This paper introduces a new approach to improve the ruggedness of now-
based kinetic determinations. After reagents and sample are mixed, the
Bow is stopped, signal vs. time data are recorded by an on-line compu
ter, and curve-fitting methods are used to compute the signal that wou
ld be measured if the reaction were monitored to equilibrium. The new
approach is evaluated using the Mo(VI) catalyzed oxidation of iodide t
o triiodide by hydrogen peroxide. Time-dependent concentrations of tri
iodide are monitored by changes in absorbance at 360 nm. Effects of ch
anges in pH as well as the concentrations of iodide and Mo(VI) are eva
luated using four different data-processing options, two of which were
expected to exhibit poor ruggedness to changes in these variables and
two of which were expected to exhibit significantly improved ruggedne
ss. All data-processing options yielded linear calibration plots and h
ad similar degrees of precision. As expected, the principal difference
among the options was the ability to reject effects of changes in exp
erimental variables. The new options are 10- to 70-fold less dependent
on changes in pH and concentrations of iodide and molybdate than the
more conventional options. (C) 1998 Elsevier Science B.V.