Eur. J. Mineral. 2011, 23, 45–61 Published online November 2010

Mineralogy from three peralkaline granitic plutons of the Late Permian Emeishan large igneous province (SW China): evidence for contrasting magmatic conditions of A-type granitoids

J. GREGORY SHELLNUTT* and YOSHIYUKI IIZUKA

Institute of Earth Sciences, Academia Sinica, 128 Academia Road Sec. 2, Nankang, Taipei 11529, Taiwan *Corresponding author, e-mail: [email protected]

Abstract: The Emeishan large igneous province contains a diverse assemblage of igneous rocks including mildly peralkaline granitic rocks of A-type affinity. The granitic rocks from the Panzhihua, Baima and Taihe plutons are temporally, spatially and chemically associated with layered mafic-ultramafic intrusions. Electron microprobe analyses of the major and accessory minerals along with major and trace element data were used to document the magmatic conditions of the three peralkaline plutons. The amphiboles show magmatic/subsolidus trends and are primarily sodic-calcic in composition (i.e., ferrorichterite or richterite). Sodic (i.e., riebeckite-arfvedsonite) amphiboles are restricted to the Panzhihua and Taihe plutons. The amphiboles from the Panzhihua and Taihe granites are very similar in composition whereas amphiboles from the Baima syenites have higher MgO wt% and lower FeOt wt% and TiO2 wt%. Whole-rock Zr saturation temperature estimates indicate the initial average magma temperatures were 940 21 C for the Panzhihua pluton, 860 17 C for the Baima pluton, and 897 14 C for the Taihe pluton. The initial Fmelt(wt%) values were calculated to be 1.1 0.1, 0.8 0.1 and 1.1 0.1 wt% for the Panzhihua, Baima and Taihe plutons, respectively. The estimated Fmelt(wt%) values are higher than what can be accounted for in the Panzhihua and Taihe plutons and indicate that they may have lost F during crystallization. In contrast the Fmelt(wt%) value for the Baima pluton can be accounted for. The presence of titanite magnetite quartz in the Baima syenites indicates oxidizing fO2 conditions whereas the presence of aenigmatite and ilmenite in the Panzhihua and Taihe granites indicate that they were relatively reducing. Although the A- type granitoids formed by the same processes (i.e., fractional crystallization of mafic magmas), their differences in major element and mineral chemistry are likely related to a combination of initial bulk magma composition and magmatic oxidation state. Key-words: ferrorichterite, peralkaline, A-type granite, Zr saturation thermometry, oxygen fugacity, aenigmatite.

1. Introduction described sodic-calcic amphiboles from silica-saturated peralkaline granitic rocks as having magmatic/subsolidus The bulk composition of a magma is ultimately responsible or primary magmatic trends. The compositional trends for the composition of minerals which crystallize within it; reflect the substitution Aliv Ca $ Si (Na, K) and the however, magmatic conditions, such as temperature, progressive decrease in Mg# ( Mg2/[Mg2 Fe2]) B F-content, oxygen fugacity (fO2) and water fugacity and Ca with an increase in Si. Sodic-calcic amphiboles (fH2O), can greatly influence their composition and type may have oxidation or late-stage reaction compositional (Mitchell, 1990; Scaillet & Macdonald, 2003, 2004, 2006a; trends as well (Strong & Taylor, 1984; Mitchell, 1990). Della Ventura et al., 2005; White et al., 2005). Within Peralkaline silica-saturated volcanic rocks are volumetri- peralkaline granitic rocks, the minerals and their composi- cally minor when compared to basalts within large igneous tion can be used to determine the likely magmatic condi- provinces but they have the potential to emit large amounts tions during their crystallization (Charles, 1975, 1977; of sulphur to the atmosphere, which may contribute to ocean Giret et al., 1980; Strong & Taylor, 1984; Wones, 1989; anoxia and mass extinctions (Scaillet & Macdonald, 2006b). Mitchell, 1990; Scaillet & Macdonald, 2001, 2004; Della The Late Permian (260 Ma) Emeishan large igneous pro- Ventura et al., 2005; White et al., 2005; Martin, 2007; vince of SW China contains numerous granitic rocks with A- Pe-Piper, 2007). For example, identifying the amphibole- type affinity (Shellnutt & Zhou, 2007; Zhong et al., 2007). type (i.e., sodic, sodic-calcic, calcic), as well as accessory The peralkaline A-type granitoids are temporally and spa- minerals amongst consanguineous peralkaline rocks, is tially associated with layered gabbroic intrusions and in helpful to determine if there is a common genetic process some cases fed sub-aerial volcanic eruptions (Shellnutt & or relationship between them (Mitchell, 1990; White et al., Jahn, 2010). Shellnutt et al. (2009a) have suggested that 2005). Strong & Taylor (1984) and Mitchell (1990) have these rocks were derived by fractional crystallization of a

eschweizerbart_xxx 0935-1221/10/0022-2073 $ 7.65 DOI: 10.1127/0935-1221/2010/0022-2073 # 2010 E. Schweizerbart’sche Verlagsbuchhandlung, D-70176 Stuttgart 46 J.G. Shellnutt, Y. Iizuka common parental magma that produced both the granitic and K]; Frost et al., 2001; Panzhihua 0.86; Baima 0.83; cumulate gabbroic rocks. The Panzhihua and Taihe granitic Taihe 0.88), (Na K)/Al (Panzhihua 1.07; Baima plutons are compositionally similar, however the Baima 1.05; Taihe 1.12), Fe* ( FeOt/[FeOt MgO]) syenitic pluton is compositionally and mineralogically dif- (Panzhihua 0.97; Baima 0.85; Taihe 0.97) and ferent (Shellnutt & Zhou, 2007). Although the bulk-rock Eu/Eu* (Panzhihua 0.73; Baima 0.79; Taihe compositions and petrography of the granitoids rocks are 0.43). Their eNd(T) isotopic values also fall within the different, they have similar normalized trace-element pat- same range (eNd(T) 1.5 to 3.2). The remarkable terns, isotopic compositions and ages. It is thought that the aspect of the Panxi peralkaline plutonic rocks is their major-element differences between the Baima syenites and similar primitive mantle and chondrite normalized trace- the Panzhihua and Taihe granites may be due to initial bulk element patterns (Shellnutt & Zhou, 2007). The normal- magma compositions, temperature, F-content, magmatic ized trace-element plots show the trace-element composi- fO2 or any combination thereof. tions are similar regardless of the bulk rock composition Originally, mantle-derived peralkaline A-type grani- and imply that the rocks formed by the same process. toids were considered to form from reduced rather than oxidized magmas (Loiselle & Wones, 1979). The rocks within the Emeishan large igneous province offer an excel- lent opportunity to examine the mineralogy from three, 3. Petrography contemporaneous, mildly peralkaline, silica-saturated granitoids which are thought to originate by the same 3.1. Baima syenites process. The purpose of this study is to examine the mineral compositions of the peralkaline granitoids in The Baima syenites are coarse grained and consist of perthi- order to determine the likely magmatic conditions of tic alkali feldspar ( 70 vol%), amphibole ( 10 vol%), their host rocks. Along with new mineral chemistry data quartz ( 10 vol%), aegirine ( 5 vol%) and accessory we use previously published major and trace element data amounts ( 5 vol %) of apatite, titanite, zircon, plagioclase, to estimate the initial magma temperatures, magmatic fO fluorite, biotite, ilmenite, magnetite and pyrite (Fig. 3a–d) 2 and F (wt%) compositions. The alkali feldspar forms large ( 1.2 cm), subhedral to melt anhedral crystals with serrate boundaries which are occa- sionally separated from other perthitic crystals by an inter- granular mixture of alkali feldspar, quartz and accessory 2. Previous work plagioclase. The intergranular alkali feldspar and quartz are small (,0.3 cm) subhedral to anhedral, rounded crys- The Late Permian (260 Ma) Emeishan large igneous pro- tals. Euhedral to subhedral, rhombic ( 0.5 cm), sodic-calcic vince of southwest China hosts a diverse assemblage of amphibole (i.e., ferrorichterite or richterite) is the most volcanic and intrusive igneous rocks (Ali et al., 2005). abundant primary mafic mineral. The amphibole crystals Although the flood basalts are volumetrically (95%) the are pleochroic (light blue to light green) and commonly largest component, there is a significant amount of within- associated with subhedral aegirine. The amphibole is inter- plate granitic rocks (1%) (Fig. 1). Many of the granitic stitial to the alkali feldspar and quartz and is often associated rocks are consistent with the classification of ‘‘A-type’’ with agglomerates of euhedral apatite, zircon, titanite, ilme- and are of peraluminous, peralkaline and metaluminous nite, magnetite, fluorite and occasionally biotite. composition (Shellnutt & Zhou, 2007). Within the Panxi region (Panzhihua-Xi Chang) of the Emeishan large 3.2. Panzhihua granites igneous province, spatially and temporally associated with the peralkaline granitic rocks, are layered gabbroic The granites are coarse grained and consist primarily of intrusions which host world-class orthomagmatic Fe-Ti-V perthitic alkali feldspar, quartz and amphibole. Known oxide deposits (Zhou et al., 2005). The association accessory minerals include apatite, zircon, ilmenite, aenig- between the peralkaline granitic rocks and ore-bearing matite, Ce-monazite and titanite. The alkali feldspars are gabbroic rocks has led to suggestions that the gabbro- subhedral, rectangular to angular in shape and comprise granite complexes each formed by fractional crystalliza- 65–70 vol % of the mode. The crystals are usually 1.5 cm tion of mantle-derived basaltic magmas (Shellnutt & Jahn, or less in length and have patchy exsolution lamellae instead 2009, 2010; Shellnutt et al., 2009a). of smooth linear lamellae (Fig. 3e, f). Quartz (1 mm) The Panzhihua and Taihe granites are nearly identical in comprises 20–25 vol % of the mode and is rounded to major-element composition although the Panzhihua gran- sub-rounded and interstitial to the alkali feldspar. Most often ites have slightly higher TiO2 and Fe2O3t while the Taihe the quartz has formed an interlocking network between the granites have slightly higher K2O (Table 1). In contrast, the feldspars, however in some cases it appears to embay the Baima pluton is syenitic rather than granitic and, as a alkali feldspar. Interstitial to the feldspar and quartz is result, has a different major-element composition (Fig. amphibole comprising 10–15 vol % of the mode. The 2). In spite of the major-element differences, the plutons amphibole has light blue-green pleochroism, is euhedral to have similar average major-element ratios such as the subhedral and has angular appearance (Fig. 3e, f). In many alumina saturation index (ASI Al/[Ca–1.67P Na cases the amphiboles are breaking down to oxide minerals

eschweizerbart_xxx Mineralogy from three Permian peralkaline granitoids, SW China 47

