Figure 3.39

"fall into line" with perfect absorptive phase. One notable exception is an aliased (folded) peak, which appears within the spectral window when it really belongs to a very different chemical shift. Its true chemical shift defines its phase error and the method fails for this peak. The two signs of an aliased peak are an uncorrectable phase error and reduced intensity due to attenuation by the audio filter.

There is one situation where any phase correction procedure can fail: It is possible to set m to a large value such that the second (nonpivot) peak is given a phase correction that is 360° too large (Fig. 3.40). This will not affect the shape of the "other" peak, but will introduce a "phase twist" to all the peaks in between. For example, a peak exactly between the pivot peak and the second peak in chemical shift will have a phase error of 180° and thus will appear upside down. Even greater phase twists (720°, 1080°, etc.) can be applied if you are not careful. Using an automatic phasing routine (e.g., Varian aph command) can give bizarre values for the first-order phase parameters (slope = m) for spectra with poor signal-to-noise ratios. If you get into this situation, set the first-order phase correction


0 0

Post a comment