Premium
Direct comparison of methods for extracting the rate from first order data with background
Author(s) -
McKin G. H.,
Backhouse C. J.,
Kalantar A. H.
Publication year - 1985
Publication title -
international journal of chemical kinetics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.341
H-Index - 68
eISSN - 1097-4601
pISSN - 0538-8066
DOI - 10.1002/kin.550170610
Subject(s) - weighting , chemistry , lag , noise (video) , statistics , standard deviation , random noise , algorithm , mathematics , artificial intelligence , computer science , physics , computer network , acoustics , image (mathematics)
Abstract Many methods involving a lag or retardation can linearize A exp(− kt ) + Z data, allowing k to be obtained from the slope. A classification scheme is found in which the efficiencies are related to the classes of methods. The methods yielding k with the smallest standard deviations (σ k ) are compared quantitatively. The resulting error factors for k and the lags are tabulated for each of the four methods compared, for use by the experimenter. This is done for various ranges, backgrounds, and kinds of random noise. Effects on k and σ k due to correlation, incorrect weighting, choice of lag, and the distribution of the plotted data are detailed. Effects at high noise levels are discussed for the (usually) best direct method, due to Guggenheim. These include the onset and extent of bias in k and failure of the method to yield any k .