Li Chang-Xiu, Wang Ya-Min, Zhang Yi-Wei, Wang Zheng
SINOPEC Research Institute of Petroleum Processing Co., Ltd., Beijing 100083, China.
Se Pu. 2024 Aug;42(8):773-782. doi: 10.3724/SP.J.1123.2024.02013.
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, -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% (=6). The analysis of a real ethanol-gasoline sample showed that the RSDs of contents of most components was less than 3% (=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.
建立了一种基于配备三根色谱柱和三个检测器的双通道气相色谱仪的方法,用于测定成品车用汽油中的各组分。使用两个自动进样器将汽油样品分别引入色谱仪的两个进样口。引入第一个进样口(通道1)的样品组分在用于汽油分析的非极性PONA柱(50 m×0.20 mm×0.5 μm)上分离,并由火焰离子化检测器(FID)检测。引入第二个进样口(通道2)的样品组分在另一根PONA柱上分离。含氧化合物(如醚类和醇类)、其他会与烃类共洗脱的非常规和禁用添加剂(如甲缩醛、碳酸二甲酯、乙酸仲丁酯和苯胺类)以及一些难分离的烃类对(如2,3,3 - 三甲基戊烷和甲苯)从PONA柱洗脱后,根据切换时间表采用迪恩斯切换程序进入DM - 624柱(30 m×0.25 mm×1.4 μm)进行进一步分离,并由FID检测。汽油样品中的常见组分3 - 甲基戊烷的峰也通过迪恩斯切换程序进入DM - 624柱用于计算。通道1中PONA柱上目标组分的峰面积通过通道1中DM - 624柱上的峰面积以及3 - 甲基戊烷在DM - 624柱和PONA柱上的峰面积,并利用两个通道之间的校正因子来计算。共洗脱组分的峰面积通过从共洗脱峰的峰面积中减去目标组分计算得到的峰面积来获得。各组分的质量百分比根据归一化方法,使用通道1中PONA柱上所有峰面积及相对响应因子来计算。测定了含氧化合物、苯胺类和各烃类的质量百分比,并根据碳数计算族组成分布。采用该程序消除了添加剂与烃类之间的分离和定量干扰。向实际的92号车用汽油样品中添加20种质量百分比约为3%的含氧化合物和非常规添加剂,并使用所提出的方法进行分析。目标组分的回收率在90.1%至118.2%之间,相对标准偏差(RSD)在0.2%至5.1%之间(n = 6)。对实际乙醇汽油样品进行分析表明,大多数组分含量的RSD小于3%(n = 6)。由于使用迪恩斯切换技术进行峰的中心切割需要保留时间的精确重复性,在对实际样品进行数千次分析后的较长时间内,研究了通道2中PONA柱上组分的保留时间重复性。在几个随机选取的92号车用汽油样品中,同一组分的保留时间变化范围为0.01至0.03分钟,这表明迪恩斯切换的合适时间表在两年内保持稳定。由于色谱仪的电动气动控制精度高以及所用色谱柱的稳定性,实现了保留时间的精确重复性。使用所建立的方法对不同辛烷值的实际成品车用汽油样品(92号汽油、95号汽油和95号乙醇汽油)进行分析,获得的结果与标准方法(GB/T 30519 - 2016、NB/SH/T 0663 - 2014和SH/T 0693 - 2000)的结果一致。如果向汽油样品中添加一些非常规添加剂(如甲缩醛),这些非常规添加剂的含量也能被检测到,这意味着所提出方法一次运行可提供相当于不同标准方法三到四次运行的结果。获取成品车用汽油各组分的信息将有助于精确汽油调合的研究。