Geochemical Journal, Vol. 36, pp.

503 to 518, 2002

Sm-Nd age and mantle source characteristics of the Dhanjori volcanic rocks, Eastern India

A. ROY ,1** A. SARKAR,2* S. JEYAKUMAR,3 S. K. AGGRAWAL3 and M. EBIHARA4

Department of Applied Geology, Indian School of Mines, Dhanbad 826 004, India Department of Geology and Geophysics, Indian Institute of Technology, Kharagpur 721 302, India 3 Fuel Chemistry Division, Bhaba Atomic Research Centre, Mumbai 40085, India 4 Tokyo Metropolitan University, Tokyo 192-03, Japan (Received August 27, 2001; Accepted May 21, 2002)

Trace, Rare Earth Element (REE), Rb-Sr and Sm-Nd isotope analyses have been carried out on selected basic-ultrabasic rocks of Dhanjori volcanic belt from the Eastern Indian Craton (EIC). The Sm-Nd isotopic data of these rocks yield an isochron age of 2072 106 Ma (MSWD = 1.56). Chondrite normalized REE plots display shallow fractionated REE pattern with LREE enrichment. In primitive mantle normalized plots also these rocks show shallow fractionated pattern with depletion of Nb and Ba and enrichment of LILE like Rb, Th and U. Depletion of Nb, Ba and Zr and enrichment of Rb, Th and U are found in N-MORB normalized plots as well. Compatible elements like Tb, Y and Yb on the other hand, show a flat pattern. Isotope, trace and REE modelling indicate that these were produced by 35% partial melting of a spinel lherzolite source. The Nd isotopic data suggest that an enriched (Nd = 2.4) mantle existed below the Dhanjori basin during ~2.1 Ga. The enrichment was possibly caused by continuous recycling of the earlier crust into the mantle whereby subducted slab derived fluid modified the surrounding mantle. The process also affected the more easily susceptible Rb-Sr systematics producing variable Sri (0.7020.717). The enriched mantle material, part of a thermal plume, pierced through the deep fractures produced due to the cooling and readjustment of the Archaean continental crust and ultimately outpoured within the Dhanjori basin. The plume magmatism was manifested by the extrusion of komatiitic/basaltic flows and basic/ultrabasic intrusives. The residence time of the plume within the upper mantle was possibly very small as no depleted signature (even in Nd isotope) has been obtained. This means a deep plume was fed by a recycled oceanic crust via globally extensive subduction process, already initiated by the end-Archaean period.

I NTRODUCTION Present day mantle is heterogeneous from mineralogical (~1 cm) to mantle (~1000 km) scale with respect to both REE, trace element composition and Rb-Sr, Sm-Nd, Pb-Pb isotopes (Dupre and Allegre, 1983; Zindler and Hart, 1986 and references therein) and to some extent Re-Os isotopic signatures as well (Lassister and Hauri, 1998; Walker et al., 1999). Such heterogeneity also existed during the Precambrian time as documented by both depleted and enriched isotopic signatures

in basic and ultrabasic rocks from various parts of the world (Patchett and Tatsumoto, 1980; Dupre and Allegre, 1983; Hart et al., 1986; Zindler and Hart, 1986; Hart, 1988; Smith and Ludden, 1989; Barling and Goldstein, 1990; Weaver, 1991; Cousens et al., 1995; Whitehouse and Neumann, 1995; Mukasa et al., 1998). A general depletion of the upper mantle with respect to Nd isotopes is observed from Precambrian to recent along with few distinct peaks during certain periods (DePaolo, 1981; McCulloch and Bennett, 1994). In addition, evidences of heterogeneity i.e., both

*Corresponding author (e-mail: sarkaranindya@hotmail.com) **Present address: Geochronology Division, Geological Survey of India, Calcutta 700 016, India

503

504

A. Roy et al.

enriched and depleted isotopic signatures during the same time period have also been found throughout the earth history. It is still being debated what geological process(es) causes this heterogeneity particularly enriched signatures in the mantle during the Precambrian. The problem is further compounded with the involvement of crustal contamination during the ascent of basic/ ultrabasic magma and post-crystallisation alteration which can erase pristine mantle signatures. Also magmatic processes in basalts often erase mantle heterogeneity over 10 km scale length (Zindler and Hart, 1986). Therefore, to understand actual mantle processes during the Precambrian it is necessary that isotopic and geochemical studies are carried out on unaltered ultramafic/ ultrabasic rocks with insignificant crustal contamination. We report here for the first time Rb-Sr, Sm-Nd systematics and complete trace, REE studies of selected and almost unaltered samples of basic volcanics and intrusives from Dhanjori volcanic belt, EIC and discuss their implications to Precambrian mantle evolution. The present investigation was undertaken with two fold purposes. Firstly to provide an absolute age for the Dhanjori volcanics and secondly to understand the nature of their mantle sources. An interesting aspect of the Dhanjori belt is the occurrence of definitive spinifex texture of komatiitic affinity, one of the few found in the EIC (Gupta et al., 1985; Majumder, 1996). GEOLOGIC SETTING The EIC or Singhbhum-Orissa craton comprises of Archaean nucleus of south Singhbhum and Proterozoic Dalma volcanic belt and Chotanagpur gneissic complex (CGC) in the north. This cratonic block is bounded by Copper thrust belt (CTB; also called Singhbhum Shear Zone) in the north, Sukinda thrust in the south, high grade metamorphic Satpura belt in the northwest and Eastern-Ghat granulite belt in the southeast (Naqvi and Rojers, 1987). The oldest rocks in this craton are the amphibolite-tonalite gneiss association called the

older metamorphic group (OMG) with an age ~3.3 Ga (Sharma et al., 1994). A major part of this craton is occupied by the Singhbhum granite batholith complex covering an area of about 10,000 km 2 . A number of shallow basins (the supracrustals) occur within and around the periphery of this granite batholith viz. iron ore basins in the west containing large economic deposits of iron ores, Simlipal-Dhanjori basin comprising volcanics and volcanoclastic sediments etc. Deposition of the Singhbhum group of metasediments and Dhanjori group of volcanoclastics followed the emplacement of Singbhum granite. Singhbhum Group is developed along the northern flank of Singhbhum granite batholith and extends to the east and southeast and terminated with the eruption of Dalma volcanics. The Dhanjori basin, resting unconformably over the Iron Ore Group (IOG) in the NE part of the craton, consists predominantly of volcanics and volcanoclastic sediments. The vast copper deposit within its low grade metavolcanic member has been extensively mined. The Dhanjori volcanosedimentary assemblage is believed to represent a greenstone cycle (Gupta et al., 1985) within south Singhbhum Proterozoics. The sequence consists a lower unit of metapelites, psammites with ultramafics and mafics (gabbro/dolerite) and an upper predominantly volcanic unit of mafic/ ultramafic tuffs, intrusives, metabasalts and tuffaceous sediments. The lower ultramafics have distinct komatiitic affinity with definitive spinifex textures (Majumder, 1996). The upper Dhanjori basaltic suite comprises alkali olivine basalts grading upwards into K-poor oceanic tholeiites. An extensive granite-granophyre complex along with rhyolite, occurs along the western margin of the main Dhanjori basin. SAMPLING Dhanjori basin is flanked by the CTB to its northern and eastern side, Mayurbhanj granite in the southeast, south and southwest and banded iron formation in the northwest (Fig. 1; Saha, 1994). The basic/ultrabasic rocks are well exposed in

