Prediction of GC RT shows impressive accuracy

Skip to Navigation


  • Published: Aug 10, 2018
  • Author: Ryan De Vooght-Johnson
  • Copyright: Image: Sebastian Duda/Shutterstock
  • Channels: Laboratory Informatics
thumbnail image: Prediction of GC RT shows impressive accuracy

Reliable methods needed to predict GC retention times

Credit: Sebastian Duda/Shutterstock.

The prediction of retention times (RT values) for GC is useful, since it can save the time and expense of carrying out a large number of experimental runs. Thermodynamic methods are typically the most accurate ways of carrying out such estimations, but can fail when equations derived from one column and instrument are applied to others.

The Edmonton researchers devised a new theoretical method of predicting retention times, applicable to different columns, instruments and temperature profiles, provided the stationary phase remains the same. Previous work using the simplex algorithm had given inconsistent results, so the new equations were carefully tested by simulation and then by using real compounds.

New theoretical GC predictions compared with experiment

Initially, a differential thermodynamic equation was derived linking the change in the distance (dx) a compound travels to the change in time (dt); the equation also contained various obtainable parameters, such as the column dimensions, pressure drop, temperature and mobile phase viscosity (in this case that of helium). The equation contained three thermodynamic constants, A, B and C. Riemann integration of the differential equation gave equations linking distance (x) to time (t), thus allowing retention times to be calculated for a known column length.

A least-squares model was used to determine the constants A, B and C for a given compound and stationary phase; at the same time the column internal diameter was calculated (this was possible if the column length was known). An optimisation algorithm, based on gradient methods, was used to find solutions to the integrated equations; it proved to be superior to the simplex algorithm used previously.

A simulation study, using MatLab R2016b, was initially carried out. Seven ‘virtual’ compounds were used, and it was found that the calculations gave plausible results. The simulation was repeated with small random errors in the retention times, which were found to have fairly large effects on the values calculated for the constants A, B and C, although the authors state that this does not necessarily mean that the predictive power of the overall method will be poor.

Retention times were calculated using the new method for a set of five compounds: 2-octanone, n-decane, 1-octanol, 2,4-dimethylphenol and 2,6-dimethylaniline. These are a fairly diverse set of compounds, one basic, one slightly acidic (the phenol) and three neutral; the compounds also vary in polarity, although none are highly polar. The column length was not determined by direct measurement; instead, methane was injected as a dead time (hold-up time) marker, and the length was calculated with the Chemstation calibration tool. The three thermodynamic constants, A, B and C, were calculated for each compound. The theoretical times were compared with those determined experimentally.

GC was carried out using an Agilent HP6890 instrument fitted with a Restek Rtx-5 column and a flame ionisation detector (FID). Helium was used as the carrier gas with a flow rate of 2.6 mL/min. Separations were carried out between 40 and 320 °C with five ramp rates: 5, 8, 10, 16 and 20 °C/min. Each run gave clear separation of the five compounds. The experimental retention times were very similar to the theoretical predictions, the differences between the values all being less than 0.1%.

New method of estimating retention times shows promise

The new method gave very good estimates of retention times for the five compounds examined. It will be extremely useful if this high accuracy proves to be general. The authors state that the paper is the first of three on this topic, so further details of its validation and applicability should be revealed shortly.

Related Links

Journal of Separation Science, Hou et al.: A simple, fast, and accurate thermodynamic‐based approach for transfer and prediction of gas chromatography retention times between columns and instruments. Part I: Estimation of reference column geometry and thermodynamic parameters.

Journal of Chromatography A 2015, 1406, 258-265. Claumann et al. "Fast and accurate numerical method for predicting gas chromatography retention time."

Journal of Chromatography A 2014, 1330, 69-73. McGinitie et al. "A standardized method for the calibration of thermodynamic data for the prediction of gas chromatographic retention times."

Article by Ryan De Vooght-Johnson

The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.

Follow us on Twitter!

Social Links

Share This Links

Bookmark and Share


Suppliers Selection
Societies Selection

Banner Ad

Click here to see
all job opportunities

Most Viewed

Copyright Information

Interested in spectroscopy? Visit our sister site

Copyright © 2018 John Wiley & Sons, Inc. All Rights Reserved