Investigation made possible by National Science Foundation Grant Ño. GP-1000. The gas chromatographic instrument used was obtained through a separate National Science Foundation Grant No. GP-2117.

No abstract

Key Words:Gas chromatography Capillary Theory Linearity between peak width and retention data Evaluation of GC equipment SummaryThe linearity of the relationship between OT,t2 and (1 + k)2 is studied with the view of using it for the evaluation of GC equipment. Based on appropriate analytical expressions it is concluded that in general this relationship is not linear and that practical proceduresfor column evaluation should be developed that take this non-linearity into account.Previous papers [1,2] show that since OT,t2 varies linearly with (1 + k)2, provided that 0~,~2 is independent of k. [OT,t2 is the variance, in (time)> units, of the peak as it appears on the chromatogram; Oe,t2 is the extracolumn contribution and aC,t2 = (1+k)2 0~,~2 , the column contribution to OT,t2].The aim of the present study is to establish the conditions under which 0~,~2 is sufficiently indipendent of k to warrant an analysis based on equation 1. Although completely new diagnostic tools are suggested by this analysis, practical procedures are not proposed; its exclusive aim is to provide explicit information on the behaviour of the equation 1. It is limited to gas chromatography in open tubes. TheoryDeparting from the almost axiomatic relationship [2,3] and the definitions A comparison of equations 1 and 6 indicates that the linearity of equation 6 is determined by the k-dependence H of the group (t)tm2 , and, since 1 and tm are not affected by k, only the k-dependence of H. This dependence is introduced explicity by substituting the Golay-equation (4) for H into equation 6. r-where Dm and Ds are the diffusion coefficient of the solute in, respectively, the mobile and the stationary phase, rt is the column radius, u the linear velocity of the mobile phase and df the thickness of the stationary phase liquid layer.The first is the relative size of ___ lg6 andf(k); obviouslyat large values of ~ lg6 (i.e. small values of (ScRe)2) it will (ScRep (ScRe)2 be relatively insensitive to changes in (l+k). The second is the extent of (l+k)2-dependence of f(k). This in turn is determined by the functional behaviour of f[(I+k)2] as well as the magnitude of X. The behaviour of these functions will now be investigated. The functions fl(k), f2(k) and f(k)The behaviour of fl(k) and f2(k) has previously been reported as functions of k [41. The present study concerns itself with the change of these functions with (l+k)2.

SummaryThe technique of controlling chromatographic selectivity by the adjustment of column temperatures in systems of seriescoupled columns is investigated by means of a general model incorporating the effects of temperature and mobile phasecompressibility. In a previous article the performance of seriescoupled columns was investigated under conditions of constant overall pressuredrop.Thepresent paper,on theother hand, considers systems in which a constant mass flow rate is maintained at all temperatures. The expressions derived for the effective partition coefficient and the retention time are compared with those of the previous paper. The numerical results for twocolumn systems exhibit the same major trends as those for constant pressure drop. It is shown that the effective partition coefficient is slightly dependent on the mass flow rate.

No abstract