Previous   |    Next

doi:10.2204/iodp.proc.314315316.201.2011

Methods

Two groups of standard minerals (C-group and K-group) were mixed: a poorly crystalline kaolinite standard, Clay Minerals Society (CMS) KGa-1, and a variety of chlorite (ripidolite), CMS standard CCa-2. We selected those source clays standards because they are easily accessible to the research community and they have been thoroughly analyzed by other research (e.g., Chipera and Bish, 2001; Vogt et al., 2002). Proportions by weight are similar for the two groups of mixtures but not identical (Table T1). Prior to blending, each standard was suspended in ~500 mL of distilled water with sodium hexametaphosphate dispersant and disaggregated using an ultrasonic cell disrupter. Particles <2 µm equivalent settling diameter were separated by centrifugation (1000 rpm for 2.4 min; ~320× g). The average concentration of each suspension was determined by extracting three aliquots and drying at 75°C to obtain dry weight of clay per unit volume of suspension, corrected for weight of dispersant. The two components were measured by pipette and their volumetric proportions converted to dry weights and weight percentages (Table T1). Oriented aggregates on glass slides were prepared using the filter-peel method and 0.45 µm membranes (Moore and Reynolds, 1997).

Working with an older analog diffractometer, Biscaye (1964) advocated the use of slow scanning rates (0.2°2θ/m; 0.01°2θ step) to create higher intensity and reproducibility of the double peaks at ~3.5 Å. We achieved optimal results with a faster scan (1°2θ/m; 0.01°2θ step) from 23° to 28°2θ. The instrument is a Scintag Pad V X-ray diffractometer with CuKα radiation (1.54 Å) and a Ni filter, set to 40 kV and 30 mA. Slits were 0.5 mm (divergence) and 0.2 mm (receiving). We processed the digital data using MacDiff software (version 4.2.5) to establish a baseline of intensity, provide smooth counts, and correct for offset of peak positions caused by misalignment of the instrument’s detector and small differences in the orientation of slides within the sample changer.

The uncorrected position of the kaolinite peak of interest ranges from 25.09° to 25.19°2θ, and the uncorrected chlorite peak ranges from 25.33° to 25.48°2θ. The peak positions were corrected to either kaolinite (002) at ~24.8°2θ (d-value = 3.58 Å) or chlorite (004) at ~25.1°2θ (d-value = 3.54 Å) (Fig. F1). The corrected peak profiles were used to locate the centers of peaks, from which the maximum intensities were determined. When both crests are resolved, angular separation between the two overlapping peaks averages 0.27°2θ (Fig. F1), which is consistent with synthetic diffractograms (e.g., Moore and Reynolds, 1997). Therefore, we always assumed that the crest of the subsidiary peak is separated from the crest of the dominant peak by 0.27°2θ when picking its point of maximum intensity on the shoulder of the composite (Fig. F1). In addition, we modeled the peaks using the Pearson II peak-fitting function of MacDiff on the corrected diffractogram to obtain the fitted area of those two peaks (Figs. F1, F2). The least residuum reduction was set to 0.001, and the largest loop number of refinement was set at 100.

On all the diffractograms generated for the standard mixtures (Fig. F3), we found that one side of the dominant peak is not distorted by interference from the minor peak. Using only the undistorted side makes it easy to compute the dominant peak’s total area by multiplying its half-peak area by 2× (Fig. F2). One important assumption here is peak symmetry, which does not necessarily hold true for natural mixtures of clay minerals (Moore and Reynolds, 1997). Nevertheless, the ratio of the dominant peak area (PA) to the total area (TA) of the composite peak should correlate with each mineral’s abundance. For mixtures in which both peak crests are distinct, each mineral’s undistorted half-peak area was measured and doubled to compute both of the respective PA values.

Top of page   |    Previous   |    Next