The pharmaceutical and bio-pharmaceutical industries rely on simulations of liquid chromatographic processes for method development and to reduce experimental cost. The use of incorrect injection profiles as inlet boundary condition for these simulations may, however, lead to inaccurate results. This study presents a novel modelling approach for accurate prediction of injection profiles for liquid chromatographic columns. The model uses the residence time distribution theory and accounts for the residence time of the sample through the injection loop, connecting tubes and heat exchangers that exist upstream of the actual chromatographic column, between the injection point and the column inlet. To validate the model, we compare simulation results with experimental injection profiles taken from the literature for 20 operating conditions. The average errors in the predictions of the mean and variance of the injection profiles result to be 8.98% and 8.52%, respectively. The model, which is based on fundamental equations and actual hardware details, accurately predicts the injection profile for a range of sample volumes and sample loop-filling levels without the need of calibration. The proposed modelling approach can help to improve the quality of in-silico simulation and optimization for analytical chromatography.
Keywords: Analytical chromatography; In-silico optimization; Injection profile; Liquid chromatography; Residence time distribution theory.
Copyright © 2023 The Authors. Published by Elsevier B.V. All rights reserved.