender (1989); Anderson & Morrison (2005) Locality Craton/orogenic belt Age (Ma) Lithologies Source rocks (lithology and age) Petrogenesis/tectonic interpretation References Ghost Lake batholith (ON, Canada) Superior Province (boundary of Wabigoon and Winnipeg River subprovinces) ∼2685, U–Pb mnz bt + crd granites to evolved ms + grt + tur leucogranites Zealand–Minnitaki–Warclub metasedimentary rocks: turbiditic wackes and mudstones. Depositional age 2733–2706 Ma Syncollisional, melting of sediments during collision of an arc (Wabigoon subprovince) and an older (∼3 Ga) microcontinent (Winnipeg River subprovince) Davis et al. (1988, 1990); Blackburn et al. (1991); Breaks & Janes (1991); Breaks & Moore (1992); Breaks et al. (2005) Sturgeon Lake granite (ON, Canada) Superior Province (Quetico subprovince) 2671 ± 2, U–Pb mnz bt + ms leucogranite, local grt, tur, cord, and brl Quetico metasedimentary rocks: greywacke protoliths deposited as forearc turbidites of the Wabigoon arc with mixed volcanic–plutonic source region. Depositional age ∼2690–2700 Ma Partial melting of Quetico metagreywackes resulting from collision of two arcs (Wabigoon and Wawa subprovinces) and burial of sediments Southwick & Sims (1979); Percival et al. (1985); Day & Weiblen (1986); Percival & Sullivan (1988); Sawyer & Barnes (1988); Percival & Williams (1989); Southwick (1991) Shannon Lake granite (MN, USA) Superior Province (Wawa subprovince) 2674 ± 5, U–Pb zrn bt + ms granite, local pegmatites with accessory grt, tur, and brl Not clearly identified. Probably local metasediments (greywackes, argillite, conglomerates) of the Wawa subprovince Post-deformational intrusion Boerboom & Zartman (1993) Preissac, Lacorne, Lamotte, Moly Hill plutons (QC, Canada) Superior Province (Pontiac subprovince and Lacorne block of Abitibi subprovince) Lamotte: 2647 ± 2, U–Pb mnz/ttn; Preissac: 2681–2660, U–Pb mnz/ttn ms + grt monzogranites Pontiac metasediments: mature turbidite wackes and pelites, interlayered with minor felsic, mafic, and ultramafic volcanic rocks. Depositional age >2685 Ma Syncollisional, remelting of mature metasedimentary rocks during continent–arc collision involving thrusting of Pontiac subprovince under Abitibi greenstone belt Dawson (1966); Feng & Kerrich (1992a, 1992b); Feng et al. (1993); Mulja et al. (1995a, 1995b); Ducharme et al. (1997) Mt Owen batholith, (WY, USA) Wyoming craton (Beartooth–Bighorn magmatic zone) 2547 ± 3, U–Pb zrn bt + ms leucogranite with local grt, pegmatitic to fine-grained zones Specific source rock not identified, but hydrous, peraluminous composition suggests formation through partial melting of pelitic or psammitic metasedimentary rocks Partial melting of quartzofeldspathic crustal source Reed & Zartman, (1973); Zartman & Reed (1998); Frost et al. (2006, 2018) Harney Peak granite (SD, USA) Trans-Hudson orogen 1715 ± 3, U–Pb mnz bt + ms ± grt granite, ms + tur ± grt granite. Associated pegmatite sills and satellite intrusions Metapelites and metagreywackes analogous to host rocks. Depositional age ∼2100–1880 Ma (for host metasedimentary rocks at surface) Partial melting of Paleoproterozoic (and possibly Archean) metasedimentary rocks during culmination of Trans-Hudson orogeny Redden et al., (1982; 1990); Duke et al. (1992); Nabelek et al., (1992a, 1992b; 1999, 2001) Hepburn intrusive suite (NWT, Canada) Wopmay orogen 1885, U–Pb zrn, upper intercept concordia bt + ms ± grt ± sil monzogranites and syenogranites Coronation Supergroup metapelites: a mixture of continental shelf, turbiditic, and deep-water sediments associated with the closure of a volcanic–sedimentary basin. Depositional age 1900 Ma Mantle-derived basalt generated in a back-arc setting, which subsequently assimilated and/or partially melted significant amounts of the sedimentary basin host rocks Lalonde (1986); Lalonde (1989); Lalonde & Bernard (1993) Silver Plume and St Vrain granites (CO, USA) SW US anorogenic Mesoproterozoic magmatism 1420–1450, Rb–Sr whole-rock and mineral isochron bt + ms ± sil granites Specific source rock not identified, but peraluminous composition and Sm–Nd isotope ratios suggest a Paleoproterozoic peraluminous quartzofeldspathic (probably dominantly metasedimentary) crustal source Anorogenic crustal melting, heat source not clearly identified Peterman et al. (1968); DePaolo (1981b); Anderson & Thomas (1985) Ak-Chin, Ruin, Oracle, Sierra Estrella granites (AZ, USA) SW US anorogenic Mesoproterozoic magmatism Ruin and Oracle granites: 1440 ± 20, U–Pb zrn bt + ms ± grt ± tur granites Specific source rock not identified, but peraluminous composition and Sm–Nd and Rb–Sr isotope ratios suggest a Paleoproterozoic peraluminous quartzofeldspathic (probably dominantly metasedimentary) crustal source Anorogenic crustal melting; heat source not clearly identified Silver et al. (1980); Farmer & DePaolo (1984); Nelson & DePaolo (1985); Anderson & Bender (1989); Anderson & Morrison (2005) Table 1 Summary of regional geology of strongly peraluminous granite localities Locality Craton/orogenic belt Age (Ma) Lithologies Source rocks (lithology and age) Petrogenesis/tectonic interpretation References Ghost Lake batholith (ON, Canada) Superior Province (boundary of Wabigoon and Winnipeg River subprovinces) ∼2685, U–Pb mnz bt + crd granites to evolved ms + grt + tur leucogranites Zealand–Minnitaki–Warclub metasedimentary rocks: turbiditic wackes and mudstones. Depositional age 2733–2706 Ma Syncollisional, melting of sediments during collision of an arc (Wabigoon subprovince) and an older (∼3 Ga) microcontinent (Winnipeg River subprovince) Davis et al. (1988, 1990); Blackburn et al. (1991); Breaks & Janes (1991); Breaks & Moore (1992); Breaks et al. (2005) Sturgeon Lake granite (ON, Canada) Superior Province (Quetico subprovince) 2671 ± 2, U–Pb mnz bt + ms leucogranite, local grt, tur, cord, and brl Quetico metasedimentary rocks: greywacke protoliths deposited as forearc turbidites of the Wabigoon arc with mixed volcanic–plutonic source region. Depositional age ∼2690–2700 Ma Partial melting of Quetico metagreywackes resulting from collision of two arcs (Wabigoon and Wawa subprovinces) and burial of sediments Southwick & Sims (1979); Percival et al. (1985); Day & Weiblen (1986); Percival & Sullivan (1988); Sawyer & Barnes (1988); Percival & Williams (1989); Southwick (1991) Shannon Lake granite (MN, USA) Superior Province (Wawa subprovince) 2674 ± 5, U–Pb zrn bt + ms granite, local pegmatites with accessory grt, tur, and brl Not clearly identified. Probably local metasediments (greywackes, argillite, conglomerates) of the Wawa subprovince Post-deformational intrusion Boerboom & Zartman (1993) Preissac, Lacorne, Lamotte, Moly Hill plutons (QC, Canada) Superior Province (Pontiac subprovince and Lacorne block of Abitibi subprovince) Lamotte: 2647 ± 2, U–Pb mnz/ttn; Preissac: 2681–2660, U–Pb mnz/ttn ms + grt monzogranites Pontiac metasediments: mature turbidite wackes and pelites, interlayered with minor felsic, mafic, and ultramafic volcanic rocks. Depositional age >2685 Ma Syncollisional, remelting of mature metasedimentary rocks during continent–arc collision involving thrusting of Pontiac subprovince under Abitibi greenstone belt Dawson (1966); Feng & Kerrich (1992a, 1992b); Feng et al. (1993); Mulja et al. (1995a, 1995b); Ducharme et al. (1997) Mt Owen batholith, (WY, USA) Wyoming craton (Beartooth–Bighorn magmatic zone) 2547 ± 3, U–Pb zrn bt + ms leucogranite with local grt, pegmatitic to fine-grained zones Specific source rock not identified, but hydrous, peraluminous composition suggests formation through partial melting of pelitic or psammitic metasedimentary rocks Partial melting of quartzofeldspathic crustal source Reed & Zartman, (1973); Zartman & Reed (1998); Frost et al. (2006, 2018) Harney Peak granite (SD, USA) Trans-Hudson orogen 1715 ± 3, U–Pb mnz bt + ms ± grt granite, ms + tur ± grt granite. Associated pegmatite sills and satellite intrusions Metapelites and metagreywackes analogous to host rocks. Depositional age ∼2100–1880 Ma (for host metasedimentary rocks at surface) Partial melting of Paleoproterozoic (and possibly Archean) metasedimentary rocks during culmination of Trans-Hudson orogeny Redden et al., (1982; 1990); Duke et al. (1992); Nabelek et al., (1992a, 1992b; 1999, 2001) Hepburn intrusive suite (NWT, Canada) Wopmay orogen 1885, U–Pb zrn, upper intercept concordia bt + ms ± grt ± sil monzogranites and syenogranites Coronation Supergroup metapelites: a mixture of continental shelf, turbiditic, and deep-water sediments associated with the closure of a volcanic–sedimentary basin. Depositional age 1900 Ma Mantle-derived basalt generated in a back-arc setting, which subsequently assimilated and/or partially melted significant amounts of the sedimentary basin host rocks Lalonde (1986); Lalonde (1989); Lalonde & Bernard (1993) Silver Plume and St Vrain granites (CO, USA) SW US anorogenic Mesoproterozoic magmatism 1420–1450, Rb–Sr whole-rock and mineral isochron bt + ms ± sil granites Specific source rock not identified, but peraluminous composition and Sm–Nd isotope ratios suggest a Paleoproterozoic peraluminous quartzofeldspathic (probably dominantly metasedimentary) crustal source Anorogenic crustal melting, heat source not clearly identified Peterman et al. (1968); DePaolo (1981b); Anderson & Thomas (1985) Ak-Chin, Ruin, Oracle, Sierra Estrella granites (AZ, USA) SW US anorogenic Mesoproterozoic magmatism Ruin and Oracle granites: 1440 ± 20, U–Pb zrn bt + ms ± grt ± tur granites Specific source rock not identified, but peraluminous composition and Sm–Nd and Rb–Sr isotope ratios suggest a Paleoproterozoic peraluminous quartzofeldspathic (probably dominantly metasedimentary) crustal source Anorogenic crustal melting; heat source not clearly identified Silver et al. (1980); Farmer & DePaolo (1984); Nelson & DePaolo (1985); Anderson & Bender (1989); Anderson & Morrison (2005) Locality Craton/orogenic belt Age (Ma) Lithologies Source rocks (lithology and age) Petrogenesis/tectonic interpretation References Ghost Lake batholith (ON, Canada) Superior Province (boundary of Wabigoon and Winnipeg River subprovinces) ∼2685, U–Pb mnz bt + crd granites to evolved ms + grt + tur leucogranites Zealand–Minnitaki–Warclub metasedimentary rocks: turbiditic wackes and mudstones. Depositional age 2733–2706 Ma Syncollisional, melting of sediments during collision of an arc (Wabigoon subprovince) and an older (∼3 Ga) microcontinent (Winnipeg River subprovince) Davis et al. (1988, 1990); Blackburn et al. (1991); Breaks & Janes (1991); Breaks & Moore (1992); Breaks et al. (2005) Sturgeon Lake granite (ON, Canada) Superior Province (Quetico subprovince) 2671 ± 2, U–Pb mnz bt + ms leucogranite, local grt, tur, cord, and brl Quetico metasedimentary rocks: greywacke protoliths deposited as forearc turbidites of the Wabigoon arc with mixed volcanic–plutonic source region. Depositional age ∼2690–2700 Ma Partial melting of Quetico metagreywackes resulting from collision of two arcs (Wabigoon and Wawa subprovinces) and burial of sediments Southwick & Sims (1979); Percival et al. (1985); Day & Weiblen (1986); Percival & Sullivan (1988); Sawyer & Barnes (1988); Percival & Williams (1989); Southwick (1991) Shannon Lake granite (MN, USA) Superior Province (Wawa subprovince) 2674 ± 5, U–Pb zrn bt + ms granite, local pegmatites with accessory grt, tur, and brl Not clearly identified. Probably local metasediments (greywackes, argillite, conglomerates) of the Wawa subprovince Post-deformational intrusion Boerboom & Zartman (1993) Preissac, Lacorne, Lamotte, Moly Hill plutons (QC, Canada) Superior Province (Pontiac subprovince and Lacorne block of Abitibi subprovince) Lamotte: 2647 ± 2, U–Pb mnz/ttn; Preissac: 2681–2660, U–Pb mnz/ttn ms + grt monzogranites Pontiac metasediments: mature turbidite wackes and pelites, interlayered with minor felsic, mafic, and ultramafic volcanic rocks. Depositional age >2685 Ma Syncollisional, remelting of mature metasedimentary rocks during continent–arc collision involving thrusting of Pontiac subprovince under Abitibi greenstone belt Dawson (1966); Feng & Kerrich (1992a, 1992b); Feng et al. (1993); Mulja et al. (1995a, 1995b); Ducharme et al. (1997) Mt Owen batholith, (WY, USA) Wyoming craton (Beartooth–Bighorn magmatic zone) 2547 ± 3, U–Pb zrn bt + ms leucogranite with local grt, pegmatitic to fine-grained zones Specific source rock not identified, but hydrous, peraluminous composition suggests formation through partial melting of pelitic or psammitic metasedimentary rocks Partial melting of quartzofeldspathic crustal source Reed & Zartman, (1973); Zartman & Reed (1998); Frost et al. (2006, 2018) Harney Peak granite (SD, USA) Trans-Hudson orogen 1715 ± 3, U–Pb mnz bt + ms ± grt granite, ms + tur ± grt granite. Associated pegmatite sills and satellite intrusions Metapelites and metagreywackes analogous to host rocks. Depositional age ∼2100–1880 Ma (for host metasedimentary rocks at surface) Partial melting of Paleoproterozoic (and possibly Archean) metasedimentary rocks during culmination of Trans-Hudson orogeny Redden et al., (1982; 1990); Duke et al. (1992); Nabelek et al., (1992a, 1992b; 1999, 2001) Hepburn intrusive suite (NWT, Canada) Wopmay orogen 1885, U–Pb zrn, upper intercept concordia bt + ms ± grt ± sil monzogranites and syenogranites Coronation Supergroup metapelites: a mixture of continental shelf, turbiditic, and deep-water sediments associated with the closure of a volcanic–sedimentary basin. Depositional age 1900 Ma Mantle-derived basalt generated in a back-arc setting, which subsequently assimilated and/or partially melted significant amounts of the sedimentary basin host rocks Lalonde (1986); Lalonde (1989); Lalonde & Bernard (1993) Silver Plume and St Vrain granites (CO, USA) SW US anorogenic Mesoproterozoic magmatism 1420–1450, Rb–Sr whole-rock and mineral isochron bt + ms ± sil granites Specific source rock not identified, but peraluminous composition and Sm–Nd isotope ratios suggest a Paleoproterozoic peraluminous quartzofeldspathic (probably dominantly metasedimentary) crustal source Anorogenic crustal melting, heat source not clearly identified Peterman et al. (1968); DePaolo (1981b); Anderson & Thomas (1985) Ak-Chin, Ruin, Oracle, Sierra Estrella granites (AZ, USA) SW US anorogenic Mesoproterozoic magmatism Ruin and Oracle granites: 1440 ± 20, U–Pb zrn bt + ms ± grt ± tur granites Specific source rock not identified, but peraluminous composition and Sm–Nd and Rb–Sr isotope ratios suggest a Paleoproterozoic peraluminous quartzofeldspathic (probably dominantly metasedimentary) crustal source Anorogenic crustal melting; heat source not clearly identified Silver et al. (1980); Farmer & DePaolo (1984); Nelson & DePaolo (1985); Anderson & Bender (1989); Anderson & Morrison (2005) Table 2 Strongly peraluminous granite samples (locality information, mineralogy, and source of data) Sample Locality Age (Ma) Location Mineralogy* Data sources† Latitude (°N) Longitude (°W) Whole- rock Mineral chemistry Geochr onology E19‐8 Ghost Lake batholith 2685 49·8161 92·9820 bt + ms 1 1 2 SP-16‐20a Ghost Lake batholith 2685 49·8152 93·0203 bt + ms 1 1 2 SP-16‐22 Ghost Lake batholith 2685 49·8152 92·9772 bt + ms 1 1 2 SP-16‐28a Ghost Lake batholith 2685 49·8018 93·0215 bt + ms + grt 1 1 2 SP-16‐23 Ghost Lake batholith 2685 49·8150 92·9504 bt + ms 1 1 2 SP-16‐24 Ghost Lake batholith 2685 49·8150 92·9496 bt + ms 1 1 2 SP-16‐34 Ghost Lake batholith 2685 49·8485 92·6937 bt + ms + grt  + tur 1 1 2 SP-16‐48 Sturgeon Lake granite 2671–2641 48·6796 90·8700 bt + ms 1 1 3 SP-16‐52 Sturgeon Lake granite 2671–2641 48·6783 91·0187 bt + ms + grt 1 1 3 SP-16‐53 Sturgeon Lake granite 2671–2641 48·6783 91·0187 bt + ms 1 1 3 SP-16‐1 Shannon Lake granite 2674±5 47·6391 92·9943 bt + ms 1 1 4 SP-16‐2b Shannon Lake granite 2674±5 47·6612 92·9386 bt + ms 1 1 4 LC-8 Lacorne block 2643–2631 See ref. 6 for map bt + ms + grt 5 5, 6 7 LC-16 Lacorne block 2643–2631 See ref. 6 for map bt + ms + grt 5 5, 6 7 LC-30 Lacorne block 2643–2631 See ref. 6 for map bt + ms + grt 5 5, 6 7 903 Preissac pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 768 Lamotte pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 797 Lamotte pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 908 Lamotte pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 616 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 634 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 652 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 672 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 692 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 707 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 26 Moly Hill pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 29 Moly Hill pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 98TI Mt. Owen batholith 2547 ± 3 43·7306 110·7806 bt + ms + grt 11 1 12 HP-6B Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 HP-7B Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 HP-21C Harney Peak granite 1715 See ref. 10 for map bt + ms 13 1, 13 14 HP-26 Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 HP 32 Harney Peak granite 1715 See ref. 10 for map bt + ms 13 1, 13 14 HP 43A Harney Peak granite 1715 See ref. 10 for map grt + ms + bt (± tur) 13 1, 13 14 HP 44A Harney Peak granite 1715 See ref. 10 for map bt + ms 13 1, 13 14 HP-23A Harney Peak granite 1715 See ref. 10 for map grt + ms + bt 13 13 14 HP-14A Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 L14 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L331 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L99 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L204 Hepburn intrusive suite 1885 Specific location not given bt + ms 15 16 17 L341A Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L125 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 S29 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 S54 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 SP-1 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-3 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-5 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-6 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-8 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SVL-3 St Vrain granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SVL-4 St Vrain granite 1450–1390 See ref. 15 for map bt + ms + sil 18 18, 19 20 SVL-5 St Vrain granite 1450–1390 See ref. 15 for map bt + ms + sil 18 18, 19 20 SVL-7 St Vrain granite 1450–1390 See ref. 15 for map bt + ms + sil 18 18, 19 20 ARU-83‐1A Ruin granite 1450–1390 See ref. 15 for map bt + ms 18 21, 19 20 AOR-83‐1 Oracle granite 1450–1390 See ref. 15 for map bt + ms 18 21, 19 20 ASE-83‐1 Sierra Estrella granite 1450–1390 See ref. 15 for map bt + ms 18 21, 19 20 AMC-83‐1 Ak-Chin granite — See ref. 18 for map bt + ms 21 21, 19 — AMC-83‐2 Ak-Chin granite — See ref. 18 for map bt + ms 21 21, 19 — Sample Locality Age (Ma) Location Mineralogy* Data sources† Latitude (°N) Longitude (°W) Whole- rock Mineral chemistry Geochr onology E19‐8 Ghost Lake batholith 2685 49·8161 92·9820 bt + ms 1 1 2 SP-16‐20a Ghost Lake batholith 2685 49·8152 93·0203 bt + ms 1 1 2 SP-16‐22 Ghost Lake batholith 2685 49·8152 92·9772 bt + ms 1 1 2 SP-16‐28a Ghost Lake batholith 2685 49·8018 93·0215 bt + ms + grt 1 1 2 SP-16‐23 Ghost Lake batholith 2685 49·8150 92·9504 bt + ms 1 1 2 SP-16‐24 Ghost Lake batholith 2685 49·8150 92·9496 bt + ms 1 1 2 SP-16‐34 Ghost Lake batholith 2685 49·8485 92·6937 bt + ms + grt  + tur 1 1 2 SP-16‐48 Sturgeon Lake granite 2671–2641 48·6796 90·8700 bt + ms 1 1 3 SP-16‐52 Sturgeon Lake granite 2671–2641 48·6783 91·0187 bt + ms + grt 1 1 3 SP-16‐53 Sturgeon Lake granite 2671–2641 48·6783 91·0187 bt + ms 1 1 3 SP-16‐1 Shannon Lake granite 2674±5 47·6391 92·9943 bt + ms 1 1 4 SP-16‐2b Shannon Lake granite 2674±5 47·6612 92·9386 bt + ms 1 1 4 LC-8 Lacorne block 2643–2631 See ref. 6 for map bt + ms + grt 5 5, 6 7 LC-16 Lacorne block 2643–2631 See ref. 6 for map bt + ms + grt 5 5, 6 7 LC-30 Lacorne block 2643–2631 See ref. 6 for map bt + ms + grt 5 5, 6 7 903 Preissac pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 768 Lamotte pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 797 Lamotte pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 908 Lamotte pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 616 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 634 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 652 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 672 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 692 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 707 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 26 Moly Hill pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 29 Moly Hill pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 98TI Mt. Owen batholith 2547 ± 3 43·7306 110·7806 bt + ms + grt 11 1 12 HP-6B Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 HP-7B Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 HP-21C Harney Peak granite 1715 See ref. 10 for map bt + ms 13 1, 13 14 HP-26 Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 HP 32 Harney Peak granite 1715 See ref. 10 for map bt + ms 13 1, 13 14 HP 43A Harney Peak granite 1715 See ref. 10 for map grt + ms + bt (± tur) 13 1, 13 14 HP 44A Harney Peak granite 1715 See ref. 10 for map bt + ms 13 1, 13 14 HP-23A Harney Peak granite 1715 See ref. 10 for map grt + ms + bt 13 13 14 HP-14A Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 L14 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L331 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L99 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L204 Hepburn intrusive suite 1885 Specific location not given bt + ms 15 16 17 L341A Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L125 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 S29 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 S54 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 SP-1 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-3 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-5 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-6 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-8 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SVL-3 St Vrain granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SVL-4 St Vrain granite 1450–1390 See ref. 15 for map bt + ms + sil 18 18, 19 20 SVL-5 St Vrain granite 1450–1390 See ref. 15 for map bt + ms + sil 18 18, 19 20 SVL-7 St Vrain granite 1450–1390 See ref. 15 for map bt + ms + sil 18 18, 19 20 ARU-83‐1A Ruin granite 1450–1390 See ref. 15 for map bt + ms 18 21, 19 20 AOR-83‐1 Oracle granite 1450–1390 See ref. 15 for map bt + ms 18 21, 19 20 ASE-83‐1 Sierra Estrella granite 1450–1390 See ref. 15 for map bt + ms 18 21, 19 20 AMC-83‐1 Ak-Chin granite — See ref. 18 for map bt + ms 21 21, 19 — AMC-83‐2 Ak-Chin granite — See ref. 18 for map bt + ms 21 21, 19 — *All samples also include quartz and two feldspars. bt, biotite; ms, muscovite; grt, garnet; tur, tourmaline; sil, sillimanite. †References: 1, this study; 2, Davis (1990); Larbi et al. (1999) (and references therein); 3, Southwick (1991); 4, Boerboom & Zartman (1993); 5, Feng & Kerrich (1992a); 6; Feng (1992); 7, Feng & Kerrich (1991); 8, Mulja (1995); 9, Mulja et al. (1995b); 10, Mulja et al. (1995a); 11, Frost et al. (2006); 12, Zartman & Reed (1998); 13, Nabelek et al. (1992a); 14, Redden et al. (1990); 15, Lalonde (1986); 16, Lalonde & Bernard (1993); 17, Hoffman & Bowring (1984); 18, Anderson & Thomas (1985); 19, L. Anderson (personal communication, 2016); 20, Peterman et al. (1968); 21, Anderson & Bender (1989). Table 2 Strongly peraluminous granite samples (locality information, mineralogy, and source of data) Sample Locality Age (Ma) Location Mineralogy* Data sources† Latitude (°N) Longitude (°W) Whole- rock Mineral chemistry Geochr onology E19‐8 Ghost Lake batholith 2685 49·8161 92·9820 bt + ms 1 1 2 SP-16‐20a Ghost Lake batholith 2685 49·8152 93·0203 bt + ms 1 1 2 SP-16‐22 Ghost Lake batholith 2685 49·8152 92·9772 bt + ms 1 1 2 SP-16‐28a Ghost Lake batholith 2685 