1-D Automatic Phase Correction
iNMR gives the user the choice between three auto-phasing algorithms, for higher flexibility. These methods are quite reliable in 1D spectroscopy. It's also true that they work perfectly in 2D, yet the user should be aware that there is no substitute for experience and knowledge, in that field. In mono-dimensional spectroscopy all new spectra are already phase corrected by iNMR, by with the global method described below, so the user should just concentrate on manual phase correction or try the other methods.
Whatever method you choose, remember that sometimes you get better results if you perform the correction twice.
The aim of this method is to rotate the whole spectrum upwards. (More technically speaking: the method keeps changing the phase while monitoring the single lowest point and reading its ordinate value; iteration stops when this value cannot grow anymore). You will easily realize it is not appropriate for experiments like DEPT, APT, NOEdiff, where peaks are expected to be both positive and negative. In the remaining cases it is exceptionally stable, being not affected at all by crowding of peaks, noise or field inhomogeneity. It is affected, however, by baseline distortions. This method is run when the user first creates one or more horizontal marks (alt-click) and then selects the command ‘Process/Autophase’. If you don't like menus, you can find an equivalent icon in the tools palette (dashboard).
If, instead, any number of vertical marks exist, the selective method is used. In this case, the aim of the program is to make symmetric the selected peaks (pointed to by the marks). Needless to say, for the method to work, the marks should fall on isolated, well-shaped peaks. To create a mark, just Cmd-click on the spectrum. Then, you can drag it on any peak. The selective method is more appropriate for spectra containing negative peaks or in the presence of a distorted baseline. It is insufficient in crowded spectra or when the sample has not been properly shimmed. It is also inferior when the digital resolution is low. In other words, it is not the method of choice.
The perfect choice when the the baseline is flat and the signals are concentrated towards the center of the spectral width. If there is a peak at the center, it is ignored (useful if it can't be phased). This is the only method that performs better than manual correction. (Default from version 2.1.4 onward) If you want to use it, check that no marks are visible before issuing the command “Process/Automatic Phase” (or clicking the corresponding icon, or using the keyboard shortcut Cmd-Shift-K).