IODP Proceedings    Volume contents     Search

doi:10.2204/iodp.proc.324.107.2010

Geochemistry

Major element oxide and trace element analyses

Forty-three samples of lavas from Hole U1350A were analyzed by inductively coupled plasma–atomic emission spectroscopy for concentrations of major element oxides and several trace elements (see "Geochemistry" in the "Methods" chapter for information on analytical procedures, instrumentation, and data quality). Samples of each of the igneous units recovered (stratigraphic Subunits IIa–IIc and Units III and IV) were included. All were taken from holocrystalline portions of the cores.

As with previous Expedition 324 analyses, total weight percentages of the major element oxides varied substantially (95.20–103.11 wt%), and we again normalized the measured values to 100 wt% totals. The normalized data are presented below the measured values in Table T4 and are used in the figures and discussion below. Before proceeding to other results, we note that surprisingly high Ba concentrations of 958 and 734 ppm were obtained for Samples 324-U1350A-17R-1 (Piece 3A, 24–26 cm) and 8R-3 (Piece 10, 81–84 cm), respectively. The other Site U1350 samples contain ≤181 ppm of Ba; most have <80 ppm, in common with igneous rocks from other Shatsky Rise drill sites. Except for their anomalously high Ba concentrations, these two samples are not unusual, although their Sr and alkali element contents are elevated slightly. The most likely explanation appears to be contamination of either the powders or solutions by Ba, perhaps by barite-rich drilling mud. However, neither we nor the ship's laboratory staff have any idea how such contamination might have occurred. Unfortunately, too little time remained in the expedition to resample and reanalyze material from these two intervals.

Weight loss on ignition varies from 0.02 to 4.40 wt% among the Site U1350 samples. This range is similar to that measured for Site U1347 (0.07 to 3.57 wt%) and, in general, the chemical effects of alteration again appear relatively modest compared to those at Sites U1346, U1348, and U1349. In a total alkalis versus SiO2 diagram, values for all but five of the Site U1350 samples fall in the field of tholeiitic basalt (Fig. F40). The exceptions appear to reflect K2O addition during alteration; in particular, several samples from the highly altered lower part of Unit IV (see "Alteration and metamorphic petrology"), from Section 324-U1350A-25R-8 through Core 324-U1350A-26R, have much higher K2O concentrations (as high as 3.08 wt%) (Fig. F41A) than those from shallower parts of the hole. In common with altered samples from other Expedition 324 sites, Na2O contents are not elevated significantly in these samples relative to the other Site U1350 rocks, although values for Site U1350 cover a slightly greater total range and, on average, tend to be slightly higher than observed for Site U1347. Much of the K2O added to the basalts in the lower part of Hole U1350A is likely to have been incorporated into feldspar, given the partial transformation of plagioclase to sanidine in this portion of the hole (see "Alteration and metamorphic petrology"). Another probable consequence of this transformation is that samples from the deepest part of the hole have the lowest CaO contents and CaO/Al2O3 ratios (Fig. F42A), indicative of Ca loss. Other elements that appear to have been affected to variable extents by alteration throughout the hole include Mn and possibly Co. Barium contents also may have been modified by alteration; however, interpretation of Ba requires caution. Given the above-mentioned possibility of Ba contamination in the two samples with anomalously high Ba contents, lesser amounts of the same type of contamination presumably could have affected other samples. Indeed, even excluding the two highest Ba samples, Ba still varies considerably, between 16 ppm (similar to values in the Site 1213 basalts) and 181 ppm. Site U1346 basalts reach the highest Ba levels measured among the sites drilled prior to Site U1350, and as a group, the Site U1346 lavas are more altered than those recovered from Site U1350 (see "Alteration and metamorphic petrology"). However, three Site U1350 samples have greater Ba concentrations than the highest value measured for Site U1346 (93 ppm). Whether affected by alteration or contamination, we do not use Ba in our interpretation of Site U1350 basement geochemistry.

Concentrations of TiO2 vary widely, from 1.31 to 2.51 wt% (Fig. F42). The range in Mg# is also large, from 51.1 to 68.5 (Mg# = 100 × Mg2+/[Mg2+ + Fe2+], assuming that Fe2O3/FeO = 0.15). Similar to Sites 1213 and U1347, the combined Mg# and TiO2 data for Site U1350 define an array that largely lies within the field of ocean-ridge basalts and shows only limited overlap with that for OJP basalts (Fig. F42B). The same is true for a number of other elements, such as Cr (Fig. F42C). At higher concentrations, it is also true for Zr (Fig. F42D); however, at low TiO2 values the Site U1350 basalts, although still within the ocean-ridge field, have slightly lower Zr concentrations than most ocean-ridge basalts. Indeed, the Zr/Ti ratio, a measure of relative incompatible-element enrichment, varies by nearly a factor of 2 overall, from 0.0075 to 0.013. For comparison, normal (N-type) and incompatible-element-enriched (E-type) ocean-ridge basalts average ~0.0097 and ~0.012, respectively (e.g., Sun and McDonough, 1989). Unlike Zr, Sr concentrations are consistently higher in the Site U1350 samples than in the great majority of ocean-ridge basalts; they also tend to be somewhat higher, for a given value of TiO2, than measured for either Sites 1213 or U1347 (Fig. F42E). This characteristic does not appear to be simply (or solely) an alteration effect (e.g., note the general increase of Sr with increasing TiO2). In contrast, V contents and V/Sc ratios tend to be lower than for Site U1347 (e.g., V/Sc = 6.3–8.9 at Site U1350 versus 8.5–11.7 for all but one Site U1347 basalt). At TiO2 concentrations greater than ~2 wt%, V contents also are somewhat lower than for the majority of ocean-ridge basalts. Similarly, Fe2O3T (total iron as ferric iron) tends to be lower at a given TiO2 value than in the Site 1213 or U1347 basalts. Instead, at low TiO2 (<1.7 wt%), Fe2O3T values are similar to those measured for basalts from ODP Leg 191 Site 1179 (9.17–11.08 and 9.30–11.17 wt%, respectively), which represents nonplateau ocean crust formed near Shatsky Rise and during the same period of time.

Interestingly, systematic downhole variations are present for a number of elements. For example, TiO2 concentrations show an overall decrease downhole (Fig. F41B). Correspondingly large downward gradients are seen for Zr and Sr. Sodium also decreases downhole, whereas Mg# generally increases with increasing depth (Fig. F41C). These gradients cannot only be a result of systematic variation in the extent of magmatic differentiation. For example, the Zr/Ti ratio is relatively insensitive to differentiation (at least prior to removal of magnetite, which is not indicated by the variation of TiO2 versus Mg#), yet Zr/Ti values markedly decrease downhole (Fig. F41E). As noted above, the highest Zr/Ti values are greater than the average for E-type ocean-ridge basalts and the lowest values are less than the average for N-type ocean-ridge basalts. This behavior is indicative of decreasing mean amounts of partial melting with time and/or variation in mantle source composition.

In addition to the general downhole trends throughout the basement section, variability is also evident within the two thickest stratigraphic subunits, IIa and IIc. Some of this variation is consistent with differences in magmatic differentiation, but some is not (e.g., in Zr/Ti). For example, within Subunit IIa, distinct differences in Cr, Ni, and Zr/Ti are present below and above ~190 mbsf (e.g., Fig. F41D). A thin, high-Cr (but not high Ni) interval is also present at the top of Subunit IIa; lavas in this interval have low TiO2 and Zr/Ti but not particularly high (or low) Mg#.

Overall, the Site U1350 basalts encompass the greatest range of chemical variation seen at any individual Shatsky Rise drill site. Careful work onshore is required to decipher the relative roles of magmatic differentiation, partial melting, and mantle source composition in producing this variation.