Introduction: Screening for volatile inhalants of abuse, as well as analyzing blood alcohol content, is commonly performed in forensic toxicology laboratories using headspace gas chromatography with flame ionization detection (HS-GC-FID). The analyses are performed using dual columns with specialized stationary phases that optimally separate these volatile compounds. While separation profiles of standard blood alcohol screening compounds are usually well characterized by column manufacturers on these application-specific columns, elution profiles of inhalants may not be as readily available. In addition, providing example chromatograms with static run conditions may not suit laboratories that want to experiment with faster run times, column dimensions, carrier gases, etc.
These issues can be solved by using computer modeling software to predict retention times of compounds of interest on a various stationary phases. In addition to the ability of the web-based software to help select a column and provide an optimized separation of compounds of interest on a specific stationary phase, the software can also be used to make changes to analytical conditions and observe the effect on elution, making it a valuable tool for method development and optimization.
Objectives: The intent of this project is to present retention time models for inhalants of abuse and blood alcohol analytes of interest on four unique stationary phases using web-based modeling software and verifying the accuracy of the models against actual analyses. This will allow for optimized separations with faster separations in addition to the evaluation of different carrier gas types.
Methods: To build a database for computer modeling of chromatographic separations, the following fused silica capillary columns were installed into an Agilent 7890A GC with a 5975C MSD: Rtx-BAC1, Rtx-BAC2, Rtx-BAC Plus 1, and Rtx-BAC Plus 2. More than 70 volatile inhalants of abuse, including solvents, refrigerants, nitrites (aka “poppers”) and their metabolites were analyzed on each column using three different temperature-programmed run conditions. Two of the analyses were used to create a retention model based on thermodynamic indices of analytes, and the third analysis was used to verify accuracy against the theoretical model. Once the models were finalized, a web-based modeler was used to optimize separations on each column, decrease analysis times, translate to different column dimensions or carrier gases, and make user input adjustments to parameters, such as carrier gas flow rate and oven ramp rates.
Results: Confirmation runs were in agreement with the theoretical modeled analysis, demonstrating acceptable accuracy of the retention time models using all four columns. Selection of various compounds of interest in the software successfully generated optimized separations on each column, allowing the user to choose the column or column set that best fits their needs. The ability to optimize the method using different carrier gases, temperatures, column flows, different column dimensions and film thicknesses was clearly demonstrated.
Discussion: Computer modeling of retention times in GC is a valuable tool to aid in column phase selection and method development/optimization. The use of this software greatly reduces the time required for manual method development since input changes update the results instantaneously. With libraries of 70+ volatile compounds on four different phases, users can select or input compounds of interest and then calculate elution profiles on each column. The software will present the number of coelutions on each column, allowing the user to select the most appropriate column for their analysis.