Sm-Nd age and of Dhanjori rocks, India

505

Fig. 1. Geological map of Eastern Indian Craton. Dhanjori, Kolhan, Mayurbhamj and Dalma Group of rocks are Proterozoic, rest are Archaean. The locations of Dhanjori basic and ultrabasic rocks are also shown (solid square).

various parts of this basin. Most of the coarse grained ultramafics are restricted to the lower level of the upper Dhanjori but some occurs both along the pelite-quartzite contact and also within the quartzite (Gupta et al., 1985). For the present study sampling was carried out in Kulamara-Kakdha (south of Rakha mines, Fig. 1) area within a distance of ~1 km on either side of a hillock. The Dhanjori basics/ultrabasic rocks here are well exposed within the phyllite and tuffaceous country rock, part of Dhanjori. Both southern and northern margins of the basics/ultrabasic rocks from the sampling area are schistose and more or less asbestiform though coarse prismatic pyroxene grains are still preserved. The central portion was mostly massive and almost unaltered. The prismatic pyroxene phenocrysts in this part are also conspicuous and often they are so large that they

occur as xenocrysts. The metabasites are fine grained, massive and occur along both the northern and southern margins of the ultrabasic rocks. No grain size variation was found in the mafic part. The SE part of the hillock i.e., near village Kakdah the ultrabasic rocks are flanked by the gabbroic rock in which large laths of plagioclase and prismatic pyroxenes are conspicuous. In this area fine grained metabasites are also present within the ultrabasic rocks as small patches. PETROGRAPHY The studied rock samples are basaltic to gabbroic in composition both containing augite and plagioclase laths. In some cases pyroxene content far exceeds that of plagioclase making them ultrabasic type rocks. Cumulates of augite

506

A. Roy et al.

and occasional olivine are also found within the fine grained plagioclase and augite groundmass. Although occasional chloritisation of some pyroxene grains is observed, both pyroxene and plagioclase are mostly unaltered. We have used the term basic/ultrabasic to include large variety of mafic/ultramafic rocks. ANALYTICAL PROCEDURE About 510 kg samples for fine grained rock and 1015 kg for coarse grained rocks were crushed into ~1 cm3 size in jaw crusher. The plates of the jaw crusher were thoroughly cleaned with compressive air jet and acetone. Before crushing any sample a part of it was crushed and rejected to avoid cross contamination. After cone-quartering one quarter of the chip-samples was repeatedly washed with deionised water in ultrasonic bath and dried under the mercury lamp. The dried chips were powdered in tungsten carbide disc grinder. Care was taken to avoid cross contamination between samples by repeated washing of the grinder with de-ionized water and acetone. For isotopic analysis about 0.5 gm of powdered sample was digested in a screw-cap TEFLON vial following the method of Todland et al. (1992). Typical sample digestion time was 34 days. Each sample solution was split into four aliquots, three of them were spiked with 84Sr (80%), 145Nd (80%) and 154Sm (99%) for Isotope Dilution analyses (IDMS) and one aliquot for isotopic compositions of Sr and Nd. In each aliquot Sr and REE were separated using BIORAD AG 50W- 8, 200 400 mesh cation exchange resin column (17 cm 0.8 cm) in 2.5N and 6N HCl media respectively (Sarkar et al., 1996). Sm and Nd were subsequently separated from the bulk REE fraction using BIO-RAD 1 2, 200400 mesh anion exchange resin column (4 cm 0.4 cm) following the method of Ramakumar et al. (1980). REE fraction, collected from the first column, was taken into 90% CH3OH + 10% 7.5 N HNO3 medium and loaded into a second column preconditioned with 90% CH3OH + 10% 7.5 N HNO3. Sm and Nd were eluted with 90% CH3OH + 10% 0.075 N HNO3

medium. The elution scheme includes rejection of first 2 column volumes followed by collection of next 7 column volume for Sm. The next 11 column volumes were discarded and final 12 column volumes were collected as Nd fraction. The entire column chemistry was carried out in the geochemistry laboratory of the Presidency College, Calcutta. Pure Sr, Sm and Nd fractions were loaded on clean rhenium filaments in nitrate medium. All isotopic analyses were carried using a double filament procedure in the Finningan MAT261 thermal ionization mass-spectrometer at Bhabha Atomic Research Centre (BARC), Mumbai. Total procedural blanks for Sr, Sm and Nd are <1 ng, <200 pg and <300 pg respectively. Our IDMS analyses of the basaltic standard BCR1 yielded concentrations of Sr, Sm and Nd as 332 ppm, 6.62 ppm and 28.82 ppm, respectively and in very good agreement with reported values (Wasserburg et al., 1981; Kagami et al., 1987; Lahay et al., 1995). Overall uncertainties in Sr, Sm and Nd concentration are <0.1%. Fractionation corrections for the Sr and Nd isotopic ratios for the unspiked samples were carried out with respect to 86Sr/ 88Sr = 0.1194 and 146Nd/144Nd = 0.7219 respectively. Internal precision for 87Sr/ 86Sr isotopic ratios was monitored by running NIST SRM 987 which gave 0.710226 16. The internal precision for 143Nd/144Nd ratios was checked with both La Jolla and MERCK Nd standard which gave 0.511867 08 and 0.511747 06 values respectively. Reliability of the entire analytical procedure was checked by ion-chromatographic separation of Sr and Nd from BCR-1 rock standard and their mass spectrometric measurements. This provided 87 Sr/ 86 Sr and 143 Nd/ 144 Nd ratios of 0.704971 23 and 0.512624 18 and are in excellent agreement with the reported values viz. 0.704998 15 and 0.512638 11 respectively (Wasserburg et al., 1981; Kagami et al., 1987; Toyoda et al., 1994). Trace element analyses were carried out by ICP-MS at National Geophysical Research Institute (NGRI), Hyderabad (Balaram et al., 1996). The REE were determined both by Instrumental Neutron Activation Analysis at Tokyo Metropolitan University, Japan (for details

