4.b.1. Porphyritic granitoids

Most of the samples (n = 9) of Jhalida porphyritic granitoids (JPG) plot within the monzogranite field of the quartz – alkali feldspar – plagioclase (QAP) diagram, whereas three samples fall within in the syenogranite field and two in the granodiorite field (Fig. 3; Table 1).

Fig. 3. Plots of mineral modal compositions (vol%) for the porphyritic granitoids of the Jhalida pluton in the QAP diagram (after Streckeisen, Reference Streckeisen1976).

A magmatic fabric is the characteristic of porphyritic granitoids that intensifies into a mylonitic fabric approaching the shear zone. The magmatic fabric in porphyritic granitoids is defined by the strong orientation of the elongation of rectangular feldspar megacrysts (Fig. 2a). The megacrysts are commonly microclines and less commonly plagioclase. The megacrysts of microcline are rarely microperthitic and may contain inclusions of finer grains of subhedral and sub-rounded slightly altered plagioclase (Fig. 4a). Plagioclases are invariably characterized by alteration to a greyish mass containing sericite needles, whereas microcline is relatively fresh. Apart from occurring as individual aggregates associated with microclines, plagioclase is present as inclusions of sub-rounded, much smaller grains within microcline commonly showing albite rims. Myrmekite has frequently developed at the contact between microcline and plagioclase. In places, thin lenticels of myrmekite have developed at the long contact of the two adjacent megacrysts of microcline (Fig. 4b). Occasionally thin vein-like myrmekite runs along the contact of coarse microclines. Tabular plagioclase in the groundmass is occasionally idiomorphic. Coarse anhedral quartz, generally mildly deformed, may contain inclusions of both subhedral microcline and sub-rounded plagioclase. In the groundmass, quartz occurs generally as both coarse and medium-sized anhedral grains and as rounded inclusion in feldspars. Biotites (pleochroic from straw yellow to dark brown) occur as thin impersistent layers, commonly forming clusters, and frequently diverge and converge around megacrysts of feldspars as already mentioned. Symplectitic intergrowth of thin to very thin lenticular quartz with coarse flakes of biotite may be present. Clusters of biotite, most of which are subparallel, warp around megacrystic feldspar (Fig. 4d). Accessory minerals such as fine-grained epidote-zoisite, apatite, and medium- to fine-grained sphene, allanite and opaque minerals are typically restricted to biotite-rich layers (Fig. 4e) with the occasional presence of fine-grained xenomorphic quartz within the clusters of biotite laths. This suggests a probability of formation of at least some biotites from Ca–Al-bearing ferromagnesian silicates, such as hornblende, by K-metasomatism related to the intrusion of porphyritic granitoid magma in the schistose amphibolite country rock (Fig. 2e, g). Sphene (weak but distinctly pleochroic from reddish-brown to almost colourless) occurs as aggregates of subhedral fine grains forming corona around magnetite (Fig. 4f), and is very commonly associated with plagioclase–biotite-rich layers in the rock. It occurs also as coarse, skeletal elongate grains which may contain poikilitic inclusions of plagioclase and quartz. The intimately associated sphene-biotite and opaque minerals frequently show general dimensional parallelism defining the foliation trend (Fig. 4g, h). Allanite (weakly pleochroic in shades of yellow and brown; the outer margin of the grain is always browner than the yellowish inner part) occurs as coarse subhedral to fine, equant grains closely associated with magnetite. Around allanite, the quartzo-feldspathic grains show radial cracks. It may also occur as coarse, triangular, subhedral grains in the interspaces of biotite and quartzofeldspathic mass. Zircon occurs as squarish and rectangular medium-sized grains in the interspaces of plagioclase and sphene (Fig. 4i), as fine elongate subhedral to idiomorphic grains in the interspaces of other minerals and as inclusions in plagioclase. Opaque minerals (magnetite) occur as coarse- to fine-sized, rectangular, squarish and rhomboid grains within both plagioclase and the plagioclase–microcline interface.

Fig. 4. Representative photomicrographs of the investigated Jhalida porphyritic granitoids (JPG) and xenolith suites. (a) Inclusions of plagioclase (antiperthite) within microcline megacryst in Jhalida porphyritic granitoid. Note the formation of smaller tabular felsic crystals along the marginal part of the microcline megacryst formed as a result of brittle deformation. Altered plagioclase, microcline and elongate quartz layers are arranged following the outline of the augen (crossed polars). (b) Thin myrmekite lens at the long contact of two megacrysts of microcline in JPG (crossed polars). (c) Coarse lenticular quartz running parallel to the foliation within recrystallized feldspar in JPG (crossed polars). (d) Subparallel biotites show wrapping around megacrystic feldspar in JPG (plane-polarized light). (e) The haphazard orientation of biotite laths in JPG. Note the intimate association of allanite, sphene and magnetite with biotite (crossed polars). (f) Corona of sphene around rhombohedral magnetite in JPG. (g) Thin stringer of sphene–biotite–magnetite assemblage showing general dimensional parallelism with the foliation trend of the JPG. (h) The intimate association of clusters of flecks and criss-cross laths of biotite with sphene in JPG; note the quartz of varied shape and size intimately associated with flaky biotite clusters at the central part. (i) Rhombohedral and tiny elongate/euhedral zircon in the interspaces of feldspars with sphene in JPG. (j) Plagioclase–hornblende schist with partial biotitization (Type-a xenolith). (k) Plagioclase–quartz–biotite (with minor hornblende) schist (Type-b xenolith). (l) Plagioclase–quartz–biotite schist (no hornblende) frequently with medium- to fine-sized quartz (Type-c xenolith).