49·8018 93·0215 bt + ms + grt 1 1 2 SP-16‐23 Ghost Lake batholith 2685 49·8150 92·9504 bt + ms 1 1 2 SP-16‐24 Ghost Lake batholith 2685 49·8150 92·9496 bt + ms 1 1 2 SP-16‐34 Ghost Lake batholith 2685 49·8485 92·6937 bt + ms + grt  + tur 1 1 2 SP-16‐48 Sturgeon Lake granite 2671–2641 48·6796 90·8700 bt + ms 1 1 3 SP-16‐52 Sturgeon Lake granite 2671–2641 48·6783 91·0187 bt + ms + grt 1 1 3 SP-16‐53 Sturgeon Lake granite 2671–2641 48·6783 91·0187 bt + ms 1 1 3 SP-16‐1 Shannon Lake granite 2674±5 47·6391 92·9943 bt + ms 1 1 4 SP-16‐2b Shannon Lake granite 2674±5 47·6612 92·9386 bt + ms 1 1 4 LC-8 Lacorne block 2643–2631 See ref. 6 for map bt + ms + grt 5 5, 6 7 LC-16 Lacorne block 2643–2631 See ref. 6 for map bt + ms + grt 5 5, 6 7 LC-30 Lacorne block 2643–2631 See ref. 6 for map bt + ms + grt 5 5, 6 7 903 Preissac pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 768 Lamotte pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 797 Lamotte pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 908 Lamotte pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 616 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 634 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 652 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 672 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 692 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 707 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 26 Moly Hill pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 29 Moly Hill pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 98TI Mt. Owen batholith 2547 ± 3 43·7306 110·7806 bt + ms + grt 11 1 12 HP-6B Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 HP-7B Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 HP-21C Harney Peak granite 1715 See ref. 10 for map bt + ms 13 1, 13 14 HP-26 Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 HP 32 Harney Peak granite 1715 See ref. 10 for map bt + ms 13 1, 13 14 HP 43A Harney Peak granite 1715 See ref. 10 for map grt + ms + bt (± tur) 13 1, 13 14 HP 44A Harney Peak granite 1715 See ref. 10 for map bt + ms 13 1, 13 14 HP-23A Harney Peak granite 1715 See ref. 10 for map grt + ms + bt 13 13 14 HP-14A Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 L14 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L331 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L99 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L204 Hepburn intrusive suite 1885 Specific location not given bt + ms 15 16 17 L341A Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L125 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 S29 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 S54 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 SP-1 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-3 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-5 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-6 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-8 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SVL-3 St Vrain granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SVL-4 St Vrain granite 1450–1390 See ref. 15 for map bt + ms + sil 18 18, 19 20 SVL-5 St Vrain granite 1450–1390 See ref. 15 for map bt + ms + sil 18 18, 19 20 SVL-7 St Vrain granite 1450–1390 See ref. 15 for map bt + ms + sil 18 18, 19 20 ARU-83‐1A Ruin granite 1450–1390 See ref. 15 for map bt + ms 18 21, 19 20 AOR-83‐1 Oracle granite 1450–1390 See ref. 15 for map bt + ms 18 21, 19 20 ASE-83‐1 Sierra Estrella granite 1450–1390 See ref. 15 for map bt + ms 18 21, 19 20 AMC-83‐1 Ak-Chin granite — See ref. 18 for map bt + ms 21 21, 19 — AMC-83‐2 Ak-Chin granite — See ref. 18 for map bt + ms 21 21, 19 — Sample Locality Age (Ma) Location Mineralogy* Data sources† Latitude (°N) Longitude (°W) Whole- rock Mineral chemistry Geochr onology E19‐8 Ghost Lake batholith 2685 49·8161 92·9820 bt + ms 1 1 2 SP-16‐20a Ghost Lake batholith 2685 49·8152 93·0203 bt + ms 1 1 2 SP-16‐22 Ghost Lake batholith 2685 49·8152 92·9772 bt + ms 1 1 2 SP-16‐28a Ghost Lake batholith 2685 49·8018 93·0215 bt + ms + grt 1 1 2 SP-16‐23 Ghost Lake batholith 2685 49·8150 92·9504 bt + ms 1 1 2 SP-16‐24 Ghost Lake batholith 2685 49·8150 92·9496 bt + ms 1 1 2 SP-16‐34 Ghost Lake batholith 2685 49·8485 92·6937 bt + ms + grt  + tur 1 1 2 SP-16‐48 Sturgeon Lake granite 2671–2641 48·6796 90·8700 bt + ms 1 1 3 SP-16‐52 Sturgeon Lake granite 2671–2641 48·6783 91·0187 bt + ms + grt 1 1 3 SP-16‐53 Sturgeon Lake granite 2671–2641 48·6783 91·0187 bt + ms 1 1 3 SP-16‐1 Shannon Lake granite 2674±5 47·6391 92·9943 bt + ms 1 1 4 SP-16‐2b Shannon Lake granite 2674±5 47·6612 92·9386 bt + ms 1 1 4 LC-8 Lacorne block 2643–2631 See ref. 6 for map bt + ms + grt 5 5, 6 7 LC-16 Lacorne block 2643–2631 See ref. 6 for map bt + ms + grt 5 5, 6 7 LC-30 Lacorne block 2643–2631 See ref. 6 for map bt + ms + grt 5 5, 6 7 903 Preissac pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 768 Lamotte pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 797 Lamotte pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 908 Lamotte pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 616 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 634 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 652 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 672 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 692 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 707 Lacorne pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 26 Moly Hill pluton 2643–2631 See ref. 8 for map bt + ms + grt 8, 9 8, 10 7 29 Moly Hill pluton 2643–2631 See ref. 8 for map bt + ms 8, 9 8, 10 7 98TI Mt. Owen batholith 2547 ± 3 43·7306 110·7806 bt + ms + grt 11 1 12 HP-6B Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 HP-7B Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 HP-21C Harney Peak granite 1715 See ref. 10 for map bt + ms 13 1, 13 14 HP-26 Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 HP 32 Harney Peak granite 1715 See ref. 10 for map bt + ms 13 1, 13 14 HP 43A Harney Peak granite 1715 See ref. 10 for map grt + ms + bt (± tur) 13 1, 13 14 HP 44A Harney Peak granite 1715 See ref. 10 for map bt + ms 13 1, 13 14 HP-23A Harney Peak granite 1715 See ref. 10 for map grt + ms + bt 13 13 14 HP-14A Harney Peak granite 1715 See ref. 10 for map bt + ms 13 13 14 L14 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L331 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L99 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L204 Hepburn intrusive suite 1885 Specific location not given bt + ms 15 16 17 L341A Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 L125 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 S29 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 S54 Hepburn intrusive suite 1885 Specific location not given bt + ms + grt 15 16 17 SP-1 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-3 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-5 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-6 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SP-8 Silver Plume granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SVL-3 St Vrain granite 1450–1390 See ref. 15 for map bt + ms 18 18, 19 20 SVL-4 St Vrain granite 1450–1390 See ref. 15 for map bt + ms + sil 18 18, 19 20 SVL-5 St Vrain granite 1450–1390 See ref. 15 for map bt + ms + sil 18 18, 19 20 SVL-7 St Vrain granite 1450–1390 See ref. 15 for map bt + ms + sil 18 18, 19 20 ARU-83‐1A Ruin granite 1450–1390 See ref. 15 for map bt + ms 18 21, 19 20 AOR-83‐1 Oracle granite 1450–1390 See ref. 15 for map bt + ms 18 21, 19 20 ASE-83‐1 Sierra Estrella granite 1450–1390 See ref. 15 for map bt + ms 18 21, 19 20 AMC-83‐1 Ak-Chin granite — See ref. 18 for map bt + ms 21 21, 19 — AMC-83‐2 Ak-Chin granite — See ref. 18 for map bt + ms 21 21, 19 — *All samples also include quartz and two feldspars. bt, biotite; ms, muscovite; grt, garnet; tur, tourmaline; sil, sillimanite. †References: 1, this study; 2, Davis (1990); Larbi et al. (1999) (and references therein); 3, Southwick (1991); 4, Boerboom & Zartman (1993); 5, Feng & Kerrich (1992a); 6; Feng (1992); 7, Feng & Kerrich (1991); 8, Mulja (1995); 9, Mulja et al. (1995b); 10, Mulja et al. (1995a); 11, Frost et al. (2006); 12, Zartman & Reed (1998); 13, Nabelek et al. (1992a); 14, Redden et al. (1990); 15, Lalonde (1986); 16, Lalonde & Bernard (1993); 17, Hoffman & Bowring (1984); 18, Anderson & Thomas (1985); 19, L. Anderson (personal communication, 2016); 20, Peterman et al. (1968); 21, Anderson & Bender (1989). Fig. 1. View largeDownload slide Map of major Archean to Mesoproterozoic basement features of central North America after Whitmeyer & Karlstrom (2007). Significant terranes, orogenic belts, and structural features are shown in different colors. Fig. 1. View largeDownload slide Map of major Archean to Mesoproterozoic basement features of central North America after Whitmeyer & Karlstrom (2007). Significant terranes, orogenic belts, and structural features are shown in different colors. Late Archean localities Most Archean localities examined in this study are from the Superior Province (Fig. 1). This comprises primarily two kinds of tectono-stratigraphic terranes: metasedimentary rock-dominated and greenstone–plutonic subprovinces that were juxtaposed during late Archean orogenic accretion (Supplementary Data Fig. A1; Card & Ciesielski, 1986; Card, 1990; Stott et al., 2010). The metasedimentary subprovinces are dominated by variably metamorphosed volcanogenic turbidites, but also contain pelagic and chemical sediments such as carbonates, chert and mudstones, and they have been interpreted as accretionary prisms (Breaks et al., 1978; van de Kamp & Beakhouse, 1979; Devaney & Williams, 1989; Percival & Williams, 1989; Card, 1990). Many of these metasedimentary rock-dominated provinces (and some of the greenstone–plutonic provinces) contain a suite of c. 2646–2691 Ma SPGs that have been interpreted as resulting from the burial and melting of the sediments during continental collision (Percival & Williams, 1989; Southwick, 1991; Breaks & Moore, 1992; Feng & Kerrich, 1992a, 1992b; Corfu et al., 1995; Larbi et al., 1999). We selected for study SPGs from the boundary between the Winnipeg River and Wabigoon subprovinces (the Ghost Lake batholith; n = 7, where n is the number of sample studied), the Quetico subprovince (Sturgeon Lake granite; n = 3), the Wawa subprovince (Shannon Lake granite; n = 2), and the Pontiac subprovince/Lacorne block (Preissac, Lamotte, Lacorne, and Moly Hill plutons; n = 15). In addition to Superior Province samples, we studied one sample from the Wyoming craton [the Mt. Owen batholith; see Frost et al. (2006, 2018) or Supplementary Data Appendix for regional geology]. We analyzed the major element chemistry of minerals from all Archean samples, except those from the Pontiac subprovince/Lacorne block, for which we used previously published data (Feng & Kerrich, 1992a; Mulja et al., 1995a). Whole-rock major and trace element data were either analyzed in this study or taken from previously published data (see Table 2 for details). Proterozoic localities Two types of Proterozoic SPG localities were examined. The first are from orogenic circum-cratonal belts of North America, including the Wopmay and Trans-Hudson orogens, and the second are ferroan granites (see definition of Frost & Frost, 2011) from the southwestern USA (Fig. 1). The Wopmay orogen is a Paleoproterozoic (1·9 Ga) collisional belt located on the western edge of the Slave craton (Hoffman & Bowring, 1984). We used previously published whole-rock and mineral data for eight samples from the syn- to post-tectonic Hepburn intrusive suite, which marked the final collision on the western flanks of the orogeny at c. 1885 Ma (Lalonde, 1989; Lalonde & Bernard, 1993). The Trans-Hudson orogen formed during the assembly of the Hearne, Wyoming, and Superior cratons into the core of Laurentia following the closure of the Manikewan Ocean (Hoffman, 1988; Lewry & Stauffer, 1990; Weller & St Onge, 2017). Although primarily located in Canada, the Trans-Hudson orogen extends into the northern USA, where the final suturing is marked by the Harney Peak granite (∼1715 Ma) and other satellite intrusions in the Black Hills of South Dakota (Redden et al., 1990; Nabelek et al., 1999). We examined previously published mineral and whole-rock data for the Harney Peak granite for nine samples (Nabelek et al., 1992a), and we re-analyzed four of those samples (provided by P. Nabelek) for major element mineral chemistry. It is important to note that a mixture of Proterozoic and Archean metasedimentary rocks have been suggested to be the source material for the samples from the Harney Peak granite considered here, based on Nd isotope data and calculated model ages (Walker et al., 1986; Nabelek et al., 1992a). However, partial melts of metasedimentary rocks may not be in Nd isotope equilibrium with their source materials owing to incomplete dissolution of apatite or monazite during anataxis (Zeng et al., 2005). Therefore, it is unclear whether the Archean model ages suggested by the Nd isotope data are actually indicative of the age of the source material. The SPGs of the southwestern USA that we studied are the Silver Plume and St Vrain granites of Colorado (n = 9) and the Ak-Chin, Ruin, Oracle, and Sierra Estrella granites of Arizona (n = 5). These granites are thought to have been derived from a quartzofeldspathic source region in the deep crust and have oxygen isotope compositions consistent with having been derived from metasedimentary material (Anderson & Thomas, 1985; Anderson & Bender, 1989; Anderson & Morrison, 2005). SAMPLE PETROGRAPHY Detailed petrographic descriptions and photomicrographs for all samples can be found in the Supplementary Data Appendix (Figs A3–A17). All samples studied petrographically in this study (n = 17) are sub-solvus granites with varying modal abundances of quartz, albitic plagioclase, K-feldspar, biotite, and muscovite. Garnet is present in five of the more evolved leucogranites from this subset of 17 samples (Table 2). Two of these leucogranites also contain tourmaline. Accessory phases include zircon, apatite, monazite, ilmenite, rutile, and pyrite. Primary muscovite is identified by textural relationships indicating early crystallization contemporaneous with biotite: the presence of euhedral to subhedral flakes and laths; the absence of ragged growth patterns either within other phases or on the rims of other phases; a relatively coarse grain size (0·5–1 mm, often occurring as euhedral books); and mineral chemistry (Miller et al., 1981; see subsequent discussion). In some samples, biotite is variably altered to chlorite along the rims and along internal cleavage planes. These grains were excluded from analysis. Samples lack foliation (either magmatic or subsolidus) and display no evidence for deformation. METHODS Whole-rock major and trace element analysis Whole-rock major and trace element concentrations were determined for 12 samples by X-ray fluorescence (XRF) using a ThermoARL Advant’XP+ automated sequential wavelength spectrometer at the GeoAnalytical Laboratory at Washington State University. Major elements (Si, Ti, Al, Fe, Mn, Mg, Ca, Na K, and P) are reported as oxide wt %. Total Fe is reported as FeO, designated as FeOT. Trace elements analyzed include Sc, V, Ni, Cr, Ba, Rb, Sr, Zr, Y, Nb, Cu, Zn, Ga, Pb, La, Ce, Th, Nd, and U, and their concentrations are reported as ppm by weight. Details of the sample preparation, analytical methods, and precision and accuracy have been given by Johnson et al. (1999). Whole-rock data are presented in Supplementary Data Table A1. Electron microprobe mineral analysis Although the focus of this study is on using the FeT/(FeT + Mg) ratios of biotite to constrain the fO2 values of SPGs, we analyzed all Fe–Mg-bearing silicates in the samples to provide context for the biotite analyses. Consequently, major and minor elements were measured in biotite, white mica, and garnet for 17 samples using the JEOL JXA-8200 electron microprobe at the California Institute of Technology. For each sample, we analyzed multiple grains of each mineral (generally >10) and multiple points on each grain to explore both intra-sample and intra-grain heterogeneity. An acceleration voltage of 15 kV, a beam current of 25 nA, and a defocused beam (10 µm diameter) were used. Counting times were 40 s for Fe, Mn, Ti, Cr, F, and Cl, and 20 s for Si, Al, Mg, Ca, Na, and K. Background counting times were half of the peak counting times. The following standards were used to calibrate elemental peaks prior to each analytical session: synthetic anorthite (SiO2, Al2O3, CaO), synthetic TiO2 (TiO2), synthetic Cr2O3 (Cr2O3), synthetic fayalite (FeO), synthetic tephroite (MnO), synthetic F-phlogopite (MgO, F), Amelia albite (Na2O), Asbestos microcline (K2O), and synthetic sodalite (Cl). Data were reduced using a modified ZAF procedure (CITIZAF, Armstrong, 1995). The detection limits for these conditions and setup were <0·01 wt % for SiO2, Al2O3, MgO, CaO, K2O, and Cl; 0·01 wt % for TiO2, FeO, Na2O, Cr2O3, and MnO; and 0·02 wt % for F. Typical 1σ uncertainties (per cent of absolute value) on biotite analyses calculated from counting statistics were as follows: Si, 0·15–0·25%; Ti, 1%; Al, 0·25–0·4%; Fe, 0·3%; Mg, 0·5–0·9%; K, 0·35%; F, 2–3%; Cl, 2%. Biotite, white mica, and garnet major element compositions are given in Supplementary Data Tables A2–A4. Data sources for each composition, whether analyzed in this study or compiled from the literature, are given in the tables. Where possible, averages of all analyses from a single sample with 2σSD are reported; for some samples, for which data were compiled from the literature, only a single ‘representative’ analysis is reported. RESULTS Whole-rock chemistry All samples are high-silica (67–76 wt % SiO2) peraluminous granodiorites to granites and are strongly peraluminous (ASI = 1·1–1·5), except for one sample from the Shannon Lake granite (SP-16–1) with an ASI of 1·04 (Fig. 2a). The FeO + MgO + TiO2 (wt %) contents in all samples are <7 wt % (Fig. 2b). For both the Archean and Proterozoic SPGs, Al2O3 concentrations are 13·3–15·8 wt % and TiO2 concentrations are 0·01–0·36 wt %. The Archean SPGs on average have higher Na2O/K2O wt % ratios [0·87 ± 0·15 (1σ)] relative to Paleoproterozoic SPGs [0·54 ± 0·11 (1σ)]. This reflects both higher Na2O contents and lower K2O contents in the Archean SPGs as compared with the Paleoproterozoic SPGs. In addition, although there is considerable variability, the P2O5 contents of the Archean SPGs [average 0·06 ± 0·04 (1σ) wt %] are on average lower than those of Proterozoic granites [average 0·14 ± 0·10 (1σ) wt %]. Molar FeT/(FeT + Mg) ratios for all samples are 0·61–0·87, with no significant differences between the range of values observed for Archean or Proterozoic samples (Fig. 2c and d). Fig. 2. View largeDownload slide Whole-rock data: (a) FeO + TiO2 + MgO vs SiO2; (b) ASI [molar Al/(Ca – 1·67P + Na + K)] vs SiO2; (c, d) histograms of molar FeT/(FeT + Mg) of Archean (c) and Proterozoic (d) samples. Fig. 2. View largeDownload slide Whole-rock data: (a) FeO + TiO2 + MgO vs SiO2; (b) ASI [molar Al/(Ca – 1·67P + Na + K)] vs SiO2; (c, d) histograms of molar FeT/(FeT + Mg) of Archean (c) and Proterozoic (d) samples. Mineral chemistry Consistent with the peraluminous nature of their host granites, biotites from the Archean and Proterozoic SPGs are more aluminous than typical biotites from calc-alkaline, metaluminous and alkaline ferroan igneous suites (which contain <16 wt % Al2O3; Abdel-Rahmen, 1994), with 16·2–20·1 wt % Al2O3 corresponding to total Al from 2·9 to 3·8 atoms per formula unit (a.p.f.u., calculated on a 24 oxygen basis). In a ternary diagram showing the Al–Fe–Mg contents of the biotites, the Al-rich nature of the biotites of the SPGs in this study is clear; they predominantly plot in more Al-rich regions than biotites from alkaline and metaluminous igneous suites, consistent with biotites from Phanerozoic peraluminous granites (Fig. 3a; Abdel-Rahmen, 1994; Bell et al., 2017). The FeT/(FeT + Mg) ratios of biotites span wide ranges for both Archean (0·59–0·81) and Proterozoic (0·52–0·86) granites, and although the distributions overlap, the two populations are distinguishable (Fig. 3b and c); Archean samples on average have biotite with higher FeT/(FeT + Mg) ratios [0·72 ± 0·14 vs 0·64 ± 0·15 (2σ)]. It should be noted that, as accurate determination of ferric iron concentrations in biotite is not feasible using electron microprobe analyses (Dyar et al., 1991; Feldstein et al., 1996), we assume that all Fe is ferrous when calculating ratios of FeT/(FeT + Mg) on the basis of stoichiometry. We do the same for muscovite and garnet compositions. Fig. 3. View largeDownload slide Average biotite compositions for Archean and Proterozoic SPG samples. (a) Al–Fe–Mg ternary (truncated). Symbols as in Fig. 2. Grey error bars are 2σSD on averages. Fields for calc-alkaline suites, alkaline suites, and peraluminous suites are from the previous compilation of biotite compositions presented by Abdel-Rahman (1994). Calc-alkaline suite data are from granitoids formed within a subduction-related environment, alkaline suites are from extensional tectonic environments, and peraluminous suites are SPGs. [See Abdel-Rahman (1994) for details of data sources.] (b, c) Histograms of FeT/(FeT + Mg) for (b) Archean samples and (c) Proterozoic samples. Fig. 3. View largeDownload slide Average biotite compositions for Archean and Proterozoic SPG samples. (a) Al–Fe–Mg ternary (truncated). Symbols as in Fig. 2. Grey error bars are 2σSD on averages. Fields for calc-alkaline suites, alkaline suites, and peraluminous suites are from the previous compilation of biotite compositions presented by Abdel-Rahman (1994). Calc-alkaline suite data are from granitoids formed within a subduction-related environment, alkaline suites are from extensional tectonic environments, and peraluminous suites are SPGs. [See Abdel-Rahman (1994) for details of data sources.] (b, c) Histograms of FeT/(FeT + Mg) for (b) Archean samples and (c) Proterozoic samples. As the focus of this study is determining the fO2 values of granites, and the Fe3+/FeT ratios of biotites are sensitive to oxygen fugacity (see discussion below), direct determination of Fe3+/FeT ratios in biotite would be ideal. For example, biotites from Phanerozoic and Proterozoic SPG granites analyzed by wet chemical titration techniques in previous studies yielded Fe3+/FeT ratios of 0·04–0·10 (de Albuquerque, 1973; Lalonde & Bernard, 1993), generally lower than biotites from I-type granitoids, which have Fe3+/FeT ratios of 0·10–0·20 (Dodge et al., 1969; Czamanske et al., 1981; Lalonde & Bernard, 1993). However, biotite Fe3+/FeT ratios in the SPGs cannot be reliably measured with bulk techniques that would require tens of milligrams of clean, pure biotite (e.g. wet chemistry or conventional Mössbauer spectroscopy) owing to the presence of chlorite alteration of the rims of biotite in some of the samples of this study. In particular, chlorite can host significant quantities of Fe3+ (e.g. up to 45 wt % Fe2O3 in ferric-chamosite; Brindley & Youell, 1953). However, as indicated by analyses of biotite from other (including Proterozoic) SPGs referenced above, Fe3+ may account for 4–10% of the total Fe in biotites measured in this study. In addition to muscovite (KAl3Si3O10[OH]2), white mica in our samples contains both paragonite (NaAl3Si3O10[OH]2) and celadonite (K[Mg, Fe]AlSi4O10[OH]2) end-members, which is typical of plutonic muscovites (Miller et al., 1981). Thus, we refer to this mineral as ‘white mica’ even though muscovite is commonly used in the literature. Na contents for white micas from the Archean and Proterozoic SPGs are 0·01–0·13 a.p.f.u. (calculated on a 24 oxygen basis) and Fe and Mg contents are 0·11–0·80 a.p.f.u. and 0·04–0·39 a.p.f.u., respectively. The range of AlT (i.e. summed over both tetrahedral and octahedral sites) is 3·85–5·65 a.p.f.u. and of Si is 6·11–6·48 a.p.f.u. FeT/(FeT + Mg) ratios for white micas from Archean SPGs are 0·51–0·87 [average 0·66 ± 0·15 (2σ)], and those from Proterozoic SPG are 0·52–0·74 [average 0·63 ± 0·16 (2σ)] (Fig. A18). The majority of samples considered in this study are not garnet-bearing, but garnets have either been reported in the literature to be present, or have been observed by us in 23 of the 59 samples studied (Table 2). Of these 23 samples, we had access to four (three Archean samples and one Paleoproterozoic sample), and we analyzed garnets from these four samples. In addition, two garnet analyses were previously reported from the Lacorne block GMG (‘garnet–muscovite granite’) series (Feng, 1992). All garnets from samples that we analyzed have homogeneous cores, and ∼20% of these have zoned rims ≤100 µm in width. We report averages for the homogeneous cores in Supplementary Data Table A4. In these cores, mole fractions of end-member garnet components are 0·01–0·11 pyrope, 0·68–0·84 almandine, 0·06–0·47 spessartine, and 0·01–0·04 grossular across all the samples, but are roughly constant for each sample (i.e. varying by less than 1 mol %). The FeT/(FeT + Mg) ratios in garnet cores are uniformly high (0·88–0·98; Supplementary Data Fig. A21). Zoned rims are characterized by increasing MnO and FeT/(FeT + Mg) ratios and decreasing MgO, FeO, and CaO toward the rim. DISCUSSION Preservation of magmatic mineral compositions The composition of magmatic biotite in granites is susceptible to modification owing to subsolidus processes such as metamorphism, deuteric alteration, and oxidation (Gilkes et al., 1972; Gilkes & Suddhiprakarn, 1979; Veblen & Ferry, 1983; Eggleton & Banfield, 1985; Jeong & Kim, 2003). In particular, the alteration of biotite to chlorite occurs through the replacement of the tetrahedral–octahedral–tetrahedral layer in biotite with brucite layers (Ferry, 1979; Veblen & Ferry, 1983). Previous studies have suggested that the chemical consequences of this sheet-silicate transformation from biotite to chlorite are a decrease in Si, Ti, and K contents, with little effect on the relative proportions of Mg, Fe, or Al (Veblen & Ferry, 1983). Although partial alteration to chlorite is thus not likely to affect the FeT/Mg ratios of the biotite, we took care during our study to avoid chlorite alteration and consider only analyses of ‘pristine’ biotite. First, we screened samples petrographically so as to exclude heavily chloritized regions from analysis, and, second, we filtered the microprobe analyses so as to exclude analyses affected by partial chloritization [as suggested by low SiO2 (<33 wt %), K2O (<8 wt %), and totals (<94 wt %)]. The percentage of analyses per sample meeting these criteria for partial chloritization and exclusion from consideration ranged from 0 to 30%. Identifying whether white mica is primary or secondary in plutonic rocks has been the focus of much research (Anderson & Rowley, 1981; Miller et al., 1981; Speer, 1984; Zen, 1988). Miller et al. (1981) provided textural and chemical criteria to identify primary magmatic white mica including the following: coarse grain size comparable with that of other magmatic phases; subhedral to euhedral crystal forms; crystals neither enclosed by nor irregularly enclosing another mineral; absence of evidence for alteration in other phases; the overall preservation of primary igneous textures. Although textural evidence is rarely definitive, Miller et al. (1981) analyzed the compositions of white micas considered either primary (i.e. magmatic) or secondary using these textural criteria and found chemical differences between the two groups, with primary white mica being consistently richer in Ti, Na, and Al and poorer in Mg and Si than secondary white mica. In addition to analyzing only white mica that appeared to be texturally primary based on Miller et al.’s (1981) criteria, we screened white mica compositions in the samples we analyzed to exclude those that differed from most analyses in a particular sample along the secondary trends identified by Miller et al. (1981). This resulted in the exclusion of analyses with AlT < 5·2 a.p.f.u., Mg > 0·25 a.p.f.u., and Si > 6·2 a.p.f.u. for samples that we directly analyzed (0–30% of analyses of white mica from a given sample). Garnets in SPGs may be a primary magmatic phase, entrained restitic material, a product of peritectic melting and/or crystallization reactions, or a xenocryst from the country rocks (Miller & Stoddard, 1978; Clarke, 1981; Stevens et al., 2007; Villaros et al., 2009; Taylor & Stevens, 2010). It is beyond the focus and scope of this work to differentiate between these possibilities, but because garnet could be a major reservoir for Fe in samples where it is present, understanding the systematics of FeT/(FeT + Mg) in garnet is important to our interpretation of biotite chemistry for such samples. As mentioned in the previous section, ∼20% of analyzed garnets have zoned rims (generally <100 µm), characterized by an increase towards the rim in Mn and Fe/(FeT + Mg) ratio and a decrease in Ca content. Although this feature is observed in metamorphic rocks and is typically taken to indicate resorption of garnet rims during subsolidus cooling (Kohn & Spear, 2000; Kohn, 2003), garnet resorption is not predicted from available data on the phase equilibria of peraluminous granites (Clemens & Wall, 1988; Clemens & Birch, 2012; Scaillet et al., 2016). Rather the zoning probably reflects evolution of the residual liquid in the partially molten granite as it crystallized (i.e. progressive Mn and FeT/Mg enrichment; Du Bray, 1988) and sufficiently low temperatures during crystallization of the rims to limit homogenization via diffusion (e.g. Dahlquist et al., 2007). However, as a precaution against including analyses affected by subsolidus exchange of Fe and Mg between garnet and biotite, we excluded from our reported biotite averages analyses of biotites directly in contact with garnet; these excluded biotites do indeed tend to have higher FeT/(FeT + Mg) ratios (by 0·01–0·02) than biotite not in contact with garnet. Importantly, we emphasize that the majority (∼61%) of the samples do not contain garnet, and for these samples, Fe–Mg exchange with garnet could not have affected their biotite Fe–Mg contents. The dependence of biotite FeT/(FeT + Mg) on fO2 ⁠, T, FeT/(FeT + Mg) of melt, and aH2O Constraints from compilation of experimental crystallization data The chemistry of mafic silicate minerals, in particular that of biotite, has been used extensively to evaluate intensive parameters during crystallization of evolved silicic magmas (e.g. Wones & Eugster, 1965; Czamanske & Wones, 1973; Czamanske et al., 1981; Ague & Brimhall, 1988a; Anderson & Bender, 1989; Lalonde & Bernard, 1993; Shabani et al., 2003). For example, at a particular fH2O ⁠, the following equilibria can be used to determine fO2 ⁠: KFe32+AlSi3O10(OH)2+O2→KAlSi3O8+Fe3O4+H2O (1) KFe32+AlSi3O10(OH)2+0·75O2→KAlSi3O8+1·5Fe2O3+H2O. (2) Increasing fO2 will shift these equilibria toward the right, resulting in more Mg-rich biotites, whereas a decrease in fO2 will stabilize more Fe2+-rich biotites (see further discussion below on this aspect). Such reactions can be used to calculate fO2 provided that the following details are known: (1) the compositions of coexisting biotite, K-feldspar, and Fe–Ti oxides (either magnetite or ilmenite); (2) activity–composition models for the aforementioned phases; (3) temperature (T), pressure (P), and fH2O ⁠; (4) the relationship between T, P, and the equilibrium constant of the reaction of interest. For some samples included in this study, but compiled from the literature, Fe–Ti oxides were described as crystallizing phases, analyzed, and previously used to obtain quantitative estimates of fO2 (e.g. the Silver Plume and St Vrain granites; Anderson & Thomas, 1985). However, for the granites collected and analyzed in this study, we were limited by the apparent absence of Fe–Ti oxides, owing to either their not being a crystallizing phase (which is plausible given the uniformly low TiO2 contents of the granites, <0·3 wt %) or their crystallization in such low modal abundances that we were unable to find them in thin section. For such granites, an equally valid redox equilibrium reaction can be written, analogous to the above equations, where the Fe3O4 or Fe2O3 component in the melt appears in place of a crystalline oxide phase: KFe3AlSi3O10(OH)2+0·75O2→KAlSi3O8+1·5Fe2O3melt+H2O. (3) We explored using equation (3) to constrain fO2 in our SPG samples by examining and thermodynamically modeling data from crystallization experiments on high-silica melts, where fO2 was controlled and biotite was a crystallizing phase. Fe–Ti oxides were present in some, but not all of the experimental runs, thus also allowing us to explore Fe partitioning between biotite and melt as a function of fO2 ⁠, both with and without the presence of co-crystallizing Fe–Ti oxides. The compiled experimental studies can be classified into three groups based on buffered fO2 ⁠. The first group comprises experiments conducted at relatively high fO2 (NNO +1·6 to NNO +3·5 (i.e. log units relative to the Ni–NiO buffer) (Dall’Agnol et al., 1999; Scaillet & Evans, 1999; Costa et al., 2004); the second group of experiments was conducted at intermediate fO2 values (NNO –0·1 to NNO +1·0) (Icenhower & London, 1995; Bogaerts et al., 2006; Mutch et al., 2016); and the third group of experiments was conducted at relatively reducing conditions (NNO –1·8 to NNO –1·3) (Dall’Agnol et al., 1999; Scaillet & Macdonald, 2001, 2003; Scaillet et al., 1995) (Fig. 4). Starting compositions included a natural tonalite, dacites, peraluminous leucogranites, and metaluminous and peralkaline granites. The experiments of Icenhower & London (1995) were different in that they partially melted synthetic metapelite compositions; however, they performed reversal experiments in which biotite was a primary crystallization product from a granitic melt. Data from these reversal (crystallization) experiments were included in our compilation of experimental results. Fig. 4. View largeDownload slide Data from crystallization experiments on hydrous high-SiO2 melts with biotite as a crystallizing phase. (a) Experimental biotite FeT/(FeT + Mg) vs experimental temperature (°C). Labels next to the data points give crystallizing phase assemblages. (b) FeT/Mg partition coefficient between biotite and melt vs oxygen fugacity of experiment (ΔNNO). It should be noted that FeT/Mg partition coefficients for every experiment could not be calculated owing to lack of melt analyses for some studies. (c) FeT/(FeT + Mg) in biotite vs bulk composition. It should be noted that data could not be shown for Icenhower & London (1995) as bulk experimental compositions were not given. Fig. 4. View largeDownload slide Data from crystallization experiments on hydrous high-SiO2 melts with biotite as a crystallizing phase. (a) Experimental biotite FeT/(FeT + Mg) vs experimental temperature (°C). Labels next to the data points give crystallizing phase assemblages. (b) FeT/Mg partition coefficient between biotite and melt vs oxygen fugacity of experiment (ΔNNO). It should be noted that FeT/Mg partition coefficients for every experiment could not be calculated owing to lack of melt analyses for some studies. (c) FeT/(FeT + Mg) in biotite vs bulk composition. It should be noted that data could not be shown for Icenhower & London (1995) as bulk experimental compositions were not given. Temperatures and pressures for the compiled experiments are 650–875°C and 0·15–0·41 GPa, and water activities (when given) are high (0·8–1·0, where a value of unity indicates a system saturated with a pure-H2O fluid). In the high- fO2 experiments on dacites (Scaillet & Evans, 1999; Costa et al., 2004) and a metaluminous monzogranite (Dall’Agnol et al., 1999), phase assemblages included melt + biotite + plagioclase + magnetite ± ilmenite ± clinopyroxene ± amphibole ± K-feldspar ± anhydrite ± titanite (Fig. 4a). In the intermediate- fO2 experiments of Icenhower & London (1995) the phase assemblages were melt + biotite + muscovite + plagioclase, whereas the phase assemblages in experiments on the granodiorite and tonalite bulk compositions of Bogaerts et al. (2006) and Mutch et al. (2016) were melt + biotite + amphibole + magnetite + ilmenite + apatite ± clinopyroxene ± orthopyroxene ± quartz ± plagioclase ± K-feldspar (Fig. 4a). In the low- fO2 experiments of Scaillet & MacDonald (2001) on a peralkaline rhyolite, phase assemblages included melt + biotite + ilmenite + quartz ± amphibole ± fayalite. The low- fO2 experiments of Scaillet et al. (1995) on a biotite + muscovite leucogranite included melt + biotite ± K-feldspar ± plagioclase phase assemblages, and their experiments on a muscovite + tourmaline leucogranite included melt + biotite + tourmaline + plagioclase ± muscovite ± K-feldspar ± quartz ± hercynite ± mullite assemblages. Finally, in the low- fO2 experiments of Dall’agnol et al. (1999) on a metaluminous granite, the phase assemblages included melt + biotite + ilmenite + hornblende + clinopyroxene ± orthopyroxene ± plagioclase ± K-feldspar + quartz (Fig. 4a). Although specific phase assemblages varied considerably among the experiments considered, an important point is that the high- and intermediate- fO2 experiments contained magnetite ± ilmenite as crystallizing phases, whereas the low- fO2 experiments contained only ilmenite or no Fe–Ti oxide phase at all. For experiments on a given bulk composition at a buffered fO2 ⁠, biotite FeT/(FeT + Mg) increases with decreasing T, as expected for ferromagnesian silicates during batch crystallization from a melt (Fig. 4a). The increase in FeT/(FeT + Mg) occurs regardless of co-crystallizing phases. As expected based on reactions (1)–(3), at a fixed T, increasing fO2 decreases the FeT/(FeT + Mg) ratio of crystallizing biotite owing to the increased stability of magnetite and the hematite-component in ilmenite or to a transfer of Fe3+ to the melt. For example, in all of the high- fO2 experiments, magnetite is a stable crystallizing phase occurring in abundances of only 1–4% by mass but containing 44–94% of the total Fe in the experiments. The strong partitioning of Fe into magnetite at high fO2 leaves little Fe available to be incorporated into biotite (which contains 1–13% of total Fe in the experiments) and thus results in relatively low FeT/(FeT + Mg) ratios (<0·4) in biotite relative to those crystallizing at the same temperature, but at lower fO2 (Fig. 4a). At the other extreme, in the low- fO2 experiments most of the Fe occurs as Fe2+ and therefore is available to be incorporated into biotite and the stability of Fe–Ti oxide decreases. In these low- fO2 experiments, magnetite is not a crystallizing phase and ilmenite, when it occurs, is ≤0·5% by mass of the experiments (Fig. 4a). Other Fe-bearing phases (tourmaline, fayalite, clinopyroxene) are also minor in abundance (≤4 wt %) and do not affect the Fe budget significantly. Importantly, in the experiments of Scaillet et al. (1995) on a biotite + muscovite leucogranite (similar to the composition of the SPGs in this study), which lack Fe–Ti oxides, FeT/(FeT + Mg) ratios in biotite (0·46–62) are elevated above those of the oxidized experiments (0·16–0·41) over the same T interval; this observation indicates that Fe enrichment in biotite at low fO2 will occur in the presence of melt whether or not an Fe–Ti oxide is co-crystallizing with biotite owing to reactions such as equation (3). The fO2-sensitive Fe partitioning behavior between biotite and melt can also be expressed by the bulk FeT/Mg distribution coefficient between biotite and melt (⁠ Kbt–meltFeT/Mg=(FeT/Mgbiotite)/(FeT/Mgmelt) ⁠) Kbt−meltFeT/Mg=FeT/MgbiotiteFeT/Mgmelt ⁠. In the high- fO2 experiments, Kbt–meltFeT/Mg is 0·14–0·36, in the intermediate- fO2 experiments it is 0·29–0·54, and in the low- fO2 experiments it is 0·39–0·71 (Fig. 4b). Similar decreases in the bulk FeT/Mg KD between olivine and melt with increasing fO2 have been described for basaltic systems and can be used to estimate fO2 in natural and experimental olivine + glass assemblages (e.g. Matzen et al., 2011). In the case of olivine, the effect is simply understood by the fact that although the Fe2+/Mg distribution coefficient, (Fe2+/Mgolivine)/(Fe2+/Mgmelt) Fe2+/MgolivineFe2+/Mgmelt ⁠, is roughly independent of T and composition, as fO2 increases, the bulk FeT/Mg KD, (Fe2+/Mgolivine)/[(Fe2++Fe3+)/Mgmelt] ⁠, decreases progressively as the increasing Fe3+ content of the melt results in an increase in the denominator relative to the Fe2+/Mg distribution coefficient. Thus, even in the absence of an Fe–Ti oxide phase at low fO2 ⁠, and even though biotite has measurable Fe3+ (which will also influence the numerator of the expression for the bulk FeT/Mg distribution coefficient between biotite and melt), biotite coexisting with melt is expected (and observed) to become progressively more magnesian with increasing fO2 at fixed T (Fig. 4b). It should be noted, however, that biotite FeT/(FeT + Mg) values will also depend on the bulk composition of the system, with higher bulk FeT/(FeT + Mg) corresponding to higher FeT/(FeT + Mg) ratios in biotite at a fixed T and fO2 (Fig. 4c). For example, in the subset of low- fO2 experiments conducted at ∼680°C and NNO –1·3 to NNO –1·6, FeT/(FeT + Mg) ratios in biotite vary between 0·59 and 0·97, with the lowest values corresponding to a bulk experimental FeT/(FeT + Mg) composition of 0·69, and the highest ones correspond to a bulk composition near unity. Similar trends are observed in the intermediate- and high- fO2 experiments (Fig. 4c). As aH2O was high in the compiled experiments (0·8–1·0), the effects of its variability on biotite FeT/(FeT + Mg) ratio cannot be assessed from this dataset. However, lower aH2O generally will result in more magnesian biotites [equations (1)–(3), Wones & Eugster, 1965), although this effect is of second order as compared with T and fO2 (Ague & Brimhall, 1988a). In addition, it has been shown that under reducing conditions, the presence of coexisting sulfides can cause ferromagnesian silicate minerals to have lower FeT/Mg ratios owing to the strong affinity between Fe and S (e.g. biotite: Tso et al., 1979; amphibole: Scaillet & Evans, 1999). Although the experiments of Costa et al. (2004) contained added S, the co-crystallizing S-bearing phase with biotite was primarily anhydrite (one experiment with co-crystallizing biotite also contained pyrrhotite). Therefore, the effects of co-crystallization of sulfide on biotite FeT/(FeT + Mg) cannot be assessed using our compiled experimental dataset. We do not discuss the effect of co-crystallization of sulfides on biotite compositions further, as sulfides either were not observed, were present in trace abundances (⁠ ≪1 modal %), or were not reported in significant abundance in the SPG samples considered here. Modeling experimental data with alphaMELTS Motivated by the evidence from the experiments suggesting that FeT/(FeT + Mg) ratios in biotite crystallizing from granitic melts are a function of fO2 ⁠, T, and bulk composition, we undertook thermodynamic modeling in an attempt to replicate the experimental biotite compositions, to explore a broader parameter space, and to develop a quantitative method to place constraints on the fO2 of crystallization of natural samples based on the FeT/Mg of biotite relative to that of the liquid from which it crystallizes. We used alphaMELTS (v.1.6; Smith & Asimow, 2005) to model batch crystallization of each compiled experiment using the bulk composition, T, P, and bulk water contents specific to each experiment. The Fe3+/FeT ratios in the melt as a function of fO2 were calculated in alphaMELTS using the parameterization of Kress & Carmichael (1991). Our choice to use alphaMELTS instead of rhyoliteMELTS (which is optimized for high-SiO2 systems) was motivated by a critical difference in the biotite solution models implemented by these two algorithms. The default biotite model included in alphaMELTS incorporates data for annite and an ideal mixing model taken from McMullin et al. (1991), whereas rhyoliteMELTS implements the biotite model of Sack & Ghiorso (1989) that overestimates the amount of the phlogopite end-member in biotite. Implementation of the revised biotite model used in alphaMELTS generates reasonable Fe abundances in biotite with the correct temperature dependence (see the alphaMELTS software manual for discussion). We took two approaches to modeling the experiments. First, we used the experimental fO2 to calculate equilibrium biotite FeT/(FeT + Mg) values using alphaMELTS and compared that with the experimental biotite FeT/(FeT + Mg) values. Figure 5a shows that calculated versus experimental biotite FeT/(FeT + Mg) values fall near a 1:1 line over a large range in FeT/(FeT + Mg) values and experimental conditions [with an average absolute deviation of 0·06 ± 0·05 (1σ)]. In fact, alphaMELTS reproduces the biotite FeT/(FeT + Mg) values of 65% of the experimental biotite compositions within 0·05. It is noteworthy that the variation in fO2 of the modeled experimental data spans about five orders of magnitude relative to NNO, with the relatively oxidized experiments at the lower left and the reduced experiments at the upper right of Fig. 5a. Second, we varied the model fO2 to reproduce the experimental biotite FeT/(FeT + Mg) ratio and compared the best-fit fO2 with that at which the experiments were conducted. Figure 5b demonstrates that 40% of experimental fO2 values were reproduced to within 0·5 log units and 70% within 1 log unit. The absolute standard deviation for all experiments from the 1:1 line is 0·82 ± 0·57 (1σ). Notably, the reproducibility of fO2 is not as good at low fO2 values where it appears that this method loses sensitivity to this modeling approach. At fO2 values between NNO –1 and NNO –2 nearly all of the Fe in the system is present as Fe2+ and the sensitivity of partitioning of FeT between biotite and the melt, as well as biotite FeT/(FeT + Mg) ratios, to fO2 is small. Thus, small uncertainties in biotite or melt composition inherent in our analyses can translate into large (and not meaningful) calculated variations in fO2 when the fO2 is low. However, the excellent agreement between model results and observed biotite chemistry shown in Fig. 5a and the capacity demonstrated in Fig. 5b to distinguish variations in fO2 to ∼0·5–1 log units over the five order of magnitude fO2 range of the experiments shown in Fig. 5b suggest that thermodynamic modeling can be used at this precision to quantify the fO2 of natural samples such as the SPGs in this study, where the compositions of the bulk-rock and biotite are known along with magmatic temperatures and water contents, provided that the fO2 is higher than NNO –1 to NNO –2. Fig. 5. View largeDownload slide alphaMELTS crystallization modeling results comparing calculated biotite compositions and fO2 with experimental data. alphaMELTS calculations were undertaken using experimental conditions (T and P) and starting compositions (including water contents, when given). Calculated vs experimental (a) biotite FeT/(FeT + Mg) and (b) fO2 ⁠. Fig. 5. View largeDownload slide alphaMELTS crystallization modeling results comparing calculated biotite compositions and fO2 with experimental data. alphaMELTS calculations were undertaken using experimental conditions (T and P) and starting compositions (including water contents, when given). Calculated vs experimental (a) biotite FeT/(FeT + Mg) and (b) fO2 ⁠. To explore this approach, we modeled more fully the peraluminous leucogranite starting composition DK89 [SiO2 = 73·04 wt %, FeT/(FeT + Mg) = 0·695] of Scaillet et al. (1995) (Fig. 6) to explore the sensitivity of biotite composition to variations in magmatic T and fO2 ⁠, which are the most significant ‘unknown’ variables in our natural samples. First, models were run at 0·4 GPa, over a temperature interval of 950–660°C, at fO2 values from NNO –3 to NNO +2, and with a bulk H2O content of 3 wt % (corresponding to aH2O = 0·6–1 over the crystallization interval). The model phase assemblages are melt + biotite + plagioclase + K-feldspar + quartz + ilmenite ± magnetite. At all fO2 values, biotite is calculated to begin crystallizing at 820–800°C. Magnetite is present in models run at fO2 > NNO and <800–825°C. Ilmenite crystallization begins at ∼780°C in the absence of magnetite, but crystallizes at 740–755°C in the higher fO2 model runs where magnetite is present. FeT/(FeT + Mg) ratios in biotite are sensitive to both T and fO2 ⁠. Over a crystallization interval of 820–660°C, biotite FeT/(FeT + Mg) ratios increase by 0·15–0·20 at all fO2 values, comparable with the increase observed in Scaillet et al.’s experiments (grey circles in Fig. 6a). Sensitivity to fO2 is strongest at higher fO2 where magnetite is a co-crystallizing phase. For example, a shift in fO2 from NNO +2 to NNO +1 results in an increase in biotite FeT/(FeT + Mg) of ∼0·15 at a fixed T (e.g. at 720°C), whereas an isothermal decrease from NNO to NNO –1 produces an increase in biotite FeT/(FeT + Mg) of only ∼0·02. At fO2 values lower than NNO –2, biotite FeT/(FeT + Mg) ratios are not strongly sensitive to variations in fO2 ⁠, approaching a maximum value that does not vary at a fixed T (Fig. 6a). As described above, this is because at these low fO2 values most of the Fe in the system occurs as Fe2+, so changes in fO2 change the Fe3+/Fe2+ ratio by only a small amount and Kbt–meltFeT/Mg essentially does not vary. Correspondingly (and as anticipated in the previous section), Kbt–bulkFeT/Mg=(FeTMg)bt/(FeTMg)bulk [where (FeT/Mg)bulk is the FeT/Mg of the bulk composition] is also sensitive to T and fO2 ⁠: Kbt–bulkFeT/Mg is lowest at elevated fO2 (e.g. 0·12 at 800°C increasing to 0·24 at 650°C at NNO +3, compared with 0·19 increasing to 0·48 at NNO +2 over the same decrease in T; Fig. 6b). Kbt–bulkFeT/Mg displays less pronounced increases with decreasing T at a higher fO2 than at a lower fO2 (e.g. a total increase of 0·12 from 800 to 650°C at NNO +3, compared with a total increase of 0·41 at NNO –1 over the same temperature interval) (Fig. 6b). Fig. 6. View largeDownload slide alphaMELTS crystallization modeling results for bulk composition DK89 (biotite + muscovite leucogranite) from Scaillet et al. (1995) with bulk FeT/(FeT + Mg) = 69·5. Initial water contents were set at 3 wt % H2O. Calculations were conducted at fixed fO2 values ranging from NNO –3 to NNO +2 (colored lines). In addition, Fe–Ti oxides were allowed to crystallize (continuous lines) and artificially suppressed (dashed lines) to explore the effect of oxygen fugacity variations on biotite compositions both with and without co-crystallizing Fe–Ti oxides. Plagioclase, K-feldspar, and quartz crystallized during the model runs. The following are plotted vs temperature: (a) biotite FeT/(FeT + Mg); (b) FeT/Mg ratios in biotite vs the bulk composition; (c) % total Fe in the system in biotite. In (a) and (b) grey circles show experimental data from DK89, which were obtained between NNO –1·3 and NNO –1·8 (indicated by italic numbers next to symbols). It should be noted that experimental data could not be shown in (c) because phase proportions were not reported for the experiments. Fig. 6. View largeDownload slide alphaMELTS crystallization modeling results for bulk composition DK89 (biotite + muscovite leucogranite) from Scaillet et al. (1995) with bulk FeT/(FeT + Mg) = 69·5. Initial water contents were set at 3 wt % H2O. Calculations were conducted at fixed fO2 values ranging from NNO –3 to NNO +2 (colored lines). In addition, Fe–Ti oxides were allowed to crystallize (continuous lines) and artificially suppressed (dashed lines) to explore the effect of oxygen fugacity variations on biotite compositions both with and without co-crystallizing Fe–Ti oxides. Plagioclase, K-feldspar, and quartz crystallized during the model runs. The following are plotted vs temperature: (a) biotite FeT/(FeT + Mg); (b) FeT/Mg ratios in biotite vs the bulk composition; (c) % total Fe in the system in biotite. In (a) and (b) grey circles show experimental data from DK89, which were obtained between NNO –1·3 and NNO –1·8 (indicated by italic numbers next to symbols). It should be noted that experimental data could not be shown in (c) because phase proportions were not reported for the experiments. We also ran crystallization models using the bulk composition DK89 from Scaillet et al. (1995) at the same conditions as described above, but in which Fe–Ti oxides were suppressed as crystallizing phases; that is, we calculated metastable equilibria in which all other parameters in the modeling were kept the same as described above, but in which the Fe–Ti oxides were not allowed to stabilize even though doing so would result in a lower Gibbs free energy; this allows us to explore the role that oxides play in Fe partitioning between biotite and melt as a function of fO2 and T [see, for example, equation (3)]. Model results (shown as dashed curves in Fig. 6) demonstrate similar trends to those where Fe–Ti oxides were allowed to stabilize (continuous curves in Fig. 6), but biotite FeT/(FeT + Mg), Kbt–bulkFeT/Mg ratios, and percentage of total Fe in biotite are shifted to higher values at a given T for most of the crystallization interval for the oxide-suppressed calculations (compare the dashed and continuous curves in Fig. 6). This is because more Fe in the bulk system is partitioned into biotite in the absence of Fe–Ti oxides. Nevertheless, biotite FeT/(FeT + Mg) ratios are still systematically lower at higher fO2 relative to low- fO2 model calculations when Fe–Ti oxides are suppressed owing to (1) the biotite solution model in alphaMELTS incorporating only Fe2+ and (2) near identical values of (Fe2+Mg)bt/(Fe2+Mg)melt (e.g. 0·22–0·25 at 800°C), over a large range in fO2 ⁠, but an increasing proportion of the total Fe in the system occurring as Fe3+ (as described above for olivine–melt equilibria). These model results demonstrate two important things. First, alphaMELTS can faithfully reproduce biotite compositions and fO2 from experimental datasets on systems that contain high-silica liquids at the conditions of the experiments. Second, biotite FeT/(FeT + Mg) ratios and Kbt–bulkFeT/Mg are sensitive to fO2 ⁠, both with and without co-crystallizing Fe–Ti oxide phases, as expected from equations (1)–(3). Modeling volcanic rock data with alphaMELTS modeling As a test of using alphaMELTS to model biotite compositions and fO2 for natural systems, we undertook equilibrium crystallization calculations for dacites and rhyolites for which bulk-rock, biotite, and Fe–Ti oxide compositions had been previously reported (Fig. 7a and b). For most samples, we calculated fO2 and T using Fe–Ti oxide thermometry and oxybarometery [based on the model of Ghiorso & Evans (2008)]. However, for the peraluminous rhyolites, where only ilmenite was present, we relied on constraints from Pichavant et al. (1988) based on hematite contents in ilmenite indicating crystallization between the quartz–fayalite–magnetite and wüstite–magnetite buffers (see Fig. 7 caption for details); we have plotted all peraluminous rhyolite data based on this constraint at NNO –2 in Fig. 7b for clarity. The compiled volcanic rock data demonstrate a range of T and fO2 of ∼650–890°C and NNO +1·1 to –2·0 (Fig. 7a and b). A key feature of this dataset is the inverse correlation between fO2 ((in terms of ΔNNO) and biotite FeT/(FeT + Mg), which is expected in light of our previous discussion (Fig. 7b). We attempted to reproduce this volcanic dataset in two ways. First, alphaMELTS was used to calculate the equilibrium partially molten assemblage corresponding to the bulk composition of each of the data points shown in Fig. 7a and b at the independently determined T and fO2 values shown in Fig. 7a; calculations were carried out at 0·3 GPa, and the bulk water contents were adjusted to reproduce the co-crystallizing mineralogy of each sample (generally between 1 and 3 wt %). As shown in Fig. 7c, alphaMELTS is able to reproduce measured biotite FeT/(FeT + Mg) ratios well (i.e. to within 0·10) over a wide range of 0·2–0·9. The average absolute deviation from the natural biotite FeT/(FeT + Mg) ratios is 0·03 ± 0·03 (1σ). Likewise, model results of Kbt–bulkFeT/Mg are mostly within 0·10 of the measured values, with an average absolute deviation of 0·06 ± 0·05 (1σ) (Fig. 7d). Second, we varied the fO2 in our alphaMELTS modeling to reproduce the natural biotite FeT/(FeT + Mg) ratios (similar to the modeling exercise used for experimental data; see Fig. 5b). Oxygen fugacity values recorded by Fe–Ti oxide oxybarometry for volcanic rocks with fO2 > NNO –1 were reproduced to within 1 log unit, with an average absolute deviation of 0·34 ± 0·28 (1σ). For the peraluminous volcanic rocks shown in Fig. 7e, there is a significant range of fO2 based on alphaMELTS modeling of biotite compositions (i.e. alphaMELTS-calculated ΔNNO ranges from –2·5 to +0·5), which, although systematically at the reducing end of the range of all the natural volcanic samples and thereby consistent with the estimates of Pichavant et al. (1988), extends to more oxidizing conditions than they estimated. However, as we noted above in connection with the similar treatment of experimental data in the previous section, the approach we have taken to quantitative estimation of fO2 becomes progressively less precise when most of the Fe in the system is present as Fe2+, and part of the deviation from the estimates of Pichavant et al. (1988) may relate to this. Despite the uncertainty associated with the most reduced of the natural volcanic rocks and experimental samples, there is still a 1:1 correlation between the alphaMELTS modeled fO2 ((in terms of ΔNNO) and that calculated or estimated for the volcanic rocks (with scatter at the level of ∼1 log unit over a >3 log unit range), suggesting that the alphaMELTS-based approach is quantitative at this level. Some additional possible explanations for the observed scatter about the 1:1 lines in Fig. 7c–e (other than for the low- fO2 peraluminous samples) are as follows: (1) the biotite solution model in alphaMELTS does not include Fe3+, even though it is known to be present at 5–30% of total Fe in natural biotites from silicic systems (see next section); (2) there was possible resetting of fO2 and T in Fe–Ti oxide pair chemistry during cooling; or (3) biotite crystallized at different magmatic conditions than the Fe–Ti oxides. Nevertheless, despite the potentially complicating effects of these factors, the similarity between the model results and natural data (Fig. 7) and experimental data (Fig. 5) suggests that alphaMELTS modeling of biotite crystallization can be used to constrain the fO2 for the volcanic rock record to an order of magnitude or better for conditions more oxidizing than NNO –2 to NNO –1 provided that other crystallization conditions (P, T, H2O) can be constrained. Fig. 7. View largeDownload slide Biotite chemistry from natural high-silica volcanic rocks (a, b) and alphaMELTS modeling of natural volcanic compositions (c, d). Data for volcanic rocks are based on the Fe–Ti two-oxide compilation of Ghiorso & Evans (2008), augmented with biotite and whole-rock major element data for biotite-bearing samples. (a) log10 fO2 vs T (°C) for volcanic samples; fO2 and T were primarily constrained using the Fe–Ti two-oxide oxygen barometer and thermometer of Ghiorso & Evans (2008) for magnetite-bearing samples (circles), except for the peraluminous rhyolites, which were based on calculations for Macusani volcanic rocks [blue region in (a)], where T was estimated by Pichavant et al. (1988) based on the water-saturated solidus of haplogranitic melts and fO2 ⁠, based on maximum hematite contents of ilmenite of 3%, placing the fO2 conditions of crystallization between the FMQ (fayalite–magnetite–quartz) and WM (wüstite–magnetite) buffers. Oxygen buffer curves are from the following calibrations: NNO, O’Neill & Pownceby (1993); FMQ, O’Neill (1988); CCO (graphite-CO-CO2 fluid), French & Eugster (1965); WM and MH (magnetite–hematite), Huebner (1971). The CCO curve shows the maximum stability of graphite in an H2O-free system. (b) FeT/(FeT + Mg) in volcanic biotite vs ΔNNO. Macusani volcanic rocks are plotted at FMQ -1.75 for simplicity. A range for FMQ between NNO –0·7 and NNO –1 is shown, as various locations for the FMQ buffer have been proposed. (c), (d), and (e) compare alphaMELTS crystallization calculations with natural data from volcanic rocks (see main text for details). alphaMELTS calculations were conducted using the bulk compositions of volcanic samples and at temperatures constrained through Fe–Ti oxide thermometry. The peraluminous samples were assumed to have crystallized at NNO –2. All calculations were carried out at a pressure of 0·3 GPa. Water contents were fixed at the lowest values necessary to allow biotite crystallization and reproduce the observed phase assemblage in natural samples, resulting in values used between 0·5 and 3·5 wt % H2O. Fig. 7. View largeDownload slide Biotite chemistry from natural high-silica volcanic rocks (a, b) and alphaMELTS modeling of natural volcanic compositions (c, d). Data for volcanic rocks are based on the Fe–Ti two-oxide compilation of Ghiorso & Evans (2008), augmented with biotite and whole-rock major element data for biotite-bearing samples. (a) log10 fO2 vs T (°C) for volcanic samples; fO2 and T were primarily constrained using the Fe–Ti two-oxide oxygen barometer and thermometer of Ghiorso & Evans (2008) for magnetite-bearing samples (circles), except for the peraluminous rhyolites, which were based on calculations for Macusani volcanic rocks [blue region in (a)], where T was estimated by Pichavant et al. (1988) based on the water-saturated solidus of haplogranitic melts and fO2 ⁠, based on maximum hematite contents of ilmenite of 3%, placing the fO2 conditions of crystallization between the FMQ (fayalite–magnetite–quartz) and WM (wüstite–magnetite) buffers. Oxygen buffer curves are from the following calibrations: NNO, O’Neill & Pownceby (1993); FMQ, O’Neill (1988); CCO (graphite-CO-CO2 fluid), French & Eugster (1965); WM and MH (magnetite–hematite), Huebner (1971). The CCO curve shows the maximum stability of graphite in an H2O-free system. (b) FeT/(FeT + Mg) in volcanic biotite vs ΔNNO. Macusani volcanic rocks are plotted at FMQ -1.75 for simplicity. A range for FMQ between NNO –0·7 and NNO –1 is shown, as various locations for the FMQ buffer have been proposed. (c), (d), and (e) compare alphaMELTS crystallization calculations with natural data from volcanic rocks (see main text for details). alphaMELTS calculations were conducted using the bulk compositions of volcanic samples and at temperatures constrained through Fe–Ti oxide thermometry. The peraluminous samples were assumed to have crystallized at NNO –2. All calculations were carried out at a pressure of 0·3 GPa. Water contents were fixed at the lowest values necessary to allow biotite crystallization and reproduce the observed phase assemblage in natural samples, resulting in values used between 0·5 and 3·5 wt % H2O. As a summary of the previous discussions, we have plotted Kbt–bulkFeT/Mg versus ΔNNO for volcanic rocks, experimental studies, and the alphaMELTS model results at different temperatures in Fig. 8. There is a broad negative correlation between Kbt–bulkFeT/Mg and ΔNNO that is consistent between the two datasets and the alphaMELTS model results. Again, the important features are that Kbt–bulkFeT/Mg increases with decreasing fO2 and decreasing T and that alphaMELTS modeling reproduces the range of Kbt–bulkFeT/Mg observed in natural and experimental data over a wide range of fO2 ⁠. Fig. 8. View largeDownload slide Kbt–bulkFeT/Mg vs fO2 (expressed as ΔNNO) for compiled volcanic rock data, experimental data, and alphaMELTS modeling. Volcanic rock data are the same as compiled in Fig. 7 and experimental data are the same as presented in Fig. 4. Experimental data and volcanic rock data are plotted as groups by temperature (either the experimental temperature or, for the volcanic rocks, as described in the caption of Fig. 7). alphaMELTS modeling is for the same bulk composition and parameters as presented in Fig. 6. Curves are shown for model runs in which Fe–Ti oxides were suppressed. Fig. 8. View largeDownload slide Kbt–bulkFeT/Mg vs fO2 (expressed as ΔNNO) for compiled volcanic rock data, experimental data, and alphaMELTS modeling. Volcanic rock data are the same as compiled in Fig. 7 and experimental data are the same as presented in Fig. 4. Experimental data and volcanic rock data are plotted as groups by temperature (either the experimental temperature or, for the volcanic rocks, as described in the caption of Fig. 7). alphaMELTS modeling is for the same bulk composition and parameters as presented in Fig. 6. Curves are shown for model runs in which Fe–Ti oxides were suppressed. Constraints from biotite compositions in the plutonic rock record The overall agreement between the alphaMELTS model results and data both from the experimental studies and from the natural felsic volcanic rocks for biotite compositions, Kbt–bulkFeT/Mg ⁠, and fO2 suggests that alphaMELTS modeling of biotite crystallization can be used to constrain the fO2 (to better than ±1 log unit) of biotite-bearing felsic igneous rocks given independent constraints on the T, P, and aH2O at which biotite crystallized from the melt. The focus of this study is quantifying the fO2 of peraluminous granites intruded before and after the ‘Great Oxidation Event’ near the Archean–Proterozoic boundary, and in this section, we apply the alphaMELTS-based approach to this end. Before proceeding with this, however, we first compare in Fig. 9 wet chemical measurements of Fe3+/FeT ratios of biotite with the FeT/(FeT + Mg) ratios of the biotites (Fig. 9a) and the bulk FeT/Mg distribution coefficient (⁠ Kbt–bulkFeT/Mg ⁠) (Fig. 9b) based on a compilation of felsic plutonic rocks from the literature (see the caption to Fig. 9 for the sources of the data). Figure 9a shows that biotites from more oxidized samples (e.g. those from arc batholiths or ‘magnetite’-series granitoids shown as red symbols) generally have Fe3+/FeT ratios between ∼0·10 and 0·25, whereas biotites from more reduced igneous suites (e.g. SPGs or ‘ilmenite’-series granites shown as blue symbols) have Fe3+/FeT ratios between 0·02 and 0·1 (Fig. 9b). This demonstrates the expected positive correlation between Fe3+/FeT ratios of biotites and fO2 ⁠, as embodied in equations (1)–(3). Although there is considerable scatter in Fig. 9a and b, what is important from our point of view is that both biotite FeT/(FeT + Mg) ratios and Kbt–bulkFeT/Mg are inversely correlated with Fe3+/FeT ratios; that is, biotites from more oxidized plutonic rocks generally have lower biotite FeT/(FeT + Mg) ratios (∼0·4–0·6) and Kbt–bulkFeT/Mg (∼0·3–0·9) than those from more reduced granitoids (∼0·6–0·8 and ∼0·8–1·2). Although it would be difficult to extract quantitative constraints on fO2 from the biotite compositions from the plutonic rocks shown in Fig. 9 owing to a lack of constraints on T, P, and water contents for these samples (which is not known for most of them), these trends nevertheless provide additional confidence for us that variations in biotite FeT/(FeT + Mg) ratios in plutonic rocks preserve information about magmatic fO2 ⁠, and that alphaMELTS calculations can be used to constrain fO2 in plutonic rocks where crystallization conditions are known (just as Fig. 7 demonstrates that this can be done for biotite-bearing felsic volcanic rocks.) Fig. 9. View largeDownload slide Biotite chemistry for granitoids. (a) FeT/(FeT + Mg) vs Fe3+/FeT in biotite from granitoids. Data sources: Sierra Nevada batholith—Dodge et al. (1969); Hercynian biotite + muscovite granites, Portugal—de Albuquerque (1973); Ilmenite and Magnetite Series of Japan—Czamanske et al. (1981); Hepburn and Bishop Intrusive Suites, NW Territories—Lalonde & Bernard (1993); Ben Nevis igneous complex—Haslam (1968). SPGs are shown as filled dark blue symbols. (b) Kbt–bulkFeT/Mg vs Fe3+/FeT in biotite from granitoids; data sources as in (a). Fig. 9. View largeDownload slide Biotite chemistry for granitoids. (a) FeT/(FeT + Mg) vs Fe3+/FeT in biotite from granitoids. Data sources: Sierra Nevada batholith—Dodge et al. (1969); Hercynian biotite + muscovite granites, Portugal—de Albuquerque (1973); Ilmenite and Magnetite Series of Japan—Czamanske et al. (1981); Hepburn and Bishop Intrusive Suites, NW Territories—Lalonde & Bernard (1993); Ben Nevis igneous complex—Haslam (1968). SPGs are shown as filled dark blue symbols. (b) Kbt–bulkFeT/Mg vs Fe3+/FeT in biotite from granitoids; data sources as in (a). Constraining T, P, and aH2O of crystallization of Archean and Proterozoic SPGs to interpret biotite FeT/(FeT + Mg) as a proxy for oxygen fugacity T and aH2O both influence biotite FeT/(FeT + Mg) ratios, so before applying our approach to determining the fO2 of Archean and Proterozoic SPGs in our study, we first had to place constraints on the crystallization T and aH2O for our samples (see Table 3 for specific constraints on each locality). Constraints on T and aH2O (and P) of crystallization were determined in the following ways. Table 3 Estimates of temperature, pressure, and aH2O during emplacement and cooling of Archean and Proterozoic SPGs Locality Solidus T (°C) P (GPa) aH2O Ghost Lake batholith 625–680 (subsolvus peraluminous granites) 0·4 (P obtained from thermobarometry in surrounding metamorphic rocks; Campion et al., 1986) >0·7 653 (+45/–19) (grt–bt Fe/Mg exchange thermometry, SP-16‐34) Sturgeon Lake granite 650–690 (subsolvus peraluminous granites) 0·25–0·38 (P obtained from thermobarometry of Quetico metasediments in vicinity; Percival & Williams, 1989) >0·7 712 (+53/–16) (grt–bt Fe/Mg exchange thermometry, SP-16‐52) Shannon Lake granite 625–680 (subsolvus peraluminous granites, assuming intrusion pressure up to 0·5 GPa) >0·22 (subsolvus granite) >0·7 Preissac, Lacorne, Lamotte, Moly Hill plutons, 2016 620–680 (subsolvus peraluminous granites) 0·44–0·58 (plag–bt–grt–ms geobarometer; Feng & Kerrich, 1990) >0·7 695 (qtz–grt O isotope thermometry; Mulja et al., 1995a) Mt Owen batholith 625–680 (subsolvus peraluminous granites, assuming intrusion pressure up to 0·5 GPa) >0·22 (subsolvus granite) >0·7 631 (+72/–53) (grt–bt Fe/Mg exchange thermometry, 98-T1) Harney Peak granite 626 (+54/–19) (grt–bt Fe/Mg exchange thermometry, HP-43A) 0·25–0·37 (based on thermobarometry from surrounding metamorphic rocks; Helms & Labotka, 1991) >∼0·3 (increasing during progressive crystallization, Nabelek & Ternes, 1997) 650 to >800 (O isotope mineral–quartz equilibration temperatures, Nabelek et al., 1992b) Hepburn intrusive suite 680–750 (grt–bt Fe/Mg exchange thermometry, Pattison et al., 1982) 0·3–0·4 (garnet–plagioclase–sillimanite–quartz barometry, Pattison et al., 1982) (note that source region may have been at ∼0·9 GPa and >800°C; Pattison et al., 1982; St Onge & King, 1987) >0·5 to 0·7 Silver Plume and St Vrain granites 740–760 (Fe–Ti oxide thermometry; Anderson & Thomas, 1985) >0·3–0·4 (lower stability of muscovite; Anderson & Thomas, 1985) 0·5–0·7 Ak-Chin, Ruin, Oracle, Sierra Estrella granites 625–680 (subsolvus peraluminous granites, assuming intrusion pressure up to 0·5 GPa) >0·27–0·32 (Anderson & Bender, 1989) >0·7 Locality Solidus T (°C) P (GPa) aH2O Ghost Lake batholith 625–680 (subsolvus peraluminous granites) 0·4 (P obtained from thermobarometry in surrounding metamorphic rocks; Campion et al., 1986) >0·7 653 (+45/–19) (grt–bt Fe/Mg exchange thermometry, SP-16‐34) Sturgeon Lake granite 650–690 (subsolvus peraluminous granites) 0·25–0·38 (P obtained from thermobarometry of Quetico metasediments in vicinity; Percival & Williams, 1989) >0·7 712 (+53/–16) (grt–bt Fe/Mg exchange thermometry, SP-16‐52) Shannon Lake granite 625–680 (subsolvus peraluminous granites, assuming intrusion pressure up to 0·5 GPa) >0·22 (subsolvus granite) >0·7 Preissac, Lacorne, Lamotte, Moly Hill plutons, 2016 620–680 (subsolvus peraluminous granites) 0·44–0·58 (plag–bt–grt–ms geobarometer; Feng & Kerrich, 1990) >0·7 695 (qtz–grt O isotope thermometry; Mulja et al., 1995a) Mt Owen batholith 625–680 (subsolvus peraluminous granites, assuming intrusion pressure up to 0·5 GPa) >0·22 (subsolvus granite) >0·7 631 (+72/–53) (grt–bt Fe/Mg exchange thermometry, 98-T1) Harney Peak granite 626 (+54/–19) (grt–bt Fe/Mg exchange thermometry, HP-43A) 0·25–0·37 (based on thermobarometry from surrounding metamorphic rocks; Helms & Labotka, 1991) >∼0·3 (increasing during progressive crystallization, Nabelek & Ternes, 1997) 650 to >800 (O isotope mineral–quartz equilibration temperatures, Nabelek et al., 1992b) Hepburn intrusive suite 680–750 (grt–bt Fe/Mg exchange thermometry, Pattison et al., 1982) 0·3–0·4 (garnet–plagioclase–sillimanite–quartz barometry, Pattison et al., 1982) (note that source region may have been at ∼0·9 GPa and >800°C; Pattison et al., 1982; St Onge & King, 1987) >0·5 to 0·7 Silver Plume and St Vrain granites 740–760 (Fe–Ti oxide thermometry; Anderson & Thomas, 1985) >0·3–0·4 (lower stability of muscovite; Anderson & Thomas, 1985) 0·5–0·7 Ak-Chin, Ruin, Oracle, Sierra Estrella granites 625–680 (subsolvus peraluminous granites, assuming intrusion pressure up to 0·5 GPa) >0·27–0·32 (Anderson & Bender, 1989) >0·7 Table 3 Estimates of temperature, pressure, and aH2O during emplacement and cooling of Archean and Proterozoic SPGs Locality Solidus T (°C) P (GPa) aH2O Ghost Lake batholith 625–680 (subsolvus peraluminous granites) 0·4 (P obtained from thermobarometry in surrounding metamorphic rocks; Campion et al., 1986) >0·7 653 (+45/–19) (grt–bt Fe/Mg exchange thermometry, SP-16‐34) Sturgeon Lake granite 650–690 (subsolvus peraluminous granites) 0·25–0·38 (P obtained from thermobarometry of Quetico metasediments in vicinity; Percival & Williams, 1989) >0·7 712 (+53/–16) (grt–bt Fe/Mg exchange thermometry, SP-16‐52) Shannon Lake granite 625–680 (subsolvus peraluminous granites, assuming intrusion pressure up to 0·5 GPa) >0·22 (subsolvus granite) >0·7 Preissac, Lacorne, Lamotte, Moly Hill plutons, 2016 620–680 (subsolvus peraluminous granites) 0·44–0·58 (plag–bt–grt–ms geobarometer; Feng & Kerrich, 1990) >0·7 695 (qtz–grt O isotope thermometry; Mulja et al., 1995a) Mt Owen batholith 625–680 (subsolvus peraluminous granites, assuming intrusion pressure up to 0·5 GPa) >0·22 (subsolvus granite) >0·7 631 (+72/–53) (grt–bt Fe/Mg exchange thermometry, 98-T1) Harney Peak granite 626 (+54/–19) (grt–bt Fe/Mg exchange thermometry, HP-43A) 0·25–0·37 (based on thermobarometry from surrounding metamorphic rocks; Helms & Labotka, 1991) >∼0·3 (increasing during progressive crystallization, Nabelek & Ternes, 1997) 650 to >800 (O isotope mineral–quartz equilibration temperatures, Nabelek et al., 1992b) Hepburn intrusive suite 680–750 (grt–bt Fe/Mg exchange thermometry, Pattison et al., 1982) 0·3–0·4 (garnet–plagioclase–sillimanite–quartz barometry, Pattison et al., 1982) (note that source region may have been at ∼0·9 GPa and >800°C; Pattison et al., 1982; St Onge & King, 1987) >0·5 to 0·7 Silver Plume and St Vrain granites 740–760 (Fe–Ti oxide thermometry; Anderson & Thomas, 1985) >0·3–0·4 (lower stability of muscovite; Anderson & Thomas, 1985) 0·5–0·7 Ak-Chin, Ruin, Oracle, Sierra Estrella granites 625–680 (subsolvus peraluminous granites, assuming intrusion pressure up to 0·5 GPa) >0·27–0·32 (Anderson & Bender, 1989) >0·7 Locality Solidus T (°C) P (GPa) aH2O Ghost Lake batholith 625–680 (subsolvus peraluminous granites) 0·4 (P obtained from thermobarometry in surrounding metamorphic rocks; Campion et al., 1986) >0·7 653 (+45/–19) (grt–bt Fe/Mg exchange thermometry, SP-16‐34) Sturgeon Lake granite 650–690 (subsolvus peraluminous granites) 0·25–0·38 (P obtained from thermobarometry of Quetico metasediments in vicinity; Percival & Williams, 1989) >0·7 712 (+53/–16) (grt–bt Fe/Mg exchange thermometry, SP-16‐52) Shannon Lake granite 625–680 (subsolvus peraluminous granites, assuming intrusion pressure up to 0·5 GPa) >0·22 (subsolvus granite) >0·7 Preissac, Lacorne, Lamotte, Moly Hill plutons, 2016 620–680 (subsolvus peraluminous granites) 0·44–0·58 (plag–bt–grt–ms geobarometer; Feng & Kerrich, 1990) >0·7 695 (qtz–grt O isotope thermometry; Mulja et al., 1995a) Mt Owen batholith 625–680 (subsolvus peraluminous granites, assuming intrusion pressure up to 0·5 GPa) >0·22 (subsolvus granite) >0·7 631 (+72/–53) (grt–bt Fe/Mg exchange thermometry, 98-T1) Harney Peak granite 626 (+54/–19) (grt–bt Fe/Mg exchange thermometry, HP-43A) 0·25–0·37 (based on thermobarometry from surrounding metamorphic rocks; Helms & Labotka, 1991) >∼0·3 (increasing during progressive crystallization, Nabelek & Ternes, 1997) 650 to >800 (O isotope mineral–quartz equilibration temperatures, Nabelek et al., 1992b) Hepburn intrusive suite 680–750 (grt–bt Fe/Mg exchange thermometry, Pattison et al., 1982) 0·3–0·4 (garnet–plagioclase–sillimanite–quartz barometry, Pattison et al., 1982) (note that source region may have been at ∼0·9 GPa and >800°C; Pattison et al., 1982; St Onge & King, 1987) >0·5 to 0·7 Silver Plume and St Vrain granites 740–760 (Fe–Ti oxide thermometry; Anderson & Thomas, 1985) >0·3–0·4 (lower stability of muscovite; Anderson & Thomas, 1985) 0·5–0·7 Ak-Chin, Ruin, Oracle, Sierra Estrella granites 625–680 (subsolvus peraluminous granites, assuming intrusion pressure up to 0·5 GPa) >0·27–0·32 (Anderson & Bender, 1989) >0·7 (1) Some samples were amenable to various geothermometers, which either we applied using mineral compositions analyzed here or had already been assessed in previous studies. For example, for analyzed garnet-bearing samples (n = 5), garnet–biotite Fe–Mg exchange thermometry (Holdaway, 2000) using unzoned average garnet core compositions and average biotite compositions was used to determine temperature. This included Archean samples from the Ghost Lake batholith, the Sturgeon Lake granite, and the Mt. Owen batholith and Proterozoic samples from the Harney Peak granite and the Hepburn intrusive suite. These samples yielded temperatures (630–750°C) similar to the water-saturated solidus in the haplogranite system (quartz + albite + orthoclase) at pressures >0·2 GPa (∼650–680°C; see Ebadi & Johannes, 1991), and no differences were observed between Archean and Proterozoic samples. Temperature constraints from previous studies include O isotope thermometry based on coexisting mineral pairs for the Harney Peak granite, which mostly suggest equilibration temperatures between 650 and 800°C (assuming isotopic equilibrium among the quartz and other coexisting minerals) (Nabelek et al., 1992b). In addition, Anderson & Thomas (1985) presented quantitative constraints for crystallization conditions of the Silver Plume and St Vrain granites based on Fe–Ti oxide and biotite compositions using equation (1). They concluded that the Silver Plume and St Vrain granites probably crystallized between 740 and 760°C, at water-undersaturated conditions. (2) For samples in which the coexisting mineralogy was not amenable to quantitative mineral chemistry-based thermometry, we used the observation that all samples for which we were able to undertake petrographic examination are sub-solvus granites, containing co-crystallized grains of both albitic plagioclase and alkali feldspar. For two feldspars to coexist with a haplogranite melt, partial pressures of water must be sufficiently high to lower the temperature of the solidus surface to intersect the feldspar solvus (Tuttle & Bowen, 1958; Luth et al., 1964; Ebadi & Johannes, 1991). For example, at 0·5 GPa, two feldspars will crystallize only from melt with aH2O≥ 0·7 (or XH2Ov > 0·6, where XH2Ov is the molar proportion of H2O in a coexisting vapor phase) (Ebadi & Johannes, 1991). At 0·5 GPa and aH2O of 0·7–1·0 in the haplogranitic system, the solidus T is between ∼700°C and 645°C (Ebadi & Johannes, 1991). Therefore, it is possible with independent pressures estimates for the SPGs in this study to place semi-quantitative constraints on both aH2O and solidus T. (3) Pressures of crystallization for the granites were constrained by several methods, including previously reported thermobarometry from surrounding metasedimentary rocks (Campion et al., 1986; Percival & Williams, 1989; Helms & Labotka, 1991) (which record the depth of intrusion and crystallization) and minimum P estimates for the low-P stability limit for muscovite (∼0·3–0·4 GPa; Anderson & Rowley, 1981; Pichavant et al., 2017) and sub-solvus granites (∼0·2 GPa; Ebadi & Johannes, 1991). Pressure estimates for all granites were between 0·2 and 0·6 GPa. Using experimentally determined P–T relationships of the haplogranite solidus at different aH2O ⁠, solidus T estimates for our granite samples are estimated to be <710°C (Fig. 7; Tuttle & Bowen, 1958; Luth et al., 1964; Ebadi & Johannes, 1991). For water-saturated peraluminous haplogranites (i.e. the system quartz–orthoclase–albite–Al2O3), the solidus T may be further decreased by 20–30°C (Holtz et al., 1992). We summarize the results of our constraints of T, P, and aH2O during crystallization in Fig. 10. For the Archean SPGs, temperatures of crystallization were constrained to <710°C, pressures of 0·2–0·6 GPa, and aH2O > 0·7. For the Proterozoic SPGs, T estimates extend from the water-saturated haplogranite solidus (∼640°C) to ∼800°C (e.g. the Harney Peak, Silver Plume, and St Vrain granites and the Hepburn Intrusive suite) at pressures of 0·3–0·4 GPa and aH2O > 0·3. Fig. 10. View largeDownload slide Summary of inferred crystallization conditions for SPGs in this study. Haplogranite solidi at different XH2Ov (grey curves) are from Ebadi & Johannes (1991), where XH2Ov is the experimentally controlled mole fraction of H2O in a coexisting vapor phase. Water-saturated peraluminous haplogranite solidus (dashed grey line) is from Holtz et al. (1992). Estimated P–T ranges for Archean SPGs are shown as blue boxes and for Proterozoic SPGs as red boxes (see Table 3). Fig. 10. View largeDownload slide Summary of inferred crystallization conditions for SPGs in this study. Haplogranite solidi at different XH2Ov (grey curves) are from Ebadi & Johannes (1991), where XH2Ov is the experimentally controlled mole fraction of H2O in a coexisting vapor phase. Water-saturated peraluminous haplogranite solidus (dashed grey line) is from Holtz et al. (1992). Estimated P–T ranges for Archean SPGs are shown as blue boxes and for Proterozoic SPGs as red boxes (see Table 3). The fO2 of Archean and Proterozoic SPGs Modeling approach The determination of fO2 for high-silica plutonic rocks using alphaMELTS is done as follows. For each granitic rock of interest, given its bulk composition and independent estimates of the P, T, and aH2O at which the magma crystallized (see previous section), the equilibrium phase relations including the composition of the equilibrium biotite are calculated using alphaMELTS. The value of fO2 is then varied systematically until a value is found at which the FeT/(FeT + Mg) ratio of the calculated equilibrium biotite matches that observed in the biotite of the natural sample. Reported errors in fO2 are based on 2σSD variations in FeT/(FeT + Mg) values in biotite observed in a sample. The bulk composition for each granite is assumed to be its major element whole-rock composition (SiO2, TiO2, Al2O3, FeO, MgO, CaO, Na2O, K2O; note that MnO and P2O5 are not included as these minor components are not accommodated realistically in phases included in alphaMELTS.) Crystallization T also exerts a strong control on biotite FeT/(FeT + Mg) ratios (see Fig. 6a), so modeling was undertaken at both maximum and minimum estimates of the temperature of magma crystallization. Figure 11 illustrates this modeling approach for sample SP-3 from the Silver Plume granite. At the highest estimated T, reproducing the highest biotite FeT/(FeT + Mg) value yields a minimum fO2 estimate (ΔNNO = –0·5), whereas at the lowest estimated T, the lowest biotite FeT/(FeT + Mg) value yields a maximum fO2 estimate (ΔNNO = –0·05). We followed this process for every sample in which biotite FeT/(FeT + Mg) values were sensitive to fO2 variations using alphaMELTS and report maximum and minimum fO2 estimates. However, as previously emphasized, for fO2 values at which nearly all of the Fe in the system is Fe2+ (e.g. >90% at NNO –2), further reduction of fO2 does not result in significant shifts in the biotite composition to more Fe-rich biotites. At NNO –2 the maximum calculated Kbt–bulkFeT/Mg approaches ∼0·9 when Fe–Ti oxides are suppressed (Fig. 6a and b). Therefore, for the samples where Kbt–bulkFeT/Mg is >0·9 we give only an upper fO2 limit of NNO –2. Fig. 11. View largeDownload slide Example of the alphaMELTS modeling approach to constrain the fO2 values of SPG samples. The bulk composition of sample SP-3 of the Silver Plume granite is used in this case. At the maximum estimated temperature of crystallization, the fO2 required to reproduce the maximum biotite FeT/(FeT + Mg) yields a minimum estimate on fO2 ⁠. Similarly, at the minimum estimated temperature of crystallization, the fO2 required to reproduce the minimum biotite FeT/(FeT + Mg) yields a maximum estimate on fO2 ⁠. For sample SP-3 the range of fO2 that can reproduce the natural data from NNO –0·05 to NNO +0·5. Fig. 11. View largeDownload slide Example of the alphaMELTS modeling approach to constrain the fO2 values of SPG samples. The bulk composition of sample SP-3 of the Silver Plume granite is used in this case. At the maximum estimated temperature of crystallization, the fO2 required to reproduce the maximum biotite FeT/(FeT + Mg) yields a minimum estimate on fO2 ⁠. Similarly, at the minimum estimated temperature of crystallization, the fO2 required to reproduce the minimum biotite FeT/(FeT + Mg) yields a maximum estimate on fO2 ⁠. For sample SP-3 the range of fO2 that can reproduce the natural data from NNO –0·05 to NNO +0·5. The T, P, and aH2O values used for modeling the Archean and Proterozoic SPGs for each sample are given in Table 4. We placed a lower limit on the crystallization T at 680°C as most of the alphaMELTS-modeled bulk compositions are below the solidus at lower T. We also chose to model aH2O values between 0·9 and 1·0 (unless otherwise indicated, as for the Silver Plume and St Vrain granites), as variations in aH2O for samples near water-saturation are not precisely known. This assumption has only a small effect on the calculated fO2 values (e.g. an increase of 0·1 in aH2O corresponds to an ∼ +0·2 log unit shift in calculated fO2 ⁠; see Supplementary Data Fig. A21 for sensitivity analysis). Table 4 Summary of alphaMELTS modeling parameters and results FeT/(FeT + Mg) FeT/Mg Temperature (°C) Calculated oxygen fugacity (ΔNNO) Sample Biotite Whole- rock Biotite Whole- rock Kd FeT/ Mg alpha MELTS modeling Min. Max. Pressure (GPa) aH2O* Min. Max. Av. Archean samples E19‐8 0·594 0·614 1·463 1·594 0·918 N — — — — — –2 SP-16‐20a 0·625 0·653 1·671 1·882 0·888 N — — — — — –2 SP-16‐22 0·614 0·650 1·592 1·854 0·859 N — — — — — –2 SP-16‐28a 0·703 0·711 2·372 2·455 0·966 N — — — — — –2 SP-16‐23 0·635 0·638 1·742 1·761 0·989 N — — — — — –2 SP-16‐24 0·627 0·645 1·682 1·810 0·929 N — — — — — –2 SP-16‐34 0·685 0·626 2·174 1·669 1·302 N — — — — — –2 SP-16‐48 0·613 0·613 1·583 1·574 1·005 N — — — — — –2 SP-16‐52† 0·705 0·764 2·396 3·228 0·742 Y 680 700 0·3 1 –0·5 0·6 0·1 SP-16‐53 0·719 0·714 2·560 2·496 1·026 N — — — — — –2 SP-16‐1† 0·751 0·810 3·027 4·272 0·708 Y 680 700 0·2 1 0·0 0·8 0·4 SP-16‐2b 0·706 0·706 2·405 2·403 1·000 N — — — — — –2 LC-8 0·726 0·659 2·654 1·937 1·370 N — — — — — –2 LC-16 0·726 0·679 2·646 2·118 1·249 N — — — — — –2 LC-30 0·759 0·679 3·156 2·113 1·494 N — — — — — –2 903 0·691 0·689 2·240 2·219 1·010 N — — — — — –2 768 0·746 0·742 2·936 2·882 1·019 N — — — — — –2 797†‡ 0·826 0·868 4·769 6·556 0·727 N 908 0·779 0·767 3·528 3·293 1·071 N — — — — — –2 616† 0·740 0·769 2·845 3·322 0·856 Y 680 700 0·5 0·9 –2 0·2 –0·9 634† 0·762 0·799 3·211 3·967 0·809 Y 680 700 0·5 0·9 <–2 0·87 –0·6 652† 0·812 0·852 4·332 5·749 0·754 Y 680 700 0·5 0·9 <–2 –1·4 –1·7 672 0·825 0·822 4·720 4·611 1·024 N — — — — — –2 692 0·789 0·731 3·741 2·720 1·375 N — — — — — –2 707† 0·821 0·874 4·586 6·959 0·659 Y 680 700 0·5 0·9 –2 0·5 –0·8 26 0·770 0·755 3·351 3·082 1·087 N — — — — — –2 29 0·689 0·684 2·210 2·164 1·022 N — — — — — –2 98TI 0·806 0·774 4·170 3·422 1·219 N — — — — — –2 Proterozoic samples HP-6B† 0·647 0·692 1·833 2·244 0·817 Y 680 700 0·3 0·95 <–2 0·4 –0·8 HP-7B† 0·621 0·755 1·639 3·085 0·531 Y 680 700 0·3 0·95–1·00 0 0·9 0·5 HP-21C† 0·627 0·675 1·684 2·080 0·810 Y 680 700 0·3 0·9–1·0 –2 0·5 –0·8 HP-26 0·664 0·685 1·981 2·178 0·910 N — — — — — –2 HP 32 0·758 0·745 3·132 2·927 1·070 N — — — — — –2 HP 43A†§ 0·644 0·872 1·808 6·818 0·265 N HP 44A† 0·646 0·794 1·826 3·864 0·473 Y 680 700 0·3 0·97–1·00 0·850 1·38 1·1 HP-23A† 0·674 0·721 2·067 2·580 0·801 Y 680 700 0·3 0·980 <–2 –1·1 –1·6 HP-14A 0·862 0·849 6·233 5·610 1·111 N — — — — — –2 L14 0·635 0·656 1·739 1·906 0·912 N — — — — — –2 L99 0·637 0·639 1·756 1·772 0·991 N — — — — — –2 L125 0·782 0·741 3·583 2·867 1·250 N — — — — — –2 L204† 0·695 0·732 2·281 2·729 0·836 Y 680 750 0·3 0·7–0·85 –2 0·55 –0·7 L331 0·682 0·703 2·144 2·366 0·906 N — — — — — –2 L341A 0·658 0·680 1·927 2·126 0·906 N — — — — — –2 S29† 0·631 0·667 1·709 2·006 0·852 Y 680 750 0·3 0·67–0·96 –2 0·1 –1·0 S54 0·628 0·654 1·758 1·893 0·929 N — — — — — –2 SP-1† 0·670 0·704 2·034 2·376 0·856 Y 740 760 0·3 0·66 <–2 –0·64 –0·6 SP-3† 0·559 0·693 1·270 2·261 0·562 Y 740 760 0·3 0·66–0·77 –0·5 0·05 –0·2 SP-5† 0·550 0·683 1·223 2·155 0·568 Y 740 760 0·3 0·66–0·77 –0·58 –0·2 –0·4 SP-6† 0·547 0·718 1·210 2·543 0·476 Y 740 760 0·3 0·66–0·78 –0·5 0·05 –0·2 SP-8† 0·562 0·689 1·284 2·214 0·580 Y 740 760 0·3 0·65–0·77 –1·1 0·18 –0·5 SVL-3† 0·678 0·724 2·108 2·622 0·804 Y 740 760 0·3 0·53–0·76 –2 –0·7 –1·4 SVL-4† 0·633 0·717 1·724 2·536 0·680 Y 740 760 0·3 0·60–0·77 –1 –0·5 –0·8 SVL-5† 0·622 0·747 1·647 2·945 0·559 Y 740 760 0·3 0·64–0·75 –0·8 –0·32 –0·6 SVL-7† 0·686 0·726 2·188 2·656 0·824 Y 740 760 0·3 0·53–0·60 –2 –0·8 –1·4 AOR-83‐1† 0·527 0·688 1·114 2·208 0·504 Y 680 700 0·3 0·95–1·0 0·55 0·75 0·7 ARU-83‐1A† 0·520 0·663 1·084 1·967 0·551 Y 680 700 0·3 0·95–1·0 0·5 0·62 0·6 ASE-83‐1† 0·539 0·667 1·171 2·006 0·584 Y 680 700 0·3 0·95–1·0 0·45 0·63 0·5 AMC-83‐1† 0·522 0·695 1·093 2·277 0·480 Y 680 700 0·3 0·95–1·0 0·5 0·82 0·7 AMC-83‐2† 0·581 0·699 1·387 2·324 0·597 Y 680 700 0·3 0·95–1·0 0·05 0·55 0·3 FeT/(FeT + Mg) FeT/Mg Temperature (°C) Calculated oxygen fugacity (ΔNNO) Sample Biotite Whole- rock Biotite Whole- rock Kd FeT/ Mg alpha MELTS modeling Min. Max. Pressure (GPa) aH2O* Min. Max. Av. Archean samples E19‐8 0·594 0·614 1·463 1·594 0·918 N — — — — — –2 SP-16‐20a 0·625 0·653 1·671 1·882 0·888 N — — — — — –2 SP-16‐22 0·614 0·650 1·592 1·854 0·859 N — — — — — –2 SP-16‐28a 0·703 0·711 2·372 2·455 0·966 N — — — — — –2 SP-16‐23 0·635 0·638 1·742 1·761 0·989 N — — — — — –2 SP-16‐24 0·627 0·645 1·682 1·810 0·929 N — — — — — –2 SP-16‐34 0·685 0·626 2·174 1·669 1·302 N — — — — — –2 SP-16‐48 0·613 0·613 1·583 1·574 1·005 N — — — — — –2 SP-16‐52† 0·705 0·764 2·396 3·228 0·742 Y 680 700 0·3 1 –0·5 0·6 0·1 SP-16‐53 0·719 0·714 2·560 2·496 1·026 N — — — — — –2 SP-16‐1† 0·751 0·810 3·027 4·272 0·708 Y 680 700 0·2 1 0·0 0·8 0·4 SP-16‐2b 0·706 0·706 2·405 2·403 1·000 N — — — — — –2 LC-8 0·726 0·659 2·654 1·937 1·370 N — — — — — –2 LC-16 0·726 0·679 2·646 2·118 1·249 N — — — — — –2 LC-30 0·759 0·679 3·156 2·113 1·494 N — — — — — –2 903 0·691 0·689 2·240 2·219 1·010 N — — — — — –2 768 0·746 0·742 2·936 2·882 1·019 N — — — — — –2 797†‡ 0·826 0·868 4·769 6·556 0·727 N 908 0·779 0·767 3·528 3·293 1·071 N — — — — — –2 616† 0·740 0·769 2·845 3·322 0·856 Y 680 700 0·5 0·9 –2 0·2 –0·9 634† 0·762 0·799 3·211 3·967 0·809 Y 680 700 0·5 0·9 <–2 0·87 –0·6 652† 0·812 0·852 4·332 5·749 0·754 Y 680 700 0·5 0·9 <–2 –1·4 –1·7 672 0·825 0·822 4·720 4·611 1·024 N — — — — — –2 692 0·789 0·731 3·741 2·720 1·375 N — — — — — –2 707† 0·821 0·874 4·586 6·959 0·659 Y 680 700 0·5 0·9 –2 0·5 –0·8 26 0·770 0·755 3·351 3·082 1·087 N — — — — — –2 29 0·689 0·684 2·210 2·164 1·022 N — — — — — –2 98TI 0·806 0·774 4·170 3·422 1·219 N — — — — — –2 Proterozoic samples HP-6B† 0·647 0·692 1·833 2·244 0·817 Y 680 700 0·3 0·95 <–2 0·4 –0·8 HP-7B† 0·621 0·755 1·639 3·085 0·531 Y 680 700 0·3 0·95–1·00 0 0·9 0·5 HP-21C† 0·627 0·675 1·684 2·080 0·810 Y 680 700 0·3 0·9–1·0 –2 0·5 –0·8 HP-26 0·664 0·685 1·981 2·178 0·910 N — — — — — –2 HP 32 0·758 0·745 3·132 2·927 1·070 N — — — — — –2 HP 43A†§ 0·644 0·872 1·808 6·818 0·265 N HP 44A† 0·646 0·794 1·826 3·864 0·473 Y 680 700 0·3 0·97–1·00 0·850 1·38 1·1 HP-23A† 0·674 0·721 2·067 2·580 0·801 Y 680 700 0·3 0·980 <–2 –1·1 –1·6 HP-14A 0·862 0·849 6·233 5·610 1·111 N — — — — — –2 L14 0·635 0·656 1·739 1·906 0·912 N — — — — — –2 L99 0·637 0·639 1·756 1·772 0·991 N — — — — — –2 L125 0·782 0·741 3·583 2·867 1·250 N — — — — — –2 L204† 0·695 0·732 2·281 2·729 0·836 Y 680 750 0·3 0·7–0·85 –2 0·55 –0·7 L331 0·682 0·703 2·144 2·366 0·906 N — — — — — –2 L341A 0·658 0·680 1·927 2·126 0·906 N — — — — — –2 S29† 0·631 0·667 1·709 2·006 0·852 Y 680 750 0·3 0·67–0·96 –2 0·1 –1·0 S54 0·628 0·654 1·758 1·893 0·929 N — — — — — –2 SP-1† 0·670 0·704 2·034 2·376 0·856 Y 740 760 0·3 0·66 <–2 –0·64 –0·6 SP-3† 0·559 0·693 1·270 2·261 0·562 Y 740 760 0·3 0·66–0·77 –0·5 0·05 –0·2 SP-5† 0·550 0·683 1·223 2·155 0·568 Y 740 760 0·3 0·66–0·77 –0·58 –0·2 –0·4 SP-6† 0·547 0·718 1·210 2·543 0·476 Y 740 760 0·3 0·66–0·78 –0·5 0·05 –0·2 SP-8† 0·562 0·689 1·284 2·214 0·580 Y 740 760 0·3 0·65–0·77 –1·1 0·18 –0·5 SVL-3† 0·678 0·724 2·108 2·622 0·804 Y 740 760 0·3 0·53–0·76 –2 –0·7 –1·4 SVL-4† 0·633 0·717 1·724 2·536 0·680 Y 740 760 0·3 0·60–0·77 –1 –0·5 –0·8 SVL-5† 0·622 0·747 1·647 2·945 0·559 Y 740 760 0·3 0·64–0·75 –0·8 –0·32 –0·6 SVL-7† 0·686 0·726 2·188 2·656 0·824 Y 740 760 0·3 0·53–0·60 –2 –0·8 –1·4 AOR-83‐1† 0·527 0·688 1·114 2·208 0·504 Y 680 700 0·3 0·95–1·0 0·55 0·75 0·7 ARU-83‐1A† 0·520 0·663 1·084 1·967 0·551 Y 680 700 0·3 0·95–1·0 0·5 0·62 0·6 ASE-83‐1† 0·539 0·667 1·171 2·006 0·584 Y 680 700 0·3 0·95–1·0 0·45 0·63 0·5 AMC-83‐1† 0·522 0·695 1·093 2·277 0·480 Y 680 700 0·3 0·95–1·0 0·5 0·82 0·7 AMC-83‐2† 0·581 0·699 1·387 2·324 0·597 Y 680 700 0·3 0·95–1·0 0·05 0·55 0·3 * FeT/Mgbiotite increases with crystallization during model run. The lower end of the range given indicates value at the maximum temperature and the higher end of the range given indicates value the minimum temperature. †Samples modeled using alphaMELTS. Y, yes; N, no. ‡alphaMELTS unable to reproduce biotite composition (possibly owing to very low MgO contents). §alphaMELTS unable to reproduce biotite composition. Table 4 Summary of alphaMELTS modeling parameters and results FeT/(FeT + Mg) FeT/Mg Temperature (°C) Calculated oxygen fugacity (ΔNNO) Sample Biotite Whole- rock Biotite Whole- rock Kd FeT/ Mg alpha MELTS modeling Min. Max. Pressure (GPa) aH2O* Min. Max. Av. Archean samples E19‐8 0·594 0·614 1·463 1·594 0·918 N — — — — — –2 SP-16‐20a 0·625 0·653 1·671 1·882 0·888 N — — — — — –2 SP-16‐22 0·614 0·650 1·592 1·854 0·859 N — — — — — –2 SP-16‐28a 0·703 0·711 2·372 2·455 0·966 N — — — — — –2 SP-16‐23 0·635 0·638 1·742 1·761 0·989 N — — — — — –2 SP-16‐24 0·627 0·645 1·682 1·810 0·929 N — — — — — –2 SP-16‐34 0·685 0·626 2·174 1·669 1·302 N — — — — — –2 SP-16‐48 0·613 0·613 1·583 1·574 1·005 N — — — — — –2 SP-16‐52† 0·705 0·764 2·396 3·228 0·742 Y 680 700 0·3 1 –0·5 0·6 0·1 SP-16‐53 0·719 0·714 2·560 2·496 1·026 N — — — — — –2 SP-16‐1† 0·751 0·810 3·027 4·272 0·708 Y 680 700 0·2 1 0·0 0·8 0·4 SP-16‐2b 0·706 0·706 2·405 2·403 1·000 N — — — — — –2 LC-8 0·726 0·659 2·654 1·937 1·370 N — — — — — –2 LC-16 0·726 0·679 2·646 2·118 1·249 N — — — — — –2 LC-30 0·759 0·679 3·156 2·113 1·494 N — — — — — –2 903 0·691 0·689 2·240 2·219 1·010 N — — — — — –2 768 0·746 0·742 2·936 2·882 1·019 N — — — — — –2 797†‡ 0·826 0·868 4·769 6·556 0·727 N 908 0·779 0·767 3·528 3·293 1·071 N — — — — — –2 616† 0·740 0·769 2·845 3·322 0·856 Y 680 700 0·5 0·9 –2 0·2 –0·9 634† 0·762 0·799 3·211 3·967 0·809 Y 680 700 0·5 0·9 <–2 0·87 –0·6 652† 0·812 0·852 4·332 5·749 0·754 Y 680 700 0·5 0·9 <–2 –1·4 –1·7 672 0·825 0·822 4·720 4·611 1·024 N — — — — — –2 692 0·789 0·731 3·741 2·720 1·375 N — — — — — –2 707† 0·821 0·874 4·586 6·959 0·659 Y 680 700 0·5 0·9 –2 0·5 –0·8 26 0·770 0·755 3·351 3·082 1·087 N — — — — — –2 29 0·689 0·684 2·210 2·164 1·022 N — — — — — –2 98TI 0·806 0·774 4·170 3·422 1·219 N — — — — — –2 Proterozoic samples HP-6B† 0·647 0·692 1·833 2·244 0·817 Y 680 700 0·3 0·95 <–2 0·4 –0·8 HP-7B† 0·621 0·755 1·639 3·085 0·531 Y 680 700 0·3 0·95–1·00 0 0·9 0·5 HP-21C† 0·627 0·675 1·684 2·080 0·810 Y 680 700 0·3 0·9–1·0 –2 0·5 –0·8 HP-26 0·664 0·685 1·981 2·178 0·910 N — — — — — –2 HP 32 0·758 0·745 3·132 2·927 1·070 N — — — — — –2 HP 43A†§ 0·644 0·872 1·808 6·818 0·265 N HP 44A† 0·646 0·794 1·826 3·864 0·473 Y 680 700 0·3 0·97–1·00 0·850 1·38 1·1 HP-23A† 0·674 0·721 2·067 2·580 0·801 Y 680 700 0·3 0·980 <–2 –1·1 –1·6 HP-14A 0·862 0·849 6·233 5·610 1·111 N — — — — — –2 L14 0·635 0·656 1·739 1·906 0·912 N — — — — — –2 L99 0·637 0·639 1·756 1·772 0·991 N — — — — — –2 L125 0·782 0·741 3·583 2·867 1·250 N — — — — — –2 L204† 0·695 0·732 2·281 2·729 0·836 Y 680 750 0·3 0·7–0·85 –2 0·55 –0·7 L331 0·682 0·703 2·144 2·366 0·906 N — — — — — –2 L341A 0·658 0·680 1·927 2·126 0·906 N — — — — — –2 S29† 0·631 0·667 1·709 2·006 0·852 Y 680 750 0·3 0·67–0·96 –2 0·1 –1·0 S54 0·628 0·654 1·758 1·893 0·929 N — — — — — –2 SP-1† 0·670 0·704 2·034 2·376 0·856 Y 740 760 0·3 0·66 <–2 –0·64 –0·6 SP-3† 0·559 0·693 1·270 2·261 0·562 Y 740 760 0·3 0·66–0·77 –0·5 0·05 –0·2 SP-5† 0·550 0·683 1·223 2·155 0·568 Y 740 760 0·3 0·66–0·77 –0·58 –0·2 –0·4 SP-6† 0·547 0·718 1·210 2·543 0·476 Y 740 760 0·3 0·66–0·78 –0·5 0·05 –0·2 SP-8† 0·562 0·689 1·284 2·214 0·580 Y 740 760 0·3 0·65–0·77 –1·1 0·18 –0·5 SVL-3† 0·678 0·724 2·108 2·622 0·804 Y 740 760 0·3 0·53–0·76 –2 –0·7 –1·4 SVL-4† 0·633 0·717 1·724 2·536 0·680 Y 740 760 0·3 0·60–0·77 –1 –0·5 –0·8 SVL-5† 0·622 0·747 1·647 2·945 0·559 Y 740 760 0·3 0·64–0·75 –0·8 –0·32 –0·6 SVL-7† 0·686 0·726 2·188 2·656 0·824 Y 740 760 0·3 0·53–0·60 –2 –0·8 –1·4 AOR-83‐1† 0·527 0·688 1·114 2·208 0·504 Y 680 700 0·3 0·95–1·0 0·55 0·75 0·7 ARU-83‐1A† 0·520 0·663 1·084 1·967 0·551 Y 680 700 0·3 0·95–1·0 0·5 0·62 0·6 ASE-83‐1† 0·539 0·667 1·171 2·006 0·584 Y 680 700 0·3 0·95–1·0 0·45 0·63 0·5 AMC-83‐1† 0·522 0·695 1·093 2·277 0·480 Y 680 700 0·3 0·95–1·0 0·5 0·82 0·7 AMC-83‐2† 0·581 0·699 1·387 2·324 0·597 Y 680 700 0·3 0·95–1·0 0·05 0·55 0·3 FeT/(FeT + Mg) FeT/Mg Temperature (°C) Calculated oxygen fugacity (ΔNNO) Sample Biotite Whole- rock Biotite Whole- rock Kd FeT/ Mg alpha MELTS modeling Min. Max. Pressure (GPa) aH2O* Min. Max. Av. Archean samples E19‐8 0·594 0·614 1·463 1·594 0·918 N — — — — — –2 SP-16‐20a 0·625 0·653 1·671 1·882 0·888 N — — — — — –2 SP-16‐22 0·614 0·650 1·592 1·854 0·859 N — — — — — –2 SP-16‐28a 0·703 0·711 2·372 2·455 0·966 N — — — — — –2 SP-16‐23 0·635 0·638 1·742 1·761 0·989 N — — — — — –2 SP-16‐24 0·627 0·645 1·682 1·810 0·929 N — — — — — –2 SP-16‐34 0·685 0·626 2·174 1·669 1·302 N — — — — — –2 SP-16‐48 0·613 0·613 1·583 1·574 1·005 N — — — — — –2 SP-16‐52† 0·705 0·764 2·396 3·228 0·742 Y 680 700 0·3 1 –0·5 0·6 0·1 SP-16‐53 0·719 0·714 2·560 2·496 1·026 N — — — — — –2 SP-16‐1† 0·751 0·810 3·027 4·272 0·708 Y 680 700 0·2 1 0·0 0·8 0·4 SP-16‐2b 0·706 0·706 2·405 2·403 1·000 N — — — — — –2 LC-8 0·726 0·659 2·654 1·937 1·370 N — — — — — –2 LC-16 0·726 0·679 2·646 2·118 1·249 N — — — — — –2 LC-30 0·759 0·679 3·156 2·113 1·494 N — — — — — –2 903 0·691 0·689 2·240 2·219 1·010 N — — — — — –2 768 0·746 0·742 2·936 2·882 1·019 N — — — — — –2 797†‡ 0·826 0·868 4·769 6·556 0·727 N 908 0·779 0·767 3·528 3·293 1·071 N — — — — — –2 616† 0·740 0·769 2·845 3·322 0·856 Y 680 700 0·5 0·9 –2 0·2 –0·9 634† 0·762 0·799 3·211 3·967 0·809 Y 680 700 0·5 0·9 <–2 0·87 –0·6 652† 0·812 0·852 4·332 5·749 0·754 Y 680 700 0·5 0·9 <–2 –1·4 –1·7 672 0·825 0·822 4·720 4·611 1·024 N — — — — — –2 692 0·789 0·731 3·741 2·720 1·375 N — — — — — –2 707† 0·821 0·874 4·586 6·959 0·659 Y 680 700 0·5 0·9 –2 0·5 –0·8 26 0·770 0·755 3·351 3·082 1·087 N — — — — — –2 29 0·689 0·684 2·210 2·164 1·022 N — — — — — –2 98TI 0·806 0·774 4·170 3·422 1·219 N — — — — — –2 Proterozoic samples HP-6B† 0·647 0·692 1·833 2·244 0·817 Y 680 700 0·3 0·95 <–2 0·4 –0·8 HP-7B† 0·621 0·755 1·639 3·085 0·531 Y 680 700 0·3 0·95–1·00 0 0·9 0·5 HP-21C† 0·627 0·675 1·684 2·080 0·810 Y 680 700 0·3 0·9–1·0 –2 0·5 –0·8 HP-26 0·664 0·685 1·981 2·178 0·910 N — — — — — –2 HP 32 0·758 0·745 3·132 2·927 1·070 N — — — — — –2 HP 43A†§ 0·644 0·872 1·808 6·818 0·265 N HP 44A† 0·646 0·794 1·826 3·864 0·473 Y 680 700 0·3 0·97–1·00 0·850 1·38 1·1 HP-23A† 0·674 0·721 2·067 2·580 0·801 Y 680 700 0·3 0·980 <–2 –1·1 –1·6 HP-14A 0·862 0·849 6·233 5·610 1·111 N — — — — — –2 L14 0·635 0·656 1·739 1·906 0·912 N — — — — — –2 L99 0·637 0·639 1·756 1·772 0·991 N — — — — — –2 L125 0·782 0·741 3·583 2·867 1·250 N — — — — — –2 L204† 0·695 0·732 2·281 2·729 0·836 Y 680 750 0·3 0·7–0·85 –2 0·55 –0·7 L331 0·682 0·703 2·144 2·366 0·906 N — — — — — –2 L341A 0·658 0·680 1·927 2·126 0·906 N — — — — — –2 S29† 0·631 0·667 1·709 2·006 0·852 Y 680 750 0·3 0·67–0·96 –2 0·1 –1·0 S54 0·628 0·654 1·758 1·893 0·929 N — — — — — –2 SP-1† 0·670 0·704 2·034 2·376 0·856 Y 740 760 0·3 0·66 <–2 –0·64 –0·6 SP-3† 0·559 0·693 1·270 2·261 0·562 Y 740 760 0·3 0·66–0·77 –0·5 0·05 –0·2 SP-5† 0·550 0·683 1·223 2·155 0·568 Y 740 760 0·3 0·66–0·77 –0·58 –0·2 –0·4 SP-6† 0·547 0·718 1·210 2·543 0·476 Y 740 760 0·3 0·66–0·78 –0·5 0·05 –0·2 SP-8† 0·562 0·689 1·284 2·214 0·580 Y 740 760 0·3 0·65–0·77 –1·1 0·18 –0·5 SVL-3† 0·678 0·724 2·108 2·622 0·804 Y 740 760 0·3 0·53–0·76 –2 –0·7 –1·4 SVL-4† 0·633 0·717 1·724 2·536 0·680 Y 740 760 0·3 0·60–0·77 –1 –0·5 –0·8 SVL-5† 0·622 0·747 1·647 2·945 0·559 Y 740 760 0·3 0·64–0·75 –0·8 –0·32 –0·6 SVL-7† 0·686 0·726 2·188 2·656 0·824 Y 740 760 0·3 0·53–0·60 –2 –0·8 –1·4 AOR-83‐1† 0·527 0·688 1·114 2·208 0·504 Y 680 700 0·3 0·95–1·0 0·55 0·75 0·7 ARU-83‐1A† 0·520 0·663 1·084 1·967 0·551 Y 680 700 0·3 0·95–1·0 0·5 0·62 0·6 ASE-83‐1† 0·539 0·667 1·171 2·006 0·584 Y 680 700 0·3 0·95–1·0 0·45 0·63 0·5 AMC-83‐1† 0·522 0·695 1·093 2·277 0·480 Y 680 700 0·3 0·95–1·0 0·5 0·82 0·7 AMC-83‐2† 0·581 0·699 1·387 2·324 0·597 Y 680 700 0·3 0·95–1·0 0·05 0·55 0·3 * FeT/Mgbiotite increases with crystallization during model run. The lower end of the range given indicates value at the maximum temperature and the higher end of the range given indicates value the minimum temperature. †Samples modeled using alphaMELTS. Y, yes; N, no. ‡alphaMELTS unable to reproduce biotite composition (possibly owing to very low MgO contents). §alphaMELTS unable to reproduce biotite composition. We divided the samples into three groups based on Kbt–bulkFeT/Mg for modeling, as follows. (1) Samples with Kbt–bulkFeT/Mg < ∼0·9. These samples were amenable to fO2 modeling as calculated biotite FeT/(FeT + Mg) ratios are sensitive to variations in fO2 in alphaMELTS. This group includes 27 (46%) of the total samples. (2) Samples with 0·9 <  Kbt–bulkFeT/Mg≤ 1. For these samples, biotite FeT/(FeT + Mg) ratios in alphaMELTS are not sufficiently sensitive to fO2 variations at the low fO2 required to produce Kbt–bulkFeT/Mg > 0·9, as the majority of the Fe in the bulk system is Fe2+. For these samples, we assume that they crystallized at an fO2≤ NNO –2, although specific fO2 values for each sample could not be calculated. This group includes 13 (22%) of total samples. (3) Samples with Kbt–bulkFeT/Mg > 1. This group includes 14 Archean and three Proterozoic samples, and based on Fig. 9b, showing data from the literature, it appears to be a common feature among more reduced granites. The observed Kbt–bulkFeT/Mg > 1 is inconsistent with alphaMELTS modeling and no constraint on fO2 using our approach is possible for samples displaying this feature. It may be explained in two ways. (1) If homogeneous biotite were the only ferromagnesian phase in a rock, the Kbt–bulkFeT/Mg would be exactly unity. Thus, for Kbt–bulkFeT/Mg to be greater than unity, as it is in some of the SPGs, there must be one or more phases with lower FeT/Mg than the biotite, and such a phase or phases must have sufficient MgO and be present in the whole-rock in sufficient quantities to decrease the denominator in the Kbt–bulkFeT/Mg expression so as to lead to a Kbt–bulkFeT/Mg greater than unity. Or (2) the whole-rock or biotite core analyses—or both—in these samples are in error or are not representative of the sample or the partially molten system from which the biotite crystallized. Regarding the first possibility, other Fe–Mg-bearing phases in the rocks include garnet, white mica, tourmaline, and chlorite. In the SPG samples in this study garnet is always more Fe-rich than biotite with Kgrt–btFeT/Mg  = 4·8–8·2, and therefore does not satisfy the criterion for a Mg-rich mineral. White mica hosts only a small fraction of the Fe and Mg budget of the whole-rocks (with FeO and MgO concentrations of white micas between 1·0 and 6·8 wt % and 0·2 and 2·3 wt %, respectively) and is not present in sufficient abundances compared with biotite, which has significantly higher FeO and MgO concentrations, to offset the high Kbt–bulkFeT/Mg > 1. Tourmaline is present in one sample that has Kbt–bulkFeT/Mg > 1 (SP-16–34). In SP-16–34, tourmaline contains 8·1 ± 0·4 wt % FeO and 4·7 ± 0·3 wt % MgO, with an average FeT/(FeT + Mg) of 0·49 ± 0·03 (values given are the average of 18 analyses with 2σSD). The tourmaline’s Mg-rich nature compared with biotite [FeT/(FeT + Mg) = 0·68] and presence in equivalent modal abundances to biotite in SP-16–34, might help to explain the Kbt–bulkFeT/Mg = 1·3, but SP-16–34 is the only sample with Kbt–bulkFeT/Mg > 1 where tourmaline is present. Chlorite alteration is present in many samples; however, chlorite, when observed and analyzed, either has FeT/(FeT + Mg) similar to biotite or higher by ∼0·1. Therefore, we do not believe that the presence of an Mg-rich phase (either primary magmatic or produced during subsolidus alteration) in the SPGs with Kbt–bulkFeT/Mg > 1 is responsible for the bulk distribution greater than unity. The second possibility, that the FeO or MgO analyses in the biotite or whole-rock may not be accurate or representative of the bulk sample, can potentially explain the observation of samples with Kbt–bulkFeT/Mg > 1. For instance, samples with Kbt–bulkFeT/Mg > 1 generally have low MgO (<0·3 wt %; Fig. 12a) and FeO (<1·6 wt %) concentrations relative to other samples considered here. It should be noted that biotite FeT/(FeT + Mg) ratios should be less susceptible to small errors in major element analyses, as FeO (18–19 wt %) and MgO (2·3–10 wt %) concentrations are relatively large for the studied samples. On the other hand, small errors in FeO and, particularly, MgO concentrations in whole-rock compositions can significantly affect FeT/(FeT + Mg) and FeT/Mg ratios. Let us consider a sample with measured whole-rock concentrations of 1·0 wt % FeO and 0·2 wt % MgO and a biotite FeT/(FeT + Mg) ratio of 0·71. This sample has a whole-rock FeT/(FeT + Mg) ratio of 0·74 and a Kbt–bulkFeT/Mg of 0·89. If the MgO concentration in the whole-rock is shifted up by 0·04 wt % (e.g. owing to small changes in bulk composition as a result of alteration or analytical error) the whole-rock FeT/(FeT + Mg) ratio will decrease to 0·70 (Fig. 12b) and the Kbt–bulkFeT/Mg will increase to 1·07 (Fig. 12c). Unfortunately, for all the samples with Kbt–bulkFeT/Mg > 1 (except tourmaline-bearing SP-16–34), whole-rock compositions were compiled from the literature (see Table 1 for sources) with no analytical uncertainties reported on FeO or MgO values. Therefore, it is difficult for us to assess the accuracy or precision associated with those whole-rock analyses and to assess whether the Kbt–bulkFeT/Mg > 1 may be attributed to these uncertainties. However, as a specific example, the samples from the Lacorne block (Abitibi subprovince, Superior Province) demonstrate a negative correlation between Kbt–bulkFeT/Mg and whole-rock MgO (wt %) (blue circles in Fig. 12a), which plausibly suggests that small variations in the MgO content of the bulk samples at minor to trace amounts (<0·3 wt %) may cause some samples to exhibit Kbt–bulkFeT/Mg > 1. Fig. 12. View largeDownload slide Whole-rock FeT/(FeT + Mg) and Kbt–bulkFeT/Mg vs MgO. (a) Kbt–bulkFeT/Mg vs whole-rock MgO (wt %) for SPG samples from this study. It should be noted that the majority of samples with Kbt–bulkFeT/Mg >1 also have low whole-rock MgO values. (b) Modeled whole-rock FeT/(FeT + Mg) ratios with variations in whole-rock MgO (wt %) at different fixed whole-rock FeO (wt %) values. (c) Modeled Kbt–bulkFeT/Mg with variation in whole-rock MgO at different fixed whole-rock FeO (wt %) values. Biotite is assumed to have FeT/(FeT + Mg) = 0·71 and FeT/Mg of 2·5. An example, as described in the text, is given in (b) and (c), where whole-rock MgO increases by 0·02, resulting in a decrease in the whole-rock FeT/(FeT + Mg) ratio of 0·04 and an increase in the Kbt–bulkFeT/Mg to a value >1·0. Fig. 12. View largeDownload slide Whole-rock FeT/(FeT + Mg) and Kbt–bulkFeT/Mg vs MgO. (a) Kbt–bulkFeT/Mg vs whole-rock MgO (wt %) for SPG samples from this study. It should be noted that the majority of samples with Kbt–bulkFeT/Mg >1 also have low whole-rock MgO values. (b) Modeled whole-rock FeT/(FeT + Mg) ratios with variations in whole-rock MgO (wt %) at different fixed whole-rock FeO (wt %) values. (c) Modeled Kbt–bulkFeT/Mg with variation in whole-rock MgO at different fixed whole-rock FeO (wt %) values. Biotite is assumed to have FeT/(FeT + Mg) = 0·71 and FeT/Mg of 2·5. An example, as described in the text, is given in (b) and (c), where whole-rock MgO increases by 0·02, resulting in a decrease in the whole-rock FeT/(FeT + Mg) ratio of 0·04 and an increase in the Kbt–bulkFeT/Mg to a value >1·0. Lastly, SPGs are commonly coarsely crystalline, which requires that large volumes of samples be processed to obtain a representative analysis of the bulk-rock. Major elements present in low abundances (e.g. MgO) are particularly sensitive to the aliquot of sample used in the whole-rock analysis. The observed Kbt–bulkFeT/Mg > 1 in some samples may therefore also be due to the choice of whole-rock aliquots that did not accurately represent the bulk-rock compositions. Although we were unable to model the samples with Kbt–bulkFeT/Mg > 1 using alphaMELTS, their biotite FeT/(FeT + Mg) ratios of 0·59–0·85 suggest low biotite Fe3+/FeT ratios and therefore crystallization under reducing conditions (Fig. 9a). Further, other SPGs not studied that have Kbt–bulkFeT/Mg > 1 also have low biotite Fe3+/FeT ratios (Fig. 9b). Therefore, we broadly constrain the crystallization fO2 of these samples at ≤NNO –2. Calculated fO2s values of Archean and Proterozoic strongly peraluminous granites The Kbt–bulkFeT/Mg value and alphaMELTS calculated fO2 for each sample are given in Table 4 and plotted in Fig. 13. A first-order difference between Archean and Proterozoic samples is observed when comparing biotite FeT/(FeT + Mg) versus whole-rock FeT/(FeT + Mg) (Fig. 13a) or distributions of Kbt–bulkFeT/Mg (Fig. 13b). As compared with Archean samples, the majority of Proterozoic samples are shifted to lower biotite FeT/(FeT + Mg) at a given whole-rock FeT/(FeT + Mg) (Fig. 13a) and therefore have lower Kbt–bulkFeT/Mg on average as compared with Archean samples [0·74 ± 0·45 vs 1·01 ± 0·21 (2σ), Fig. 13b]. This difference in the partitioning of iron between biotite and the whole-rock for peraluminous Archean vs Proterozoic granites is a robust finding of our study and strongly suggests crystallization of the Proterozoic SPGs at lower fO2 values and/or higher T than the Archean SPGs. Although subject to the uncertainties emphasized above, quantitative modeling of fO2 using alphaMELTS confirms this. Of the 28 Archean SPGs studied, only seven were amenable to alphaMELTS modeling (i.e. Kbt–bulkFeT/Mg <0·9), but this subset of the samples we studied yielded calculated fO2 values between NNO –1·7 and NNO +1·0. However, although we could not assign specific fO2 values to the remainder of Archean SPGs (n = 21), and many have Kbt–bulkFeT/Mg >1, our overall assessment based on available experiments and alphaMELTS modeling is that the occurrence of such Fe-rich biotites is most consistent with an fO2 ≤ NNO –2. In contrast, 21 out 31 (68%) of Proterozoic SPGs had Kbt–bulkFeT/Mg <0·9 and thus were amenable to alphaMELTS fO2 modeling, yielding an fO2 between NNO –1·0 and NNO +1·1. Fig. 13. View largeDownload slide Summary of biotite and whole-rock compositions, Kbt–bulkFeT/Mg ⁠, and calculated fO2 for Archean and Proterozoic SPGs. (a) Biotite FeT/Mg vs whole-rock FeT/Mg for Archean and Proterozoic SPGs. Logarithmic axes should be noted. Error bars on biotite composition represent 2σSD on averages of biotite analyses for each sample. (b) Histograms of Kbt–bulkFeT/Mg for Archean and Proterozoic samples. Samples in lighter blue and red indicate those not amenable to modeling via alphaMELTS (i.e. Kbt–bulkFeT/Mg > 0·9). (c) alphaMELTS calculated log10(⁠ fO2 ) vs temperature for Archean and Proterozoic SPG samples. Error bars on symbols represent the bounds of maximum and minimum calculated fO2 (see text for discussion) and range of temperatures used in the calculations. Buffers are the same as in Fig. 7a (calculated at 0·3 GPa). In addition, the maximum stability of graphite in equilibrium with a C–O–H fluid is shown at 0·2, 0·3, 0·4, and 0·5 GPa. The continuous lines indicate maximum stability in a pure C–O fluid and the dashed lines indicate the maximum stability when the activity of H2O is at a maximum. Curves were calculated using the Perple_X thermodynamic software of Connolly (1990), implementing the modified Redlich–Kwong equation of state presented by Connolly & Cesare (1993). Also shown is the boundary between S2– and SO42– predominance of dissolved sulfur in andesitic to dacitic hydrous melts (after Carroll & Rutherford, 1988). (d) Histograms of average fO2 values for Archean and Proterozoic samples. When fO2 values could not be uniquely calculated using alphaMELTS, we plotted the sample at ΔNNO = –2, which represents an upper limit to our fO2 estimates. Fig. 13. View largeDownload slide Summary of biotite and whole-rock compositions, Kbt–bulkFeT/Mg ⁠, and calculated fO2 for Archean and Proterozoic SPGs. (a) Biotite FeT/Mg vs whole-rock FeT/Mg for Archean and Proterozoic SPGs. Logarithmic axes should be noted. Error bars on biotite composition represent 2σSD on averages of biotite analyses for each sample. (b) Histograms of Kbt–bulkFeT/Mg for Archean and Proterozoic samples. Samples in lighter blue and red indicate those not amenable to modeling via alphaMELTS (i.e. Kbt–bulkFeT/Mg > 0·9). (c) alphaMELTS calculated log10(⁠ fO2 ) vs temperature for Archean and Proterozoic SPG samples. Error bars on symbols represent the bounds of maximum and minimum calculated fO2 (see text for discussion) and range of temperatures used in the calculations. Buffers are the same as in Fig. 7a (calculated at 0·3 GPa). In addition, the maximum stability of graphite in equilibrium with a C–O–H fluid is shown at 0·2, 0·3, 0·4, and 0·5 GPa. The continuous lines indicate maximum stability in a pure C–O fluid and the dashed lines indicate the maximum stability when the activity of H2O is at a maximum. Curves were calculated using the Perple_X thermodynamic software of Connolly (1990), implementing the modified Redlich–Kwong equation of state presented by Connolly & Cesare (1993). Also shown is the boundary between S2– and SO42– predominance of dissolved sulfur in andesitic to dacitic hydrous melts (after Carroll & Rutherford, 1988). (d) Histograms of average fO2 values for Archean and Proterozoic samples. When fO2 values could not be uniquely calculated using alphaMELTS, we plotted the sample at ΔNNO = –2, which represents an upper limit to our fO2 estimates. In particular, among the Proterozoic samples, the SPGs of Arizona and Colorado have on average the lowest Kbt–bulkFeT/Mg (0·61 ± 0·25) and highest calculated fO2 values (ΔNNO = –0·2 ± 0·7). Supporting this assessment, Anderson & Thomas (1989) quantitatively constrained the T and fO2 of the Silver Plume and St Vrain granites at 740–760°C and 10–15·2 to 10–15·6 (NNO –0·1 to NNO –0·5) using Fe–Ti oxide thermometry and oxybarometry. In addition, the peraluminous granites of the SW USA are the only magnetite-bearing granites in this study, whereas most SPGs contain only ilmenite if any Fe–Ti oxide phase is present. For the 21 Archean and nine Proterozoic samples to which we assign an fO2 at or below NNO –2, this upper limit to fO2 corresponds to crystallization near the maximum stability of graphite in equilibrium with a C–O–H fluid in which the activity of H2O is at a maximum (although not equal to unity) over an array of upper- to mid-crustal pressures (0·2–0·5 GPa) (Fig. 13c). These upper limits to our fO2 estimates are consistent with water-rich granitic melts derived through partial melting of sediments containing reduced carbon (Ohmoto & Kerrick, 1977; Connolly & Cesare, 1993). Controls on the fO2 of Archean and Paleoproterozoic SPGs The fO2 of Phanerozoic SPGs or granites strongly contaminated by metasedimentary rocks is generally assumed to reflect the fO2 of their metasedimentary source region or assimilant, which is typically dominated by the presence of reduced carbon (Flood & Shaw, 1977; Wyborn et al., 1981; Ague & Brimhall, 1988a; Blevin & Chappell, 1992). Although post-segregation processes such as H degassing can drive oxidation of granitic magmas (Czamanske & Wones, 1973; Pichavant et al., 1996, 2016), we do not think that this is the primary control on the fO2 recorded in the Archean and Proterozoic SPGs, because oxidation of peraluminous leucogranite magmas is generally manifested by the destabilization of biotite and the crystallization of muscovite and tourmaline, which are favored at high fO2 values (Pichavant et al., 1996, 2016). All of the SPG samples considered in this study contain biotite, and samples from a single locality display similar calculated fO2 values (except for the Harney Peak granite, Fig. 13c), suggesting that degassing-related oxidation was not a primary factor controlling fO2 for most samples. Overall, the biotite geochemistry and calculated fO2 of granites from this study indicate that the sources of Archean SPGs were on average more reduced than their Proterozoic counterparts. Importantly, however, >80% of the Archean and ∼40% of the Proterozoic SPGs appear to have crystallized at a low fO2 with an upper limit consistent with a graphite-saturated C–O–H fluid (Fig. 13c and d), suggesting the possibility that the presence of reduced carbon in the sources of these particular samples controlled their redox budgets. This would be in agreement with the observation that distributions of total organic carbon are similar for both Archean and Proterozoic/Phanerozoic shales (Dimroth & Kimberley, 1976; Holland, 1984; Lyons et al., 2014). Sedimentary sequences accreted and buried during collisional orogenies are generally deposited in shallow water environments with high sedimentation rates, such as marginal or foreland basins or continental slope environments. Given that total organic carbon contents in these sediments are generally high (e.g. Müller & Suess, 1979; Calvert & Pedersen, 1992), it would not be surprising if reduced carbon exerted the dominant control on the redox budgets of many of the SPGs derived from burial and partial melting of such sediments. As a detailed example, the host Proterozoic schists and graywackes of the Harney Peak granite, which are probably representative of at least some of the source material for the granite (Nabelek et al., 1992a), contain disseminated graphite as a common accessory mineral (Huff & Nabelek, 2007). Although our results are only consistent with control of fO2 in the most reduced of our samples by graphite-bearing sources and/or graphite-saturated C–O–H-fluids—and we emphasize that not all of our samples are sufficiently reducing to have crystallized under such conditions—if the upper limit we have deduced for the fO2 of these samples were to be shown to be the actual value rather than the upper limit, it would reinforce the result from this and other studies that Proterozoic to Phanerozoic SPGs and peraluminous volcanic rocks are generally reduced relative to metaluminous granites and volcanic rocks and sometimes contain graphite from their metasedimentary source rocks (Chappell & White, 1992; Zeck, 1992; Lalonde & Bernard, 1993; Ishihara, 2004; Acosta-Vigil et al., 2007). Many of the Proterozoic SPGs (primarily those from the SW USA) and some of the Archean SPGs appear to have crystallized at an fO2 above that expected through derivation from a graphite-dominated sedimentary source rock (Fig. 13c and d), suggesting that at least in these samples, factors other than saturation with graphite or graphite-saturated fluids controlled the redox state of Fe in the magmas and their sedimentary source rocks. As described in the Introduction, the appearance of oxidized species (such as sulfate and/or Fe3+-bearing minerals) in sediments in the Proterozoic owing to the GOE could have offset the reducing power of organic carbon. In particular, partial melting of sedimentary sulfate, oxidized Fe-bearing clays, or Fe-oxides and/or hydroxides would be an effective way to increase the fO2 of the Proterozoic SPGs. Although relatively reduced SPGs from both before and after the GOE are present in our sample suite, they dominate among the Archean samples and are less abundant in the Proterozoic suite. This result is consistent with a role for the change in near-surface, atmospheric fO2 that occurred near the Archean–Proterozoic boundary in controlling the fO2 of SPGs, most probably through its effect on the oxidation states of Fe and/or S in their sedimentary sources. If sedimentary sulfate played a significant role in the oxidized sources of Proterozoic SPGs, this may be observable as elevated 34S/32S ratios in granites derived from such sources relative to those derived from sedimentary sources in which sulfide dominates (see review by Canfield, 2001). Ultimately, the fO2 that SPGs record arises from the aggregate oxidation state of their sedimentary sources, which may contain both reduced carbon-bearing and more oxidized sediments. The GOE may have served to enhance the proportion and extent of oxidized material in the sedimentary sources and thus may be responsible for the observed increase in the fO2 of some of the Proterozoic SPGs. CONCLUSIONS Biotite FeT/(FeT + Mg) ratios are sensitive to fO2 during crystallization of granitic melts. We have developed an approach using thermodynamic modeling based on the alphaMELTS software to constrain fO2 values of high-silica plutonic rocks, where biotite and whole-rock compositions are known and estimates of P, T, and aH2O can be obtained. Using this method to constrain the fO2 of crystallization of SPGs formed from the partial melting of sediments deposited before and after the GOE, we find that most of the Archean SPGs included in our study crystallized at an fO2 ≤ NNO –2, consistent with their fO2 having been buffered by graphite plus a C–O–H-rich fluid in their source regions. However, although some of the Paleoproterozoic SPGs we studied are also consistent with low fO2 values that reflect graphite + fluid saturation in their sources, most crystallized at higher fO2 (up to ΔNNO = +1·1). This observation can be placed into the context of the GOE, which is expressed in the sedimentary record by the onset of oxidative weathering dominated by an increase in the amount of minerals containing oxidized Fe and S. Overall, however, the observation that many SPGs on either side of the GOE (including Phanerozoic ones) are still sufficiently reduced to be near the maximum stability of graphite suggests that the oxidation state of their metasedimentary source rocks reflects a balance between the oxidation state and relative abundances of the various metasedimentary rocks in their sources. In the context of this hypothesis, the fO2 of the sources of Phanerozoic SPGs appears to have been dominated by the presence of a significant amount of organic material. The generally reduced nature of Archean SPGs plausibly reflects the presence of organic material and relatively reduced metasedimentary rocks prior to the GOE. Although most Proterozoic SPGs are sufficiently oxidized to possibly reflect the presence of post-GOE, oxidized metasedimentary rocks, there are some that are as reduced as the bulk of the Archean SPGs, which may be attributed to sufficient amounts of organic matter in their source regions to overwhelm the more oxidized nature of the metasedimentary rocks after the GOE. ACKNOWLEDGEMENTS We would like to thank Peter Nabelek for sharing samples and data for the Harney Peak granite, Carol Frost and Ron Frost for providing a sample of the Mount Owen batholith, and Lawford Anderson for providing original datasets of mineral and whole-rock data for the SW USA peraluminous granites. We thank Chi Ma for support with the electron microprobe analyses. Conversations with Oliver Shorttle on these topics were highly constructive and insightful. Three insightful reviews by Ron Frost, Bruno Scaillet, and Dan Frost helped to improve the paper. FUNDING This work was supported through an O. K. Earl Postdoctoral Fellowship awarded to C. Bucholz at Caltech. REFERENCES Abdel-Rahman A. F. M. ( 1994 ). Nature of biotites from alkaline, calc-alkaline, and peraluminous magmas . Journal of Petrology 35 , 525 – 541 . Google Scholar Crossref Search ADS Acosta-Vigil A. , Cesare B. , London D. , Morgan G. B. ( 2007 ). Microstructures and composition of melt inclusions in a crustal anatectic environment, represented by metapelitic enclaves within El Hoyazo dacites, SE Spain . Chemical Geology 237 , 450 – 465 . Google Scholar Crossref Search ADS Ague J. J. , Brimhall G. H. ( 1988a ). Magmatic arc asymmetry and distribution of anomalous plutonic belts in the batholiths of California: effects of assimilation, crustal thickness, and depth of crystallization . Geological Society of America Bulletin 100 , 912 – 927 . Google Scholar Crossref Search ADS Ague J. J. , Brimhall G. H. ( 1988b ). Regional variations in bulk chemistry, mineralogy, and the compositions of mafic and accessory minerals in the batholiths of California . Geological Society of America Bulletin 100 , 891 – 911 . Google Scholar Crossref Search ADS Anderson J. L. , Bender E. E. ( 1989 ). Nature and origin of Proterozoic A-type granitic magmatism in the southwestern United States of America . Lithos 23 , 19 – 52 . Google Scholar Crossref Search ADS Anderson J. L. , Morrison J. ( 2005 ). Ilmenite, magnetite, and peraluminous Mesoproterozoic anorogenic granites of Laurentia and Baltica . Lithos 80 , 45 – 60 . Google Scholar Crossref Search ADS Anderson J. L. , Rowley M. C. ( 1981 ). Synkinematic intrusion of peraluminous and associated metaluminous granitic magmas, Whipple Mountains, California . Canadian Mineralogist 19 , 83 – 101 . Anderson J. L. , Thomas W. M. ( 1985 ). Proterozoic anorogenic two-mica granites: Silver Plume and St. Vrain batholiths of Colorado . Geology 13 , 177 – 180 . Google Scholar Crossref Search ADS Armstrong J. T. ( 1995 ). CITZAF—a package of correction programs for the quantitative electron microbeam X-ray-analysis of thick polished materials, thin-films, and particles . Microbeam Analysis 4 , 177 – 200 . Bell E. A. , Boehnke P. , Harrison T. M. , Wielicki M. M. ( 2017 ). Mineral inclusion assemblage and detrital zircon provenance . Chemical Geology 477 , 151 – 160 . Google Scholar Crossref Search ADS Blackburn C. E. , Johns G. W. , Ayer J. , Davis D. W. ( 1991 ). The Wabigoon Subprovince . In: Thurston P. C. (ed.) Geology of Ontario . Ontario Geological Survey, Sudbury, ON, pp. 303–382 . Blackburn C. E. , Young J. B. , Breaks F. W. ( 2008 ). Precambrian geology of the Separation Lake greenstone belt—west sheet . Preliminary Map P.2673, Scale 1:20,000. Ontario Geological Survey, Sudbury, ON . Blevin P. L. , Chappell B. W. ( 1992 ). The role of magma sources, oxidation states and fractionation in determining the granite metallogeny of eastern Australia . Earth and Environmental Science Transactions of the Royal Society of Edinburgh 83 , 305 – 316 . Boerboom T. J. , Zartman R. E. ( 1993 ). Geology, geochemistry, and geochronology of the central Giants Range batholith, northeastern Minnesota . Canadian Journal of Earth Sciences 30 , 2510 – 2522 . Google Scholar Crossref Search ADS Bogaerts M. , Scaillet B. , Auwera J. V. ( 2006 ). Phase equilibria of the Lyngdal granodiorite (Norway): implications for the origin of metaluminous ferroan granitoids . Journal of Petrology 47 , 2405 – 2431 . Bowen N. L. ( 1928 ). The Evolution of Igneous Rocks . New York : Dover; Princeton, NJ: Princeton University Press . Breaks F. W. , Janes D. A. ( 1991 ). Granite-related mineralization of the Dryden Area, Superior Province of northwestern Ontario . Geological Association of Canada, Toronto'91 Organizing Committee. Sudbury, Ontario. 71 p. Field Trip B7 Guidebook . Breaks F. W. , Moore J. M. ( 1992 ). The Ghost Lake Batholith, Superior Province of northwestern Ontario: a fertile, S-type, peraluminous granite–rare-element pegmatite system . Canadian Mineralogist 30 , 835 – 875 . Breaks F. W. , Bond W. D. , Stone D. ( 1978 ). Preliminary geological synthesis of the English River subprovince: northwestern Ontario and its bearing upon mineral exploration. Ontario Geological Survey Miscellaneous Paper 72 , 55 pp. Breaks F. W. , Selway J. B. , Tindle A. G. ( 2005 ). Fertile peraluminous granites and related rare element pegmatites, Superior province of Ontario . In: Linnen R. L. , Samson I. M. (eds) Rare Element Geochemistry and Mineral Deposits. Geological Association of Canada, Short Course Notes 17, 87–125 . Brindley G. W. , Youell R. F. ( 1953 ). Ferrous chamosite and ferric chamosite . Mineralogical Magazine and Journal of the Mineralogical Society 30 , 57 – 70 . Google Scholar Crossref Search ADS Brounce M. N. , Kelley K. A. , Cottrell E. ( 2014 ). Variations in Fe3+/∑Fe of Mariana arc basalts and mantle wedge fO2 ⁠. Journal of Petrology 55 , 2513 – 2536 . Google Scholar Crossref Search ADS Brown L. , Klein J. , Middleton R. , Sacks I. S. , Tera F. ( 1982 ). 10Be in island-arc volcanoes and implications for subduction . Nature 299 , 718 – 720 . Google Scholar Crossref Search ADS Calvert S. E. , Pedersen T. F. ( 1992 ). Geochemistry of recent oxic and anoxic marine sediments: implications for the geological record . Marine Geology 113 , 67 – 88 . Google Scholar Crossref Search ADS Campion M. E. , Perkins D. , Roob C. ( 1986 ). Contrasting contact zones between the English River Subprovince of Ontario–Manitoba and greenstone belts to the north and south . Geological Society of America, Abstracts with Program 18 , 556 . Canfield D. E. ( 2001 ). Biogeochemistry of sulfur isotopes . In: Valley J. W. , Cole D. (eds) Stable Isotope Geochemistry. Mineralogical Society of America and Geochemical Society, Reviews in Mineralogy and Geochemistry 43 , 607 – 636 . Card K. D. ( 1990 ). A review of the Superior Province of the Canadian Shield, a product of Archean accretion . Precambrian Research 48 , 99 – 156 . Google Scholar Crossref Search ADS Card K. D. , Ciesielski A. ( 1986 ). Subdivisions of the Superior Province of the Canadian Shield . Geoscience Canada 13 , 5 – 13 . Carmichael I. S. E. ( 1991 ). The redox states of basic and silicic magmas: a reflection of their source regions? Contributions to Mineralogy and Petrology 106 , 129 – 141 . Google Scholar Crossref Search ADS Carroll M. R. , Rutherford M. J. ( 1988 ). Sulfur speciation in hydrous experimental glasses of varying oxidation state: results from measured wavelength shifts of sulfur X-rays . American Mineralogist 73 , 845 – 849 . Chandler F. W. ( 1988 ). Diagenesis of sabkha-related, sulphate nodules in the early Proterozoic Gordon Lake formation, Ontario, Canada . Carbonates and Evaporites 3 , 75 – 94 . Google Scholar Crossref Search ADS Chappell B. W. , White A. J. R. ( 1974 ). Two contrasting granite types . Pacific Geology 8 , 173 – 174 . Chappell B. W. , White A. J. R. ( 1992 ). I- and S-type granites in the Lachlan Fold Belt . Earth and Environmental Science Transactions of the Royal Society of Edinburgh 83 , 1 – 26 . Google Scholar Crossref Search ADS Clarke D. B. ( 1981 ). The mineralogy of peraluminous granites: a review . Canadian Mineralogist 19 , 3 – 17 . Clemens J. D. , Birch W. D. ( 2012 ). Assembly of a zoned volcanic magma chamber from multiple magma batches: the Cerberean Cauldron, Marysville Igneous Complex, Australia . Lithos 155 , 272 – 288 . Google Scholar Crossref Search ADS Clemens J. D. , Vielzeuf D. ( 1987 ). Constraints on melting and magma production in the crust . Earth and Planetary Science Letters 86 , 287 – 306 . Google Scholar Crossref Search ADS Clemens J. D. , Wall V. J. ( 1988 ). Controls on the mineralogy of S-type volcanic and plutonic rocks . Lithos 21 , 53 – 66 . Google Scholar Crossref Search ADS Clémente B. , Scaillet B. , Pichavant M. ( 2004 ). The solubility of sulphur in hydrous rhyolitic melts . Journal of Petrology 45 , 2171 – 2196 . Google Scholar Crossref Search ADS Collins W. J. ( 1996 ). Lachlan Fold Belt granitoids: products of three-component mixing . Earth and Environmental Science Transactions of the Royal Society of Edinburgh 87 , 171 – 181 . Google Scholar Crossref Search ADS Connolly J. A. D. ( 1990 ). Multivariable phase diagrams: an algorithm based on generalized thermodynamics . American Journal of Science 290 , 666 – 718 . Google Scholar Crossref Search ADS Connolly J. A. D. , Cesare B. ( 1993 ). C–O–H–S fluid composition and oxygen fugacity in graphitic metapelites . Journal of Metamorphic Geology 11 , 379 – 388 . Google Scholar Crossref Search ADS Corfu F. , Stott G. M. , Breaks F. W. ( 1995 ). U–Pb geochronology and evolution of the English River Subprovince, an Archean low P–high T metasedimentary belt in the Superior Province . Tectonics 14 , 1220 – 1233 . Google Scholar Crossref Search ADS Costa F. , Scaillet B. , Pichavant M. ( 2004 ). Petrological and experimental constraints on the pre-eruption conditions of Holocene dacite from Volcan San Pedro (36°S, Chilean Andes) and the importance of sulphur in silicic subduction-related magmas . Journal of Petrology 45 , 855 – 881 . Google Scholar Crossref Search ADS Czamanske G. K. , Wones D. R. ( 1973 ). Oxidation during magmatic differentiation, Finnmarka Complex, Oslo Area, Norway: Part 2, the mafic silicates . Journal of Petrology 14 , 349 – 380 . Google Scholar Crossref Search ADS Czamanske G. K. , Ishihara S. , Atkin S. A. ( 1981 ). Chemistry of rock-forming minerals of the Cretaceous–Paleocene batholith in southwestern Japan and implications for magma genesis . Journal of Geophysical Research: Solid Earth 86 , 10431 – 10469 . Google Scholar Crossref Search ADS Dahlquist J. A. , Galindo C. , Pankhurst R. J. , Rapela C. W. , Alasino P. H. , Saavedra J. , Fanning C. M. ( 2007 ). Magmatic evolution of the Peñón Rosado granite: petrogenesis of garnet-bearing granitoids . Lithos 95 , 177 – 207 . Google Scholar Crossref Search ADS Dall’Agnol R. , Scaillet B. , Pichavant M. ( 1999 ). An experimental study of a lower Proterozoic A-type granite from the eastern Amazonian Craton, Brazil . Journal of Petrology 40 , 1673 – 1698 . Google Scholar Crossref Search ADS Davis D. W. , Sutcliffe R. H. , Trowell N. F. ( 1988 ). Geochronological constraints on the tectonic evolution of a late Archaean greenstone belt, Wabigoon Subprovince, Northwest Ontario, Canada . Precambrian Research 39 , 171 – 191 . Google Scholar Crossref Search ADS Davis D. W. , Pezzutto F. , Ojakangas R. W. ( 1990 ). The age and provenance of metasedimentary rocks in the Quetico Subprovince, Ontario, from single zircon analyses: implications for Archean sedimentation and tectonics in the Superior Province . Earth and Planetary Science Letters 99 , 195 – 205 . Google Scholar Crossref Search ADS Dawson K. R. ( 1966 ). A comprehensive study of Preissac–Lacorne batholith . Geological Survey of Canada Bulletin 142 , 175 . Day W. C. , Weiblen P. W. ( 1986 ). Origin of late Archean granite: geochemical evidence from the Vermilion granitic complex of northern Minnesota . Contributions to Mineralogy and Petrology 93 , 283 – 296 . Google Scholar Crossref Search ADS de Albuquerque C. ( 1973 ). Geochemistry of biotites from granitic rocks, northern Portugal . Geochimica et Cosmochimica Acta 37 , 1779 – 1802 . Google Scholar Crossref Search ADS DePaolo D. J. ( 1981a ). Trace element and isotopic effects of combined wallrock assimilation and fractional crystallization . Earth and Planetary Science Letters 53 , 189 – 202 . Google Scholar Crossref Search ADS DePaolo D. J. ( 1981b ). Neodymium isotopes in the Colorado Front Range and crust–mantle evolution in the Proterozoic . Nature 291 , 193 – 196 . Google Scholar Crossref Search ADS Devaney J. R. , Williams H. R. ( 1989 ). Evolution of an Archean subprovince boundary: a sedimentological and structural study of part of the Wabigoon–Quetico boundary in northern Ontario . Canadian Journal of Earth Sciences 26 , 1013 – 1026 . Google Scholar Crossref Search ADS Dimroth E. , Kimberley M. M. ( 1976 ). Precambrian atmospheric oxygen: evidence in the sedimentary distributions of carbon, sulfur, uranium, and iron . Canadian Journal of Earth Sciences 13 , 1161 – 1185 . Google Scholar Crossref Search ADS Dodge F. C. W. , Smith V. C. , Mays R. E. ( 1969 ). Biotites from granitic rocks of the central Sierra Nevada batholith, California . Journal of Petrology 10 , 250 – 271 . Google Scholar Crossref Search ADS Du Bray E. A. ( 1988 ). Garnet compositions and their use as indicators of peraluminous granitoid petrogenesis—southeastern Arabian Shield . Contributions to Mineralogy and Petrology 100 , 205 – 212 . Google Scholar Crossref Search ADS Ducharme Y. , Stevenson R. K. , Machado N. ( 1997 ). Sm–Nd geochemistry and U–Pb geochronology of the Preissac and Lamotte leucogranites, Abitibi Subprovince . Canadian Journal of Earth Sciences 34 , 1059 – 1071 . Google Scholar Crossref Search ADS Duke E. F. , Papike J. J. , Laul J. C. ( 1992 ). Geochemistry of a boron-rich peraluminous granite pluton; the Calamity Peak layered granite–pegmatite complex, Black Hills, South Dakota . Canadian Mineralogist 30 , 811 – 833 . Dyar M. D. , Colucci M. T. , Guidotti C. V. ( 1991 ). Forgotten major elements: hydrogen and oxygen variation in biotite from metapelites . Geology 19 , 1029 – 1032 . Google Scholar Crossref Search ADS Ebadi A. , Johannes W. ( 1991 ). Beginning of melting and composition of first melts in the system Qz–Ab–Or–H2O–CO2 . Contributions to Mineralogy and Petrology 106 , 286 – 295 . Google Scholar Crossref Search ADS Eggleton R. A. , Banfield J. F. ( 1985 ). The alteration of granitic biotite to chlorite . American Mineralogist 70 , 902 – 910 . Farmer G. L. , DePaolo D. J. ( 1984 ). Origin of Mesozoic and Tertiary granite in the western United States and implications for Pre-Mesozoic crustal structure: 2. Nd and Sr isotopic studies of unmineralized and Cu- and Mo-mineralized granite in the Precambrian Craton . Journal of Geophysical Research: Solid Earth 89 , 10141 – 10160 . Google Scholar Crossref Search ADS Feldstein S. N. , Lange R. A. , Vennemann T. , O’Neil J. R. ( 1996 ). Ferric–ferrous ratios, H2O contents and D/H ratios of phlogopite and biotite from lavas of different tectonic regimes . Contributions to Mineralogy and Petrology 126 , 51 – 66 . Google Scholar Crossref Search ADS Feng R. ( 1992 ). Tectonic juxtaposition of the Archean Abitibi Greenstone Belt and Pontiac Subprovince: evidence from geobarometry, geochemistry, and Ar–Ar geochronology of metasedimentary rocks and granitoids . PhD dissertation, University of Saskachewan, Saskatoon . Feng R. , Kerrich R. ( 1990 ). Geobarometry, differential block movements, and crustal, structure of the southwestern Abitibi greenstone belt, Canada . Geology 18 , 870 – 873 . Google Scholar Crossref Search ADS Feng R. , Kerrich R. ( 1991 ). Single zircon age constraints on the tectonic juxtaposition of the Archean Abitibi greenstone belt and Pontiac subprovince, Quebec, Canada . Geochimica et Cosmochimica Acta 55 , 3437 – 3441 . Google Scholar Crossref Search ADS Feng R. , Kerrich R. ( 1992a ). Geochemical evolution of granitoids from the Archean Abitibi Southern Volcanic Zone and the Pontiac subprovince, Superior Province, Canada: implications for tectonic history and source regions . Chemical Geology 98 , 23 – 70 . Google Scholar Crossref Search ADS Feng R. , Kerrich R. ( 1992b ). Geodynamic evolution of the southern Abitibi and Pontiac terranes: evidence from geochemistry of granitoid magma series (2700–2630 Ma) . Canadian Journal of Earth Sciences 29 , 2266 – 2286 . Google Scholar Crossref Search ADS Feng R. , Kerrich R. , Maas R. ( 1993 ). Geochemical, oxygen, and neodymium isotope compositions of metasediments from the Abitibi greenstone belt and Pontiac Subprovince, Canada: evidence for ancient crust and Archean terrane juxtaposition . Geochimica et Cosmochimica Acta 57 , 641 – 658 . Google Scholar Crossref Search ADS Ferry J. M. ( 1979 ). Reaction mechanisms, physical conditions, and mass transfer during hydrothermal alteration of mica and feldspar in granitic rocks from south–central Maine, USA . Contributions to Mineralogy and Petrology 68 , 125 – 139 . Google Scholar Crossref Search ADS Flood R. H. , Shaw S. E. ( 1977 ). Two ‘S-type’ granite suites with low initial 87Sr/86Sr ratios from the New England Batholith, Australia . Contributions to Mineralogy and Petrology 61 , 163 – 173 . Google Scholar Crossref Search ADS French B. M. , Eugster H. P. ( 1965 ). Experimental control of oxygen fugacities by graphite–gas equilibriums . Journal of Geophysical Research 70 , 1529 – 1539 . Google Scholar Crossref Search ADS Frost B. R. , Frost C. D. ( 2008 ). A geochemical classification for feldspathic igneous rocks . Journal of Petrology 49 , 1955 – 1969 . Google Scholar Crossref Search ADS Frost C. D. , Frost B. R. ( 2011 ). On ferroan (A-type) granitoids: their compositional variability and modes of origin . Journal of Petrology 52 , 39 – 53 . Google Scholar Crossref Search ADS Frost B. R. , Barnes C. G. , Collins W. J. , Arculus R. J. , Ellis D. J. , Frost C. D. ( 2001 ). A geochemical classification for granitic rocks . Journal of Petrology 42 , 2033 – 2048 . Google Scholar Crossref Search ADS Frost B. R. , Frost C. D. , Cornia M. , Chamberlain K. R. , Kirkwood R. ( 2006 ). The Teton–Wind River domain: a 2·68–2·67 Ga active margin in the western Wyoming Province . Canadian Journal of Earth Sciences 43 , 1489 – 1510 . Google Scholar Crossref Search ADS Frost B. R. , Swapp S. M. , Frost C. D. , Bagdonas D. A. , Chamberlain K. R. ( 2018 ). Neoarchean tectonic history of the Teton Range: record of accretion against the present-day western margin of the Wyoming Province . Geosphere 14 , 1008 – 1030 . Google Scholar Crossref Search ADS Ghiorso M. S. , Evans B. W. ( 2008 ). Thermodynamics of rhombohedral oxide solid solutions and a revision of the Fe–Ti two-oxide geothermometer and oxygen-barometer . American Journal of Science 308 , 957 – 1039 . Google Scholar Crossref Search ADS Gilkes R. J. , Suddhiprakarn A. ( 1979 ). Biotite alteration in deeply weathered granite. I. Morphological, mineralogical, and chemical properties . Clays and Clay Minerals 27 , 349 – 360 . Google Scholar Crossref Search ADS Gilkes R. J. , Young R. C. , Quirk J. P. ( 1972 ). Oxidation of ferrous iron in biotite . Nature 236 , 89 – 91 . Grandstaff D. ( 1980 ). Origin of uraniferous conglomerates at Elliot Lake, Canada and Witwatersrand, South Africa: implications for oxygen in the Precambrian atmosphere . Precambrian Research 13 , 1 – 26 . Google Scholar Crossref Search ADS Gray C. M. ( 1984 ). An isotopic mixing model for the origin of granitic rocks in southeastern Australia . Earth and Planetary Science Letters 70 , 47 – 60 . Google Scholar Crossref Search ADS Gualda G. A. , Ghiorso M. S. , Lemons R. V. , Carley T. L. ( 2012 ). Rhyolite-MELTS: a modified calibration of MELTS optimized for silica-rich, fluid-bearing magmatic systems . Journal of Petrology 53 , 875 – 890 . Google Scholar Crossref Search ADS Guo Q. , Strauss H. , Kaufman A. J. , Schröder S. , Gutzmer J. , Wing B. , Baker M. A. , Bekker A. , Jin Q. , Kim S. T. , Farquhar J. ( 2009 ). Reconstructing Earth’s surface oxidation across the Archean–Proterozoic transition . Geology 37 , 399 – 402 . Google Scholar Crossref Search ADS Haslam H. W. ( 1968 ). The crystallization of intermediate and acid magmas at Ben Nevis, Scotland . Journal of Petrology 9 , 84 – 104 . Google Scholar Crossref Search ADS Healy B. , Collins W. J. , Richards S. W. ( 2004 ). A hybrid origin for Lachlan S-type granites: the Murrumbidgee Batholith example . Lithos 78 , 197 – 216 . Google Scholar Crossref Search ADS Helms T. S. , Labotka T. C. ( 1991 ). Petrogenesis of Early Proterozoic pelitic schists of the southern Black Hills, South Dakota: constraints on regional low-pressure metamorphism . Geological Society of America Bulletin 103 , 1324 – 1334 . Google Scholar Crossref Search ADS Hoffman P. F. ( 1988 ). United Plates of America, the birth of a craton; early Proterozoic assembly and growth of Laurentia . Annual Review of Earth and Planetary Sciences 16 , 543 – 603 . Google Scholar Crossref Search ADS Hoffman P. F. , Bowring S. A. ( 1984 ). Short-lived 1·9 Ga continental margin and its destruction, Wopmay orogen, northwest Canada . Geology 12 , 68 – 72 . Google Scholar Crossref Search ADS Holdaway M. J. ( 2000 ). Application of new experimental and garnet Margules data to the garnet–biotite geothermometer . American Mineralogist 85 , 881 – 892 . Google Scholar Crossref Search ADS Holland H. D. ( 1984 ). The Chemical Evolution of the Atmosphere and Ocean . Princeton, NJ : Princeton University Press . Holland H. D. ( 2006 ). The oxygenation of the atmosphere and oceans . Philosophical Transactions of the Royal Society of London, Series B 361 , 903 – 915 . Google Scholar Crossref Search ADS Holtz F. O. , Johannes W. ( 1991 ). Genesis of peraluminous granites I. Experimental investigation of melt compositions at 3 and 5 kb and various H2O activities . Journal of Petrology 32 , 935 – 958 . Google Scholar Crossref Search ADS Holtz F. , Johannes W. , Pichavant M. ( 1992 ). Effect of excess aluminium on phase relations in the system Qz–Ab–Or: experimental investigation at 2 kbar and reduced H2O-activity . European Journal of Mineralogy 4 , 137 – 152 . Google Scholar Crossref Search ADS Huebner J. S. ( 1975 ). Oxygen fugacity values of furnace gas mixtures . American Mineralogist 60 , 815 – 823 . Huff T. A. , Nabelek P. I. ( 2007 ). Production of carbonic fluids during metamorphism of graphitic pelites in a collisional orogen—an assessment from fluid inclusions . Geochimica et Cosmochimica Acta 71 , 4997 – 5015 . Google Scholar Crossref Search ADS Icenhower J. , London D. ( 1995 ). An experimental study of element partitioning among biotite, muscovite, and coexisting peraluminous silicic melt at 200 MPa (H2O) . American Mineralogist 80 , 1229 – 1251 . Google Scholar Crossref Search ADS Ishihara S. ( 1977 ). The magnetite-series and ilmenite-series granitic rocks . Mining Geology 27 , 293 – 305 . Ishihara S. ( 2004 ). The redox state of granitoids relative to tectonic setting and earth history: the magnetite–ilmenite series 30 years later . Earth and Environmental Science Transactions of The Royal Society of Edinburgh 95 , 23 – 33 . Jeong G. Y. , Kim H. B. ( 2003 ). Mineralogy, chemistry, and formation of oxidized biotite in the weathering profile of granitic rocks . American Mineralogist 88 , 352 – 364 . Google Scholar Crossref Search ADS Johnson D. M. , Hooper P. R. , Conrey R. M. , Laboratory G. ( 1999 ). XRF analysis of rocks and minerals for major and trace elements on a single low dilution Li-tetraborate fused bead . Advances in X-Ray Analysis 41 , 843 – 867 . Johnson J. E. , Gerpheide A. , Lamb M. P. , Fischer W. W. ( 2014 ). O2 constraints from Paleoproterozoic detrital pyrite and uraninite . Geological Society of America Bulletin 126 , 813 – 830 . Google Scholar Crossref Search ADS Kelley K. A. , Cottrell E. ( 2009 ). Water and the oxidation state of subduction zone magmas . Science 325 , 605 – 607 . Google Scholar Crossref Search ADS PubMed Kelley K. A. , Cottrell E. ( 2012 ). The influence of magmatic differentiation on the oxidation state of Fe in a basaltic arc magma . Earth and Planetary Science Letters 329–330 , 109 – 121 . Google Scholar Crossref Search ADS Kohn M. J. ( 2003 ). Geochemical zoning in metamorphic minerals . In: Rudnick R. L. (ed.), Treatise on Geochemistry, The Crust, Elsevier , (v. 3) pp. 229 – 261 . Kohn M. J. , Spear F. ( 2000 ). Retrograde net transfer reaction insurance for pressure–temperature estimates . Geology 28 , 1127 – 1130 . Google Scholar Crossref Search ADS Kress V. C. , Carmichael I. S. ( 1991 ). The compressibility of silicate liquids containing Fe2O3 and the effect of composition, temperature, oxygen fugacity and pressure on their redox states . Contributions to Mineralogy and Petrology 108 , 82 – 92 . Google Scholar Crossref Search ADS Lalonde A. E. ( 1986 ). The intrusive rocks of the Hepburn metamorphic–plutonic zone of the central Wopmay orogen , N.W.T. Montreal. PhD thesis, McGill University, Montreal, QC . Lalonde A. E. ( 1989 ). Hepburn intrusive suite: peraluminous plutonism within a closing back-arc basin, Wopmay orogen, Canada . Geology 17 , 261 – 264 . Google Scholar Crossref Search ADS Lalonde A. E. , Bernard P. ( 1993 ). Composition and color of biotite from granites: two useful properties in the characterization of plutonic suites from the Hepburn Internal Zone of the Wopmay Orogen, Northwest Territories . Canadian Mineralogist 31 , 203 – 217 . Larbi Y. , Stevenson R. , Breaks F. , Machado N. , Gariépy C. ( 1999 ). Age and isotopic composition of late Archean leucogranites: implications for continental collision in the western Superior Province . Canadian Journal of Earth Sciences 36 , 495 – 510 . Google Scholar Crossref Search ADS Lewry J. , Stauffer M. ( 1990 ). The Early Proterozoic Trans-Hudson orogen of North America. Geological Association of Canada, Special Paper 37 . Luth W. C. , Jahns R. H. , Tuttle O. F. ( 1964 ). The granite system at pressures of 4 to 10 kilobars . Journal of Geophysical Research 69 , 759 – 773 . Google Scholar Crossref Search ADS Lyons T. W. , Reinhard C. T. , Planavsky N. J. ( 2014 ). The rise of oxygen in Earth’s early ocean and atmosphere . Nature 506 , 307 – 315 . Google Scholar Crossref Search ADS PubMed Matzen A. K. , Baker M. B. , Beckett J. R. , Stolper E. M. ( 2011 ). Fe–Mg partitioning between olivine and high-magnesian melts and the nature of Hawaiian parental liquids . Journal of Petrology 52 , 1243 – 1263 . Google Scholar Crossref Search ADS McMullin D. W. , Berman R. G. , Greenwood H. J. ( 1991 ). Calibration of the SGAM thermobarometer for pelitic rocks using data from phase-equilibrium experiments and natural assemblages . Canadian Mineralogist 29 , 889 – 908 . Miller C. F. , Stoddard E. F. ( 1978 ). Origin of garnet in granitic rocks; an example of the role of Mn from the Old Woman–Piute range, California . Geological Society of America, Abstracts with Programs 10 , 456 . Miller C. F. , Stoddard E. F. , Bradfish L. J. , Dollase W. A. ( 1981 ). Composition of plutonic muscovite: genetic implications . Canadian Mineralogist 19 , 25 – 34 . Morris J. D. , Leeman W. P. , Tera F. ( 1990 ). The subducted component in island-arc lavas—constraints from Be isotopes and B–Be systematics . Nature 344 , 31 – 36 . Google Scholar Crossref Search ADS PubMed Mulja T. ( 1995 ). Magmatic Processes in Rare-Element Granite–Pegmatite Systems: the Preissac–Lacorne Batholith, Quebec, Canada. PhD thesis, McGill University, Montreal, QC . Mulja T. , Williams-Jones A. E. , Wood S. A. , Boily M. ( 1995a ). The rare-element-enriched monzogranite–pegmatite–quartz vein systems in the Preissac–Lacorne Batholith, Quebec; I, Geology and mineralogy . Canadian Mineralogist 33 , 793 – 815 . Mulja T. , Williams-Jones A. E. , Wood S. A. , Boily M. ( 1995b ). The rare-element-enriched monzogranite–pegmatite–quartz vein systems in the Preissac–Lacorne Batholith, Quebec; II, Geochemistry and petrogenesis . Canadian Mineralogist 33 , 817 – 833 . Müller P. J. , Suess E. ( 1979 ). Productivity, sedimentation rate, and sedimentary organic matter in the oceans—I. Organic carbon preservation . Deep-Sea Research, Part A 26 , 1347 – 1362 . Google Scholar Crossref Search ADS Mutch E. J. F. , Blundy J. D. , Tattitch B. C. , Cooper F. J. , Brooker R. ( 2016 ). An experimental study of amphibole stability in low-pressure granitic magmas and a revised Al-in-hornblende geobarometer . Contributions to Mineralogy and Petrology 171 , 85 . Google Scholar Crossref Search ADS Nabelek P. I. , Ternes K. ( 1997 ). Fluid inclusions in the Harney Peak Granite, Black Hills, South Dakota, USA: implications for solubility and evolution of magmatic volatiles and crystallization of leucogranite magmas . Geochimica et Cosmochimica Acta 61 , 1447 – 1465 . Google Scholar Crossref Search ADS Nabelek P. I. , Russ-Nabelek C. , Denison J. R. ( 1992a ). The generation and crystallization conditions of the Proterozoic Harney Peak leucogranite, Black Hills, South Dakota, USA: petrologic and geochemical constraints . Contributions to Mineralogy and Petrology 110 , 173 – 191 . Google Scholar Crossref Search ADS Nabelek P. I. , Russ-Nabelek C. , Haeussler G. T. ( 1992b ). Stable isotope evidence for the petrogenesis and fluid evolution in the Proterozoic Harney Peak leucogranite, Black Hills, South Dakota . Geochimica et Cosmochimica Acta 56 , 403 – 417 . Google Scholar Crossref Search ADS Nabelek P. I. , Sirbescu M. , Liu M. ( 1999 ). Petrogenesis and tectonic context of the Harney Peak Granite, Black Hills, South Dakota . Rocky Mountain Geology, University of Wyoming 34 , 165 – 181 . Google Scholar Crossref Search ADS Nabelek P. I. , Liu M. , Sirbescu M. L. ( 2001 ). Thermo-rheological, shear heating model for leucogranite generation, metamorphism, and deformation during the Proterozoic Trans-Hudson orogeny, Black Hills, South Dakota . Tectonophysics 342 , 371 – 388 . Google Scholar Crossref Search ADS Nelson B. K. , DePaolo D. J. ( 1985 ). Rapid production of continental crust 1·7 to 1·9 b.y. ago: Nd isotopic evidence from the basement of the North American mid-continent . Geological Society of America Bulletin 96 , 746 – 754 . Google Scholar Crossref Search ADS Ohmoto H. , Kerrick D. M. ( 1977 ). Devolatilization equilibria in graphitic systems . American Journal of Science 277 , 1013 – 1044 . Google Scholar Crossref Search ADS O’Neil J. R. , Shaw S. E. , Flood R. H. ( 1977 ). Oxygen and hydrogen isotope compositions as indicators of granite genesis in the New England Batholith, Australia . Contributions to Mineralogy and Petrology 62 , 313 – 328 . Google Scholar Crossref Search ADS O’Neill H. S. C. ( 1988 ). Systems Fe–O and Cu–O: thermodynamic data for the equilibria Fe–’FeO’, Fe–Fe3O4, ‘FeO’–Fe3O4, Fe3O4–Fe2I3, Cu–Cu2O, and Cu2O–CuO from emf measurements . American Mineralogist 73 , 470 – 486 . O’Neill H. S. C. , Pownceby M. I. ( 1993 ). Thermodynamic data from redox reactions at high temperatures. I. An experimental and theoretical assessment of the electrochemical method using stabilized zirconia electrolytes, with revised values for the Fe–’FeO’, Co–CoO, Ni–NiO and Cu–Cu2O oxygen buffers, and new data for the W–WO2 buffer . Contributions to Mineralogy and Petrology 114 , 296 – 314 . Google Scholar Crossref Search ADS Pan Y. , Breaks F. W. ( 1996 ). Metamorphic P–T–t–fluid history of the English River Subprovince, Separation Lake area, Ontario (abstract) . Geological Association of Canada—Mineralogical Association of Canada, Programs 21 , A.72 . Pan Y. , Therens C. , Breaks F. W. ( 1997 ). Geochemistry of granitoids in the English River subprovince, Separation Lake area. Paper presented at the Western Superior Lithoprobe Transect 3rd Annual Workshop, Western Superior Lithoprobe Transect, Toronto, ON , pp. 55 – 61 . Parkinson I. J. , Arculus R. J. ( 1999 ). The redox state of subduction zones: insights from arc-peridotites . Chemical Geology 160 , 409 – 423 . Google Scholar Crossref Search ADS Patiño Douce A. E. , Johnston A. D. ( 1991 ). Phase equilibria and melt productivity in the pelitic system: implications for the origin of peraluminous granitoids and aluminous granulites . Contributions to Mineralogy and Petrology 107 , 202 – 218 . Google Scholar Crossref Search ADS Pattison D. R. M. , Carmichael D. M. , St-Onge M. R. ( 1982 ). Geothermometry and geobarometry applied to early proterozoic “S-type” granitoid plutons, Wopmay Orogen, Northwest Territories, Canada . Contributions to Mineralogy and Petrology 79 , 394 – 404 . Google Scholar Crossref Search ADS Pearce J. A. , Peate D. W. ( 1995 ). Tectonic implications of the composition of volcanic arc magmas . Annual Review of Earth and Planetary Sciences 23 , 251 – 285 . Google Scholar Crossref Search ADS Percival J. A. , Sullivan R. W. ( 1988 ). Age constraints on the evolution of the Quetico belt, Superior Province , Canada. In: deWit M. J. , Ashwal L. D. (eds) Workshop on Tectonic Evolution of Greenstone Belts. Lunar and Planetary Institute. Houston, Texas. pp. 88 – 82 . Percival J. A. , Williams H. R. ( 1989 ). Late Archean Quetico accretionary complex, Superior province, Canada . Geology 17 , 23 – 25 . Google Scholar Crossref Search ADS Percival J. A. , Stern R. A. , Digel M. R. ( 1985 ). Regional geological synthesis of western Superior Province, Ontario . Geological Survey of Canada Paper 85–1A , 385 – 397 . Peterman Z. E. , Hedge C. E. , Braddock W. A. ( 1968 ). Age of Precambrian events in the Northeastern Front Range, Colorado . Journal of Geophysical Research 73 , 2277 – 2296 . Google Scholar Crossref Search ADS Pichavant M. , Kontak D. J. , Valencia Herrera J. , Clark A. H. ( 1988 ). The Miocene–Pliocene Macusani Volcanics, SE Peru. I: Mineralogy and magmatic evolution of a two-mica aluminosilicate-bearing ignimbrite suite . Contributions to Mineralogy and Petrology 100 , 300 – 324 . Google Scholar Crossref Search ADS Pichavant M. , Hammouda T. , Scaillet B. ( 1996 ). Control of redox state and Sr isotopic composition of granitic magmas: a critical evaluation of the role of source rocks . Earth and Environmental Science Transactions of the Royal Society of Edinburgh 87 , 321 – 329 . Google Scholar Crossref Search ADS Pichavant M. , Villaros A. , Deveaud S. , Scaillet B. , Lahlafi M. ( 2016 ). The influence of redox state on mica crystallization in leucogranitic and pegmatitic liquids; mica crystallization in leucogranitic and pegmatitic liquids . Canadian Mineralogist 54 , 559 – 581 . Google Scholar Crossref Search ADS Rasmussen B. , Buick R. ( 1999 ). Redox state of the Archean atmosphere: evidence from detrital heavy minerals in ca. 3250–2750 Ma sandstones from the Pilbara Craton, Australia . Geology 27 , 115 – 118 . Google Scholar Crossref Search ADS Redden J. ( 1993 ). Industrial minerals project, Sioux Lookout–Dryden area, parts of Kenora and Patricia Mining Divisions. Ontario Geological Survey, Open File Report 5766 , 140 pp. Redden J. A. , Norton J. J. , McLaughlin R. J. ( 1982 ). Geology of the Harney Peak Granite, Black Hills, South Dakota. US Geological Survey , Open-File Report 82–481 . Redden J. A. , Peterman Z. E. , Zartman R. E. , DeWitt E. ( 1990 ). U–Th–Pb zircon and monazite ages and preliminary interpretation of the tectonic development of Precambrian rocks in the Black Hills, South Dakota . In: Lewry J. F. , Stauffer M. R. (eds) The Trans-Hudson Orogen. Geological Association of Canada Special Paper 37 , 229 – 251 . Reed J. C. , Zartman R. E. ( 1973 ). Geochronology of Precambrian rocks of the Teton Range, Wyoming . Geological Society of America Bulletin 84 , 561 – 582 . Google Scholar Crossref Search ADS Sack R. O. , Ghiorso M. S. ( 1989 ). Importance of considerations of mixing properties in establishing an internally consistent thermodynamic database: thermochemistry of minerals in the system Mg2SiO4–Fe2SiO4–SiO2 . Contributions to Mineralogy and Petrology 102 , 41 – 68 . Google Scholar Crossref Search ADS Sawyer E. W. , Barnes S. J. ( 1988 ). Temporal and compositional differences between subsolidus and anatectic migmatite leucosomes from the Quetico metasedimentary belt, Canada . Journal of Metamorphic Geology 6 , 437 – 450 . Google Scholar Crossref Search ADS Scaillet B. , Evans B. W. ( 1999 ). The 15 June 1991 eruption of Mount Pinatubo. I. Phase equilibria and pre-eruption P–T– fO2 fH2Oconditions of the dacite magma . Journal of Petrology 40 , 381 – 411 . Google Scholar Crossref Search ADS Scaillet B. , Macdonald R. ( 2001 ). Phase relations of peralkaline silicic magmas and petrogenetic implications . Journal of Petrology 42 , 825 – 845 . Google Scholar Crossref Search ADS Scaillet B. , Macdonald R. ( 2003 ). Experimental constraints on the relationships between peralkaline rhyolites of the Kenya Rift Valley . Journal of Petrology 44 , 1867 – 1894 . Google Scholar Crossref Search ADS Scaillet B. , Pichavant M. , Roux J. ( 1995 ). Experimental crystallization of leucogranite magmas . Journal of Petrology 36 , 663 – 705 . Google Scholar Crossref Search ADS Scaillet B. , Holtz F. , Pichavant M. ( 2016 ). Experimental constraints on the formation of silicic magmas . Elements 12 , 109 – 114 . Google Scholar Crossref Search ADS Shabani A. A. , Lalonde A. E. , Whalen J. B. ( 2003 ). Composition of biotite from granitic rocks of the Canadian Appalachian orogen: a potential tectonomagmatic indicator? Canadian Mineralogist 41 , 1381 – 1396 . Google Scholar Crossref Search ADS Shand S. J. ( 1927 ). Eruptive Rocks . New York : J. Wiley and Sons . Silver L. T. , Williams I. S. , Woodhead J. A. ( 1980 ). Uranium in granites from the southwestern United States: actinide parent–daughter systems, sites and mobilization . Report of the US Department of Energy DOE-GJBX-45(81), 379 . Smith P. M. , Asimow P. D. ( 2005 ). Adiabat_1ph: a new public front-end to the MELTS, pMELTS, and pHMELTS models . Geochemistry, Geophysics, Geosystems 6 , Q02004 . Google Scholar Crossref Search ADS Southwick D. L. ( 1991 ). On the genesis of Archean granite through two-stage melting of the Quetico accretionary prism at a transpressional plate boundary . Geological Society of America Bulletin 103 , 1385 – 1394 . Google Scholar Crossref Search ADS Southwick D. L. , Sims P. K. ( 1979 ). The Vermilion Granitic Complex—a new name for old rocks in northern Minnesota . US Geological Survey, Professional Papers 1124A , A1 – A11 . Speer J. A. ( 1984 ). Micas in igneous rocks . In: Bailey S. W. (ed.) Micas . Mineralogical Society of America, Reviews in Mineralogy 13, 299–356 . Stevens G. , Villaros A. , Moyen J.-F. ( 2007 ). Selective peritectic garnet entrainment as the origin of geochemical diversity in S-type granites . Geology 35 , 9 – 12 . Google Scholar Crossref Search ADS Stott G. M. , Corkery M. T. , Percival J. A. , Simard M. , Goutier J. ( 2010 ). A revised terrane subdivision of the Superior Province . In: Summary of Field Work and Other Activities 2010. Ontario Geological Survey, Open File Report 6260 , 20–1 – 20–10 . St-Onge M. R. , King J. E. ( 1987 ). Evolution of regional metamorphism during back-arc stretching and subsequent crustal shortening in the 1.9 Ga Wopmay Orogen, Canada . Philosophical Transactions of the Royal Society London 321 , 199 – 218 . Google Scholar Crossref Search ADS Sverjensky D. A. , Lee N. ( 2010 ). The great oxidation event and mineral diversification . Elements 6 , 31 – 36 . Google Scholar Crossref Search ADS Sylvester P. J. ( 1998 ). Post-collisional strongly peraluminous granites . Lithos 45 , 29 – 44 . Google Scholar Crossref Search ADS Taylor H. P. Jr ( 1980 ). The effects of assimilation of country rocks by magmas on 18O/16O and 87Sr/86Sr systematics in igneous rocks . Earth and Planetary Science Letters 47 , 243 – 254 . Google Scholar Crossref Search ADS Taylor J. , Stevens G. ( 2010 ). Selective entrainment of peritectic garnet into S-type granitic magmas: evidence from Archaean mid-crustal anatectites . Lithos 120 , 277 – 292 . Google Scholar Crossref Search ADS Thirlwall M. F. , Graham A. M. , Arculus R. J. , Harmon R. S. , Macpherson C. G. ( 1996 ). Resolution of the effects of crustal assimilation, sediment subduction, and fluid transport in island arc magmas: Pb–Sr–Nd–O isotope geochemistry of Grenada, Lesser Antilles . Geochimica et Cosmochimica Acta 60 , 4785 – 4810 . Google Scholar Crossref Search ADS Tso J. L. , Gilbert M. C. , Craig J. R. ( 1979 ). Sulfidation of synthetic biotites . American Mineralogist 64 , 304 – 316 . Tuttle O. F. , Bowen N. L. ( 1958 ). Origin of granite in the light of experimental studies in the system NaAlSi3O8–KAlSi3O8–SiO2–H2O. Geological Society of America, Memoirs 74 , 153 pp. van de Kamp P. C. , Beakhouse G. P. ( 1979 ). Paragneisses in the Pakwash Lake area, English River Gneiss Belt, Northwest Ontario . Canadian Journal of Earth Sciences 16 , 1753 – 1763 . Google Scholar Crossref Search ADS Veblen D. R. , Ferry J. M. ( 1983 ). A TEM study of the biotite–chlorite reaction and comparison with petrologic observations . American Mineralogist 68 , 1160 – 1168 . Villaros A. , Stevens G. , Buick I. S. ( 2009 ). Tracking S-type granite from source to emplacement: clues from garnet in the Cape Granite Suite . Lithos 112 , 217 – 235 . Google Scholar Crossref Search ADS Walker R. J. , Hanson G. N. , Papike J. J. , O’Neil J. R. ( 1986 ). Nd, O and Sr isotopic constraints on the origin of Precambrian rocks, southern Black Hills, South Dakota . Geochimica et Cosmochimica Acta 50 , 2833 – 2846 . Google Scholar Crossref Search ADS Weller O. M. , St Onge M. R. ( 2017 ). Record of modern-style plate tectonics in the Palaeoproterozoic Trans-Hudson orogen . Nature Geoscience 10 , 305 – 311 . Google Scholar Crossref Search ADS Whalen J. B. , Chappell B. W. ( 1988 ). Opaque mineralogy and mafic mineral chemistry of I- and S-type granites of the Lachlan fold belt, Southeast Australia . American Mineralogist 73 , 281 – 296 . White A. J. R. , Chappell B. W. ( 1977 ). Ultrametamorphism and granitoid genesis . Tectonophysics 43 , 7 – 22 . Google Scholar Crossref Search ADS White W. M. , Dupré B. ( 1986 ). Sediment subduction and magma genesis in the Lesser Antilles: isotopic and trace element constraints . Journal of Geophysical Research 91 , 5927 . Google Scholar Crossref Search ADS Whitmeyer S. J. , Karlstrom K. E. ( 2007 ). Tectonic model for the Proterozoic growth of North America . Geosphere 3 , 220 – 259 . Google Scholar Crossref Search ADS Wones D. R. , Eugster H. P. ( 1965 ). Stability of biotite: experiment, theory and application . American Mineralogist 50 , 1228 – 1272 . Wyborn D. , Chappell B. W. , Johnston R. M. ( 1981 ). Three S-type volcanic suites from the Lachlan Fold Belt, southeast Australia . Journal of Geophysical Research: Solid Earth 86 , 10335 – 10348 . Google Scholar Crossref Search ADS Zartman R. E. , Reed J. C. Jr ( 1998 ). Zircon geochronology of the Webb Canyon Gneiss and the Mount Owen Quartz Monzonite, Teton Range, Wyoming: significance to dating Late Archean metamorphism in the Wyoming craton . Mountain Geologist 35 , 71 – 77 . Zeck H. P. ( 1992 ). Restite–melt and mafic–felsic magma mixing and mingling in an S-type dacite, Cerro del Hoyazo, southeastern Spain . Earth and Environmental Science Transactions of the Royal Society of Edinburgh 83 , 139 – 144 . Google Scholar Crossref Search ADS Zen E.-A. ( 1988 ). Phase relations of peraluminous granitic rocks and their petrogenetic implications . Annual Review of Earth and Planetary Sciences 16 , 21 – 51 . Google Scholar Crossref Search ADS Zeng L. , Saleeby J. B. , Asimow P. ( 2005 ). Nd isotope disequilibrium during crustal anatexis: a record from the Goat Ranch migmatite complex, southern Sierra Nevada batholith, California . Geology 33 , 53 – 56 . Google Scholar Crossref Search ADS © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - A Comparison of Oxygen Fugacities of Strongly Peraluminous Granites across the Archean–Proterozoic Boundary
JF - Journal of Petrology
DO - 10.1093/petrology/egy091
DA - 2018-11-01
UR - https://www.deepdyve.com/lp/oxford-university-press/a-comparison-of-oxygen-fugacities-of-strongly-peraluminous-granites-hU8d0uZ9i9
SP - 2123
VL - 59
IS - 11
DP - DeepDyve
ER -