Previous   |   Next

doi:10.2204/iodp.proc.324.105.2010

Geochemistry

Major and trace element analyses

Thirteen samples from Hole U1348A were analyzed for concentrations of major element oxides and several trace elements (Table T6) by ICP-AES (see "Geochemistry" in the "Methods" chapter for information on analytical procedures, instrumentation, and data quality). Two samples were bulk specimens from clay-rich intervals in stratigraphic Unit II; one of these samples was from a light green interval (Sample 324-U1348A-12R-1, 17–18 cm), and the other was from a yellow-green interval (Sample 324-U1348A-12R-CC, 6–8 cm) (see "Sedimentology"). The remaining eleven samples were from totally altered material within stratigraphic Units III, V, and VI (see "Alteration and metamorphic petrology"); six were bulk samples of volcaniclastic rock and five were volcanic clasts that we separated from their matrix. Each of the clasts was crushed to roughly millimeter-size chips, and the chips were hand-picked under a binocular microscope to obtain the least-altered material.

As with the Site U1346 and U1347 data, total weight percentages for the major element oxides vary significantly, from 98.57 to 104.22 wt%. We again normalized the raw major element values to 100 wt% totals in order to facilitate comparison of our results with one another and with data from the literature. The normalized values are presented below the raw data in Table T6 and are used in the figures and in the discussion below.

The two samples from Unit II have moderate (for sedimentary rocks) weight loss on ignition (LOI) values (11.68 and 9.12 wt% in Samples 324-U1348A-12R-1, 17–18 cm, and 12R-CC, 6–8 cm, respectively) and very low CaO concentrations (below the ~0.08 wt% detection limit), indicating that very little, if any, carbonate is present. An absence of phosphate minerals is indicated by P₂O₅ contents below the ~0.06 wt% detection limit. The chemical effects of diagenesis are difficult to disentangle from original (i.e., syndepositional) composition in these samples. However, the samples have high SiO₂ (62.87 and 58.75 wt%; Fig. F49) for their Al₂O₃ contents (19.45 and 14.16 wt%), relatively low MgO (3.80 and 4.95 wt%), high K₂O (3.34 and 2.81 wt%), very low TiO₂ (0.19 and 0.12 wt%), and high Zr (210 and 281 ppm). No radiolarian microfossils or other biogenic materials were observed in the layers from which the samples were taken (see "Paleontology"). Therefore, the two clay-rich layers may contain a significant proportion of wind- and/or water-borne magmatic arc– or continental crust–derived material. However, the presence of silica cement elsewhere in Unit II and of chert in Unit I (see "Sedimentology") suggests that some of the SiO₂ in these two layers may have a postdepositional origin through circulating silica-rich solutions derived from dissolution of siliceous microfossils in overlying or underlying beds.

Despite our efforts to avoid alteration in the volcanic material of Units III, V, and VI, alteration remained substantial in all of the samples selected for analysis. Very high LOI values of 25.09 and 27.08 wt% were obtained for bulk Samples 324-U1348-14R-1, 18–20 cm, and 18R-3, 49–50 cm, respectively, as well as very high CaO contents of 38.97 and 29.93 wt%. In contrast, SiO₂ concentrations are low (33.63 and 36.81 wt%). Together, these results indicate that a large amount of carbonate was present in the analyzed splits of these two samples, and we do not discuss them further. For the other nine samples, LOI values are lower but still very high (Fig. F50A), between 6.22 and 12.86 wt%, except for one sample with 3.69 wt% (cf. LOI values of 0.07 to 3.57 wt% for the Site U1347 basalts or <1 wt% for unaltered tholeiitic basalts). Total alkali contents are as high as 7.87 wt% (Fig. F49), even higher than for the Site U1346 samples; the highest values were obtained from clasts. Concentrations of K₂O, in particular, appear to have been elevated substantially, though variably, by alteration. Concentrations of SiO₂ range between 45.06 and 52.51 wt%. Allowing for alteration-related modification of SiO₂, this range suggests the samples are probably all basaltic, or at least were originally. In comparison, SiO₂ values in the basaltic volcaniclastic rocks of Ontong Java Plateau ODP Leg 192 Site 1184 lie outside the range of basalt compositions only for the most altered samples, which also have the highest total alkali contents (Mahoney, Fitton, Wallace, et al., 2001; Shafer et al., 2004). Variation diagrams reveal significant alteration-related disturbance of many other elements, including CaO, MgO, MnO, and Fe₂O₃^T (e.g., Fig. F50B). Phosphate contents are highly variable, ranging from below the detection limit in Section 324-U1348-26R-2 to 5.98 wt% in Section 17R-3 (Fig. F50C). A positive correlation between P₂O₅ and CaO is present, and five samples have P₂O₅ concentrations >1 wt%, indicative of phosphatization during alteration. In addition, phosphatization has disturbed Y and Sr contents significantly, as evidenced by positive correlations of these elements with P₂O₅ (Fig. F51). Concentrations of Ba, Ni, Cr, and Zn, and to a lesser extent Sc, Cu, and V, are also likely to at least partly reflect the influence of the pervasive alteration, as these elements define rather scattered distributions in variation diagrams.

Thus, we are left with few elements with which to infer original composition. Two of the most generally alteration-resistant elements during low-temperature and hydrothermal alteration are Zr and Ti (e.g., Cann, 1970; Humphris and Thompson, 1978; Patino et al., 2003). In a diagram of TiO₂ versus Zr (Fig. F50D), the Site U1348 data define an array very similar to that of the vastly less altered Site U1347 basalts (even including an off-array high-Zr sample resembling the evolved segregation of Site U1347). On this basis, the Site U1348 volcaniclastic rocks appear to have been derived from basalts broadly similar in composition to those of Site U1347.

Top of page   |   Previous   |   Next