In the deformation zones, the most conspicuous fabric of the porphyritic granitoids is gneissic, defined by the preferred alignment of ellipsoidal and augen-shaped feldspar megacrysts and stretched felsic layers, along with the finer centimetre-thick impersistent layers of mafic (mainly biotite) quartzofeldspathic minerals defining the foliation trend (Fig. 2c). The stretched felsic layers are composed of aggregates of finer subhedral microcline, plagioclase and quartz. Stretching (and granulation) of the feldspar megacrysts into thin elongate augens running parallel to the gneissosity suggest the continuity of the shearing movement even after the consolidation of the porphyritic granitoid magma. The biotite-bearing groundmass layers show divergence and convergence at the ends of augens of felsic megacrysts. In places, the augens are composed mainly of aggregates of quartz and feldspar instead of feldspar only. The augen-shaped megacrysts of microcline have, in places, developed aggregates of smaller tabular crystals along the marginal part due to brittle deformation. Where the shearing effect is intense, many groundmass feldspars have been elongated parallel to the foliation trend and, in places, show marginal granulation. In high-grade sheared rocks, lenticular coarse quartz runs parallel to the foliation trend and, together with other groundmass minerals, wraps around the elliptical or eye-shaped megacrysts of feldspars, mainly microcline (Fig. 4c). Biotites (pleochroic from straw yellow to dark brown) frequently diverge and converge around augens of feldspars. The secondary epidotization, muscovitization and chloritization are more restricted in the highly deformed samples. Secondary muscovite laths, occasionally present with altered (sericitized) plagioclase, replace biotite in places. Epidote occurs as aggregates of subhedral to anhedral grains; it is an alteration product of plagioclase, commonly in contact with biotite laths. Chlorite occurs as medium-sized laths, simulating the shape of the biotite that it replaces.

4.b.2. Generic interpretation of fabric of porphyritic granitoids

The presence of porphyritic texture, subidiomorphic microcline and plagioclase grains, euhedral zircon, euhedral sphene, magnetite/ilmenite, etc. all suggest a magmatic origin. The preferred E–W orientation of the K-feldspar megacrysts, ferromagnesian minerals such as biotite and opaques, and bending of the (flow) foliation of the porphyritic granitoids around the xenoliths are interpreted as a result of the magmatic flowage. The occasional presence of large ribbon or lenticular quartz grains or augen-shaped feldspar indicates that these have been deformed by shearing under sub-solidus conditions. The preferred orientation of feldspars and biotite is largely a result of magmatic flowage. Moreover, stronger preferred orientation of the microcline megacrysts and their augens or large ribbon or lenticular quartz grains are the result of shear strain. It appears that the magmatic fabric becomes much more intense approaching the shear zone, which is highly consistent with emplacement while the shear zone was active.

4.b.3. Mafic xenolithic suite

Mafic xenoliths present in porphyritic granitoids are classified into three distinct types based on mineral assemblages and texture: Type-a, plagioclase–hornblende schist with minor biotite; Type-b, plagioclase–quartz–biotite schist with minor hornblende; and Type-c, plagioclase–quartz–biotite schist with no hornblende.

Type-a xenoliths are coarse-grained schistose amphibolites, composed dominantly of hornblende and plagioclase with subordinate biotite and accessory quartz, epidote, allanite, apatite and rare opaque minerals. Schistosity is defined by the subparallel orientation of hornblende and plagioclase along a particular direction (Fig. 4g). The schistosity is accentuated by the presence of laths or clusters of laths of biotite. Hornblende occurs mostly as coarse, prismatic and also platy grains forming clusters in the interspaces of plagioclase. Poikiloblastic hornblendes contain inclusions of one or more of the following fine-grained minerals: quartz, plagioclase, epidotes, sphene and opaque minerals. Plagioclase occurs as coarse, subidioblastic elongate grains running parallel to the schistosity, and also as aggregates of medium to coarse, subidioblastic grains in the interspaces of hornblendes. In places, plagioclase has been mylonitized to aggregates of fine-sized subidiomorphic grains. Myrmekite is rarely present. Biotite (yellow to dark brown) occurs as coarse, occasionally unusually elongate laths, both parallel and transverse to hornblende schistosity. It replaces hornblende, releasing fine xenomorphic quartz as a by-product. It also occurs as clusters of laths at the marginal parts of hornblendes in contact with plagioclase. Minor quartz occurs generally as medium to fine, xenoblastic grains in the interspaces of plagioclase, hornblende–biotite and hornblende–biotite–plagioclase. Epidote occurs as fine-grained aggregates, intimately associated with hornblende–hornblende, hornblende–plagioclase and occasionally hornblende–biotite contacts. It also occurs as fine-grained inclusions in hornblende. Allanite occurs as coarse- and medium-sized, subidiomorphic yellow-brown grains with radiating cracks in the surrounding plagioclase and hornblende. Yellow-brown subidiomorphic allanite is surrounded by radial growth of epidotes in contact with surrounding plagioclase. Such epidotes show radial cracks originating from allanite. Apatite and sphene are present as medium to fine subidiomorphic grains, both stumpy and elongate as well as equant grains in the interspaces of plagioclase, ferromagnesian minerals and as inclusions in hornblende and biotite. Opaque minerals occur as xenomorphic grains associated with both biotite and hornblende.

The following petrographic changes have been noted from Type-a→ Type-b→Type-c mafic xenolith suites.

1. (i) Type-a xenolith is characterized by the first appearance of biotite in a minor amount in coarse-grained schistose amphibolite protolith, now represented by hornblende–plagioclase–biotite accessory quartz–epidote–allanite–apatite opaques.

2. (ii) Type-b and Type-c xenoliths are characterized by a gradual increase in the amount of biotite with a decrease of hornblende. In type-b xenoliths, quartz, apart for forming aggregates of fine-grained granoblasts with plagioclase, also occurs as sporadic coarse- to medium-sized lenticles running parallel to the schistosity. In Type-c xenoliths, it is significant to note that both quartz and plagioclase, and also opaques, show general elongation parallel to the trend of schistosity.

3. (iii) An increase of opaques, sphene and apatite from Type-a→ Type-b→ Type-c is observed.

4. (iv) The more frequently preferred orientation of finer plagioclase, quartz and biotite gives rise to stronger schistosity.

4.c.1. Feldspar

Representative EPMA of plagioclase and alkali feldspar from porphyritic granitoid rocks and plagioclase from plagioclase–quartz–biotite schist (mafic xenolith suite Type-c) and cationic proportions are given in Tables 2 and 3, respectively. In the unaltered megacrystic plagioclase of JPG, the An content in the core and the rim ranges from 26.90 to 21.90 wt% and 26.10 to 18.30 wt%, respectively (Fig. 5). The An content of plagioclase from plagioclase–quartz–biotite schist in the core and rim ranges from 38.70 to 34.10 wt% and 32.50 to 29.70 wt%, respectively (Fig. 5).

Fig. 5. Classification of feldspars (after Deer et al. Reference Deer, Howie and Zussman1992) from porphyritic granitoids and plagioclase–quartz–biotite schists (Type-c xenolith).