Sm-Nd age and of Dhanjori rocks, India

507

see Roy et al., 1997) and ICP-MS in NGRI, Hyderabad mainly to check the reliability of measured elemental concentrations. Analytical errors for the ICP-MS analyses (including Rb) and INAA are <10% and <5% respectively. RESULTS AND DISCUSSION As substantial amount of major element data on different basic/ultrabasic units of the Dhanjori sequence have already been reported (Gupta et al., 1985; Sarkar et al., 1992; Ghosh and Ray, 1994; Majumder, 1996; Deb, 1999), only few major element analysis has been carried out in the present study. On the basis of major element chemistry (SiO2: 4650% and 4952%, MgO: 1325% and 38%, CaO: 410%, and 8.6%, Na2O: 0.21.9% and 2.5% for the peridotitic/basaltic komatiites and pillowed tholeiitic basalts respectively) these have been tentatively interpreted as bimodal suites emplaced in ocean-island (Gupta et al., 1985; Sarkar et al., 1992; Majumder, 1996; Deb, 1999) or transitional back-arc (Ghosh and Ray, 1994) settings. However, no isotopic data are available for any of these units; trace element/REE data are also scanty to critically evaluate the genesis of these rocks vis-a-vis tectonic setting. Apart from assessing the nature of subcrustal mantle, their tentative Proterozoic age, if true, poses interesting problems regarding the origin of such belt of komatiitic volcanism during this period. Few major elements along with trace element and REE data for basics/ultrabasic rocks within the Dhanjori basin are given in Table 1. Chondrite normalized REE plots (Chondrite values taken from McDonough and Sun, 1995, Fig. 2(a)) display shallow fractionated REE pattern with light rare earth element (LREE) enrichment ((La/ Sm) N > 2; Table 1). Deb (1999) reported REE analyses of few basaltic rocks of upper Dhanjori sequence from Mosaboni area which have been plotted along with ours in Fig. 2(b) for comparison. It can be seen from Fig. 2(b) both the data sets overlap and the REE patterns are quite similar. In primitive mantle normalized (primitive mantle data taken from McDonough and Sun,

Table 1. Major, trace and REE data for basic/ultrabasic rocks from the Dhanjori basin
Sample name Na 2 O* K2 O* CaO* Cr I NiI CoI Rb Ba Sr + Nb Hf Zr Y Th U La I Ce I Pr # Nd# SmI EuI Gd# TbI Dy# Ho# Er # Tm# YbI LuI (Ce/Yb) N (Ba/La) N (La/Sm) N Th/Nb U/Nb Th/Yb Ce/Yb Ba/Rb T2 1.03 0.38 7.08 1132 252 68.80 20.18 60.03 120.20 1.82 0.99 27.25 11.93 1.23 0.26 6.39 13.30 1.35 8.16 1.72 0.35 2.42 0.30 2.25 0.55 1.15 0.24 1.46 0.20 2.39 0.92 2.32 0.68 1.35 0.84 9.11 2.97 T3 1.62 0.62 6.79 438 332 69.60 38.73 126.20 218.13 5.67 1.84 94.07 24.85 3.06 0.97 15.39 32.80 3.02 20.13 4.49 1.37 3.48 0.78 3.31 0.73 1.64 0.28 2.08 0.25 4.14 0.80 2.14 0.54 2.43 1.47 15.77 3.26 R5 0.71 0.24 5.36 1934 459 78.70 18.23 31.12 63.50 1.64 1.60 36.23 9.53 1.15 0.41 7.17 14.80 1.54 7.17 1.50 0.37 1.45 0.28 1.09 0.31 0.86 0.19 0.89 0.16 4.36 0.43 2.98 0.70 1.99 1.29 16.63 1.71 R7 0.54 0.20 7.79 1298 308 81.40 1.50 20.04 120.11 3.30 1.92 33.19 14.08 1.77 0.31 12.20 24.00 2.09 14.59 2.89 1.02 2.55 0.54 2.61 0.45 1.02 0.17 1.36 0.18 4.63 0.16 2.64 0.54 22.13 1.30 17.65 13.36

*Concentration are in %; +measured by IDMS; IINAA data; # measured in ICP-MS. All trace element concentrations are in ppm. N = chondrite normalized; T2 and T3: gabbroic with slightly higher modal clinopyroxene; R5 and R7: basaltic but R7 contains higher modal plagioclase feldspar.

1995) plots also (Fig. 3) these rocks show shallow fractionated pattern with depletion of Nb and Ba and enrichment of large ion lithophile elements (LILE) like Rb, Th and U. Depletion of Nb, Ba and Zr and enrichment of Rb, Th and U are indicated also in N-MORB normalized plots (NMORB data from McDonough and Sun, 1995; Fig.

508

A. Roy et al.

Fig. 2. (a) Chondrite normalized REE plot for Dhanjori basics and ultrabasic rocks. (b) Chondrite normalized REE plot for Dhanjori basic and ultrabasic rocks including data of Deb (1999; G-series) and present work.

Fig. 3. Primitive mantle normalized trace element patterns for Dhanjori basics and ultrabasic rocks.

4), whereas compatible elements like Tb, Y and Yb show a flat pattern. Chondrite normalized REE pattern along with 2 < (Ce/Yb)N < 5 (Table 1) of these rocks indicates that they might have formed at a relatively shallower depth close to spinel

stabilization depth (Frey, 1982). Noticeable negative anomalies of Ba, Nb and slight depletion of Zr, found in primitive mantle normalized plot are indicative of either crustal contamination during ascent of the magma or source signature. As aver-

Sm-Nd age and of Dhanjori rocks, India

509

Fig. 4. N-MORB normalized trace element patterns for Dhanjori basics and ultrabasic rocks.

age continental crust contains high amount of Na and K, crustal contamination during the ascent of the basic/ultrabasic magma results relatively higher Na and K but mostly unaffecting the compatible elements like Ni, Cr and Co which occur in a very low concentration within the crust. It can also be argued that the slightly elevated LREE pattern with absence of Eu anomaly (as observed) might be due to the assimilation of the Archean TTG (Tonalite-Trondhjemite-Granodiorite; Taylor and Mclennan, 1985). However, Low Na and K, high Ca, Cr and Ni content (Table 1) along with fractionated REE pattern with no Eu anomaly and (Ba/La)N < 1 of these basic/ultrabasic rocks, indicate least chance of crustal contamination (Polat et al., 1997). Hence depletion of Ba, Nb and to some extent Zr in these rocks are not due to the crustal contamination rather they represent the source signature. N-MORB normalized trace element plots (Fig. 4) also display general enrichment of the incompatible elements, slight depletion of compatible elements and Ba but a strong depletion of Nb in addition to Zr. Isotopic data for these rocks are given in Table 2. The 87Rb/86Sr and 87Sr/ 86Sr ratios of these rocks vary between 0.036 and 0.833 and 0.71770

and 0.73400 respectively. 147Sm/ 144Nd and 143Nd/ 144 Nd ratios vary between 0.0607 and 0.1502 and 0.510652 and 0.511859 respectively. While RbSr data show large scatter, Sm-Nd data show a reasonably good linear correlation (Fig. 5). Interpreted as an isochron this line corresponds to an age of 2072 106 Ma (MSWD = 1.56) and an initial ratio (INd) of 0.509829 0.000082 (Nd = 2.4). The large error in age calculation is possibly due to the limited spread in Sm/Nd ratio of the samples. The low Sm/Nd ratios (e.g., 0.0607 etc.) in some samples are due to relatively higher modal plagioclase feldspar. At present, establishing a link between the genesis of these rocks and specific tectonic settings, based on trace element and isotopic compositions, seems to be difficult task. More than one single mechanism can produce the observed patterns and concentrations which are discussed below. The negative Ba and Nb anomalies with high Th/Nb (Table 1) can be explained by metasomatism of overlying mantle wedge at source induced by the fluid coming out from some subducted crust. It is possible that due to the subduction of the earlier oceanic (mafic) crust, fluid (silicate melt) was squeezed out from the slab

