Thermodynamic-based retention time predictions of endogenous steroids in comprehensive two-dimensional gas chromatography

Abstract

This work evaluates the application of a thermodynamic model to comprehensive two-dimensional gas chromatography (GC × GC) coupled with time-of-flight mass spectrometry for anabolic agent investigation. Doping control deals with hundreds of drugs that are prohibited in sports. Drug discovery in biological matrices is a challenging task that requires powerful tools when one is faced with the rapidly changing designer drug landscape. In this work, a thermodynamic model developed for the prediction of both primary and secondary retention times in GC × GC has been applied to trimethylsilylated hydroxyl (O-TMS)- and methoxime-trimethylsilylated carbonyl (MO-TMS)-derivatized endogenous steroids. This model was previously demonstrated on a pneumatically modulated GC × GC system, and is applied for the first time to a thermally modulated GC × GC system. Preliminary one-dimensional experiments allowed the calculation of thermodynamic parameters (ΔH, ΔS, and ΔC p ) which were successfully applied for the prediction of the analytes' interactions with the stationary phases of both the first-dimension column and the second-dimension column. The model was able to predict both first-dimension and second-dimension retention times with high accuracy compared with the GC × GC experimental measurements. Maximum differences of -8.22 s in the first dimension and 0.4 s in the second dimension were encountered for the O-TMS derivatives of 11β-hydroxyandrosterone and 11-ketoetiocholanolone, respectively. For the MO-TMS derivatives, the largest discrepancies were from testosterone (9.65 ) for the first-dimension retention times and 11-keto-etiocholanolone (0.4 s) for the second-dimension retention times.

Theoretical modeling and machine learning-based data processing workflows in comprehensive two-dimensional gas chromatography—A review

“…McGinitie et al worked with a single standard mixture of alkanes, alcohols, and ketones for which average errors of 0.64 % for 1 t r and 2.22 % for 2 t r predictions were obtained [17]. Silva et al on the other hand reported average retention time errors of 3.6 s for 1 D and 0.2 s for 2 D for endogenous steroids [82]. Regardless of the chemical composition of the used samples, most of the published works reported higher modeling errors for the 2 t r predictions in comparison with those of the 1 t r .…”

Application of comprehensive 2D chromatography in the anti-doping field: Sample identification and quantification

“…GCxGC predicted retention times for steroids have been modelled based on thermodynamic properties. [43] Silva et al reported the advantage of this is that these thermodynamic models are independent of the operating conditions and that the use would aid in the identification of unknowns, as the authors suggest this is still a limiting factor in GCxGC due to the "thousands of peaks". Following on from previous work, this was based on the use of a thermally modulated system.…”

Top-Down Approach to Retention Time Prediction in Comprehensive Two-Dimensional Gas Chromatography–Mass Spectrometry

“…For instance, maximum 2 t r deviations of 1.2 and 0.4 s along with a 2 t r relative deviation of 7% were reported in the literature for cryogenically modulated systems. 9,11,13 Contrastingly, average 2 t r relative errors (REs) of 2.22 and 2.09% were reported for flow-modulated systems. 10,18 The lack of direct user pressure control in cryogenically modulated systems, except for the pressure control at the inlet, was argued to be one of the main contributors to this difference.…”

“…On the other hand, they help shift the analytical chemist’s focus to the more onerous, tedious, and time-consuming data-processing step. In this respect, considerable modeling efforts were dedicated to studying the GC × GC separation mechanisms, most of which have patently focused on modeling the thermodynamics of the separation. − …”

“…However, reaching the same modeling precision for the second-dimension retention times ( 2 t r ) continues to be a significant challenge, especially for systems using cryogenic modulation due to insufficient assessment of the impact of the analytes’ sorption and desorption on their retention behavior in the modulator. For instance, maximum 2 t r deviations of 1.2 and 0.4 s along with a 2 t r relative deviation of 7% were reported in the literature for cryogenically modulated systems. ,, Contrastingly, average 2 t r relative errors (REs) of 2.22 and 2.09% were reported for flow-modulated systems. , The lack of direct user pressure control in cryogenically modulated systems, except for the pressure control at the inlet, was argued to be one of the main contributors to this difference . Multiple efforts were dedicated to decreasing the 2 D retention time modeling errors.…”