Table 2. Electron probe microanalyses of plagioclase and cationic calculation on the basis of 32 oxygen from porphyritic granitoids and biotite schist (Type-c xenolith). Ab – albite mol%; An – anorthite mol%; Or – orthoclase mol%.

Table 3. EPMA of microcline and cationic calculation on the basis of 32 oxygen from porphyritic granitoids.

Microcline crystals from the JPG have Or contents in the core and rim ranging between 92.40 and 85.80 wt% and 96.80 and 92.40 wt% (Fig. 5).

4.c.2. Biotite

Representative EPMA of biotite from porphyritic granitoids and plagioclase–quartz–biotite schist (mafic xenolith suite Type-c) and cationic proportions are given in Table 4.

Table 4. Electron Probe Microanalyses of biotite and cationic calculation on the basis of 22 oxygen from porphyritic granitoids and biotite schist (Type-c xenolith).

According to the classification scheme of Deer et al. (Reference Deer, Howie and Zussman1992), the analytical data plot within the biotite field for both porphyritic granitoids and plagioclase–quartz–biotite schist (Fig. 6a). TiO₂ content in the plagioclase–quartz–biotite schist (mafic xenolith suite Type-c) ranges from 3.56 wt% (core) to 3.41 wt% (rim), while TiO₂ concentrations in biotites in the porphyritic granitoids are slightly lower than those in the xenoliths, ranging over 3.11–1.95 wt% (core) and 3.11–1.86 wt% (rim). Similarly, Fe²⁺/(Fe²⁺+Mg) ratios in biotite from the plagioclase–quartz–biotite schist (xenolith) are higher (average 0.6) than those from the porphyritic granitoids (average 0.53).

Figure 6. (a) Classification of biotites from porphyritic granitoids and plagioclase–quartz–biotite schist (Type-c xenolith) in Fe/(Fe+Mg) versus Al^IV diagram (after Deer et al. Reference Deer, Howie and Zussman1992). (b) Plots of biotite compositions of the porphyritic granitoids of Jhalida in the MgO–FeO–Al₂O₃ discrimination diagram (after Abdel-Rahman, Reference Abdel-Rahman1994). (c) Plots of biotite compositions of the porphyritic granitoids of Jhalida in the Mg (pfu) versus Al (total) discrimination diagram (after Nachit et al. Reference Nachit, Razafimahefa, Stussi and Carron1985). Symbols: as in Figure 5.

Biotites of Jhalida granitoid pluton are aluminous (Total Al = 2.70–2.98 apfu) and plot in the calc-alkaline field in the MgO–FeO–Al₂O₃ triangular diagram (after Abdel-Rahman, Reference Abdel-Rahman1994) (Fig. 6b). The plots of biotites in the Mg (apfu) versus Al^T (apfu) diagram proposed by Nachit et al. (Reference Nachit, Razafimahefa, Stussi and Carron1985) also lie in the calc-alkaline field (Fig. 6c).

4.c.3. Magnetite

Representative EPMA of magnetite from porphyritic granitoid rocks and plagioclase–quartz–biotite schist (mafic xenolith suite Type-c) and cationic proportions are given in Table 5. Magnetites from both the porphyritic granitoids and plagioclase–quartz–biotite schist (mafic xenolith suite Type-c) are nearly pure.

Table 5. EPMA of magnetite and cationic calculation on the basis of 4 oxygen from porphyritic granitoids and biotite schist (Type-c xenolith).

4.c.4. Sphene

Representative EPMA of sphene from porphyritic granitoid rocks and cationic proportions are given in Table 6. The Ti content in the core and rim of sphene from the porphyritic granitoids is in the range of 0.89–0.0.94 apfu (5 oxygen basis), while Ca content ranges between 1.03 (core) and 0.99 (rim) apfu. Al concentration is also low (0.09–0.07 apfu), which is consistent with the typical magmatic nature of the sphene (< 0.28 apfu when total cation = 3) in granitoids (Enami et al. Reference Enami, Suzuki, Liou and Bird1993).

Table 6. EPMA of sphene and cationic calculation on the basis of 5 oxygen from porphyritic granitoids

4.d.1. Porphyritic granitoids

SiO₂ contents of the porphyritic granitoids range over 66.16–75.62 wt%. The Na₂O+K₂O values of these granitoids range over 8.21–10.4 wt%. In a SiO₂ versus (Na₂O+K₂O) diagram (after Cox et al. Reference Cox, Bell and Pankhurst1979), the studied JPG samples are plotted in the granite, alkali granite and syenite fields and the three xenolith samples are plotted in the gabbro field (Fig. 7a). In the SiO₂ versus K₂O diagram (after Peccerillo & Taylor, Reference Peccerillo and Taylor1976), five samples fall in the field of shoshonite series while only four samples fall in the field of high-K calc-alkaline series (Fig. 7b). In the Na₂O–K₂O diagram (after Turner et al. Reference Turner, Amand, Rogers, Hawkesworth, Harris, Kelley, Calsteren and Deng1996), four samples of Jhalida granitoids are classified as shoshonitic, four samples plot within the ultra-potassic series and one sample plots at the boundary of the shoshonitic–ultrapotassic series (Fig. 7c). The K₂O/Na₂O ratios of these granitoids range over 1.14–2.43. These granitoids are metaluminous to weakly peraluminous (Fig. 7 d), with A/CNK values ranging over 0.96–1.13. The agpaitic index (AI = molar (Na₂O + K₂O)/Al₂O₃) of these granitoids ranges over 0.73–0.89 (Table 7).

Fig. 7. Plots of porphyritic granitoids and three xenolith rock samples from this study in the classification diagrams: (a) total alkali versus silica diagram of Cox et al. (Reference Cox, Bell and Pankhurst1979); (b) SiO₂–K₂O diagram (after Peccerillo & Taylor, Reference Peccerillo and Taylor1976); (c) Na₂O versus K₂O diagram (after Turner et al. Reference Turner, Amand, Rogers, Hawkesworth, Harris, Kelley, Calsteren and Deng1996); (d) Shand’s Index diagram (after Maniar & Piccoli, Reference Maniar and Piccoli1989); A/NK = mol Al₂O₃/(Na₂O+K₂O); A/CNK = mol Al₂O₃/(CaO+Na₂O+K₂O); (e) SiO₂ versus MALI (modified alkali lime index; Na₂O+K₂O–CaO) (after Frost et al. Reference Frost, Barmes, Collins, Arcutus, Ellis and Frost2001); (f) SiO₂ versus Fe* number [FeO(t)/(FeO(t)+MgO)] diagram (after Frost et al. Reference Frost, Barmes, Collins, Arcutus, Ellis and Frost2001); (g) Ta/Yb versus Ce/Yb diagram of Pearce (Reference Pearce and Thorp1982); (h) Ba–Rb–Sr triangular plot of Tarney & Jones (Reference Tarney and Jones1994).

