Modulation ratio in comprehensive two-dimensional gas chromatography

Comprehensive two-dimensional chromatography employs a serially coupled two-column arrangement where effluent from the first column is collected or sampled and then introduced to the second column according to a chosen modulation period. This is effected by use of a modulator at or near the column junction. One of the considerations in applying the technique is the period of the modulator, which determines the sampling duration of the first column effluent. Here, we propose that the sampling rate can be most effectively described by a new term, called the modulation ratio (M-R). This is defined as the ratio of 4 times the first column peak standard deviation (4 sigma) divided by the modulation period (P-M) or 1.6985 times the half-height width of the peak (w(h)): M-R = 4 sigma/ P-M = w(b)/P-M = (w(h) x 1.6985)/P-M. The 4 sigma value is more commonly recognized as the peak base width (w(b)). The use of 4 sigma as the numerator is preferred to simply sigma because when the PM value used for an experiment is equal to sigma, then the MR value is calculated to be 4, implying that the primary peak will be modulated similar to 4 times as is normally recommended for a comprehensive multidimensional separation. The less well-defined term of modulation number (N-M) has been previously used and proposed as the number of modulations per peak and, therefore, is intended to convey the manner in which the primary column peak is sampled; this is a subjective and not well-characterized value. The use of M-R should provide users with a meaningful and strictly defined value when reporting experimental conditions. The utility of M-R is demonstrated through a mathematical model of the modulation process for both Gaussian and tailing peaks, supported by an experimental study of the modulation ratio. It is shown that for the analysis of trace compounds where precise quantitative measurements are being made, the experiment should be conducted with an M-R of at least 3. Conversely, for semiquantitative methods or the analysis of major components, an M-R of similar to 1.5 should suffice.