102o 00’ E 103o 00’ E Basalts N Xichang Fe-Ti-V-oxide-bearing intrusions Taihe Peralkaline Gabbro-Granite Syenite/Granite Complex

Emeishan LIP Jurassic-Quaternary strata Triassic strata

Palaeozoic strata o 27o 20’ N 27 20’ N Neoproterozoic igneous -metamorphic complex Mesozoic granite Baima Igneous Faults Complex

80° 90 100° 110° 130° 120°

Paleo-Asian Miyi Orogenic Belt 40°

Beijing 40° North China Craton Panzhihua Kunlun-Tibet Domain o o 26 40’ N Gabbro-Granite 26 40’ N 30° Yangtze Block Complex lock 30°

Cathaysian B Hong Kong Panzhihua Emeishan LIP

102o 00’ E

Fig. 1. Simplified geological map of the Panzhihua- Xi Chang region (Panxi) showing the location and distribution of late Permian granitic rocks. including hematite. Minor amounts of subhedral aegirine Panzhihua, 5 syenites (GS03-097, -101, GS04-004, -015, are also present but they are heavily altered. Accessory -016) from Baima and 5 granites (GS04-054, -055, -057, zircon, Ce-monazite, aenigmatite, titanite and apatite -058, -059) from Taihe, were selected for analyses. (,,1%) are occasionally found. Mineralogical investigation was carried out by a field emis- sion electron probe micro analyzer (FE-EPMA: JEOL JXA- 3.3. Taihe granites 8500F) equipped with five wave-length dispersive spectro- meters (WDS) at the Institute of Earth Sciences, Academia The Taihe granites are texturally and mineralogically similar Sinica in Taipei. Secondary- and back-scattered electron to the Panzhihua granites. The rocks are granular, coarse images were used to guide the analysis on target positions grained and consist mostly of alkali feldspar, quartz and of minerals. A 2 mm defocused beam was operated for quan- amphiboles with accessory titanite, zircon, fluorite, Ce-mon- titative analysis at an acceleration voltage of 12 kV with a azite, aenigmatite and rare biotite (Fig. 3g, h). The alkali beam current of 5 nA. The measured X-ray intensities were feldspar is euhedral to subhedral with perthitic texture and is corrected by ZAF method using the standard calibration of typically 1–2 cm in size and comprise 65–70 vol% of bulk synthetic chemical-known standard minerals with various mineralogy. The quartz is 0.2 cm, is rounded and interstitial diffracting crystals, as follows: wollastonite for Si with TAP to the alkali feldspar and comprises 20–25 vol% of the mode. crystal and Ca with PET crystal, rutile for Ti (PET), corun- There are sodic-calcic and sodic amphiboles present. The dum for Al (TAP), chromium oxide for Cr (PET), fayalite for ferrorichterite (10–15 vol%) is distinguished from the rie- Fe with LiF crystal, tephroite for Mn (PET), pyrope for Mg beckite/arfvedsonite by its colour and pleochroism. Some (TAP), albite for Na (TAP), adularia for K (PET), apatite for P amphiboles are partially altered to a translucent red-brown (PET), sodalite for Cl (PET) and fluorite for F (TAP). Peak mineral likely to be hematite. Small quantities (5 vol%) of counting for each element and both upper and lower baselines subhedral aegirine, aenigmatite, anhedral zircon and ilmenite are counted for 10 and 5 s, respectively. For the analysis of (rare) are also present. apatite, the conditions were modified as follows: a 2 mm defocused beam, an acceleration voltage of 12 kV with a beam current of 6 nA and, for the standards and diffracting 4. Method and data processing crystals, fluorite for F with LDE1 crystal, apatite for Ca and P with PET crystals, rutile for Ti (PET), hematite for Fe with A total of 13 samples from the three peralkaline plutons, LiF crystal, jadeite for Na (TAP), celestite (SrSO4)forSr which include 3 granites (GS03-007, -011, -014) from (PET), anhydrite for S (PET), and monazite for La (LiF) and

eschweizerbart_xxx 48 Table 1. Whole-rock data of the samples analyzed for this study.

Sample GS04-054 GS04-055 GS04-057 GS04-058 GS04-059 GS03-007 GS03-011 GS03-014 GS03-097 GS03-101 GS04-004 GS04-015 GS04-016 Pluton Taihe Taihe Taihe Taihe Taihe Panzhihua Panzhihua Panzhihua Baima Baima Baima Baima Baima

SiO2 (wt%) 70.88 72.4 70.78 69.8 71.88 70.67 70.22 69.93 64.53 66.02 64.52 63.86 65.4 TiO2 0.48 0.32 0.44 0.65 0.44 0.25 0.48 0.56 0.81 0.82 0.67 1.02 0.69 Al2O3 11.62 11.64 11.45 11.21 10.94 11.3 10.96 11.07 16.31 15.34 15.73 14.52 15.71 Fe2O3 5.89 4.78 5.8 6.89 5.83 6.32 6.53 6.11 3.57 4 3.92 5.66 3.95 MnO 0.17 0.14 0.16 0.2 0.14 0.17 0.16 0.13 0.18 0.2 0.16 0.26 0.17 MgO 0.13 0.13 0.11 0.13 0.11 0.16 0.09 0.45 0.74 0.63 0.51 0.78 0.51 CaO 0.69 0.67 0.64 0.75 0.54 0.66 0.67 0.71 0.92 0.57 0.71 1.13 0.79 Na2O 4.9 4.71 4.78 5.18 4.74 5.16 5.14 4.57 6.8 7.04 6.54 6.41 6.56 K2O 4.83 4.85 4.78 4.39 4.72 4.17 4.37 4.31 4.5 4.87 5.15 4.91 5.07 P2O5 0.03 0.03 0.03 0.03 0.02 0.03 0.01 0.03 0.16 0.13 0.1 0.32 0.11 LOI 0.25 0.32 0.16 0.65 0.45 0.52 1.04 1.36 0.66 0.31 0.67 0.64 0.61 TOTAL 99.87 99.98 99.14 99.88 99.81 99.42 99.67 99.24 99.15 99.92 98.69 99.51 99.56 ASI 0.80 0.82 0.81 0.77 0.79 0.80 0.76 0.82 0.93 0.86 0.90 0.83 0.89 Fe* 0.98 0.97 0.98 0.98 0.98 0.98 0.99 0.93 0.83 0.86 0.88 0.88 0.89 Sc (ppm) 8 8 8 7 7 14 12 12 11 11 12 12 11 V 3 2 2 3 2 1 1 2 15 12 121 17 130 Cr 24274 33344143 Co 11110 00011121 Iizuka Y. Shellnutt, J.G. Ni 46476 51134944 Cu 6 89 8 55 14 6 3 4 8 6 5 7 4 Zn 219 204 217 283 219 282 235 348 166 165 137 185 153 eschweizerbart_xxx Ga 35.7 35.3 37.1 36.3 36 39.7 39.7 41.4 33.1 37.5 40 34 35 Rb 128 130 140 124 141 90 95 85 175 111 123 113 126 Sr 15 20 17 19 14 31 33 13 320 132 89 87 95 Y 95 82 103 98 82 80 90 110 44 36 53 46 51 Zr 585 710 675 421 608 1022 1155 1400 465 361 1075 1103 1204 Nb 96 99 108 120 105 91 95 125 92 81 132 132 112 Cs 0.4 0.4 0.4 0.6 0.5 0.2 0.3 0.2 1.4 0.4 0.4 0.2 0.3 Ba 326 344 369 382 295 496 538 301 1446 843 388 346 391 La 108 99 149 113 103 108 107 127 49 81 436 102 74 Ce 273 198 355 305 254 270 280 342 110 170 819 220 171 Pr 26.8 24.8 35 28 25.4 27.9 28 33.7 14.2 20.4 74.5 27.7 22.3 Nd 103 95 133 106 93 109 110 134 56 76 248 104 87 Sm 20 18.3 25.2 20.6 18.3 21.5 22 27.9 10.9 13 32.5 18.8 16.8 Eu 3.16 2.78 3.89 3.19 2.76 4.66 4.85 5.93 4 3.55 3.98 2.84 2.81 Gd 17.5 15.5 21.1 17.8 15.8 18.9 19.9 24.8 10.3 12.2 24.8 14.2 12.9 Tb 2.95 2.61 3.53 3.04 2.66 2.91 3.18 3.93 1.5 1.62 3.06 2.17 2.17 Dy 17.8 15.9 20.7 18.8 16 16.7 18.3 22.9 7.6 7.6 13.1 10.8 11.3 Ho 3.63 3.28 4.11 3.8 3.26 3.23 3.57 4.49 1.5 1.4 2.33 1.98 2.12 Er 10.7 9.93 12 11 9.76 8.99 9.87 12.53 4.55 3.96 6.62 5.48 5.8 Tm 1.53 1.46 1.66 1.59 1.41 1.23 1.33 1.7 0.68 0.54 0.89 0.76 0.79 Yb 10.1 9.7 10.9 10.4 9.39 7.68 8.19 10.4 4.62 3.62 6.09 5.34 5.24 Lu 1.46 1.45 1.6 1.54 1.38 1.1 1.14 1.47 0.64 0.54 0.93 0.87 0.75 Hf 16 19 18.7 11.6 16.4 20.2 22.2 27.6 14 9 28.5 24.3 28.8 Ta 5.57 6.25 6.77 6.43 6.77 6.27 6.73 7.41 7.33 4.08 8.22 6.89 7.45 Th 16 19.3 20.8 17.4 19.5 15.4 15.8 19 18.7 8 29.4 6.8 7.9 U 3.62 4.52 4.33 3.16 4.05 3.11 3.21 4.35 2.81 1.42 3.87 1.79 2.38 Eu/Eu* 0.50 0.49 0.50 0.50 0.48 0.69 0.69 0.67 1.14 0.85 0.41 0.51 0.56

Data from Shellnutt & Zhou (2007) and Shellnutt et al. (2009a). ASI Al/(Ca 1.67P Na K). Fe* FeOt(FeOt MgO). Eu/Eu* (2*EuN/[SmNGdN]); N chondrite-normalized to values of Sun & McDonough (1989). Mineralogy from three Permian peralkaline granitoids, SW China 49