In the SiO₂ (wt%) versus MALI (modified alkali lime index, (Na₂O+K₂O−CaO) wt%) diagram (after Frost et al. Reference Frost, Barmes, Collins, Arcutus, Ellis and Frost2001) the porphyritic granitoid samples are spread over the calc-alkalic, alkali-calcic and alkalic fields (Fig. 7e). In the SiO₂ versus Fe-number diagram of Frost et al. (Reference Frost, Barmes, Collins, Arcutus, Ellis and Frost2001), six samples plot in the magnesian field and three samples plot within the ferroan field (Fig. 7f). In the Ta/Yb versus Ce/Yb diagram of Pearce (Reference Pearce and Thorp1982), two samples of the porphyritic granitoids fall in the field of shoshonitic series and other samples plot close to the field of shoshonitic series (Fig. 7g). These granitoids are characterized by high concentrations of Ba (936.82–2045.4 ppm) and Sr (325.35–654.06 ppm) and relatively low concentrations of Rb (105.43–210.9 ppm); hence most of them lie in the high Ba–Sr (Ba > 500 ppm, Sr > 300 ppm) field in the ternary Rb–Sr–Ba diagram (Fig. 7h) of Tarney & Jones (Reference Tarney and Jones1994).

On the Harker variation diagrams (Fig. 8a–p), the general decrease of MgO, TiO_2, Fe₂O₃(t) and Al₂O₃ values with increasing SiO₂ indicates the fractional crystallization of biotite, Fe–Ti oxides and sphene. The general decreasing trend of CaO and Al₂O₃ with increasing SiO₂ indicates plagioclase fractionation. The CaO and TiO₂ values show a distinctly positive correlation in these granitoids, which supports sphene fractionation. P₂O₅ content of the samples decreases with increasing SiO₂, which suggests apatite fractionation, a diagnostic character of I-type granitoids (Chappell, Reference Chappell1999). When plotted in the Rb/Sr versus Sr logarithmic diagram (Fig. 8l), porphyritic granitoid samples define trends with a negative covariation, suggesting K-feldspar accumulation. The high total alkali (K₂O+Na₂O = 8.21 to 10.4 wt%), intermediate to high SiO₂ (66.16–75.62%) and high FeO(t)/MgO ratios (1.8–7.4), together with the almost complete absence of amphibole, suggests a significant role of fractional crystallization in the evolution of Jhalida granitoids.

Fig. 8. Plots of compositions of porphyritic granitoid samples and three xenolith rock samples of Jhalida pluton in Harker variation diagrams. Experimentally determined isotherms for apatite saturation temperature (Green & Watson, Reference Green and Watson1982) are also shown in the SiO₂ versus P₂O₅ diagram.

In the primitive-mantle-normalized spidergrams (Fig. 9a), most of the Jhalida samples show enrichment in Rb, Ba and Zr, which may be related to the accumulation of biotite, and zircon, sphene and allanite. However, one sample (SD/A/4) shows troughs in Th and Zr that are to the result of the removal of Zr. All the Jhalida samples show more prominent negative Nb, Ti, P and Sr anomalies. Negative Nb and Ti anomalies are characteristics of magmatism of convergent plate margins, and negative P is to the result of the removal of apatite during fractional crystallization.

Fig. 9. (a) MORB-normalized (normalization values after Pearce, Reference Pearce, Hawkesworth and Norry1983) spider diagram and (b) MORB-normalized (normalization values after Pearce, Reference Pearce, Hawkesworth and Norry1983) spider diagram xenolith rocks of Jhalida pluton. (c) chondrite-normalized REE diagram (normalization values after McDonough & Sun, Reference McDonough and Sun1995) for the porphyritic granitoid samples.

All the samples of Jhalida granitoid show high REE abundances (Table 7). In the chondrite-normalized REE diagram (Fig. 9b), these granitoids show fractionated light (LREE) and heavy rare earth element (HREE) patterns, with (La/Yb)_N ratios varying over 31.38–216.06. Nine samples show moderately negative to slightly negative Eu anomalies (Eu* = 0.18–0.39; Table 7 and Fig. 9b). Only one sample (sample 83) shows no Eu-anomaly.

4.d.2. Mafic xenolith

In general, the mafic xenoliths have lower SiO₂, Na₂O and K₂O and higher TiO₂, FeO(t), MgO, MnO, CaO and P₂O₅ than the porphyritic granitoids (Table 8; Fig. 8).

Some of the elements behave compatibly in basic rocks, while some other concentrate in granitoid rocks. It is therefore convenient to divide the trace elements into three groups when comparing xenoliths and porphyritic granitoids. Group 1 (Cr, Co, Ni, V and Sc) consists of compatible elements that tend to be present at higher concentrations in mafic xenoliths than in porphyritic granites. Concentrations of Group 2 elements (Ba, Ta, Th and Zr) should be present in higher concentrations in granitoids; however, these elements occur in higher concentrations in the xenoliths than in the porphyritic granitoids. Group 3 elements (Nb, Y and U) includes high-field-strength elements (HFSE) that should occur at higher levels in the granitoids, but show higher concentrations in xenoliths (Tables 7, 8; Fig. 8).

K/Rb ratios in the xenoliths are lower than in the porphyritic granitoid (Tables 7, 8), which is consistent with the higher presence of modal biotites in the xenoliths than the porphyritic granitoids (Table 1) (e.g. Barbey & Cuney, Reference Barbey and Cuney1982).

Ba is most concentrated in K-feldspar of granitic rocks, whereas the K/Ba ratio is approximately half that of coexisting biotite (e.g. Barbey & Cuney, Reference Barbey and Cuney1982). Variable K/Ba ratios (17.8–97.5) in xenolith rocks are consistent with the abundance of modal biotite in them.

