[Determination of individual components in finished motor gasoline by dual-channel three-column gas chromatography]
Chang-Xiu Li,Ya-Min Wang,Yi-Wei Zhang,Zheng Wang
DOI: https://doi.org/10.3724/SP.J.1123.2024.02013
Abstract:A method based on a dual-channel gas chromatograph equipped with three columns and three detectors was established for the determination of individual components in finished motor gasoline. The gasoline samples were separately introduced into the two injection ports of the chromatograph using two autosamplers. The components of the sample introduced into the first injection port (channel 1) were separated on a nonpolar PONA column (50 m×0.20 mm×0.5 μm) for gasoline analysis and detected by an flame ionization detector (FID). The components of the sample introduced into the second injection port (channel 2) were separated on another PONA column. Oxygenates (e.g., ethers and alcohols), other unconventional and prohibited additives that would co-elute with the hydrocarbons (e.g., methylal, dimethyl carbonate, sec-butyl acetate, and anilines), and some difficult-to-separate hydrocarbon pairs (e.g., 2,3,3-trimethylpentane and toluene) eluted from the PONA column and entered a DM-624 column (30 m×0.25 mm×1.4 μm) to achieve further separation according to the switch timetable using the Deans switch procedure and detected by an FID. The peak of 3-methylpentane, a common component in gasoline samples, also entered the DM-624 column by the Deans switch procedure for calculation purposes. The peak areas of target components on the PONA column in channel 1 were calculated using the peak areas on the DM-624 column as well as those of 3-methylpentane on both the DM-624 and PONA columns in channel 1 with a calibration factor between the two channels. The peak areas of co-eluted components were obtained by subtracting the calculated peak areas of the target components from those of the co-eluted peaks. The mass percentages of the individual components were calculated according to the normalization method using all peak areas on the PONA column in channel 1 with relative response factors. The mass percentages of the oxygenates, anilines, and individual hydrocarbons were determined, and the group-type distribution was calculated according to the carbon number. Separation and quantitation interferences between the additives and hydrocarbons were eliminated using this procedure. Twenty oxygenates and unconventional additives, each with a mass percentage of approximately 3%, were added to a real motor gasoline-92 sample and analyzed using the proposed method. The recoveries of the target components were between 90.1% and 118.2% with relative standard deviations (RSDs) between 0.2% and 5.1% (n=6). The analysis of a real ethanol-gasoline sample showed that the RSDs of contents of most components was less than 3% (n=6). Because the heart-cut of peaks using Deans switch technique requires the precise repeatability of retention times, the retention-time repeatability of components on the PONA column in channel 2 was investigated over an extended period of time after thousands of runs of real-sample analysis. The retention times of the same component in several randomly selected motor gasoline-92 samples varied from 0.01 to 0.03 min, indicating that the proper timetable for the Deans switch remained stable for two years. The precise repeatability of retention times was achieved owing to the high precision of the electric pneumatic control of the chromatograph and the stability of the column used. Real finished motor gasoline samples with different octane numbers (gasoline-92, gasoline-95, and ethanol gasoline-95) were analyzed using the developed method, and the results acquired were consistent with those of standard methods (GB/T 30519-2016, NB/SH/T 0663-2014, and SH/T 0693-2000). If some unconventional additives (such as methylal) were added to gasoline samples, the contents of these unconventional additives could also be detected, which means one run of the proposed method could provide results corresponding to three or four runs of different standard methods. The acquisition of information on the individual components of finished motor gasoline will assist in research on precise gasoline blending.