19 The sodic-calcic amphiboles generally have similar Panzhihua comenditic trachyte Taihe SiO2, MnO and K2O, however there are large differences 17 Baima with respect to the FeOt, MgO and TiO2 and to a lesser extent low ƒO2 melts Al2O3, Na2O and CaO (Fig. 4). Figure 4a shows a negative 15 correlation between Mg2 and Fe2 Fe3 suggesting a substitution relationship. The trend also reveals three distinct 13 pantelleritic 2 , trachyte groups of the richterites. The first group has low Mg ( 0.4 2 3

O (wt%) 211 3 apfu) and high Fe Fe (.3.5 apfu), the second group Al has moderate Mg2 (1.0–1.5 apfu) and Fe2 Fe3 (3.8–3.1 comendite 9 apfu) and the third group has high Mg2 (1.8–2.8 apfu) and low Fe2 Fe3 (3.1–1.8 apfu). The three groups corre- 7 pantellerite spond to their respective host rock, Group I amphiboles are from sample GS03-097 of the Baima pluton, Group II amphi- 5 024681012boles are from all samples of the Baima pluton excluding FeOt (wt%) GS03-097 and Group III amphiboles are from the Panzhihua and Taihe plutons. Titanium can also subdivide the sodic- Fig. 2. The classification scheme of peralkaline rocks from calcic amphiboles. The sodic-calcic amphiboles from sample Macdonald (1974). The compositional range of low fO2 melts is shown in grey (Scaillet & Macdonald, 2003). Data from Shellnutt GS03-097 have the lowest Ti (apfu) whereas the sodic-calcic & Zhou (2007) and Shellnutt et al. (2009a). amphiboles from the Panzhihua and Taihe granites have higher Ti than those from the Baima syenites (Fig. 4b). Ce (LiF). Peak counting and both upper and lower baselines The richterite and winchite analyses are restricted to are counted for 20 s and 5 s on F and S, for 10 s and 10 s on Ca, sample GS03-097. The compositions of the richterite ana- Na, Fe, and Ti, for 20 s and 10 s on La and Ce, and for 30 s and lyses have similar concentrations for all major elements 10 s on Sr, respectively. SiO2 (49.9–52.8 wt%), TiO2 (0.2–0.8 wt%), Al2O3 The WinAmphcal program, based on the nomenclature and (1.4–2.7 wt%), FeOt (17.6–23.1 wt%), MnO (0.9–2.0 classification scheme of amphiboles by the International wt%), MgO (8.2–11.6 wt%), CaO (3.1–7.1 wt%), Na2O Mineralogical Association (IMA), was used to classify and (4.6–7.2 wt%), K2O (1.0–1.4 wt%) and F (1.5–2.5 wt%). process the amphibole data (Leake et al., 2004; Yavuz, 2007). The winchite analyses are very similar to each other. The The structural formulae are calculated based on minimum ferrowinchite analyses are restricted to the Taihe granites Fe3, 13eCNK, tetrahedral Si and Al 8, and 23 oxygen and do not show significant chemical variation. atoms. All data are freely available in the supplementary material on GSW site of the journal at http://eurjmin. 5.3. Aegirine-augite geoscienceworld.org Aegirine-augite is observed in all rocks and is the second most abundant mafic mineral in the plutons. The aegirine- augite from sample GS03-097 (Baima) has higher MgO 5. Results (1.0–1.8 wt%) and CaO (4.6–7.0 wt%) and lower TiO2 (,0.6 wt%), Fe2O3t (29.1–31.2 wt%) and Na2O 5.1. Sodic amphiboles (9.0–10.4 wt%) than the aegirine-augite from the Panzhihua and Taihe granites (Table 3). Although the Ninety-seven spot analyses of sodic amphiboles are from the aegirine-augite from the Panzhihua and Taihe granites Taihe and Panzhihua granites (Taihe 76; Panzhihua are similar, the crystals from the Taihe granites tend to 21). The results show that the sodic amphiboles are riebeck- have higher MgO, TiO2 and CaO contents (Fig. 5). ite (37), arfvedsonite (50) and ferroeckermannite (10). The majority (66%) of spot analyses have a modifier such as 5.4. Aenigmatite fluorian, potassian or titanian whereas the others do not. Chemically all the analyses have similar concentrations of Aenigmatite is only observed within the Panzhihua and SiO2, Al2O3, MnO, CaO, Na2O, K2O and FeOt but TiO2 can Taihe granites, it shows limited compositional variation. be variable (Table 2). The concentration of F is 1.6 wt%. The crystals from the Panzhihua granites tend to have higher TiO2 than the crystals from the Taihe granites although the MnO content is higher within the Taihe crys- 5.2. Sodic-calcic amphiboles tals (Table 4). Generally, both groups of aenigmatite have higher TiO2 and lower Al2O3 than results reported in other The sodic-calcic amphiboles are the dominant amphibole studies (White et al., 2005; Grew et al., 2008). in all three plutons. The amphibole analyses are classified as ferrorichterite (95 analyses), richterite (25 analyses), 5.5. Titanite ferrowinchite (7 analyses) and winchite (3 analyses). The most common modifiers of the sodic-calcic amphiboles are Titanite is observed in all rocks, however it is more abundant ferrian and/or fluorian as the F content is 2.7 wt%. within the Baima syenites. Nearly all the titanite analyses

eschweizerbart_xxx 50 J.G. Shellnutt, Y. Iizuka

GS03-097[A] (a) GS03-097[A]

(b)

GS04-004[A] (c) GS04-016[A]

(d)

GS03-011[A] GS03-007[A]

(e) (f)

GS04-055[B] GS04-058B-A2

(g) (h)

Fig. 3. Backscattered electron images taken from the peralkaline granitoids. (a) GS03-097[A] (Baima) showing titanite, magnetite, apatite, quartz, ilmenite and richterite. (b) GS03-097 (Baima) showing individual zircon crystals surrounded by alkali feldspar. (c) GS04-004 (Baima) showing ferrorichterite with zircon, ilmenite and apatite inclusions surrounded by alkali feldspar. (d) GS04-016 (Baima) showing titanite, ferrorichterite and zircons surrounded by alkali feldspar. (e) GS03-011[A] (Panzhihua) showing aenigmatite surrounded by ferrorichterite. (f) GS03-007[A] (Panzhihua) a typical ferrorichterite surrounded by alkali feldspar. (g) GS04-055[B] (Taihe) ferrorichterite with inclusions of Ce-monazite breaking down to aegirine surrounded by quartz and alkali feldspar. (h) GS04-058-A2 (Taihe) showing the relationship between ferrorichterite and aegirine. Mineral symbols from Siivola & Schmid (2007). Ab – albite; Aen – aenigmatite; Agt – aegirine-augite; Ap – apatite; Bt – biotite; Fl – fluorite; Ilm – ilmenite; Mag – magnetite; Mnz – monazite; Or – orthoclase; Qtz – quartz; Rit – richterite; Spn – titanite; Zrn – zircon. sum to ,97 wt%, which is attributed to the amount of trace few ppm to a few wt% in titanite and preliminary results elements (Table 5). It is known that titanite contains REE, suggest that the titanite from the Baima syenites can contain especially the LREE, and Y in detectable quantities (Sahama, up to 10 wt% ÆREE Y (Shellnutt, 2007). 1946; Smith, 1970; Higgins & Ribbe, 1976; Ribbe, 1980; The amount of total iron, expected to be mostly Fe3, is Hughes et al., 1997). Typically the ÆREE can vary from a typically ,3 wt% in titanite, however the results here indicate

eschweizerbart_xxx Mineralogy from three Permian peralkaline granitoids, SW China 51

Table 2. Compositions of the sodic-calcic and sodic amphibole from the peralkaline plutons.

Sample B-62 B-64 A-1 C-58 A-34 B-31 D-75 C-27 B-18 A-13 A-29 ex2-79 Pluton GS03-097 GS03-097 Baima Baima Panzhihua Panzhihua Taihe Taihe Panzhihua Panzhihua Taihe Taihe

SiO2 (wt%) 52.09 51.58 49.99 51.71 49.80 49.36 48.97 49.72 50.47 51.77 50.86 53.72 TiO2 0.67 0.46 1.33 1.58 1.73 1.63 1.74 1.15 2.01 1.63 1.76 0.36 Al2O3 1.63 2.49 1.57 1.54 0.42 0.35 1.03 1.21 0.19 0.47 0.23 0.55 Cr2O3 0.16 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 FeO 21.66 18.49 27.37 26.09 33.77 33.74 33.11 32.69 34.31 33.02 34.99 36.17 MnO 1.58 0.97 1.19 1.29 1.17 0.82 0.98 1.27 0.91 0.69 0.90 0.48 MgO 9.16 11.25 5.40 5.56 0.04 0.04 0.95 0.64 0.04 1.45 0.23 0.29 CaO 4.81 6.21 5.21 4.47 3.09 3.08 3.53 3.28 1.22 1.74 1.67 0.37 Na2O 5.43 4.79 5.55 5.51 7.10 8.33 6.09 6.54 8.73 7.43 6.86 6.35 K2O 1.21 1.16 1.27 1.17 1.47 1.61 1.62 1.54 1.48 0.95 1.61 0.23 F 2.02 2.04 1.63 1.14 1.11 0.67 0.60 0.79 0.96 0.80 0.40 0.00 Cl 0.01 0.04 0.05 0.00 0.04 0.03 0.01 0.05 0.02 0.00 0.00 0.01 H2O 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 O F 0.85 0.86 0.69 0.48 0.47 0.28 0.25 0.33 0.40 0.34 0.17 0.00 O Cl 0.00 0.01 0.01 0.00 0.01 0.01 0.00 0.01 0.01 0.00 0.00 0.00 Total 100.73 99.75 101.13 100.77 100.81 100.91 100.48 100.59 101.42 101.72 101.37 100.52 Cations on the basis of 23 O Si 7.715 7.569 7.714 7.729 7.921 7.934 7.806 7.777 7.965 7.915 7.958 7.905 Ti 0.074 0.050 0.154 0.177 0.207 0.197 0.209 0.135 0.238 0.187 0.207 0.040 Al 0.285 0.431 0.286 0.271 0.079 0.066 0.194 0.223 0.035 0.085 0.042 0.095 Cr 0.018 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Fe3 0.796 0.646 0.474 0.420 0.000 0.000 0.337 0.145 0.017 0.491 0.602 1.476 Fe2 1.887 1.623 3.058 2.841 4.492 4.535 4.077 4.131 4.510 3.731 3.976 2.974 Mn 0.199 0.120 0.156 0.163 0.158 0.112 0.132 0.168 0.121 0.089 0.119 0.060 Mg 2.022 2.462 1.242 1.239 0.009 0.010 0.226 0.149 0.009 0.330 0.054 0.063 Ca 0.763 0.977 0.861 0.716 0.527 0.530 0.603 0.550 0.207 0.285 0.280 0.058 Na 1.561 1.362 1.661 1.596 2.190 2.596 1.882 1.983 2.670 2.203 2.081 1.811 K 0.228 0.217 0.250 0.224 0.298 0.330 0.329 0.307 0.299 0.185 0.321 0.044 F 0.948 0.947 0.795 0.538 0.558 0.341 0.302 0.391 0.478 0.387 0.198 0.000 Cl 0.001 0.009 0.013 0.000 0.011 0.008 0.003 0.013 0.005 0.000 0.000 0.003 Total 15.548 15.456 15.856 15.376 15.881 16.310 15.801 15.569 16.072 15.502 15.641 14.526

that some titanite contain up to 5.75 wt% Fe2O3 (Ribbe, 1980; from sample GS03-097 has MnO wt% values between 6.3 Deer et al., 1997a; Della Ventura & Bellatreccia, 1999; Frost and 11.1 wt% whereas the remaining Baima syenites have et al., 2000). The titanite from the Taihe granites are Na2O- ilmenite with MnO values between 2.7 and 5.0 wt%. The rich and have less Fe2O3 and F than the titanites from the ilmenite from the Panzhihua granites have moderate MnO Panzhihua and Baima plutons. However, in general, the tita- content between 4.6 and 6.0 wt%, in contrast the ilmenite nite is Fe2O3-rich and Al2O3-poor. The Cl content is negligi- from Taihe granites have the highest MnO contents ble but the F-content is 1.2 wt%. (MnO 9.0–15.9 wt%). The MnO content of ilmenite is unusually high, especially for the Taihe granites and sam- 5.6. Apatite ple GS03-097 (Baima). Magnetite was found only within the Baima syenites. It , The apatite is fluorapatite containing between 3.6 and 4.5 is nearly pure end-member magnetite with 1.2 wt% TiO2 and no evidence of oxidation exsolution lamellae. There wt% F with between 0.4 and 3.8 wt% Ce2O3 (Table 6). Due to the high F content, the apatites do not have calculated are very low (0.2 wt%) concentrations of Al2O3, MgO, MnO and Cr2O3 (Table 8). XOH. Some of the apatite crystals have high FeOt (.1.0 wt%) concentration. There does not appear to be a sub- stantial compositional difference between the apatites from the three plutons for the exception of sample GS03- 6. Discussion 097 which has higher SrO and lower FeOt, La2O3 and Ce2O3 than the others. 6.1. Magmatic/subsolidus trends of the amphiboles

5.7. Ilmenite and magnetite The composition/type of amphibole (e.g., sodic or sodic- calcic) is related to temperature, oxygen fugacity (fO2), Ilmenite is observed in all three plutons and can be sub- abundance of volatiles and alkalinity (Charles, 1977; divided on the basis of MnO content into four groups Mitchell, 1990; King et al., 2000; Scaillet & Macdonald, which correspond to their host rock (Table 7). The ilmenite 2001, 2003). The differences in amphibole composition

eschweizerbart_xxx 52 J.G. Shellnutt, Y. Iizuka

3 Baima syenites are different. This relationship is mirrored with the major-element data. The main differences between the Baima amphiboles and those from the Taihe and Panzhihua granites are the Fe2 Fe3 (apfu), Mg (apfu) 2 and Ti (apfu) values (Fig. 4). The amphiboles with the highest Mg (apfu) are richterite from sample GS03-097 of the Baima Group I pluton. Overall, amphiboles from the Panxi plutons show

Mg (apfu) similar compositional trends as sodic and sodic-calcic amphi- 1 boles from other silica-saturated peralkaline plutons (Fig. 6). Panzhihua The amphiboles generally define magmatic/subsolidus Taihe Group II trends, however there is evidence which indicates an oxida- Baima GS03-097 Group III tion or a late-stage reaction trend in the Panzhihua and Taihe granites, suggesting they may have had a different post- 0 1234 5magmatic history from the Baima syenites (Giret et al.,

2+ 3+ 1980; Strong & Taylor, 1984; Mitchell, 1990). Fe +Fe (apfu) Alternatively, the oxidation trends may be a feature of post- 0.5 emplacement deuteric alteration. Sodic-Calcic The magmatic/subsolidus trends observed in the amphi- Panzhihua boles suggest that there was a common evolution within each 0.4 Taihe Baima pluton in spite of their major-element differences. The com- GS03-097 positional differences, in particular the MgO content between

0.3 Sodic the amphiboles, are possibly related to the oxidation state of (apfu) the magmas (Wones, 1989). Therefore the compositional Ti differences between the syenites and the granites may be 0.2 due to the original starting composition and initial magmatic conditions (e.g., temperature, F-content, pressure and fO2). 0.1

0.0 23456.2. Magma temperature estimates

2+ 3+ Fe +Fe (apfu) The initial magmatic temperature of some granitic rocks Fig. 4. Subdivision of richterites from GS03-097 and ferrorichterites can be estimated using zircon saturation thermometry from the Panzhihua, Taihe and Baima plutons using (a) Mg (apfu) (Watson & Harrison, 1983; Hanchar & Watson, 2003; versus Fe2 Fe3 (apfu), and the sodic-calcic amphiboles from the Miller et al., 2003). Obtaining meaningful temperature sodic amphiboles using (b) Ti (apfu) versus Fe2 Fe3 (apfu). estimates for peralkaline rocks is problematic because peralkalinity increases Zr solubility whereas F, a common may therefore indicate localized changes in magmatic or constituent in peralkaline rocks, can suppress zircon crys- post-magmatic conditions (e.g., temperature, fO2, oxida- tallization and thus the host magma may not become satu- tion/reduction reactions) or merely a difference in the rated, producing meaningless temperature estimates parental magma composition (Neumann, 1976; Deer (Watson, 1979; Keppler, 1993; Miller et al., 2003). et al., 1997b). Watson (1979) pointed out that the crystallization of Amphibole is the most abundant mafic mineral within the alkali-rich pyroxene (e.g., aegirine) and alkali-rich amphi- peralkaline plutons although there are variable quantities bole (e.g., ferrorichterite) may allow magmas to become (5%) of aegirine phenocrysts within the Baima syenites. saturated in Zr. Whole-rock elemental chemistry of the peralkaline plutons It is clear that the Zr content in the Panxi peralkaline is shown in Table 1. It is clear that the Panzhihua and Taihe granitoids is not necessarily higher in the samples which granites are similar in composition and that the Baima sye- have high (Na K)/Al values, indicating that Zr saturation nites are different (Fig. 2). The abundance of quartz is the likely occurred (Fig. 7). The calculated M ([Na K major mineralogical difference between the Baima syenites 2Ca]/[Al*Si]) values, of the Panzhihua (M 1.74 0.09), and the granites of Panzhihua and Taihe. The Baima syenites Baima (M 1.88 0.04) and Taihe (M 1.74 0.05) have 10 vol% modal quartz whereas the Panzhihua and plutons are within the experimental range specified by Taihe granites have 20–25 vol% modal quartz which Watson & Harrison (1983) (Fig. 8). The presence of mag- accounts for the differences in their bulk SiO2 wt%. There matic zircon within all of the granitoids and their M values are other differences in their mineralogy, in particular higher suggest the calculated Zr saturation temperatures may be amount of fluorite, aegirine, titanite, apatite and zircon and reasonable (Hanchar & Watson, 2003). absence of aenigmatite in the Baima pluton. The Zr concentrations of the three plutons have a wide The amphibole compositions of the Panzhihua and Taihe range (Panzhihua 635–1820 ppm; Baima 182–1256 granites are very similar whereas the amphiboles of the ppm; Taihe 421–1400 ppm). Previous studies involving

eschweizerbart_xxx Mineralogy from three Permian peralkaline granitoids, SW China 53

Table 3. Compositions of the aegirine-augite from the peralkaline plutons.

Sample Y-99 A-19 A-20 EE-55 B-39 ex-51 055-A-2 C-82 Pluton GS03-097 GS03-097 GS03-097 Panzhihua Panzhihua Taihe Taihe Taihe

SiO2 (wt%) 52.12 52.16 52.47 52.75 52.63 54.43 53.07 52.06 TiO2 0.00 0.46 0.45 0.09 1.79 1.80 0.98 3.86 Al2O3 0.33 0.27 0.29 0.42 0.25 0.26 0.66 0.14 Cr2O3 0.00 0.00 0.06 0.00 0.05 0.11 0.00 0.00 Fe2O3 30.06 29.58 30.42 32.67 30.06 30.96 30.37 30.73 MnO 0.75 1.04 0.89 0.86 0.93 0.28 1.09 0.63 MgO 1.31 1.69 1.62 0.07 0.00 0.05 0.56 0.02 CaO 6.69 6.37 5.93 0.14 0.52 0.58 1.44 0.18 Na2O 9.44 9.09 9.48 13.85 14.07 12.76 13.49 13.23 K2O 0.02 0.01 0.00 0.00 0.00 0.00 0.00 0.00 P2O5 0.08 0.03 0.00 0.05 0.02 0.00 0.04 0.00 F 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Cl 0.02 0.00 0.01 0.02 0.00 0.00 0.04 0.03 Total 100.80 100.71 101.61 100.92 100.32 101.23 101.73 100.88 Cations on the basis of 6 O Si 1.982 1.981 1.977 2.006 2.006 2.036 1.996 1.972 Ti 0.000 0.013 0.013 0.003 0.051 0.051 0.028 0.110 Al 0.015 0.012 0.013 0.019 0.011 0.011 0.029 0.006 Cr 0.000 0.000 0.002 0.000 0.001 0.003 0.000 0.000 Fe3 0.860 0.845 0.862 0.935 0.862 0.871 0.859 0.875 Mn 0.024 0.033 0.029 0.028 0.030 0.009 0.035 0.020 Mg 0.074 0.096 0.091 0.004 0.000 0.003 0.031 0.001 Ca 0.272 0.259 0.239 0.006 0.021 0.023 0.058 0.007 Na 0.695 0.669 0.692 1.021 1.039 0.925 0.983 0.971 K 0.001 0.001 0.000 0.000 0.000 0.000 0.000 0.000 P 0.002 0.001 0.000 0.002 0.001 0.000 0.001 0.000 Total 3.925 3.911 3.918 4.022 4.023 3.933 4.021 3.963