510

A. Roy et al.

Table 2. Rb-Sr and Sm-Nd isotopic data for basic/ultrabasic rocks from the Dhanjori Basin
Sample No. Sm (ppm) Nd (ppm) 147 Sm/1 4 4 Nd fSm/Nd 143 Nd/1 4 4 Nd Nd (2.072) TDM (Ma) Rb (ppm) Sr (ppm) 87 Rb/8 6 Sr 87 Sr/8 6 Sr Sr i T2 2.00 8.07 0.1502 0.24 0.511859 10 2.91 2832 20.18 120.20 0.486 0.71923 19 0.70457 T3 5.34 23.07 0.1400 0.29 0.511741 07 2.51 2696 38.73 218.13 0.514 0.717701 18 0.70220 R5 1.74 9.02 0.1165 0.41 0.511448 11 2.02 2494 18.23 63.50 0.833 0.734002 16 0.70890 R7 3.20 18.05 0.1072 0.46 0.511317 08 2.11 2462 1.50 120.12 0.036 0.718660 18 0.71757 R8 3.95 23.79 0.1003 0.49 0.511163 08 3.29 2520 nd nd nd nd nd R9 3.43 34.18 0.0607 0.69 0.510652 08 2.80 2372 nd nd nd nd nd

nd = not determined. f Sm/Nd = [(147Sm/ 144Nd) sample/( 147Sm/ 144Nd) CHUR] 1. CHUR (chondrite uniform reservoir): 147 Sm/144 Nd = 0.1967, 143 Nd/144Nd = 0.512638. Nd (T) = Nd (0) 25.09TfSm/Nd. Used in model age calculation, DM (depleted mantle): 147Sm/ 144Nd = 0.2137, 143Nd/ 144Nd = 0.51315. = decay constant of 147Sm = 6.54 1012. All errors quoted are 2 error. R8 and R9: basaltic.

Fig. 5. 147 Sm/ 144Nd- 143Nd/ 144Nd whole rock isochron diagram of the Dhanjori intrusives. The isochron age is calculated using ISOPLOT program (model 1 solution; Ludwig, 1988).

and incorporated into the overlying mantle wedge. As this fluid generally contains H 2O, CO 2 and chloride ions, it will be enriched in less acidic elements like Th and U (particularly in slightly oxidised condition) and depleted in very strong acidic element like Nb (Meen et al., 1989; Keppler, 1996)

which can result in the high Th/Nb and U/Nb ratios (Keppler, 1996; Kelemen et al., 1993; Davies et al., 1989; Karsten et al., 1996; Tatsumi and Kogiso, 1997; Kogiso et al., 1997). It also enriches moderately strong acidic elements like Rb, Ba and Sr. Among Rb, Ba and Sr, Rb is of least acidic character. Furthermore, solubility and mobility of Rb in chloride medium is higher compared to Ba and Sr (op.cit.). This could explain the relative depletion of Ba and Sr compared to Rb. However, Ba depletion has also been reported to be due to the presence of Ba-depleted sediment subductioncomponent at source (Smith et al., 1986). The high Ce/Yb and Th/Yb ratios (Table 1) relative to the primitive mantle of these rock suites also indicate selective enrichment of Th and Ce due to the invasion of fluid coming out from subducted slab (Hawkesworth et al., 1984; Peltonen, 1995). Depletion of Zr in N-MORB normalized plot (Fig. 4) can be explained by the fractionation of accessory minerals like zircon and rutile in the source (Green, 1994). Furthermore, negative Nd value (2.4) indicates the presence of an enriched source. The TDM-147Sm/144Nd plot (Fig. 6) of these rock suites which shows a crude hyperbolic mix-

Sm-Nd age and of Dhanjori rocks, India

511

Nd isotope and REE modeling The Nd values can be used to estimate the 147 Sm/ 144Nd ratio of the mantle source of these rocks during 2.072 Ga by using the following relationship (DePaolo, 1988):
source fSm / Nd =

Nd Q(TS TX )

Fig. 6. TDM-147Sm/ 144Nd plot of basic and ultrabasic rocks from Dhanjori basin. Note a hyperbolic mixing relation at the source.

where, T S = model age of the magma source and T X = crystallisation age of the rocks. Considering T S = 4.55 Ga where from accretion of the earth started, T X = 2.1 Ga, Nd = 2.4 (taken from Isochron) and Q = 25.09 Ga1 (calculated using present day values of (147Sm/144Nd)CHUR = 0.1967 and (143 Nd/144Nd)CHUR = 0.512638; Jacobsen and
source Wasserburg, 1984), the fSm / Nd value has been cal-

ing relation, attests a short term enrichment in the source. The two asymptotes of this hyperbolic mixing line indicate that an enriched component was mixed up with a relatively depleted component (~147Sm/144Nd = 0.15) during ~2.3 Ga i.e., just before the emplacement of these rock suites at around 2.1 Ga. It is also to be noted that Sri values (calculated back to 2.1 Ga) of these rocks considerably vary from 0.702 to 0.717 where as Nd values do not. This indicates two possibilities. Either the Sr isotopic signature has been disturbed by low-grade metamorphism and/or post crystallisation alteration (albeit insignificant) without affecting Nd-isotope systematics or the source itself was heterogeneous (mixing of different source components) at least in terms of RbSr systematics. It should be mentioned in this context that for the present work Rb was measured by the ICP-MS method. The analytical error (monitored by the analyses of rock standard JB2) for this is ~<5%. Even considering the interaliquot heterogeneity and analytical error together (~15%), no correlation has been observed between 87 Sr/ 86Sr and Rb/Sr ratios and hence the scatter indeed is real. The variation in Sri (calculated back to 2.1 Ga) is, therefore, also real. The Nd isotopic data, however, indicate a near-uniform but slightly enriched source.

culated to be ~0.04. Again,

fSm / Nd

( =

147

Sm/ 144 Nd

source 144 147

147

Sm/ 144 Nd
CHUR

Sm/

Nd

CHUR

source Using fSm / Nd = 0.04 in the above equation, 147 144 ( Sm/ Nd) source = 0.189 has been obtained which is about ~4% lower than the average chondrite. source rock Taking fSm / Nd = 0.04 and fSm / Nd = 0.43 (average of the six samples, Table 2) values, fractionation factor () has been calculated using the following relationship of DePaolo (1988):

Sm / Nd =

rock 1 + fSm / Nd source 1 + fSm / Nd

which gives the value of Sm/Nd = 0.594, i.e., ~41% depletion in the Sm/Nd ratio of the precursor of these rocks at source. This requires very low degree of partial melting (DePaolo, 1988) of the spinel lherzolite source which was already enriched at a scale of 4% with respect to Sm-Nd isotope.

