[Simulation of gas chromatographic separation based on random diffusion]
Yinlu Sun,Lin Wang,Zhiyu Yin,Jianwei Zhao
DOI: https://doi.org/10.3724/SP.J.1123.2021.10011
2022-03-08
Abstract:Understanding the diffusion behavior of particles during chromatographic analysis is critical for optimizing the operation conditions, improving the chromatographic performance, and designing a new separation device. Most of the existing simulations focus on chromatographic thermodynamics, while very few consider the overall diffusion and separation process. Herein, a new simulation method for gas chromatography separation was developed based on the random diffusion theory in microscale restricted space. This method retained the key information for controlling the diffusion behavior, simplified the interaction between the particles to be separated, and enlarged the time scale of each step one molecule walks. Thus, the operational efficiency could be significantly improved due to reduced calculation, and the entire diffusion process in the gas chromatography capillary column could be simulated. In this model, the capillary column was represented by a two-dimensional long and narrow rectangle where the particles to be separated randomly diffused. Besides, a directional velocity along the axis of the chromatographic column exerted on the particle represented the driving force of the mobile phase. If a particle collided with the inner wall of the column, it would remain at the collision position even after some time lapsed. When desorption occured, the particle would flow along with the mobile phase until its next adsorption on the surface. The interaction between the particle and the inner wall was expressed by the equivalent adsorption time. By dynamically tracking the trajectories of particles, the statistical distribution of time for the residence of the particles in the chromatographic column could be obtained, that is, the detection signal (retention time). Based on the previous simulation studies on the separation of n-alkane homologues, combined with the Kovats Retention Index, the functional relationships between adsorption steps and temperature as well as carbon number were established. As a result, the separation parameter systems for various homologues at different temperatures were set up. The separation of alcohol homologues at different temperatures was considered as an example to verify the reliability of the simulation method. The relative errors between the measured and simulated values were within 5% for the retention time and 0.75%-60% for the peak width. The reasons for the large relative errors in the peak width are summarized as follows. On the one hand, parameterization of alcohol homologues was performed on the basis of a previous study on the separation law of n-alkane. Given the limitations of the current computing capability, the insufficient iteration in the parameterized process led to large errors. In addition, the errors at different temperatures further accumulated in extrapolated approximations. On the other hand, the strong hydrogen bond force between the alcohol molecules was simplified with the elastic collision, which increased the magnitude of the errors. Although the simulation method proposed in this paper can accurately predict the retention time and reasonably describe the morphological characteristics of chromatographic peaks, there is still room for improvement. In particular, the description of the detailed interactions between molecules must be improved. For example, the method of molecular mechanics may be assistant with the investigation of the functional relationship between interaction potential and adsorption steps. The interaction potential calculated on the basis of molecular mechanics replaces the parameterized adsorption steps, yielding more accurate simulation results. In general, the simulation method proposed in this study is a valuable reference for the optimization of chromatographic operating conditions and for the development of new chromatographic techniques.