Detector numerical deconvolution

Super-resolution in chromatography is defined as the ability to detect the presence of two or more components when their mean elution tines fall within two standeurd peak widths of each other. Numerical deconvolution techniques for achieving superresolution eu e considered for the case of a single channel detector (7ZD), a multichannel detector (HS) and a two dimensional detector (video fluorometer). It is shown that the degree of ambiguity of the deconvolution decreases in going from a monochannel detector to a multicdiannel detector to a multidimensional detector. Three types of analytical situations are considered for each type of detector ... 

Before comparing with experiment, however, the theoretical results at an energy of 40 meV (equivalent to room temperature) were convoluted with the energy resolution function of the detector used for the measurement. This procedure was adopted because deconvolution of the experimental data was found to be numerically unstable. The convoluted theoretical data were then normalized to the experimental data at zero Doppler shift to yield the results shown. The agreement between the convoluted theoretical results and experiment is extraordinarily good, extending as it does over more than three orders of magnitude. These results also reveal... 

Typically, in measurements of time-resolved luminescence in the time regime of tens of picoseconds, data obtained from 10 to 20 laser shots are averaged to improve the signal-to-noise ratio and to minimize the effects of shot-to-shot variations in the laser pulse energy and shape. Once the reliability of the data has been ensured by application of the corrections described above and made necessary by detector-induced distortions, the time-resolved fluorescence data is analyzed in terms of a kinetic model which assumes that the emitting state is formed with a risetime, xR, and a decay time, Tp. Deconvolution of the excitation pulse from the observed molecular fluorescence is performed numerically. The shape of the excitation pulse to be removed from the streak camera data is assumed to be the same as the prepulse shape, and therefore the prepulse is generally used for the deconvolution procedure. Figure 6 illustrates the quality of the fit of the time-dependent fluorescence data which can be achieved. 

---