512

A. Roy et al.

As a first approximation shallow fractionated pattern in chondrite normalized REE plot indicates its derivation from partial melting of a less fractionated REE source. In the trace element modelling, it is difficult to get a unique solution. This is mainly due to the improper estimation of the initial source composition. Using the measured REE concentrations, Nd isotope signatures and partial melting model of Allegre and Minster (1978), the probable source composition of these basic/ultrabasic rocks and the extent of melting have been assessed. Partition coefficient (D) values, used in this model, are taken from Shaw (1970) and McKenzie (1995). The modelling shows that ~35% partial melting of a slightly enriched source with REE concentration of (~2.5 chondrite LREE values and 2 chondrite HREE values) and spinel lherzolite composition (Ol:Opx:Cpx = 60:25:15) can give rise to Dhanjori basaltic or gabbroic end products. Weight fraction of liquid contributed by each phase during melting is Ol:Opx:Cpx = 10:35:55. Since modal percentage of pyroxenes is higher and (Ce/Yb)N > 1, initial rock composition and melting mode have been considered as discussed above (for details see Roy et al., 1997). This estimation is consistent with the Nd isotope modelling of these rocks. Source enrichment The foregoing discussion indicates that at 2.1 Ga i.e., early Proterozoic, the mantle below the Dhanjori basin was enriched which had a relatively short residence time whereas it was heterogeneous in terms of Rb-Sr isotopic character. Trace element signatures also indicate that the mantle enrichment could be the result of invasion of fluids originated from subducted slab. This resulted in low Nb, Ba with enriched LREE pattern in chondrite normalized REE plot, high Th/Nb ratio and enriched Nd isotopic signature. However, apart from the fluid-invasion (subducted slab of altered oceanic crust? Smith et al., 1986), the lower Nb i.e., high Th/Nb ratio and higher Rb of these basics and ultrabasic rocks can as well be explained by the contribution of delaminated continental crust at the source itself

(Rudnick and Fountain, 1995). In the first case, fluid coming out from the subducting oceanic crust (hydrated), enriched in LILE like Rb, Th and U and LREE and depleted in HFSE like Nb, Ta and Zr and occasionally Ba (if Ba depleted sediment is subducted; Ben Othman et al., 1989; Smith et al., 1986; Jahn et al., 1999), is mixed up with upper mantle wedge resulting Nb and Ba depleted but LREE and LILE enriched upper mantle (Schiano et al., 1995). The second case i.e., delamination of continental crust occurs when the continental crust is thicker and differentiated (Rudnick and Fountain, 1995). With progressive crust building activities the metamorphosed lower part of thicker continental crust becomes denser. If the density of lower crust exceeds that of the immediate neighbouring upper mantle, density inversion i.e., sinking of a part of the lower crust within the upper mantle occurs. The upper mantle material instead occupies the volume left behind by the detached lower crust. After a considerable time, mixing of the delaminated lower crust with the mantle results a hybridised upper mantle with enriched Nd isotopic and variable Sr isotopic signatures (McKenzie and ONions, 1983) along with LILE enrichment and Nb depletion. Though it is difficult to choose either of these two processes, higher Th/Nb, Th/Yb, Ce/Yb, low but variable Ba/Rb ratios and variable Sri of these rocks favour more a source affected by subduction induced fluid rather than delamination of continental crust. If slightly higher amount of low Ba bearing sediments, which are generally characteristic of hemipelagic sequence, are subducted, the fluid coming out from these sediments will also be slightly enriched in U and Th but depleted in Ba, Nb and Zr, resulting in Ba, Nb and Zr depleted upper mantle (Smith et al., 1986; Ben Othman, 1989). This of course does not suggest any tectonic setting (subduction or otherwise) for basic/ ultrabasic rocks of Dhanjori but merely suggests a process of modification of the source. It is not possible to comment about the exact time of source enrichment, however, Nd isotope data indicate it to be prior to ~2.3 Ga i.e., lower Proterozoic time. The modified mantle material might come up

Sm-Nd age and of Dhanjori rocks, India

513

Fig. 7. Nb/Y-Zr/Y dicriminant plot for the Dhanjori basics/ultrabasic rocks (present work) Dalma intrusives and tholeiites (data taken from Bose et al., 1989; Saha, 1994; Roy, 1998); Primitive mantle and average N-MORB data from McDonough and Sun (1995); Average lower, middle and upper crust data from Rudnick and Fountain (1995); note that both the Dhanjori and Dalma basic/ultrabasic data fall within the plume array (for details see text).

through a number of tectonic settings including plumes which, according to many, was responsible for generating all the komatiitic (or high MgO basaltic) melts like those of Dhanjoris (McKenzie and Bickle, 1988; Campbell and Griffith, 1992; Arndt et al., 1997). In this regard, Nb, a parameter which is insensitive to the effects of variable degrees of mantle melting, source depletion through melt extraction, crustal contamination of the magmas and post crystallisation alterations (Fitton et al., 1997) can be used for source characterisation. Since both continental crust and NMORB are depleted in Nb, the low abundance of Nb in the upper mantle cannot be explained by simple mass balance calculation involving extraction of continental crust at the expense of upper mantle. The lost Nb, possibly stored in the subducted oceanic crust, is recycled back through mantle plumes (Saunders et al., 1988). This produces relatively higher Nb concentration in plume magma compared to that of N-MORB. On a Nb/ Y-Zr/Y plot the present day plume basalts from Iceland have been shown to fall on an array which runs exactly parallel to the N-MORB array but

with significantly higher Nb/Y ratios (Fitton et al., 1997). Crustal contamination, in this plot, is easily discernible from low Nb/Y-high Zr/Y array deviating from either plume or N-MORB array. Nb (=1.74 + log(Nb/Y) 1.92log(Zr/Y)) with positive and negative values indicate plume and MORB sources respectively (op.cit.). Owing to the non-availability of precise trace element data for majority of Dhanjori komatiitic lavas it has not been possible to use Nb/Y-Zr/Y discriminant diagram (a potential diagram to identify plume related volcanism) and compare (and compile) the Nb/Y-Zr/Y ratios between komatiites and basic/ ultrabasic rocks under investigation. However, when plotted on this discriminant diagram two samples of Dhanjori basic/ultrabasic rocks (T2 and R7) have been found to lie exactly within the plume array close to primitive mantle but away from crustal contamination whereas other two samples (T3 and R5) are slightly deviating (Fig. 7). While the sample T3 falls very close to the lower boundary of the plume array (Nb ~ 0), Nb for R5 is ~0.14 occurring quite close to the NMORB along the N-MORB-Crust array. The nega-

