3. The Geochemical Methodology

Specimens of outcrop rock were analysed at the Open University using Spectrace TN9000 PXRF equipment. Stone implements were analysed using curved and weathered surfaces, with corrections made for Fe back-scatter peaks, calibrations for offsets and other acquisitioned errors (Markham 2000). PXRF measurements made on an ideal sample surface were statistically compared with those made on non-ideal surfaces to establish if there was any difference in XPRF accuracy. The main rock type encountered in this work was gabbro. In order to investigate the accuracy of the XPRF when measuring gabbro, a selection of gabbroic rocks was obtained from exposures on Carrock Fell and prepared for the analysis in the form of pressed powder pellets and glass discs. All PXRF measurements were taken on the same day to minimise the potential for any time-related bias of the PXRF equipment. Four non-overlapping PXRF measurements were taken on each rock specimen and stone implement in order to achieve a sample precision of better than 5% relative standard deviation of the mean, as suggested by Potts et al. (1997). At least two PXRF analyses of the exposure pellets were made; the number of analyses was justified by the grain size of each rock specimen and the results averaged. A dataset produced by measuring control pellets was used to examine the ongoing accuracy of the XPRF, and to examine the precision obtained through repeated measurements of the same pellets.

Seven 'fresh' Carrock Fell gabbro specimens and one fine-grained granophyre specimen were selected for laboratory analysis by ICP, at Imperial College, London. The ARL 34000 simultaneous spectrometer was calibrated using 12 standards, and two international standards (MRG-1 and PM-S) were run with the batch to monitor data quality. Powder rock samples were dry weighed (0.2500g ± 0.0001g) and mixed with flux. The samples were fused in an electric furnace at 1000 degrees C for 30 minutes. The red-hot bead was added to a solution of 25ml 2 M nitric acid and 150ml of deionised water, and mixed with a magnetic stirrer for 30 minutes until dissolved. The solution was then diluted to 250ml and filtered. A loss-on-ignition (LOI) was performed to complete the major element analysis.

PREVIOUS NEXT CONTENTS HOME