A critical evaluation of multicomponent quantitative analysis by time-correlated single photon counting fluorometry
Abstract
An approach to multi-component analysis of fluorescence decay curves by multiple linear regression is presented with an emphasis on decreased variance. A formula to compute the signal-to-noise ratio of truncated single exponential decays is developed in terms of a shifted time axis, lifetime and pre-exponential factor of the decay. The signal-to-noise ratio for single exponential decays can be maximized to a value equivalent to that possible with time-filtered detection. The signal-to-noise ratio in the multi-component case can be optimized to a value slightly less than that possible for the single component case by this time-axis shift. It is noted that the variance-covariance matrix is minimized for an offset value which optimizes both or all three variables in the matrix at once. This is equivalent to minimizing the variance of the intensity values returned from the regression. Colinearity in the functions used in the multiple linear regression prevented the simultaneous determination of three components by this method if the decay at one wavelength was used. If the decays at several wavelengths were used, the covariance in the terms was reduced. An equation is derived to predict the minimum signal required for an emitter of one lifetime to be detected in a background of a different lifetime in the truncated region. Mixtures of two decays with lifetimes differing by 50% were resolved and mixtures of three emitters were resolved if the lifetimes of the three spanned an order of magnitude.
Degree
Ph.D.
Advisors
Lytle, Purdue University.
Subject Area
Analytical chemistry
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.