514

A. Roy et al.

tive Nb of R5 might be due to the mixing between the N-MORB and crust as both have negative Nb values. Based on few major and compatible element concentrations we earlier indicated minimum chance of crustal contamination. Additionally, depletion of Ba compared to the other incompatible element like Rb along with high Th/ Yb and Ce/Yb in Dhanjori basics/ultrabasic rocks also negates crustal contamination during emplacement. The geochemistry of these rocks rather indicates other geological process(es) operative at source. We have also plotted in Fig. 7 Nb/Y-Zr/Y ratios of mid-Proterozoic (~1.6 Ga; Roy et al., 1999, 2002a) Dalma basalts, komatiites and ultrabasic intrusives all of which have been found to be depleted mantle products (in terms of both REE and Nd, data taken from Bose et al., 1989; Saha, 1994; Roy, 1998, Roy et al., 2002a). Interestingly, all these units of Dalma belt also fall within the plume array and together they indicate that the plume magmatism was quite active during the Proterozoic of the EIC. The fundamental difference between Dhanjori and Dalma plumes are that the former possibly represents tapping of an enriched part of a plume while the later represents a depleted part. This is consistent with the prevalent idea of large compositional spectrum encompassed by different kinds of plume (MacDougall, 1988). It is, therefore, concluded that the Dhanjori basic/ultrabasic rocks are genetically connected to a plume which was fed by fluid derived from an ancient subducted altered oceanic crust and not related to crustal delamination process. Since the Dhanjori basic/ultrabasic rocks are uniformly enriched (both geochemically and isotopically), it is tempting to speculate that possibly the plume originated at deep crust-mantle boundary and rapidly pierced through the crust without any contamination (and stalling or residence) with shallow depleted upper mantle (Fitton et al., 1997). A deep seated plume with enriched signature can show depleted components only if it travels slowly or stalls within the depleted upper mantle for a long time by acquiring an envelope of depleted component (Kerr et al., 1995).

IMPLICATIONS It has been well documented that by early Proterozoic the crust in the south Singhbhum nucleus was considerably thick due to the emplacement of vast Singhbhum granite (Saha, 1994). Hence, the Dhanjori basalts including komatiites (and gabbros), if related to plume magmatism, must have had access to some weak zones through which they were extruded/emplaced. It is envisaged that number of small isolated basins like Dhanjori, Simlipal etc. were formed within the Singhbhum nucleus towards the early Proterozoic time possibly due to the readjustment in protocrustal stability and resetting of tectonomagmatically active margins (Gupta et al., 1985). During the end-Archaean period cooling down of the vast intrusion of granite batholith (Singhbhum Granite) took place which possibly induced an isostatic readjustment. Local tensional regimes and deep seated fractures, thus produced, might have culminated in the formation of small isolated basins. Subsequent upwelling of an enriched plume from the deeper mantle through these weak zones accentuated the basin evolution resulting in outpouring of basaltic/komatiitic magma with basic and ultrabasic intrusives. It is also noticeable that an enriched mantle (in terms of trace elements, REE and Nd isotopes) existed below the Dhanjori basin during ~2.1 Ga or early Proterozoic period. The enrichment was possibly caused by the continuous recycling of the earlier crust into the mantle whereby subducted slab derived fluid modified the surrounding mantle, a process which started much earlier below the EIC as evident from the Nd values of 2.6 Ga old ultramafic rocks associated with Newer Dolerite dykes (NDD; Roy, 1998; Roy et al., 2002b). The Nd isotopic data of these Archaean dyke (the data are being published separately) bodies indicate that the mantle below the EIC was heterogeneous displaying both early depleted and later enriched components (op.cit.). The 2.6 Ga time is a period of major global crust building activities and initiation of recycling of enriched

Sm-Nd age and of Dhanjori rocks, India

515

component(s) to the mantle via subduction (McCulloch and Bennett, 1994). As the crustal recycling continued, the underlying mantle became more and more enriched and homogeneous by 2.1 Ga as manifested by uniformly enriched Nd values of Dhanjori basic/ultrabasic rocks. The process also affected the more easily susceptible Rb-Sr systematics producing variable Sri of the basic/ultrabasic rocks from Dhanjori basin. The enriched mantle material, part of a thermal plume, not only produced the basic/ultrabasic rocks of the Dhanjori group but also aided extrusion of high temperature komatiitic flows. That the komatiites are possibly product of plume magmatism has been convincingly demonstrated (Kerr et al., 1995; Arndt et al., 1997). The residence time of the Dhanjori plume within the upper mantle was possibly very small as no depleted signature (even in Nd isotope) has been obtained. This means a deep plume was fed by recycled crust via globally extensive subduction process already initiated during the end-Archaean period (op.cit.). CONCLUSIONS (1) The Sm-Nd isotopic analyses of basicultrabasic rocks of Dhanjori basin from the EIC yield an isochron age of 2072 106 Ma (MSWD = 1.56) indicating their age as early Proterozoic. (2) Chondrite normalized REE plots display shallow fractionated REE pattern with LREE enrichment. Also in primitive mantle normalized plots these rocks show shallow fractionated pattern with depletion of Nb and Ba and enrichment of LILE like Rb, Th and U. Depletion of Nb, Ba and Zr with enrichment of Rb, Th and U are found in N-MORB normalized plots as well. Compatible elements, like Tb, Y and Yb, on the other hand, show a flat pattern. (3) Isotope, trace and REE modeling indicate that these are produced by 35% partial melting of spinel lherzolite source. The data suggest that an enriched (Nd = 2.4) mantle existed below the Dhanjori basin during ~2.1 Ga which had a relatively short residence time in the mantle. The en-

richment was possibly caused by the continuous recycling of the earlier crust into the mantle whereby subducted slab derived fluid modified the surrounding mantle. The process also affected the more easily susceptible Rb-Sr systematics producing variable Sri (0.7020.717). (4) The enriched mantle material, part of a thermal plume, pierced through the deep fractures produced due to the cooling and readjustment of the Archaean continental crust and ultimately outpoured within the Dhanjori basin. The plume magmatism was manifested by the extrusion of komatiitic/basaltic flows and basic/ultrabasic intrusives. The residence time of the plume within the upper mantle was possibly very small as no depleted signature (even in Nd isotope) has been obtained. This means a deep plume was fed by a recycled oceanic crust via globally extensive subduction process already initiated by the endArchaean period.
Acknowledgments We thank Prof. B. B. Bhattacharya, Director, Indian School of Mines, Dhanbad, Dr. D. K. Paul, Calcutta and Prof. H. C. Jain, BARC, Mumbai for continuous encouragement. This paper forms part of the Ph.D. thesis of AR, who thanks Department of Atomic Energy, Govt. of India for a fellowship. We also thank Dr. A. Gupta, GSI for stimulating discussion and Dr. K. L. Ramakumar, BARC for continuous help during the operation of mass spectrometer. AR acknowledges Mr. B. Mukhopadhyay, GSI, CHQ, Geodata for preparing geological map. AS thanks JSPS for an exchange program fellowship during which a part of the INAA data were generated. We gratefully remember late Prof. A. K. Saha who inspired us to undertake the isotope studies of mafic-ultramafic rocks from Eastern Indian craton.

