Extending the power transform approach for recovering areas of overlapping peaks

Power functions, pn(x)=[f(x)]n , are embedded in some modern chromatography detectors and software which not only alter the linear dynamic range of such detectors but also improve the cosmetic aspects of the chromatograms. These aspects include a reduction in baseline noise, improved peak symmetry, and better resolution. However, after raising the electronic output of a detector to a selected power (n > 1), the original peak area information is lost as are very small peaks. Recent advances in the peak processing protocol allow us to recover peak areas from overlapping peak areas, even in noisy environments using power functions. One of the primary protocol requirements of this approach was to have resolution factors ≥0.9. An increasing positive bias in the recovered area was observed as the resolution factors decreased below 0.9. In this work, we extend the capabilities of power function to lower resolution values. This bias is addressed by considering the increasing contributions in peak height coming from adjacent peaks. It is shown that the power function can now be used as long as the peak maxima are visible, making Rs = 0.5, the lower treatable resolution factor for two peaks of similar areas.