Retention time calculation gradient elution

The BSF was further analysed with HPLC-DAD and compared to reference palytoxin from P toxica, by using a mobile phase of water acidified to pH 2.5 with trifluoroacetic acid (solvent A) and pure acetouitrile (solvent B). A linear gradient was apphed from 30% to 70% of solvent B over 45 minutes. The BSF HPLC screening revealed two distinct peaks, which were eluted at approximately 38% acetonitrile with retention times very close to that of reference palytoxin. Both peaks also showed two UV absorption maxima, at 233 and 263 mn, while the ratio between their absorbance (233 versus 263 tun) was identical to that calculated for reference palytoxin. The peaks were collected separately and the toxic compounds pmified from the BSF were called mascarenotoxin-A (McTx-A) and mas-carenotoxin-B (McTx-B) (Lenoir et al. 2004). 

Gradient elution in RPC, NPC or lEC systems can be optimised using principally the same strategies as in isocratic chromatography, which are briefly described in Section 1.4.7. Simultaneous optimisation of gradient time (steepness), initial concentration and — if necessary — gradient shape can use Eqs. (1.32)-(1.37) for predictive calculations of the retention and of the resolution of the individual pairs of sample compounds from the isocratic data acquired in a few mobile phases of different composition or in a few initial gradient-elution runs. 

Comparison of these values with the ones calculated from peaks eluting at similar retention times in the analysis of extracts could be a fast way to verify the identity and to check the peak purity. Analysis of the semipurified taxine mixture by gradient HPLC-DAD revealed the presence of seven taxanes with 10-deacetyl baccatin III (10-DAB III) among them. All the nine expected taxines were positively identified (Table 1). 

Figure 18.14 Window diagram. The separation was performed with a linear gradient from 0 to 45% B and the optimum runtime needs to be found out. (a) Gradient in 15 min (b) gradient in 45 min with some elution orders reversed (c) window diagram calculated from the initial two experiments with a linear relationship between retention time and %B assumed the plot shows the resolution / of the peak pair which is critical under the respective conditions it is necessary to use long gradient runtimes to obtain a good resolution (d) optimized chromatogram with 0-45% B in 80 min, but separation is already finished after 45 min and 25% B.

To scale up from an analytical run to a preparative run, we first calculate the concentration of ACN that elutes the peptide of interest and use that value to program a gradient for preparative HPLC. With the analytical HPLC used in our laboratory, the time for solvent to pass from the inlet to the detector is 2 min. The retention time (r.t.) of the peptide of interest is used to calculate the concentration of ACN that is required to elute the desired product at a preparative scale ... 

The order of the columns was then reversed so that the solute entered the 15 cm phenyl phase column and eluted from the 10 cm column. A forty minute gradient from 30% to 70% methanol was run. The resulting data are presented In Table IX. Again, excellent agreement between the actual and calculated retention times occurred for all five compounds. 

In a gradient, both (dependant on the course of the gradient) and k (dependant only on %B at the moment of exit from the column) must be calculated [11]. So on the face of it the retention factor k corresponds to the integral (simplified the average ) of the individual retention factors over the entire gradient under the selected conditions, while k is the retention factor at the moment of elution/exit from the column. As with the above-mentioned analogy, the retention time is calculated from k and the peak width from k. In isocratic separations, k and k coincide. Ihe gradient slope G is explained below (Section 3.2). 

In theory, it is possible to derive the optimal conditions for a gradient elution with a sodium hydroxide eluent from the functional dependence of log (Vms — Vd)/Vd from log R. However, as mentioned before, this applies only to simple linear gradients with the initial eluent ion concentration of zero, which is rarely used for practical purposes. Much shorter analysis times are obtained when the gradient run starts at a higher eluent ion concentration than zero. Furthermore, gradient programs with different ramps, sometimes combined with isocratic periods, have to be developed to obtain optimal selectivity and speed of analysis. A mathematical description of the retention is impossible in all these cases, because the resulting equation for the calculation of the retention volume would be far too complex. 

---