Section 6: Precision and Accuracy
6.1 Nine detergent samples assaying from 3.3% to 4.7% P2O5 were analyzed in quadruplicate by the procedure. The results are shown in Table 1 (see appendix). The pooled standard deviation was 0.10% P2O5 and the pooled coefficient of variation was 2.5%
6.2 The phosphorus content of the above nine detergents was determined by taking the "titration solution" after the procedure as described above, diluting to 500 ml and measuring the phosphate content by an automated molybdenum-blue colorimetric method similar to that previously described[1]. The manifold used is shown in Figure 9. Each solution was analyzed in duplicate, the results are also shown in Table 1. The pooled standard deviation was 0.08% P2O5 and the pooled coefficient of variation was 2.0%.
6.3 The difference between the results by these two methods was an average of 0.10% P2O5, the titration procedure giving the higher result. Treating the 18 mean results by the two methods as paired observations, the standard error of the mean difference was calculated. This calculation shows that the above difference of 0.10% P2O5 is a statistically significant difference at the 95% confidence level. It is believed that the colorimetric determination is a better measurement of the 'true' phosphorus content of the detergent and hence that there is a bias of approximately ± 0.10% P2O5 in the titrimetric determination. However in view of the standard deviation of 0.10% P2O5 in the titrimetric procedure based on the quadruplicate determinations, the 'true' result is included in the confidence limits of the result obtained by the titrimetric procedure.
6.4 The phosphorus content of the detergents was determined in quadruplicate by the ASTM procedure as written. These results are shown in Table 1. In all cases the ASTM procedure produces a result that is higher than the 'true' value.
Page details
- Date modified: