SBIR-STTR Award

Alkyl-silica columns (e.g., C8, C18) are presently the most popular by far for use in analyzing samples by means of high-performance liquid chromatography (HPLC). Unfortunately, column manufacturers are currently unable to guarantee that different production batches of nominally equivalent column packing will provide the same separation of different samples, while user surveys indicate that column irreproducibility is a serious problem that can result in major delays and added costs in routine HPLC analysis. The present proposal is intended to solve this problem in two ways. First, a review and extension of previously published studies suggests that it should be possible to characterize different columns in terms of the fundamental column properties that determine sample retention and separation, by the use of a suitable standard sample (mixture of test compounds) and test procedure. It may also be possible to modify these column properties by suitable adjustments in other separation conditions (mobile phase, temperature, flow rate), so as to reduce (and hopefully eliminate) minor differences in column retention for any sample. Second, it is proposed to develop computer software that will allow the user to conveniently predict separation as a function of small, simultaneous variations in any number of experimental conditions (e.g., mobile phase pH, column temperature, etc.) The software would also automatically correct for column-to-column differences in retention by recommending suitable adjustments in separation conditions. This would minimize the impact of column irreproducibility, as well as allow unattended correction for any change in the column during use. This capability is currently needed for high-throughput applications of HPLC, as are found in combinatorial synthesis. Only uses of this new software are described in the proposal. PROPOSED COMMERCIAL APPLICATION: The proposed software and accompanying protocols can be used by (a) column manufacturers to better control column manufacture and testing and (b) chromatographers to develop HPLC methods, to troubleshoot assay procedures and to adapt procedures for column irreproducibility.

Manufacturers of alkyl-silica columns (e.g., C8, C18) for high- performance liquid chromatography (HPLC) are currently unable to guarantee that different production batches of nominally equivalent column packing will provide the same retention and separation of different samples. This in turn can lead to a number of related problems during the development and/or use of an HPLC method. Software (RP-COLUMN) that we propose to develop will (a) allow differences in column retention to be minimized by a predictable change in one or more separation conditions (e.g., temperature, pH, etc.), (b) correct for changes in the column during use, (c) facilitate method transfer between laboratories, (d) allow the more reliable duplication of HPLC procedures from the literature, (e) diagnose and correct problems due to error in the selection of experimental conditions (troubleshooting), (f) improve the resolution and robustness of a previously developed procedure, (g) allow the user to characterize any reversed-phase column in terms of its selectivity, (h) provide a data base for the selectivity characteristics of columns of different type and source and (i) select separation conditions such that quite different columns will perform adequately when used with a given procedure (HPLC method). These proposed nine applications of RP- COLUMN each represent a solution to a significant problem in the development or use of HPLC procedures. Applications (a-f) are based on the fact that retention as a function of simultaneous change in several variables (conditions) can be predicted on the basis of a small number of experiments. Applications (g) and (h) require an expanded insight into the retention process for reversed-phase HPLC, while application (i) addresses the problem of column variability in an entirely different fashion. PROPOSED COMMERCIAL APPLICATIONS: The proposed product RP-COLUMN will be regarded as an important addition to virtually every laboratory carrying out HPLC analysis. Manufacturers of HPLC equipment will probably want to incorporate the features of this software into their own data systems and system controllers. We expect initial sale-of $2-5 MM/year.