In the mid-ocean-ridge-basalt-normalized (N-MORB) multicationic spider diagram, both the three xenolith samples and porphyritic granitoids show more or less the same patterns (Fig. 9a, b). Significantly, both the xenoliths and porphyritic granitoids show a strong negative Ta–Nb anomaly, characteristic of subduction zone magma and continental crust. Although both have negative Nb–Ta anomalies, the Nb/Ta ratios are notably different. Whereas Nb/Ta ratios in porphyritic granitoids range over 1.36–29 (average 8.17), these ratios are higher in the xenolithic rocks (53–150). These geochemical features suggest that the sources of the porphyritic granite are significantly different from the mafic xenoliths occurring within them. At the same time, xenolith rocks do not any show negative P and Ti anomalies that are very prominent in granitoids (Fig. 9). Higher P and Ti contents of the xenoliths are consistent with the higher modal abundances of apatite and opaques and sphene in them.

4.e.1. Temperature

From high-temperature experimental studies, Watson & Harrison (Reference Watson and Harrison1983) obtained the following equation for Zr-saturation temperature (T _Zr):

$${T_{{\rm{Zr}}}} = {\rm{ }}{{12900{\rm{ }}} \over {\ln {\rm{(}}496000{\rm{ }}496000{\rm{Z}}{{\rm{r}}_{{\rm{melt}}}}{\rm{Z}}{{\rm{r}}_{{\rm{melt}}}}{\rm{)}} + 0.85M + 2.95}},$$

where M = (2Ca+K+Na)/(Al×Si) and whole-rock Zr content approximates the Zr content in the melt.

The Zr-saturation temperatures for Jhalida granitoids range between 798 and 891°C (Table 9). Since most of the zircons of Jhalida pluton are euhedral and magmatic, and interpreted to be early crystallizing, the Zr-saturation temperature can be interpreted as the near-liquidus temperature of the magma.

Table 9. Estimated physicochemical parameters of crystallization for the Jhalida porphyritic granitoids

Apatite saturation temperature can be calculated using the equation of Harrison & Watson (Reference Harrison and Watson1984):

$$\ln D_{\rm{P}}^{{\rm{apatite/melt}}} = \left[ {8400 + {{({\rm{Si}}{{\rm{O}}_2} - 0.5) \times 2.64 \times {{10}^4}} \over T}} \right] - \left\{ {3.1 + \left[ {12.4 \times ({\rm{Si}}{{\rm{O}}_2} - 0.5)} \right]} \right\}$$

where SiO₂ is the weight fraction of silica in the melt and T is the absolute temperature. The results obtained for Jhalida granitoids are represented graphically in Fig. 8f. Whole-rock apatite saturation temperatures in the Jhalida granitoids vary from c. 800°C to > 950°C.

According to Jung & Pfänder (Reference Jung and Pfaänder2007), Al₂O₃/TiO₂ ratios of orogenic granitoid rocks can be used to estimate liquidus temperature using the following equation:

$$T = {{309901} \over {\left( {{{{\rm{A}}{{\rm{l}}_2}{{\rm{O}}_3}} \mathord{\left/ {\vphantom {{{\rm{A}}{{\rm{l}}_2}{{\rm{O}}_3}} {{\rm{Ti}}{{\rm{O}}_2}}}} \right. {{\rm{Ti}}{{\rm{O}}_2}}}} \right) + 309}}.$$

Using this linear equation and assuming an igneous source composition, Jhalida granitoids yield a temperature range of 797–920°C. This temperature range is quite consistent with the zircon-saturation temperature (798–891°C) obtained from these rocks. The zircon- and apatite-saturation temperatures suggest that these temperatures are around normal for granitic magmas, not notably high, which further indicates their crustal origin.

4.e.2. Pressure

In the Q–Ab–Or diagram (for hydrous granitic system), samples of Jhalida porphyritic granitoids plot close to the minima on the quartz–feldspar cotectic for pressures of c. 2 kbar (Fig. 10b).

Fig. 10. (a) Normative Q–Ab–Or plot. Ternary cotectic curves and eutectic minima are after Winter (Reference Winter2001). (b) Temperature versus log(fO₂) diagram. Methods for determination of the temperature and oxygen fugacity values are described in the text. The NNO and NNO+2 curves are taken from O’Neill & Pownceby (Reference O’Neill and Pownceby1993). (c) Plots of biotite compositions in the Al^VI+Al^IV versus Fe/(Fe+Mg) diagram of Anderson et al. (Reference Anderson, Barth, Wooden and Mazdab2008).

4.e.3. Oxygen fugacity

The assemblage Mag–Spn–Aln in Jhalida porphyritic granitoids suggests high oxygen fugacity (above quartz–fayalite–magnetite (QFM) buffer) of the melt (Wones, Reference Wones1989; Chesner & Ettlinger, Reference Chesner and Ettlinger1989). Enami et al. (Reference Enami, Suzuki, Liou and Bird1993) suggested that the occurrences of euhedral magnetite as an early crystallizing phase suggest high oxygen fugacity conditions. Magnetite-bearing granitoids (‘magnetite series’ of Ishihara, Reference Ishihara1977) crystallize under high oxygen fugacity.

Czamanske & Wones (Reference Czamanske and Wones1973) showed that the Mg/(Mg+Fe) ratios of mafic silicates can indicate the redox conditions of the magmatic system. The opaques in the Jhalida granitoids are exclusively magnetite, which suggests the I-type nature of the magma (Chappell & White, Reference Chappell and White1992; Ishihara, Reference Ishihara1977). Moreover, the presence of sphene and magnetite, together with the presence of allanite, in Jhalida pluton suggests that the magma was relatively oxidized as expected in calc-alkaline melts (Wones, Reference Wones1989; Chesner & Ettlinger, Reference Chesner and Ettlinger1989). Wones (Reference Wones1989) gave the following equation to estimate the oxygen fugacity (log unit) for the equilibrium relationship titanite + magnetite + quartz = hedenbergite + ilmenite:

$$\log f{{\rm{O}}_2} = - {{30939} \over T} + 14.98 + {{0.142 \times \left( {P - 1} \right)} \over T},$$

where T is temperature in degrees in Kelvin and P is pressure in bars.

