A modified mean normalization method to reduce the effect of peak overlap in two-dimensional correlation spectroscopy

Two-dimensional correlation spectroscopy (2D-COS) has been widely used to separate overlapped spectroscopic bands. However, band overlap may sometimes cause misleading results in the 2D-COS spectra, especially if one peak is embedded within another peak by the overlap. Based on the analysis of Lorentzian-type spectra, a new data normalization method has been proposed to overcome such overlap effect. The normalization factor is expressed as the square of the mean value divided by the peak value of each spectrum, Fmean2/Fmax. This expression may be viewed as a modification to the mean normalization method originally put forward by Ozaki and co-workers. Computing simulation demonstrated that the new method works well with simulated dynamic spectra.