Sm-Nd analyses have been made on 12 greenstone samples from the Russian Karelia. This work is part of the co-operation between Dr. Sergei Svetov (Institute of Geology, Karelian Research Centre, Petrozavodsk, Russia) and GSF (Huhma H., Luukkonen et al.). The emphasis of isotopic work has been on the characterization of the well studied Archaean sequences in Koikary and Palaselga domains, where rocks are considered to be better preserved than in Finland.
The samples have been chosen to show the range of MgO compositions ( Koikary domain- MgO from 7% to 20%; Palaselga domain - MgO from 4% to 31%) and to give represent of each rock type. 5 samples are from spinifex zones in lava flows, 5 samples are from massive zones, 1 from cumulate zone in lava flow and 1 from pillow lava.
Samples were taken with a hammer and processed to exclude secondary veins and altered zones. The concentrations of major elements as well as , Cr, Ni, Sc, V, Cu, Pb, Zn, Bi, Mo, S, As, Rb, Ba, Sr, Ga, Nb, Zr, Y, Th, and U were determined by the XRF method (sequential x-ray spectrometer system Philips PW1480), and Rare earth elements (REEs) were analyzed by ICP-MS method in the Geochemical Laboratory of the Geological Survey of Finland. Sm-Nd methods are described in Peltonen et al (1996). The samples were dissolved in Savillex screw cap teflon beakers. The concentrations were measured using a mixed 150Nd/149Sm-spike without aliquoting. The measurements have been made using a VG SECTOR 54 mass-spectrometer. The estimated error in 147Sm/144Nd is 0.4%. 143Nd/144Nd ratio is normalized to 146Nd/144Nd=0.7219. During the measurements of the analyses the average value for La Jolla standard was 143Nd/144Nd = 0.511851 +- 6 (std. n=15).
Geochemistry and Sm-Nd system
To geochemically study and classify rocks of komatiite-tholeiite associations, data on slightly mobile components were used to rule out the effect of metamorphic processes on rock composition. Based on chemical composition, Vedlozero-Segozero volcanics can be subdivided into high-MgO basalts as well as basaltic, pyroxenitic and peridotitic komatiites. Their MgO content varies from 5-10% in high-MgO basalts to 32-35% in cumulate varieties of peridotitic komatiites, TiO2 content being 0.2-0.7% for komatiites and 0.5-1.5% for basalts. The Vedlozero-Segozero komatiites could be classified as aluminium-undepleted type (Munro type) on the basis of Al2O3/TiO2 ration (Koikary domain-13-22 ; Palaselga domain-13-25) which is close to condritic. However, the CaO/Al2O3 parameter are not universal in many cases because CaO is redistributed during the metamorphism of archean komatiitic associations. For example, Vedlozero-Segozero komatiites, whereas CaO/Al2O3 ration is more than 1 for 5-10% of all samples and less than 1 for 90-95% of the samples. In samples with CaO/Al2O3 >1 CaO was possibly added as a result of metamorphic alterations (carbonatization). In the rest samples, the systematics can be used to define Vedlozero-Segozero komatiites as an Al-undepleted type. This is in good agreement with the age of 2,9-3,0 Ga komatiites from Central Karelia (Russia). It is known that Al-depleted komatiites (Barberton type) dominate in the age interval 3.1-3.4 Ga and Al-undepleted (Munro- type) komatiites prevail over the age interval 2.7-3.1 Ga (Arndt, 1986,1992).
The trace elements rations Ti/Y (Koikary-230-290; Palaselga- 170-210) and Ti/Zr (Koikary-90-100; Palaselga- 60-87) show characteristics of the Al-undepleted komatiites. Pyroxenitic komatiites (350-61;90-1) and high-Mg basalts (847-1; 849-1) samples from Koikary domain have a close to chondrite REE distribution- (La/Sm)n =0.8-1.1, (Gd/Yb)n=0.9-1.2, (Ce/Yb)n=0.9-1.1; Pyroxenitic komatiites from Koikary domain (2- 1, 9-50) and peridotitic, piroxenitic komatiites from Palaselga domain (275-16; 5410- 6; 60-9; 60-8a) have slightly LREE-depleted distribution - (La/Sm)n=0.4-0.7, (Gd/Yb)n=0.89-1.4, (Ce/Yb)n=0.6-0.8; In contrast, basaltic samples from Palaselga domain (5460-5, 5414-1) have LREE- eriched distribution with chondrite HREE pattern.
The Sm-Nd analyses are technically fairly good and are shown in Table and Figure. According to the age estimates above the epsilon in the table has been calculated using the age of 2940 Ma. The analysed rocks from the Palaselga domain include four LREE depleted komatiites and two basalts, which are strongly enriched in LREE. Three of the Palaselga komatiites have positive þ-values whereas one (275-16) is really off. One likely reason for this deviation could be a slight Proterozoic Sm/Nd fractionation, which has been observed frequently e.g. in Finland. Similarly one of the four LREE depleted Koikari komatiites deviates significantly from the others, which provide an e (2940) of ca. +1.7. The anomalous sample (90-1) has also a very low CaO content (0.56%!). The Koikary Mg-basalts have flat REE patterns and positive e-values, which is also the case with the Palaselga LREE enriched basalts, when the age of 2.9-3.0 Ga is used. One can thus try to calculate the age relationship for the analysed samples.
The five Palaselga samples (275-16 excluded) provide an age estimate of 2893 +/-110 Ma (e =+1.2, MSWD=5), and the five Koikary samples (90-1 excl) an age of 2944 +/-170 Ma (e=+1.7, MSWD=2). It should be noted that the age calculation assumes that the samples are cogenetic i.e. they did originate simultaneously and had the same initial Nd isotopic composition. The requirements are really strong, and difficult to verify for a set of whole rocks (What might be e.g. the genetic link between Palaselga komatiites and LREE enriched basalts?). There are several examples (e.g. Kambalda), where whole rock age estimates are much too old, because LREE- enriched samples have been strongly contaminated by older crustal material.
In fact, whole rock Sm-Nd dating is generally very suspect, but the data are useful when other age constraints are available. Provided that both Koikary and Palaselga are cogenetic the calculation gives an age of 2921 +/-55 Ma (e=+1.5, MSWD=5, n=10). The MSWD values exceeding one suggest that there should be some scatter in excess of the analytical errors, which could be due to unrelated samples, but also due to late postmagmatic effects as discussed above. Nevertheless, the age range fits the estimate derived from other data. If the age of the rocks is 2.9-3.0 Ga the Sm- Nd data show that also the LREE enriched basalts originated from the depleted mantle without interaction with much older crustal material. The Sm-Nd data also fit the model where komatiites were 2.8 Ga old (e ca. +2) and the LREE enriched basalts (e ca. 0) were then slightly contaminated by older crustal material.
Центр коллективного пользования
Проект договора на выполнение работ и оказания услуг для проведения научных исследований, а так же осуществления экспериментальных разработок