Mg modelling from previous studies (Clemens et al., 1986;

GS03-097 Creaser & White, 1991; White et al., 2005; Shellnutt & Taihe Panzhihua Jahn, 2009, 2010; Shellnutt et al., 2009a). The high tem- perature estimates would suggest that the magmas were dry as evident from the hypersolvus feldspar although rocks of similar composition have lower temperatures and higher H2O contents, thus casting doubt on the results Ca Ti (Scaillet & Macdonald, 2003, 2006a). However, the calcu- Fig. 5. Ca–Mg–Ti (apfu) ternary diagram of the aegirine-augites lated temperature estimates are close to the range esti- from the Baima (GS03-097), Taihe and Panzhihua plutons. mated by Pang et al. (2008a and b) for the end of oxide–silicate equilibration in the Panzhihua layered mafic intrusion (950C). The oxide–silicate equilibra- radiometric age dating of zircons and Hf isotopic studies tion temperature is important because it likely represents indicate that there are few inherited zircons within the the time when the felsic magmas began to consolidate after plutons (Xu et al., 2008; Shellnutt et al., 2009b). The fractionation of the mafic minerals in the Panzhihua intru- sion. Considering the basaltic origins of the peralkaline estimated Zr saturation temperatures (TZr C) were calcu- lated for all three plutons (Fig. 9). The Baima syenites granitoids and experimental studies, magmatic tempera- produced the highest range of estimated temperatures ture estimates are not unreasonable (Clemens et al., 1986; between 755 and 950C with the lowest mean (860 Price et al., 1999). 17C) (Fig. 9a). The Panzhihua granites produced the smallest range of estimated temperatures between 871 6.3. Initial magmatic fluorine concentrations and 1052C with the highest mean (940 21C) (Fig. 9b). The Taihe granites produced an intermediate Using the fluorite stability calculation of Scaillet & range (848–1005 C) with an intermediate mean of 897 Macdonald (2004) and the TZr( C) estimates, we have 14 C (Fig. 9c). calculated the Fmelt(wt%) composition of the granitoids. The temperature estimates for the Panxi plutons are The Panzhihua granites are calculated to have Fmelt(wt%) within the expected temperature range of some A-type composition of 1.1 0.1 wt% which is similar to the Taihe granitic rocks (870–950C) and the results of MELTS (F 1.1 0.1 wt%) granites but higher than the Baima

eschweizerbart_xxx 54 J.G. Shellnutt, Y. Iizuka

Table 4. Compositions of the aenigmatite from the Panzhihua and Taihe plutons.

Sample A-10 A-19 A-2 A-1 057_A-13 057_A-14 B-32 B-25 Pluton Panzhihua Panzhihua Panzhihua Panzhihua Taihe Taihe Taihe Taihe

SiO2 (wt%) 41.38 42.08 41.85 42.06 42.90 42.40 41.11 39.37 TiO2 10.01 9.59 9.41 9.32 9.77 9.67 8.58 8.48 Al2O3 0.41 0.71 0.54 0.71 0.06 0.12 0.13 0.60 Cr2O3 0.00 0.00 0.00 0.05 0.00 0.00 0.00 0.00 FeO 41.46 41.82 41.13 41.60 40.06 39.73 39.74 42.01 MnO 1.15 0.75 0.96 1.07 1.87 2.15 1.42 1.66 MgO 0.06 0.13 0.07 0.05 0.16 0.08 0.10 0.13 CaO 0.78 0.71 0.54 0.56 0.22 0.12 0.07 0.52 Na2O 7.42 7.94 7.11 7.67 8.50 8.85 7.06 7.35 K2O 0.02 0.03 0.00 0.03 0.03 0.02 0.00 0.04 P2O5 0.00 0.00 0.08 0.00 0.11 0.00 0.00 0.00 F 0.02 0.00 0.00 0.00 0.03 0.00 0.00 0.19 Cl 0.01 0.01 0.01 0.01 0.03 0.00 0.02 0.00 Total 102.72 103.76 101.69 103.12 103.74 103.13 98.22 100.34 Cations on the basis of 20 O Si 5.816 5.841 5.905 5.872 5.933 5.913 6.009 5.749 Ti 1.058 1.000 0.998 0.978 1.016 1.014 0.943 0.930 Al 0.068 0.115 0.090 0.117 0.010 0.019 0.022 0.104 Cr 0.000 0.000 0.000 0.006 0.000 0.000 0.000 0.000 Fe2 4.872 4.853 4.852 4.855 4.631 4.632 4.856 5.129 Mn 0.137 0.088 0.114 0.126 0.219 0.254 0.176 0.205 Mg 0.013 0.027 0.014 0.010 0.032 0.016 0.021 0.028 Ni 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Ca 0.118 0.106 0.081 0.084 0.032 0.017 0.011 0.081 Na 2.020 2.135 1.944 2.074 2.279 2.390 1.998 2.080 K 0.004 0.005 0.000 0.006 0.005 0.004 0.000 0.007 P 0.000 0.000 0.010 0.000 0.013 0.000 0.000 0.000 Total 14.104 14.171 14.008 14.128 14.169 14.260 14.036 14.313

Table 5. Compositions of the titanite from the peralkaline plutons.

Sample Y-86 B-51 A-31 A-1 A-15 A-5 Ex57 ex58 Pluton GS03-097 GS03-097 Baima Baima Panzhihua Panzhihua Taihe Taihe

SiO2 (wt%) 29.16 30.33 29.13 28.81 31.88 30.19 28.91 28.89 TiO2 30.66 35.09 33.63 33.46 32.52 35.53 35.81 37.88 Al2O3 0.47 0.52 0.32 0.31 1.13 0.71 0.34 0.40 Fe2O3 5.75 4.36 4.09 3.87 5.49 3.31 2.86 1.70 MnO 0.00 0.27 0.22 0.00 0.18 0.03 0.00 0.00 MgO 0.09 0.04 0.01 0.05 0.16 0.08 0.06 0.04 CaO 25.11 26.15 25.19 24.56 25.66 27.24 27.51 27.82 Na2O 0.38 0.36 0.70 0.63 0.33 0.15 0.15 0.55 K2O 0.02 0.00 0.01 0.00 0.05 0.05 0.00 0.00 Cr2O3 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 Cl 0.04 0.03 0.02 0.01 0.03 0.03 0.00 0.00 F 0.62 0.67 1.24 0.79 0.13 0.44 0.00 0.00 Total 92.31 97.82 94.55 92.49 97.56 97.77 95.65 97.28 Cations on the basis of 4 Si Si 4.000 4.000 4.000 4.000 4.000 4.000 4.000 4.000 Ti 2.948 3.374 3.234 3.218 3.127 3.417 3.444 3.643 Al 0.071 0.078 0.048 0.047 0.170 0.107 0.051 0.060 Fe3 0.554 0.419 0.393 0.372 0.528 0.319 0.275 0.164 Mn 0.000 0.029 0.024 0.000 0.019 0.003 0.000 0.000 Mg 0.017 0.008 0.002 0.010 0.030 0.015 0.011 0.008 Ca 3.440 3.583 3.451 3.365 3.516 3.732 3.769 3.812 Na 0.094 0.089 0.174 0.156 0.082 0.037 0.037 0.136 K 0.003 0.000 0.002 0.000 0.008 0.008 0.000 0.000 Cr 0.000 0.000 0.000 0.000 0.000 0.000 0.002 0.000 Total 11.128 11.581 11.328 11.167 11.481 11.638 11.589 11.822

eschweizerbart_xxx Mineralogy from three Permian peralkaline granitoids, SW China 55

Table 6. Compositions of the apatite from the peralkaline plutons.

Sample 097-5 097-i18 004_Z19 004_Z22 014_1 014_2 054_2 057_12 Pluton GS03-097 GS03-097 Baima Baima Panzhihua Panzhihua Taihe Taihe

TiO2 (wt%) 0.00 0.00 0.18 0.00 0.55 0.02 0.04 0.06 FeO 0.00 0.10 0.44 0.54 1.03 0.59 0.54 1.10 SrO 1.26 3.58 0.18 0.07 0.07 0.02 0.05 0.00 CaO 51.34 49.94 50.15 49.90 49.03 48.59 51.37 50.36 Na2O 0.38 0.16 0.38 0.47 0.25 0.45 0.44 0.47 P2O5 42.11 41.09 39.91 40.89 38.56 39.74 41.49 41.01 La2O3 0.38 0.43 1.13 0.61 0.89 0.83 0.97 0.71 Ce2O3 0.83 0.87 2.60 2.48 2.58 2.06 1.69 1.83 SO3 0.30 0.00 0.01 0.01 0.06 0.00 0.00 0.00 F 3.90 4.46 4.22 4.31 3.88 3.69 3.90 3.78 Total 100.50 100.63 99.20 99.28 96.90 95.99 100.49 99.32 Cations on the basis of 26 (O,OH,F,Cl) Ti 0.000 0.000 0.020 0.000 0.070 0.000 0.010 0.010 Fe2 0.000 0.015 0.065 0.081 0.155 0.090 0.079 0.162 Sr 0.129 0.368 0.018 0.007 0.007 0.002 0.005 0.000 Ca 9.678 9.477 9.514 9.537 9.439 9.536 9.601 9.509 Na 0.130 0.055 0.130 0.163 0.087 0.160 0.149 0.161 P 6.273 6.162 5.983 6.175 5.866 6.163 6.127 6.119 La 0.020 0.030 0.070 0.040 0.060 0.060 0.060 0.050 Ce 0.050 0.060 0.170 0.160 0.170 0.140 0.110 0.120 Total 16.280 16.170 15.970 16.160 15.850 16.150 16.140 16.130 S 0.040 0.000 0.001 0.001 0.008 0.000 0.000 0.000 F 2.170 2.498 2.363 2.431 2.205 2.138 2.151 2.107

Table 7. Compositions of the ilmenite from the peralkaline plutons.

Sample A-7 C-39 A-6 A-9 C-37 A-30 057_A-12 ex2-54 Pluton GS03-097 GS03-097 Baima Baima Panzhihua Panzhihua Taihe Taihe

SiO2 (wt%) 0.33 0.07 0.11 0.03 0.61 0.25 0.44 0.01 TiO2 51.20 49.60 49.68 48.97 50.37 49.55 52.04 53.69 Al2O3 0.08 0.00 0.03 0.06 0.21 0.01 0.03 0.01 Cr2O3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 FeOtot 35.41 41.33 44.38 44.69 41.98 41.92 29.79 37.73 MnO 11.13 6.24 4.95 2.70 4.65 6.00 15.86 9.03 MgO 0.00 0.05 0.01 0.10 0.00 0.08 0.00 0.05 CaO 0.13 0.12 0.04 0.06 0.05 0.04 0.71 0.00 Total 98.27 97.42 99.21 96.6 97.86 97.83 98.87 100.52 Fe2O3 0.45 3.57 5.19 3.99 0.81 3.69 0.00 0.00 FeO 35.00 38.12 39.72 41.09 41.25 38.60 29.79 37.74 Total 98.32 97.78 99.73 97.00 97.95 98.20 98.87 100.53 Si 0.009 0.002 0.003 0.001 0.016 0.006 0.011 0.000 Al 0.003 0.000 0.001 0.002 0.006 0.000 0.001 0.000 Fe3 0.009 0.069 0.099 0.078 0.016 0.071 0.000 0.000 Ti 0.986 0.963 0.947 0.959 0.973 0.958 0.994 1.013 Cr 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Fe2 0.750 0.823 0.842 0.895 0.886 0.830 0.633 0.792 Mn 0.241 0.137 0.106 0.060 0.101 0.131 0.341 0.192 Mg 0.000 0.002 0.001 0.004 0.000 0.003 0.000 0.002 Ca 0.004 0.003 0.001 0.002 0.001 0.001 0.019 0.000 Total 2.000 2.000 2.000 2.000 2.000 2.000 2.000 2.000

syenites (F 0.8 0.1 wt%). The Fmelt(wt%) values are and monazite (F 7.0 wt%) comprise the bulk of the not exceptionally high but the Panzhihua granites have F-bearing minerals. Based on the observed modal miner- higher F content than what can be accounted for in the alogy, the bulk F content for the Panzhihua granites should bulk mineralogy. For example, fluorite was not observed be 0.2 wt%. The calculated Fmelt(wt%) values for the and the amphibole (F 1.2 wt%), titanite (F 0.40 wt%) Taihe granites are also too high as 15 vol% amphibole

eschweizerbart_xxx 56 J.G. Shellnutt, Y. Iizuka

Table 8. Compositions of the magnetite from samples GS03-097 (Baima).

Sample A-37 A-35 A-36 A-1 A-33 A-34 A-38 Pluton GS03-097 GS03-097 GS03-097 GS03-097 GS03-097 GS03-097 GS03-097

SiO2 (wt%) 0.67 0.40 0.13 0.10 0.07 0.07 0.04 TiO2 0.05 0.52 0.40 0.46 0.06 0.24 0.23 Al2O3 0.18 0.00 0.03 0.00 0.02 0.03 0.06 Cr2O3 0.00 0.02 0.11 0.00 0.01 0.00 0.00 FeOtot 90.59 89.34 91.55 92.09 93.78 90.75 91.05 MnO 0.00 0.00 0.00 0.10 0.00 0.13 0.00 MgO 0.15 0.05 0.00 0.00 0.00 0.00 0.00 Total 91.63 90.33 92.22 92.74 93.94 91.21 91.37 Fe2O3 65.96 64.84 67.01 67.51 69.26 66.88 67.05 FeO 31.24 30.99 31.25 31.34 31.46 30.57 30.72 Total 98.24 96.82 98.94 99.50 100.88 97.91 98.09 Si 0.026 0.016 0.005 0.004 0.003 0.003 0.002 Al 0.008 0.000 0.001 0.000 0.001 0.001 0.003 Fe3 1.937 1.937 1.962 1.966 1.990 1.979 1.981 Ti 0.001 0.015 0.012 0.013 0.002 0.007 0.007 Cr 0.000 0.001 0.003 0.000 0.000 0.000 0.000 Fe2 1.019 1.029 1.017 1.014 1.004 1.005 1.008 Mn 0.000 0.000 0.000 0.003 0.000 0.004 0.000 Mg 0.008 0.003 0.000 0.000 0.000 0.000 0.000 Total 3.000 3.000 3.000 3.000 3.000 3.000 3.000

(F 0.5 wt%), 1 vol% titanite (F 0.5 wt%), 0.05 vol% assemblages of the Panzhihua and Taihe granites are very apatite (F 5.0 wt%), 1 vol% fluorite (F 43 wt%) and similar and since they both contain aenigmatite and ilme- 0.05 vol% monazite (F 7.0 wt%) would be equal to 0.5 nite, estimates can be made on their likely fO2 conditions wt% F. It is uncertain why there are discrepancies between (White et al., 2005). The Baima syenites, in comparison, the calculated Fmelt(wt%) values and the modal F contents are mineralogically and chemically different which may of the Panzhihua and Taihe granites, but it could be related indicate a different magmatic oxidation state. to deuteric alteration, devolatilization during crystalliza- The Baima syenites do not contain aenigmatite but the tion or high calculated TZr( C) estimates. If the TZr( C) presence of titanite, quartz and magnetite indicates fO2 condi- temperature estimates are incorrect, then magma tempera- tions at or above the FMQ buffer (Charles, 1977; Wones, 1981, tures must be between 600C and 650C in order to have 1989; Platt & Woolley, 1986; Xirouchakis & Lindsley, 1998). the Fmelt(wt%) values equal to the calculated F values. Applying the equilibrium equation of Wones (1989), estimated However, F is typically lost during the subsolidus break- log fO2 values for the Baima syenites fall along the TMQ-HD down of amphibole to aegirine. There is textural evidence IL curve (Fig. 10). The calculated values are assumed to be the for such a reaction in both plutons and it is likely that the maximum because Xirouchakis & Lindsley (1998) have sug- Panzhihua and Taihe granites lost F during subsolidus gested that the titanite–magnetite–quartz assemblage is not reactions or devolatilization, rather than incorrect tempera- necessarily higher than the FMQ buffer. ture estimates (Scaillet & Macdonald, 2001, 2006a). The aenigmatite–ilmenite assemblage within the In contrast, the calculated Fmelt(wt%) content of the Panzhihua and Taihe granites indicate that they formed Baima syenites are roughly equal to the observed mode under lower fO2 conditions than the Baima syenites (Famphibole 1.5 wt%, Ftitanite 1.0 wt%, Fapatite 5.0 (Lindsley, 1971; Marsh, 1975; Borley, 1976). The ilmenites wt%, Ffluorite 45 wt%) of the F-bearing minerals (Fmelt from the Panzhihua and Taihe granites have higher TiO2 and 0.8 0.1 wt%, Fmode 0.7 wt%). There is some textural lower FeOt than the Baima syenites. The ilmenite from the evidence for the breakdown of amphibole to aegirine but Taihe granites consistently has higher MnO content (MnO both minerals appear as separate phenocrysts thus the bulk 9.0–15.9 wt%) than the other plutons. White et al. (2005) F content is accounted for in the Baima syenites. have shown that fO2 conditions can be estimated if aenigma- tite and ilmenite are present within silica-saturated peralkaline lavas. Using the equation log fO2 24,698/T 10.293 6.4. Magma oxidation states 0.215 (P 1)/T 2log(aHem) log(aIlm) where tempera- ture is in kelvins, pressure in bars and aHem and aIlm are the The oxygen fugacity (fO2) of igneous rocks may be deter- activities of the Fe2O3 and FeTiO3 components within ilme- mined by analyzing mineral pairs of ilmenite and magne- nite, we have attempted to estimate the fO2 of the Taihe and tite and also fayalite and clinopyroxene and ilmenite and Panzhihua granites (White, pers. comm.). Due to the high aenigmatite (Buddington & Lindsley, 1964; Anderson pyrophanite (Mn) and ilmenite (Ti) components within ilme- et al., 1993; White et al., 2005). Generally, the mineral nite from the Taihe and Panzhihua granites, aHem cannot be

eschweizerbart_xxx Mineralogy from three Permian peralkaline granitoids, SW China 57

3 1.3 (a) Ta Panzhihua 1.2

Magmatic/Subsolidus T 2 Ba 1.1 Kt Ca Al<=> Na

IV

iv Peralkaline 1.0 Fe2+ Si <=> Fe 3+ Al IV Si rends

Na+K/Al Ca+Al Metaluminous 1 Wi Rt 0.9

0.8 Panzhihua Oxidation Trends Rb A A 2+ 3+ Baima K Na Fe <=> Fe Ar Taihe 0 0.7 8 9 10 11 12 0 500 1000 1500 2000 Si+Na+K Zr (ppm) 3 (b) Fig. 7. (Na K)/Al versus Zr (ppm) of the peralkaline plutons. Data Ta Baima from Shellnutt & Zhou (2007) and Shellnutt et al. (2009a). GS03-097 Magmatic/Subsolidus Trends Kt 2 Ba Ca Al<=> Na Si 1000 IV

iv Panzhihua Baima Taihe Ca+Al 800 Fe2+ Si <=> Fe 3+ Al IV Rt 1 Wi 600

Oxidation Trends Rb KA Na A Fe 2+ <=> Fe 3+ Ar Zr (ppm) 400 0 M = 1.3-1.9 8 9 10 11 12 Si+Na+K 200 3 (c) M = 1.6 Ta Taihe 0 Magmatic/Subsolidus Trends 700 800 900 1000 Kt 2 Ba Ca Al<=> Na Si TZr (C)°

IV

iv Fig. 8. Zirconium (ppm) versus zirconium-saturation thermometry estimate (TZr C) of the peralkaline plutons. The M value (Na K Ca+Al Fe2+ Si <=> Fe 3+ Al IV Rt 2Ca/[Al*Si]). The calibrated range for M values is shown in grey 1 Wi (Miller et al., 2003). Data from Shellnutt & Zhou (2007) and Shellnutt et al. (2009a).

Oxidation Trends KA Na A Fe 2+ <=> Fe 3+ Rb Ar accurate the fO path is. The estimated fO path does pass 0 2 2 8 9 10 11 12 through the ferrorichterite stability field at lower tempera- Si+Na+K tures, which would be consistent with late-stage crystalliza- tion and the high Mn content of the ilmenite. For example, Fig. 6. Compositional trends of the amphiboles from the (a) Panzhihua Mn-rich ilmenites from adamellites of California are inter- pluton, (b) Baima pluton and (c) Taihe pluton plotted according to Aliv preted as representing late-stage crystallization (Snetsinger, Ca versus Si Na K. Nomenclature based on Leake et al. (2004). 1969). It is likely that, given the similar bulk-rock composi- Compositional fields based on Giret et al. (1980) and the magmatic/ subsolidus and oxidation trends from Strong & Taylor (1984). tions and mineralogy, the estimated fO2 path may be applic- Experimental amphiboles from Scaillet & Macdonald (2003, 2006a) able for the Taihe granites. Furthermore, the low Mg- are also plotted for reference. The amphiboles from Scaillet & amphibole in the granitic plutons is consistent with experi- , , Macdonald (2003) are grey stars at fO2 FMQ–1. The amphiboles mentally derived amphiboles from reducing (fO2 FMQ–1) from Scaillet & Macdonald (2006a) are white stars at fO2 FMQ–1.3. conditions (Wones, 1989; Scaillet & Macdonald, 2003, 2006a) (Fig. 6). calculated for the exception of one analysis from Panzhihua The fO2 estimates for all of the plutons are in agreement (C-37). The result is plotted on Fig. 10 and indicates a lower with their amphibole and bulk-rock compositions. fO2 path than for the Baima syenites. Since only one pair of Amphiboles with higher-Mg contents are expected to crys- aenigmatite–ilmenite produced results, it is uncertain how tallize in oxidized conditions whereas lower-Mg

eschweizerbart_xxx 58 J.G. Shellnutt, Y. Iizuka

5 o 1020o C Baima (a) (a) Mean = 860 C 3500 o 930 C N= 35 860o C 4 S.D. = 50.5 3000

2500 3 2000

1500 2

1000 1 500

0 0 1.0 1.2 1.4 1.6 1.8 2.0 750 800 850 900 950 1000 o (Na+K+2Ca)/(Al*Si) TZr ( C) 4

o Panzhihua (b) o 3500 1020 C (b) Mean = 940 C 930o C N= 19 o 3000 860 C 3 S.D. = 45.2

2500

2000 2

1500

Frequency

1000 1

500

Zr (ppm) required for saturation 0 1.0 1.2 1.4 1.6 1.8 2.0 850 900 950 1000 1050 1100 o (Na+K+2Ca)/(Al*Si) TZr ( C) 5 o 1020o C Taihe (c) (c) Mean = 897 C 3500 930o C N= 24 860o C 4 S.D. = 34.7 3000

2500 3 2000

1500 2

1000 1 500

0 1.0 1.2 1.4 1.6 1.8 2.0 800 850 900 950 1000 1050 o (Na+K+2Ca)/(Al*Si) TZr ( C)

Fig. 9. Zircon-saturation temperature estimates for the (a) Baima, (b) Panzhihua and (c) Taihe plutons.

amphiboles crystallize in reducing conditions (Wones, of a mafic parental magma with or without crustal 1989; Scaillet & Macdonald, 2003, 2006a). The major- contamination, crustal melting, or hybrid magmas element chemistry is also in agreement with the fO2 esti- derived from the crust or mafic rocks (Martin, 2007). mates as the compositions of the Panzhihua and Taihe Studies of the temporal, chemical and geological rela- plutons overlap with low fO2 melts from Scaillet & tionships of the Panxi peralkaline plutons and their Macdonald (2003) (Fig. 2). associated layered gabbroic intrusion suggest that the gabbros and granitoids originated by fractional crystal- lization of a mafic parental magma (Shellnutt & Jahn, 6.5. Magmatic history of the peralkaline rocks 2009, 2010; Shellnutt et al., 2009a). The mineral com- positions however, indicate the magmatic conditions Peralkaline silica-saturated granitoids from extensional were likely different between the Baima syenites and settings may be the result of fractional crystallization the Panzhihua and Taihe granites.

eschweizerbart_xxx Mineralogy from three Permian peralkaline granitoids, SW China 59

suggested that A-type granitoids derived by fractionation -5 HM of mafic magmas are indicative of reduced conditions (Loiselle & Wones, 1979). Our results suggest that some A-type granitoids are derived from oxidized magmas. TMQ HD Il -10 Layered 2 Fr basaltic magmas 7. Conclusion Magnetite Granites

log ƒO

Ox 37 GC -15 Rd C- The principal magmatic amphibole in the Panxi peralka- line granitoids rocks is ferrorichterite. The compositions of the ferrorichterite are nearly identical in the Panzhihua and Taihe granites but the Baima syenites contain high-Mg -20 Ilmenite Granites ferrorichterite and richterite. The mineralogical and geo- Ferrorichterite FMQ Stability chemical evidence suggests that the initial magma tem- peratures of the Panxi peralkaline granitic rocks were high 600 800 1000 1200 (860 to 940 C) and their initial Fmelt(wt%) were 1 TCo wt% whereas their magmatic fO2 conditions were differ- ent. The higher fO2 condition is the likely reason why the Fig. 10. Estimated oxygen fugacity (fO2) and temperature ( C) con- ditions of the Baima (solid circles) and Panzhihua (C-37) plutons Baima syenites have Mg-rich amphibole and aegirine- (modified from Wones, 1981). GC graphite–CO2 equilibrium; augite as well as the quartz–titanite–magnetite assemblage FMQ fayalite–magnetite–quartz buffer; HD Il hedenbergite–il- whereas the Panzhihua and Taihe granites have low-Mg menite; TMQ titanite–magnetite–quartz; HM hematite–magne- amphiboles and aenigmatite and ilmenite. Our results sug- tite buffer. Rd reduction of magma by the addition of carbon; Ox gest that some A-type granitoids are formed under oxidiz- oxidation of magma by addition of oxygen or hydrogen loss; Fr ing conditions whereas others are formed under more closed-system fractionation trend of layered basaltic magmas. Stability of ferrorichterite is taken from Mitchell (1990). reducing conditions.

The mineral assemblage (e.g., magnetite, titanite, quartz, Acknowledgements: We would like to thank Sandro fluorite) and amphibole compositions from the Baima syenites Conticelli, Bruno Scaillet and an anonymous reviewer for suggest the parental magma was likely oxidized, had high their constructive reviews and John White for his assis- initial Fmelt(wt%) and magma temperature. These features tance with the aenigmatite–ilmenite calculations. This are consistent with experimental results of some A-type gran- study was supported by Academia Sinica through a post- itoids. Price et al. (1999) have suggested that granitic rocks doctoral fellowship to JGS. This study was partially sup- with primary titanite fluorite indicate high-temperature ported by National Science Council of Taiwan (grant #97- conditions, shallow-crystallization environments, moderate 2116-M-001-008 to YI). fO2, subaluminous (ANCK , 1.1) compositions and F (wt%) . 1 wt%. Shellnutt et al. (2009a) have suggested melt References shallow depths (1.0 kbar) and fO2 FMQ . 0 for the genesis of the Baima syenites. Furthermore, relatively Mg- rich amphiboles are expected in higher fO2 environments Ali, J.R., Thompson, G.M., Zhou, M.-F., Song, X.Y. (2005): (Wones, 1989). Since the Baima syenites are not exceptionally Emeishan large igneous province, SW China. Lithos, 79, peralkaline ([Na K]/Al 1.05 0.02), it is likely that early 475–489. crystallization of aegirine prevented the magma from becom- Anderson, D.J., Lindsley, D.H., Davidson, P.M. (1993): QUILF: a ing too peralkaline and thus allowing Zr saturation (Scaillet & Pascal program to assess equilibria among Fe-Mg-Mn oxides, Macdonald, 2001, 2003; Della Ventura et al., 2005). pyroxenes, olivine, and quartz. Com. Geosci., 19, 1333–1350. The Panzhihua and Taihe granites likely had different mag- Borley, G.D. (1976): Aenigmatite from an aegirine-riebeckite gran- ite, Liruei complex, Nigeria. Mineral. Mag., 40, 595–598. matic conditions. Although the estimated Fmelt(wt%) and Buddington, A.F. & Lindsley, D.H. (1964): Iron-titanium oxide TZr( C) are slightly higher than Baima syenites, the calculated values and mineralogy (e.g., titanite fluorite) are still con- minerals and synthetic equivalents. J. Petrol., 5, 310–357. Charles, R.W. (1975): The phase equilibria of richterite and ferrori- sistent with the geological conditions described by Price et al. chterite. Am. Mineral., 60, 367–374. (1999). Shellnutt & Jahn (2009, 2010) modelled the Panzhihua — (1977): The phase equilibria of intermediate compositions of the and Taihe granites with similar emplacement depths (e.g., pseudobinary Na2CaMg5Si8O22(OH)2-Na2CaFe5Si8O22(OH)2. 1.0 kbar) as the Baima pluton but with lower fO2 (FMQ Am. J. Sci., 277, 594–625. 1). The calculated log fO2 value of the Panzhihua pluton, Clemens, J.D., Holloway, J.R., White, A.J.R. (1986): Origin of an A- based on ilmenite and aenigmatite mineral pairs, suggests a type granite: experimental constraints. Am. Mineral., 71, 317–324. lower fO2 path and possibly reduced conditions. Creaser, R.A. & White, A.J.R. (1991): Yardea dacite- large-volume The difference in oxidation state between the Panxi high temperature felsic volcanism from the middle Proterozoic peralkaline plutons is significant because it was originally of South Australia. Geology, 19, 48–51.