REFERENCES
Allegre, C. J. and Minster, J. F. (1978) Quantitative models of trace element behaviour in magmatic processes. Earth Planet. Sci. Lett. 38, 125. Arndt, N. T., Kerr, A. C. and Tarney, H. (1997) Dynamic melting of plume heads: the formation of Gorgona komatiites and basalts. Earth Planet. Sci. Lett. 146, 289301. Balaram, V., Ramesh, S. L. and Anjaiah, K. V. (1996) New trace element and REE data in thirteen GSF reference samples by ICP-MS. Geostand. Newslett. 20,

516

A. Roy et al. Gupta, A., Basu, A. and Singh, S. K. (1985) Stratigraphy and petrochemistry of Dhanjori greenstone belt, Eastern India. Quart. J. Geol. Min. Met. Soc. Ind. 57, 248263. Hart, S. R. (1988) Heterogeneous mantle domains: signatures, genesis and mixing chronologies. Earth Planet. Sci. Lett. 90, 273296. Hart, S. R., Gerlach, D. C. and White, W. M. (1986) A possible Sr-Nd-Pb mantle array and consequences for mantle mixing. Geochem. Cosmochim. Acta 50, 15511557. Hawkesworth, C. J., Rogers, N. W., Van Calsteren, P. W. C. and Menzies, M. A. (1984) Mantle enrichment processes. Nature 311, 331335. Jacobsen, S. B. and Wasserburg, G. J. (1984) Sm-Nd isotopic evolution of chondrites and achondrites-II. Earth Planet. Sci. Lett. 67, 137150. Jahn, B., Fuyuan, Wu, Ching-Hua Lo and Chin-Ho Tsai (1999) Crust-mantle interaction induced by deep subduction of the continental crust: geochemical and SrNd isotopic evidence from post-collisional maficultramafic intrusions of the northern Dabie complex, Central China. Chem. Geol. 157, 119146. Kagami, H., Iwata, M., Sano, S. and Honma, H. (1987) Sr and Nd isotopic compositions and Rb, Sr, Sm and Nd concentrations of standard samples. Technical Rep. ISEI Okayama Univ. Ser. B 4, 16 pp. Karsten, J. L., Klein, E. M. and Sherman, S. B. (1996) Subduction zone geochemical characteristics in ocean ridge basalts from the southern Chile Ridge: Implications of modern ridge subduction systems for the Archean. Lithos 37, 143161. Kelemen, P. B., Shimizu, N. and Dunn, T. (1993) Relative depletion of niobium in some arc magmas and the continental crust: partitioning of K, Nb, La and Ce during melt/rock reaction in the upper mantle. Earth Planet. Sci. Lett. 120, 11134. Keppler, H. (1996) Constraints from partitioning experiment on the composition of subduction-zone fluids. Nature 380, 237240. Kerr, A. C., Arndt, N. T., Tarney, J., Nivia, A., Saunders, A. D., Storey, M. J. and Duncan, R. A. (1995) The petrogenesis of komatiites, picrites, and basalts from the Isle of Gorgona, Colombia: new field, petrographic and geochemical constaints. Lithos 37, 245 260. Kogiso, T., Tatsumi, Y. and Nakano, S. (1997) Trace element transport during dehydration processes in the subducted oceanic crust: 1. experiments and implications for the origin of ocean island basalts. Earth Planet. Sci. Lett. 148, 193205. Lahay, Y., Arndt, N., Byerly, G., Chauval, C., Fourcade, S. and Gruau, G. (1995) The influence of alteration on the trace element and Nd-isotopic composition of

7178. Barling, J. and Goldstein, S. L. (1990) Extreme isotopic variations in Heard island lavas and the nature of mantle reservoir. Nature 348, 5962. Ben Othman, D., White, W. M. and Patchett, J. (1989) The geochemistry of marine sediments, island arc magma genesis and crust-mantle recycling. Earth Planet. Sci. Lett. 94, 121. Bose, M. K., Chakrabarti, M. K. and Saunders, A. D. (1989) Petrochemistry of lavas from Proterozoic volacanic belt, Singhbhum, Eastern India. Geol. Rundsch. 78, 633648. Campbell, I. H. and Griffith, R. W. (1992) The changing nature of mantle hotspot through time: implication for the chemical evolution of the mantle. J. Geol. 92, 497523. Cousens, B. L., Allan, J. F., Leybourne, M. I., Chase, R. L. and Wagoner, N. V. (1995) Mixing of magmas from enriched and depleted mantle sources in the north east Pacific: west valley segment, Juan de Fuca Ridge. Contrib. Mineral. Petrol. 120, 337357. Davies, G. R., Norry, M. J., Gerlach, D. C. and Cliff, R. A. (1989) A combined chemical and Pb-Sr-Nd isotope study of the Azores Cape Verde hot-spots: the geodynamic implications. Magmatism in Ocean Basins (Saunders, A. D. and Norry, M. J., eds.), Geol. Soc. Specl. Publ. 42, 231255. Deb, G. K. (1999) Evolution of the Dhanjori volcanosedimentary rocks in the North Singhbhum fold belt: A reappraisal. Ind. Minerals 53(1 and 2), 2536. DePaolo, D. J. (1981) Nd isotopes in the Colorado front range and crust mantle evolution in the Proterozoic. Nature 291, 193196. DePaolo, D. J. (1988) Nd Isotope Geochemistry: An Introduction. Springer, Berlin, Heidelburg, New York, 187 pp. Dupre, B. and Allegre, C. J. (1983) Pb-Sr isotope variation in Indian Oceanic basalts and mixing phenomenon. Nature 303, 142146. Fitton, J. G., Saunders, A. D., Norry, M. J. and Hardarson, B. S. (1997) Thermal and chemical structure of the Iceland plume. Earth Planet. Sci. Lett. 153, 197208. Frey, F. A. (1982) Rare earth element abundance in the upper mantle rocks. Rare Earth Element Geochemistry (Handerson, P., ed.) 153204, Elsevier, Amsterdam. Ghosh, K. and Ray, J. (1994) Some observations on geochemistry and tectonic setting of Dhanjori metavolcanics of the Eastern Indian shield. Ind. J. Geol. 66, 279295. Green, T. H. (1994) Experimental studies of trace element partitioning applicable to igneous petrogenesisSedona 16 years later. Chem. Geol. 117, 136.

Sm-Nd age and of Dhanjori rocks, India komatiites. Chem. Geol. 126, 4364. Lassister, J. C. and Hauri, E. H. (1998) Osmium-isotope variations in Hawaiian lavas: evidence for recycled oceanic lithosphere in the Hawaiian plume. Earth Planet. Sci. Lett. 164, 483496. Ludwig, K. R. (1988) ISOPLOT: A plotting and regression program for radiogenic isotopic data. U.S. Geol. Surv., Open-file Rep. 91-445. MacDougall, J. D. (ed.) (1988) Continental Flood Basalts. Dordrecht, Kluwer, The Netherlands, 341 pp. Majumder, S. K. (1996) Precambrian geology of peninsular India. Ind. Minerals 50(3), 139174. McCulloch, M. T. and Bennett, V. C. (1994) Progressive growth of the earths continental crust and depleted mantle: Geochemical constrains. Geochim. Cosmochim. Acta 58, 47174738. McDonough, W. F. and Sun, S.-s. (1995) The composition of the earth. Chem. Geol. 120, 223253. McKenzie, D. (1995) The source regions of ocean island basalts. J. Petrol. 36, 133159. McKenzie, D. and Bickle, M. J. (1988) The volume and composition of melt generation by extension of continental lithosphere. J. Petrol. 29, 625679. McKenzie, D. and ONion, R. K. (1983) Mantle reservoirs and ocean island basalts. Nature 301, 229231. Meen, J. K., Eggler, D. H. and Ayers, J. C. (1989) Experimental evidence for a very low solubility of rare earth elements in CO2-rich fluids at mantle condition. Nature 340, 301303. Mukasa, S. B., Wilson, A. H. and Carlson, R. W. (1998) A multielement geochronologic study of the great dyke, Zimbabwe: singnificance of the robust and reset ages. Earth Planet. Sci. Lett. 164, 353369. Naqvi, S. M. and Rogers, J. J. M. (1987) Precambrian Geology of India. Clarendon, Oxford, 223 pp. Patchett, P. J. and Tatsumoto, M. (1980) Hafnium isotopic variations in oceanic basalts. Geophys. Res. Lett. 7, 10771080. Peltonen, P. (1995) Petrogenesis of ultramafic rocks in the Vammala Nickel Belt: implication for crustal evolution of the early Proterozoic Svecofennian arc terrain. Lithos 34, 253274. Polat, A., Kerrich, R. and Casey, J. F. (1997) Geochemistry of Quarternary basalts erupted along the east Anatolian and Dead Sea fault zones of southern Turkey: implication for mantle source. Lithos 40, 5568. Ramakumar, K. L., Aggarwal, S. K., Kavimandan, V. D., Raman, V. A., Khodade, P. S., Jain, H. C. and Mathews, C. K. (1980) Separation and purification of Magnesium, Lead and Neodymium from dissolver solution of irradiated fuel. Separation Science and Technology 15(7), 14711481. Roy, A. (1998) Isotopic evolution of the Precambrian

517

mantle: a case study of Singhbhum Orissa craton. Ph.D. Thesis, ISM, Dhanbad, 137 pp. Roy, A., Sarkar, A., Bhattacharaya, S. K., Ozaki, H. and Ebihara, M. (1997) Rare earth element geochemistry of selected mafic-ultramafic units from Singhbhum Craton: implications to source heterogeneity. J. Geol. Soc. Ind. 50, 717726. Roy, A., Sarkar, A., Jeyakumar, S., Aggrawal, S. K. and Ebihara, M. (1999) Mid-Proterozoic plume related thermal event in Eastern Indian craton: evidence from trace elements, REE geochemistry and Sr-Nd isotope systematics of mafic-ultramafic intrusives from Dalma volcanic belt. 8th National Symposium of Mass-Spectrometry, p. 51, Hydrebad. Roy, A., Sarkar, A., Jeyakumar, S., Aggrawal, S. K. and Ebihara, M. (2002a) Mid-Proterozoic plume related thermal event in Eastern Indian craton: evidence from trace elements, REE geochemistry and Sr-Nd isotope systematics of mafic-ultramafic intrusives from Dalma volcanic belt. Gondwana Res. 5, 133146. Roy, A., Sarkar, A., Jeyakumar, S., Aggrawal, S. K. and Ebihara, M. (2002b) Late Archaean mantle metasomatism below Eastern Indian craton: evidence from trace elements, REE geochemistry and Sr, Nd and O isotope systematics of ultramafic dykes. Int. Jour. Earth Sci. (submitted). Rudnick, R. L. and Fountain, D. M. (1995) Making continental crust. Nature 378, 571574. Saha, A. K. (1994) Crustal evolution of North Singhbhum-Orissa, eastern India. Geol. Soc. India Memo. 27, 341 pp. Sarkar, A., Roy, A., Ghatak, G. S. and Bhattacharya, S. K. (1996) Strontium isotope study of Krol-Tal carbonates: implication to the strontium isotope flux of Himalayan rivers. Ind. J. Geol. 68, 255262. Sarkar, S. C., Gupta, A. and Basu, A. (1992) North Singbhum Proterozoic mobile belt, eastern India: Its character, Evolution and Metallogeny. Metallogeny Related to Tectonics of the Proterozoic Mobile Belt (Sarkar, S. C., ed.), 271307, Oxford and IBH, New Delhi. Saunders, A. D., Norry, M. J. and Tarney, J. (1988) Origin of MORB and chemically depleted mantle reservoirs: trace element constraints. J. Petrol. (Special Lithosphere issue), 415445. Schiano, P., Clocchiatti, R., Shimizu, N., Maury, R. C., Jochum, K. P. and Hofmann, A. W. (1995) Hydrous, silica-rich melts in the sub-arc mantle and their relationship with mantle and their relationship with erupted arc lavas. Nature 377, 595600. Sharma, M., Basu, A. R. and Ray, S. L. (1994) Sm-Nd isotopic and geochemical study of the Archean Tonalite-Amphibolite association from the eastern Indian Craton. Contrib. Mineral. Petrol. 117, 4555.

518

A. Roy et al. Geochemical correlation with hotspots in the south atlantic and implication for the mantle source. Earth Planet. Sci. Lett. 126, 315331. Walker, R. J., Storey, M., Kerr, A. C., Terney, J. and Arndt, N. T. (1999) Implication of 187Os isotopic heterogeneities in a mantle plume: evidences from Gorgona Island and Curacao. Geochim. Cosmochim. Acta 63, 713728. Wasserburg, G. J., Jacobsen, S. B., DePaolo, D. J., McCulloch, M. T. and Wen, T. (1981) Precise determination of Sm/Nd ratios, Sm and Nd isotopic abundances in standard solutions. Geochim. Cosmochim. Acta 45, 23112323. Weaver, B. L. (1991) The origin of oceanic island basalt end-member compositions: trace element and isotopic constraints. Earth Planet. Sci. Lett. 104, 381397. Whitehouse, M. J. and Neumann, E. R. (1995) Sr-NdPb isotope data for ultramafic xenoliths from Hirro, Canary Islands: melt infiltration processes in the upper mantle. Contrib. Mineral. Petrol. 119, 239 246. Zindler, A. and Hart, S. (1986) Chemical geodynamics. Ann. Rev. Earth Planet. Sci. Lett. 14, 493571.

Shaw, D. M. (1970). Trace element fractionation during anatexsis. Geochim. Cosmochim. Acta 34, 237 243. Smith, A. D. and Ludden, J. N. (1989) Nd evolution of the Precambrian mantle. Earth Planet. Sci. Lett. 93, 1422. Smith, T. E., Thirlwall, M. F. and Macpherson, C. (1986) Trace elements and isotope geochemistry of volcanic rocks of Bequia, Grenadine islands, Lesser Antilles arc: a study of subduction enrichment and intra-crustal contamination. J. Petrol. 37, 117143. Tatsumi, Y. and Kogiso, T. (1997) Trace element transport during dehydration processes in the subducted oceanic crust: 2. origin of chemical and physical characteristics in arc magmatism. Earth Planet. Sci. Lett. 148, 207221. Taylor, S. R. and McLennan, S. M. (1985) The Continental Crust: Its Composition and Evolution. Beckwell, Oxford, 312 pp. Todland, M., Jarvis, I. and Jarvis, K. E. (1992) An assessment of dissolution techniques for the analysis of geological samples by plasma spectroscopy. Chem. Geol. 95, 3562. Toyoda, K., Horiuchi, H. and Tokonami, M. (1994) Dupal anomaly of Brazilian carbonatites: