Peak measurements and area conlputations provide a substantial portion of the working man-hours required in gas-liquid chromatography, GLC, work. Modern electronic digital computers not only speed simple routine calculations but make it feasible to employ a series of systematic correction factors designed to increase the over-all accuracy of the analytical restllts in GLC work.
Basic to most computation procedures is the assumption that there exists a simple relationship between peak area and sample component weight. For component i:
Wti k (peak area)~ (I)
Bartlet and Smith t) have indicated that GLC peaks closely approximate Gaussian probability curves whose areas can be computed from the equation:
where H is the peak height and er is the standard deviation. Furthermore, they report that a is proportional to the retention time expressed either in time or chart distance units. Equation constants become more general when retention time is expressed as a retention ratio, RR, in which the retention time for each component is divided by the retention time for methyl stearate. A combination of these considerations leads to the relationship frequently used to compute routine GLC results:
where k is a proportionality constant. The utility of this equation is increased by the fact that in many cases retention ratios are more easily and more precisely measurable than are the standard deviations required by equation ( 2). The work reported here demonstrates that k ofeq. (3) is not a constant, but is a systematic function of retention time and injected sample size. The determination and use of this function in routine calculations can help to avoid analytical errors of 10-407/,, in the case of certain components.
Experimentnl
The methyl esters of caprylic (8:0), capric (10:0), lauric (12:0), myristic (14:0), palmitic (16:0), stearic (18:0), arachidic (20:0), and erucic (22: 1) * Institute of Comparative Medicine Publication Number 666