Adopting zircon-saturation temperature (798–891°C) and pressure of c. 2 kbar, fO₂ conditions of crystallization were estimated to be above the Ni–NiO buffer condition (about 10^–11 to 10^–12 bar between 798 and 891°C) (Fig. 10c).

Based on the calibration of Wones (Reference Wones1981), Anderson et al. (Reference Anderson, Barth, Wooden and Mazdab2008) plotted approximate fO₂ of granitoid magma relative to the QFM buffer (assuming P_H2O = P_total) in a binary Al^IV+Al^VI versus Fe/(Fe+Mg) composition diagram of biotite (Fig. 10d). Plots of the biotite composition of Jhalida samples on Anderson’s diagram show that these granites crystallized between QFM buffers of +0.8 and +1.6 (Fig. 10d), which is quite high and consistent with the calc-alkaline nature of the magma.

4.e.4. Water content

Wones (Reference Wones1981) gave the following equation relating oxygen and water fugacities and activities of annite in biotite, sanidine in K-feldspar and magnetite in opaques:

$$\log f{{\rm{H}}_2}{\rm{O}} = {{4819} \over T} + 6.69 + 0.5\log f{{\rm{O}}_2} + \log {a_{{\rm{ann}}}} - \log {a_{{\rm{Kf}}}} - \log {a_{{\rm{mag}}}} - {{0.011\left( {P - 1} \right)} \over T},$$

where T is temperature in degrees Kelvin and activities are given as molar fractions; a _ann (the activity of annite in biotite) can be calculated from different ideal activity models (after Holland & Powell, Reference Holland and Powell1990; Nash, Reference Nash1993; Patiño Douce, Reference Patiño Douce1993); a _Kf can be taken as 0.6 for magmatic systems (Czamanske & Wones, Reference Czamanske and Wones1973); and a _mag is c. 1.

The water activity (a _H2O) of magma is defined:

$${a_{{\rm{H}}2{\rm{O}}}} = {{f{{\rm{H}}_2}{\rm{O}}} \over {{f_0}{{\rm{H}}_2}{\rm{O}}}},$$

where f ₀H₂O is the standard state water fugacity.

Based on the water solubility models of Burnham (Reference Burnham and Yoder1979) and Burnham & Nekvasil (Reference Burnham and Nekvasil1986), it is possible to convert the a _H2O, f _H2O to X _H2O, which can then be expressed in wt%. We use the experimental data of Moore et al. (Reference Moore, Vennemann and Carmichael1998) to determine the H₂O (wt%) from fH₂O (vapour) in magmas. This model can be applied to any natural silicate liquid at a temperature of 700–1200°C and pressure of 1–3000 bar. For the conversion of fH₂O to H₂O (wt%), the following regression equation is derived from the experimental data of Moore et al. (Reference Moore, Vennemann and Carmichael1998):

$${{\rm{H}}_2}{\rm{O}}\;({\rm{wt}}\% ) = - 0.000\,000\,2{\left( {f{{\rm{H}}_2}{\rm{O}}} \right)^2} + 0.002\left( {f{{\rm{H}}_2}{\rm{O}}} \right) + 1.499.$$

The fH₂O of Jhalida pluton is estimated to be c. 5347–9632 bar, corresponding to 2.2–6.5 wt% of the water in the magma (Table 9). These values are consistent with the typical values of calc-alkaline magma crystallization in an arc-related environment (Ridolfi et al. Reference Ridolfi, Renzulli and Puerini2010).

4.g.1. Petrogenesis of porphyritic granitoids

The negative correlation between P₂O₅ and SiO₂, low P₂O₅ content and the metaluminous to weakly peraluminous nature of most of the samples with high oxygen fugacity (ΔQFM ranging between +0.8 and +1.6) of the Jhalida granitoids suggest their I-type character.

The Jhalida porphyritic granitoid samples show highly fractionated REE patterns ((La/Yb)_N > 31.4–216.1) (Fig. 9b). Strong negative Sr and Ti anomalies, but weak negative to positive Eu-anomalies (Eu/Eu* = 0.54–1,19; average 0.8; Fig. 9b) and strong positive Ba anomalies in spider diagram (Fig. 9a), suggest that hornblende and plagioclase and a Ti-rich mineral such as ilmenite or titanite fractionated while K-feldspar accumulated.

Nb/U and Ce/Pb ratios can provide information about the source of igneous melts because these ratios do not change significantly during partial melting or fractional crystallization (e.g. Hofmann, Reference Hofmann1988). However, these ratios can change during subduction dehydration and crust-mantle interaction (McCulloch & Gamble, Reference McCulloch and Gamble1991; Hofmann, Reference Hofmann2003; Kelley et al. Reference Kelley, Plank, Ludden and Staudigel2003).

The average Nb/U ratios (8.5) of Jhalida porphyritic granitoids are slightly higher than those estimated for the middle continental crust (7.7; Rudnick & Gao, Reference Rudnick, Gao, Holland and Turekian2003) and much lower than the ratios obtained from the lower continental crust (Nb/U c. 25; Rudnick & Gao, Reference Rudnick, Gao, Holland and Turekian2003) and ocean-island basalt (OIB; 47 ± 10; Hofmann et al. Reference Hofmann, Jochum, Seufert and White1986). Further, the average Ce/Pb ratio (9.0) of the Jhalida porphyritic granitoids are higher than those of lower continental crust (c. 5; Rudnick & Gao, Reference Rudnick, Gao, Holland and Turekian2003) but much lower than those of MORB and OIB (Ce/Pb = 25 ± 5; Hofmann et al. Reference Hofmann, Jochum, Seufert and White1986). The observed Nb/U and Ce/Pb ratios of the Jhalida granitoid samples therefore probably reveal the values of the source rock and, hence, crustal rocks.

Partial melting experiments have shown that granitoid magmas can be generated from a wide range of crustal rocks at geologically reasonable pressures and temperatures, and exhibit discrete chemical composition that can be used to identify the actual source rock of the magma.

Tepper et al. (Reference Tepper, Nelson, Bergantz and Irving1993), Wolf & Wyllie (Reference Wolf and Wyllie1994) and Patiño Douce (Reference Patiño Douce, Castro, Fernandez and Vigneresse1999) have experimentally shown that partial melting of crustal metabasalt can produce a variety of granitoid rocks under different water activities. The required heat for partial melting of the crust can be supplied from the underplating of basaltic magmas, or from an upwelling of the mantle (see Bonin, Reference Bonin2007).

Granitic magma generated by the partial melting of the amphibolitic (basaltic) continental crust should have low Mg no. values (<40) (Rapp & Watson, Reference Rapp and Watson1995). The Mg no. values of the Jhalida porphyritic granitoids are low (average Mg no. = 15) and vary from 10 to 23. These values are consistent with their derivation from the melting of the crustal source. In the SiO2 versus Mg no. diagram compiled by Jiang et al. (Reference Jiang, Jia, Liu, Liao, Zhao and Xhou2013), the samples of Jhalida granites plot within the experimentally determined field of crustal partial melts (Fig. 12a).

Fig. 12. Source rock of the Jhalida porphyritic granitoids. (a) SiO₂ (wt%) versus Mg no. ((Mg×100)/(Mg+FeO(t))). The fields of pure crustal partial melts determined in experimental studies on dehydration melting were compiled by Jiang et al. (Reference Jiang, Jia, Liu, Liao, Zhao and Xhou2013). (b) (Al₂O₃+MgO+FeO(t)+TiO₂) versus Al₂O₃/(MgO+FeO(t)+TiO₂) and (c) (Na₂O+K₂O+FeO(t)+MgO+TiO₂) versus (Na₂O+K₂O)/(FeO(t)+MgO+TiO₂) diagrams to show the composition of Jhalida porphyritic granitoids compared with melts produced by experimental dehydration-melting of various kinds of metasediments and amphibolites. Reference fields after Patiño Douce (Reference Patiño Douce, Castro, Fernandez and Vigneresse1999). (d) SiO₂ (wt%) versus K₂O (wt%) and (e) SiO₂ (wt%) versus Na₂O (wt%) diagrams with published data for experimental melts as compiled by Chen et al. (Reference Chen, Yang, Zhang, Sun and Wilde2013) from the literature. Reference fields: amphibolites (Beard & Lofgren, Reference Beard and Lofgren1991); metabasalt (Rapp & Watson, Reference Rapp and Watson1995); quartz amphibolites and biotite gneiss (Patiño Douce & Beard, Reference Patiño Douce and Beard1995); medium- to high-K basaltic rocks (Sisson et al. Reference Sisson, Ratajeski, Hankins and Glazner2005). (f) Plots of compositions of Jhalida porphyritic granitoids in the An–Ab–Kfs normative triangle for melts produced from dehydration-melting reactions and water-present-melting reactions of distinct source rocks (diagram modified after Weinberg & Hasalova, Reference Weinberg and Hasalova2015). (g) Plots of compositions of JPG in the Q–Or–(Ab+An) diagram. Note that most of the samples of JPG plot within the field produced by the melting of high-K high-alumina basalt (amphibolite) at 7 kbar (Sisson et al. Reference Sisson, Ratajeski, Hankins and Glazner2005).

Patiño Douce (Reference Patiño Douce, Castro, Fernandez and Vigneresse1999) showed that compositional variations of magmas generated by partial melting of common crustal rocks such as amphibolites, metagreywackes and metapelites can be discriminated using major oxide ratios. He drew the fields of granitic melts generated from the partial melting of amphibolites, metagreywackes and felsic pelites in (i) (Al₂O₃+FeO(t)+MgO+TiO₂) versus Al₂O₃/(FeO(t)+MgO+TiO₂) and (ii) (Na₂O+K₂O+FeO(t)+MgO+TiO₂) versus (Na₂O+K₂O)/(FeO(t)+MgO+TiO₂) diagrams. The Jhalida porphyritic granitoids exhibit low Al₂O₃/(FeO(t)+MgO+TiO₂) and intermediate (Na₂O+K₂O)/(FeO(t)+MgO+TiO₂) ratios, suggesting their derivation from amphibolite and/or metagreywacke sources (Fig. 12b,c). Since the Jhalida porphyritic granitoids are metaluminous to slightly peraluminous, we can discard the pelitic rocks as their potential source.

In the SiO₂ versus K₂O and the SiO₂ versus Na₂O diagrams (compiled by Chen et al. Reference Chen, Yang, Zhang, Sun and Wilde2013), the Jhalida porphyritic granitoid samples plot well within the experimentally derived melt field, drawn from the melting of medium- to high-K high-alumina basaltic source rock, at around 7 kbar and 825–950°C (Sisson et al. Reference Sisson, Ratajeski, Hankins and Glazner2005) (Fig. 12d,e). The medium- to high-K high-alumina basalts are characteristic of subduction zone environment.

Weinberg & Hasalova (Reference Weinberg and Hasalova2015) reported that a great diversity of granitoid magma can be obtained from the melting of similar protoliths under different activity levels of water. In the normative An–Ab–Or diagram of Weinberg & Hasalova (Reference Weinberg and Hasalova2015), the samples of Jhalida porphyritic granitoids plot in the fields of dehydration melting of quartz amphibolite and biotite–gneiss and water-present melting of hornblende shoshonitic high-alumina basalt (Sisson et al. Reference Sisson, Ratajeski, Hankins and Glazner2005) (Fig. 12g). However, the metaluminous character of Jhalida porphyritic granitoids strongly supports an amphibolite (basaltic) source for these rocks.

The high estimated H₂O content (2.2–6.5 wt%) together with liquidus temperature near 800°C (zircon-saturation temperature) suggests that the parental magma of Jhalida porphyritic granitoids was not produced by the melting of the dry granulitic lower crust; in contrast, dehydration melting of an amphibolite-grade middle crust is the most possible source.

Sisson et al. (Reference Sisson, Ratajeski, Hankins and Glazner2005) experimentally produced granitic melt from moderately hydrous (1.7–2.3 wt% H₂O), medium-to-high K basaltic rocks at 7 kbar pressure, 825–975°C temperature and relatively high oxygen fugacity (ΔNi–NiO = –1.3 to +4). The compositional range of granitic melt derived from their study was plotted on the CIPW normative quartz (Qtz), orthoclase (Or) and albite+anorthite triangular diagram (Fig. 12h). Many of the samples of Jhalida porphyritic granitoids are plotted within the compositional field of granite melt derived in these experiments (Fig. 12h). Medium-to-high K basaltic is therefore a potential source rock for the granitoids of this study.

Further, the Jhalida porphyritic granitoids show strong REE fractionation with slightly concave intermediate REE patterns in the chondrite-normalized REE diagram (Fig. 9b), suggesting that amphibole was retained in the source as a residual phase. Moreover, depletion in Sr and Eu in the primitive-mantle-normalized spider diagram (Fig. 9a, b) of the granitoids of this study strongly suggests a source rock within the plagioclase stability field or fractionation of plagioclase. High-K basaltic source rocks which metamorphosed under amphibolite facies would therefore be a potential source rock for the generation of the Jhalida porphyritic granites. The medium- to high-K high-alumina basalt was enriched in LILE and LREEs but depleted in high-field-strength elements (HFSE), suggesting that these were emplaced during a previous subduction process and later metamorphosed under amphibolite facies.

4.g.2. Batch melting model for porphyritic granite magma

The REE patterns in the Jhalida porphyritic granite can be modelled following the experiments of Sisson et al. (Reference Sisson, Ratajeski, Hankins and Glazner2005) by assuming a high-K amphibolite parent rock had been enriched and emplaced at the middle crust during earlier subduction. In the absence of direct information on the composition of high-K amphibolite xenoliths within porphyritic granitoids, we have chosen amphibole granulites from the NPSZ as the possible source rock (detailed analyses of these amphibole granulites are given in online Supplementary Table S1, available at http://journals.cambridge.org/geo). Mineral-melt partition coefficients for andesite were used after Rollinson (Reference Rollinson1993).

It may also be mentioned that the relative enrichment in LREEs, the relatively flat HREE pattern and relative depletion of middle REEs (MREEs; Fig. 9b) can be interpreted as the result of partial melting of these amphibole granulites. Fractionation of Eu depends on the proportion of plagioclase and amphibole involved in melting. To obtain significant enrichment in LREEs, the melting of a significant amount of pyroxene is necessary.

The REE melting model suggests that the parental melt of the porphyritic granite can be produced by the 10–20% batch melting of the amphibole granulite source. The modelled proportions of minerals in Figure 13 are similar to those obtained from the average of eight hornblende granulite samples of the NPSZ.

Fig. 13. Modelled rare earth element (REE) compositions derived from partial melting (batch melting) of enriched metabasalt. Starting material (assumed metabasalt) is an incompatible-element-enriched amphibole granulite occurring in the North Purulia Shear Zone. Mineral-melt partition coefficients from Rollinson (Reference Rollinson1993).

4.g.3. Role of migmatization–granitization in the petrogenesis of porphyritic granitoids

Saha et al. (Reference Saha, Bhattacharyya and Lakshmipathy1976) suggested that the migmatization of amphibolite by injection of granite magma near Tulin (9.2 km west of Jhalida town) has given rise to porphyritic granite gneiss locally. The injected veins of granite are generally pink-colored, coarse-grained hololeucocratic rocks. Alkali feldspar megacrysts and biotite in the locally neoformed porphyritic granite gneiss were related to K-metasomatism and transformation of hornblende into biotite.

In this work, it has been concluded that the porphyritic granite magma was generated by partial melting of high-K high-alumina amphibolites. In the present area, the field relations indicate that the porphyritic granitoids were injected into the host amphibolite and, as a result, the hornblende of the latter have been biotitized in varying degrees, culminating in biotite schist in which hornblende has completely disappeared, giving rise to an abundance of biotite with other mineralogical and textural changes. We have described in Section 4.a that at least some biotites in porphyritic granitoids have been derived from the highly biotitized mafic xenoliths. A comparison of the chemical analysis of biotite from a Type-c xenolith (SD/B/6) and those from Jhalida porphyritic granitoids (Table 4) shows that the cationic Fe²⁺/(Fe²⁺+Mg) ratios of the biotites for the porphyritic granitoids (0.48–0.62) are very similar to those of the xenolith rocks (0.60). The xenocrystic mineral grains incorporated within a granitic magma will tend to change their composition as per the composition of the same minerals crystallizing at that time from the magma to attain equilibrium (Clarke, Reference Clarke2007).

4.g.4. Petrogenesis of xenolith suites

Restitic xenoliths within granitoid rocks contain refractory mineral phases (Didier & Barbarin, Reference Didier and Barbarin1991). However, xenoliths devoid of any refractory mineral are shown as restitic based on Nd-isotope studies, zircon morphology and age dating of zircon (Chen et al. Reference Chen, Song, Niu and Wei2014). We discuss the nature (restitic or not) of the xenolith suite in porphyritic granitoids in the light of the experimental work of Zharikov & Khodorevskaya (Reference Zharikov and Khodorevskaya2006).

Zharikov & Khodorevskaya (Reference Zharikov and Khodorevskaya2006) experimentally reproduced the granitization of amphibolites at 5–7 kbar fluid pressures and temperatures of 700–950°C. They showed the results of chemical changes of amphibolites in an (Al₂O₃+SiO₂)–(Na₂O+K₂O)–(MgO+FeO+CaO) diagram as a result of metasomatic alteration by the infiltrating granitic melt. Zharikov & Khodorevskaya (Reference Zharikov and Khodorevskaya2006) showed in a trajectory (Fig. 14) that feldspathization and granitization are the major changes as a result of the metasomatic alteration of amphibolite protolith. All the Type-a, Type-b and Type-c amphibolites plot on the trajectory defined by Zharikov & Khodorevskaya (Reference Zharikov and Khodorevskaya2006), suggesting that the amphibolite xenoliths of this study are crustal xenoliths that have undergone metasomatic alteration during the emplacement of porphyritic granitoid magma. The amphibolite xenoliths have been incorporated into the porphyritic granitoid magma during emplacement into its current position.

Fig. 14. Plots of chemical compositions of porphyritic granites of Jhalida and xenolithic rocks in the (Al₂O₃+SiO₂)–(Na₂O+K₂O)–(MgO+FeO+CaO) diagram (after Zharikov & Khodorevskaya, Reference Zharikov and Khodorevskaya2006). Note the plots of chemical compositions of parental amphibolite and granitic melt derived by the melting of parental amphibolite in the experiments of Zharikov & Khodorevskaya (Reference Zharikov and Khodorevskaya2006). Also, note the alteration of amphibolite by invading granitic melt, as given by Zharikov & Khodorevskaya (Reference Zharikov and Khodorevskaya2006).