eschweizerbart_xxx 60 J.G. Shellnutt, Y. Iizuka

Deer, W.A., Howie, R.A., Zussman, J. (1997a): Rock forming Mitchell, R.H. (1990): A review of the compositional variation of minerals volume 1A: orthosilicates. Geol. Soc., London, 919 p. amphiboles in alkaline plutonic complexes. Lithos, 26, 135–156. —, —, — (1997b): Rock forming minerals. 2B. Double-chain sili- Neumann, E.-R. (1976): Compositional relations among pyroxenes, cates. Geol. Soc., London, 764 p. amphiboles and other mafic phases in the Oslo region plutonic Della Ventura, G. & Bellatreccia, F. (1999): Zr- and LREE-rich rocks. Lithos, 9, 85–109. titanite from Tre Croci, Vico Volcanic complex (Latium, Pang, K.-N., Zhou, M.-F., Lindsley, D., Zhao, D., Malpas, J. (2008a): Italy). Mineral. Mag., 63, 123–130. Origin of Fe-Ti oxide ores in mafic intrusions: evidence from the Della Ventura, G., Iezzi, G., Redhammer, G.J., Hawthorne, F.C., Panzhihua intrusion, SW China. J. Petrol., 49, 295–313. Scaillet, B., Novembre, D. (2005): Synthesis and crystal-chem- Pang, K.-N., Li, C., Zhou, M.-F., Ripley, E. M. (2008b): Abundant istry of alkali amphiboles in the system Na2O-MgO-FeO-Fe2O3- Fe-Ti oxide inclusions in olivine from the Panzhihua and SiO2-H2O as a function of fO2. Am. Mineral., 90, 1375–1383. Hongge layered intrusions, SW China: evidence for early satura- Frost, B.R., Chamberlain, K.R., Schumacher, J.C. (2000): Sphene tion of Fe-Ti oxides in ferrobasaltic magma. Contrib. Mineral. (titanite): phase relations and role as a geochronometer. Chem. Petrol., 156, 307–321. Geol., 172, 131–148. Pe-Piper, G. (2007): Relationship of amphibole composition to host- Frost, B.R., Barnes, C.G., Collins, W.J., Arculus, R.J., Ellis, D.J., rock geochemistry: the A-type gabbro-granite Wentworth pluton, Frost, C.D. (2001): A geochemical classification for granitic Cobequid shear zone, eastern Canada. Eur. J. Mineral., 19, 29–38. rocks. J. Petrol., 42, 2033–2048. Platt, R.G. & Woolley, A.R. (1986): The mafic mineralogy of the Giret, A., Bonin, B., Leger, J.-M. (1980): Amphibole compositional peralkaline syenites and granites of the Mulanje complex, trends oversaturated and undersaturated alkaline plutonic ring- Malawi. Mineral. Mag., 50, 85–99. complexes. Can. Mineral., 18, 481–495. Price, J.D., Hogan, J.P., Gilbert, M.C., London, D., Morgan VI, G.B. Grew, E.S., Halenius, U., Pasero, M. (2008): The crystal-chemistry (1999): Experimental study of titanite-fluorite equilibria in the of aenigmatite revisited: electron microprobe data, structure A-type Mount Scott granite: implications for assessing F con- refinement and Mo¨ssbauer spectroscopy of aenigmatite from tents of felsic magma. Geology, 27, 951–954. Vesteroya (Norway). Eur. J. Mineral., 20, 983–991. Ribbe, P.H. (1980): Titanite. Rev. Mineral., 5, 137–154. Hanchar, J.M. & Watson, E.B. (2003): Zircon saturation thermome- Sahama, Th. G. (1946): On the chemistry of the mineral titanite. try. Rev. Mineral. Geochem., 53, 89–112. Bull. Comm. Geol. Fin., 24, 88–120. Higgins, J.B. & Ribbe, P.H. (1976): The crystal chemistry and space Scaillet, B. & Macdonald, R. (2001): Phase relations of peralkaline groups of natural and synthetic titanites. Am. Mineral., 61, silicic magmas and petrogenetic implications. J. Petrol. 42, 878–888. 825–845. Hughes, J.M., Bloodaxe, E.S., Hanchar, J.M., Foord, E.E. (1997): —, — (2003): Experimental constraints on the relationships between Incorporation of rare earth elements in titanite: stabilization of peralkaline rhyolites of the Kenya rift valley. J. Petrol., 44, the A2/a dimorph by creation of antiphase boundaries. Am. 1867–1894. Mineral., 82, 512–516. —, — (2004): Fluorite stability in silicic magmas. Contrib. Mineral. Keppler, H. (1993): Influence of fluorine on the enrichment of high Petrol., 147, 1867–1894. field strength trace elements in granitic rocks. Contrib. Mineral. —, — (2006a): Experimental constraints on pre-eruption conditions Petrol., 114, 479–488. of pantelleritic magmas: evidence from the Eburru complex, King, P.L., Hervig, R.L., Holloway, J.R., Delaney, J.S., Dyar, M.D. Kenya rift. Lithos, 91, 95–108. 3 (2000): Partitioning of Fe /Fetotal between amphibole and basa- —, — (2006b): Experimental and thermodynamic constraints on the nitic melt as a function of oxygen fugacity. Earth Planet. Sci. sulphur yield of peralkaline and metaluminous silicic flood Lett., 178, 97–112. eruptions. J. Petrol., 47, 1413–1437. Leake, B.E., Woolley, A.R., Arps, C.E.S., Birch, W.D., Burke, E.A.J., Shellnutt, J.G. (2007): A-type granites of the Permian Emeishan Ferraris, G., Grice, J.D., Hawthorne, F.C., Kisch, H.J., large igneous province: implications for the formation of the Krivovichev, V.G., Schumacher, J.C., Stephenson, N.C.N., giant magmatic oxide deposits. PhD thesis, The University of Whittaker, E.J.W. (2004): Nomenclature of amphiboles: additions Hong Kong, Hong Kong, Hong Kong SAR, 437 p. and revisions to the International Mineralogical Association’s Shellnutt, J.G. & Jahn, B.-M. (2009): Formation of the Late Permian amphibole nomenclature. Eur. J. Mineral., 16, 191–196. Taihe layered gabbro-granite complex: a common regional pro- Lindsley, D.H. (1971): Synthesis and preliminary results on the cess for the formation of the Fe-Ti oxide deposits of the stability of aenigmatite (Na2Fe5TiSi6O20). Carnegie Inst. Emeishan large igneous province (SW China). Eos Trans. Washington, Ann. Rept. Dir. Geophys. Lab., 1969–70, 188–190. AGU, Suppl., Abstract V21-1969. Loiselle, M.C. & Wones, D.R. (1979): Characteristics and origin of —, — (2010): Formation of the Late Permian Panzhihua plutonic- anorogenic granites. Geol. Soc. Am. Abs. Prog., 11, 468. hypabyssal-volcanic igneous complex: implications for the gen- Macdonald, R. (1974): Nomenclature and petrochemistry of the esis of Fe–Ti oxide deposits and A-type granites of SW China. peralkaline oversaturated extrusive rocks. Bull. Volc., 39, Earth Planet. Sci. Lett., 90, 509–519. 498–516. Shellnutt, J.G. & Zhou, M.-F. (2007): Permian peralkaline, peralu- Marsh, J.S. (1975): Aenigmatite stability in silica-undersaturated minous and metaluminous A-type granites in the Panxi district, rocks. Contrib. Mineral. Petrol., 41, 23–29. SW China: their relationship to the Emeishan mantle plume. Martin, R.F. (2007): Amphiboles in the igneous environment. Rev. Chem. Geol., 242, 286–316. Mineral. Geochem., 67, 323–357. Shellnutt, J.G., Zhou, M.-F., Zellmer, G.F. (2009a): The Permian Miller, C.F., McDowell, S.M., Mapes, R.W. (2003): Hot and cold Baima Igneous Complex (SW China): the role of Fe-Ti oxide granites? Implications of zircon saturation temperatures and crystallization in the formation of A-type granitoids with impli- preservation of inheritance. Geology, 31, 529–532. cation for the Daly gap. Chem. Geol., 259, 204–217.

eschweizerbart_xxx Mineralogy from three Permian peralkaline granitoids, SW China 61

Shellnutt, J.G., Wang, C.Y., Zhou, M.-F., Yang, Y. (2009b): Zircon Wones, D.R. (1989): Significance of the assemblage titanite Lu–Hf isotopic compositions of metaluminous and peralkaline magnetite quartz in granitic rocks. Am. Mineral., 74, 744–749. A-type granitic plutons of the Emeishan large igneous province — (1981): Mafic silicates as indicators of intensive variables in (SW China): constraints on the mantle source. J. Asian Earth granitic magmas. Min. Geol., 31, 191–212. Sci., 35, 45–55. Xirouchakis, D. & Lindsley, D.H. (1998): Equilibria among titanite, Siivola, J. & Schmid, R.A (2007): Systematic nomenclature for meta- hedenbergite, fayalite, quartz, ilmenite, and magnetite: experi- morphic rocks: 12. list of mineral abbreviations. Recommendations ments and internally consistent thermodynamic data for titanite. by the IUGS Subcommission on the Systematics of Metamorphic Am. Mineral., 83, 712–725. Rocks., web version of 01.02.2007. Xu, Y.-G., Luo, Z.-Y., Huang, X.-L., He, B., Xiao, Xie, L.-W., Shi, Smith, A.L. (1970): Sphene, perovskite and coexisting Fe-Ti oxide Y.-R. (2008): Zircon U-Pb and Hf isotope constraints on crustal minerals. Am. Mineral., 55, 264–269. melting associated with the Emeishan mantle plume. Geochim. Snetsinger, K.G. (1969): Manganoan ilmenite from a Sierran ada- Cosmochim. Acta, 72, 3084–3104. mellite. Am. Mineral., 54, 431–436. Yavuz, F. (2007): WinAmphcal: a Windows program for the IMA-04 Strong, D.F. & Taylor, R.P. (1984): Magmatic-subsolidus and oxidation amphibole classification. Geochem. Geophys. Geosystems, 8 (1), trends in composition of amphiboles from silica-saturated peralka- doi: 10.1029/2006GC001391. line igneous rocks. Tschermaks Min. Petr. Mitt., 32, 211–222. Zhong, H., Zhu, W.G., Song, X.Y., He, D.F. (2007): SHRIMP U-Pb Sun, S.S. & McDonough, W.F. (1989): Chemical and isotopic systema- zircon geochronology, geochemistry, and Nd-Sr isotopic study tics of oceanic basalts: implications for mantle composition and of contrasting granites in the Emeishan large igneous province, processes. in ‘‘Magmatism in the Ocean Basins’’, A.D. Saunders & SW China. Chem. Geol., 236, 112–137. M.J. Norry, eds., Geol. Soc. London Sp. Publ., 42, 313–435. Zhou, M.-F., Robinson, P.T., Lesher, C.M., Keays, R.R., Zhang, Watson, E.B. (1979): Zircon saturation in felsic liquids: experimen- C.J., Malpas, J. (2005): Geochemistry, petrogenesis and metal- tal data and application to trace element geochemistry. Contrib. logenesis of the Panzhihua gabbroic layered intrusion and asso- Mineral. Petrol., 70, 407–419. ciated Fe-Ti-V oxide deposits, Sichuan Province, SW China. Watson, E.B. & Harrison, T.M. (1983): Zircon saturation revisited: J. Petrol., 46, 2253–2280. temperature and composition effects in a variety of crustal magma types. Earth Planet. Sci. Lett., 64, 295–304. White, J.C., Ren, M., Parker, D.F. (2005): Variation in mineralogy, Received 27 July 2010 temperature, and oxygen fugacity in a suite of strongly peralkaline Modified version received 16 September 2010 lavas and tuffs, Pantelleria, Italy. Can Mineral., 43, 1331–1347. Accepted 16 September 2010

eschweizerbart_xxx