An experimental study was carried out for the separation of a series of homologous 2-ketoalkanes by reversed-phase gradient elution. Both linear and non-linear gradients were used. Computer simulation was applied to these same separations, and comparisons were made between experimental and predicted results (values of retention time, retention time difference and bandwidth). Errors in the predicted separation were generally small and similar for both linear and non-linear gradients. Somewhat larger errors (retention time differences) were found in non-linear gradient separations for bands eluting after a change in gradient steepness. This can be attributed to the dispersion (rounding) of the gradient as it passes through the high-performance liquid chromatographic equipment. Computer simulation was demonstrated to provide predictions that are adequate for the purposes of developing an optimized gradient separation.

N,N′-1,2-Ethylenediylbis-l-cysteine dietyl ester (ECD) is a chiral compound and is a key component of Neurolite, a brain imaging agent. In order for the material to be used in the manufacture of Neurolite, it must meet optical rotation and other purity specifications. The optical rotation of ECD is affected by the presence of oxidation-related impurities of the parent material. Prior to this work, the optical rotation was used as a gross indication of these impurities. During product development, information regarding the impurity profile became necessary in order to understand and monitor ECD degradation. A gradient elution high-performance liquid chromatographic method compatible with liquid chromatography—mass spectrometry was developed and optimized using Drylab G software. System suitability of the method was assessed by adding l-methionine ethyl ester and acetophenone to the ECD standard as resolution markers. Comparison of the resolution between each marker and ECD with previous measures of resolution ensures sufficient zone capacity to resolve all potential impurities.

The use of computer simulation software for high-performance liquid chromatographic (HPLC) method development is considered. In particular, gradient elution data entered into DryLab G/plus are used to predict isocratic retention times. The motivation was to establish whether data generated in a research-grade, gradient environment might be used to simulate accurately isocratic HPLC conditions applicable to a process monitoring operation. Good agreement between experimentally obtained and computer-predicted retention times for a homologous series of alkylketones was found for the conversion from gradient to isocratic elution conditions.

DryLab G software was used for the optimization of gradient elution programs in reversed-phase high-performance liquid chromatography applied to the chromatographic analysis of flavonoids present in marsh-mallow (Althaea officinalis). The optimization experiments were carried out for a mixture of eight standard solutes (isolated previously from the plant) and then applied to an extract from the flowers of marsh-mallow. Computer simulation experiments allowed convenient analytical conditions to be chosen. The use of two modifiers, methanol and acetonitrile, made the identification of separated components more certain. Good agreement between simulated and experimental chromatograms was obtained.

Rapid development of robust and reliable high-performance liquid chromatographic methods for routine quality control of Ginkgo biloba is possible with computer simulation. The goal is to reduce method development time and to increase transparency of the complex composition of plant extracts. With only two basic experiments and a peak tracking process based on the total area ratio compared to the individual peak-area ratios a robust method with more than 50 simulated experiments was completed in 8 h. The best method has been verified experimentally. The correlation between the best simulated run and the final experiment was satisfactory.

The use of computer simulation for developing optimized gas chromatographic (GC) separations is illustrated for several samples. Resolution maps (plots of Rsvs. heating rate r) for different starting temperatures provide a means for the rapid exploration of separation as a function of the temperature program in the case of programmed-temperature GC. Isothermal separations are easily developed by trial and error, because of the speed of computer simulation. More complex samples may require multi-ramp temperature programs, t