基于单体烃分析的辛烷值测定方法中模糊聚类技术的应用研究

介绍了一种应用模糊聚类技术构建辛烷值预测模型的新方法。该模型用来由汽油单体烃数据预测汽油辛烷值,通过提取单体烃分析谱图中的140个工艺特征组分以及对辛烷值贡献大的组分为特征值进行模糊聚类。实际应用时,在待测汽油样品共同参与聚类的条件下,用与待测汽油样品为同一类、并有最小欧氏距离(<1.5)的3~10个样本作为构建辛烷值预测模型的样本。这种按新的建模样本选择方式得到预测模型的方法具有更好的辛烷值预测精度、更广的适用范围和更高的数据资源的利用率。     

一种新的汽油辛烷值的气相色谱测定方法

建立了一种新的气相色谱结果关联计算汽油辛烷值的方法。采用高分辨毛细管柱对汽油的组成进行测定,根据汽油单体烃组分的含量和纯组分辛烷值乘积的大小,将单体烃组分分为两组,每一组为一个变量,建立实测辛烷值与两个变量间的回归模型。实际分析时,根据不同的样品类型选择不同的模型进行关联计算即可获得色谱分析样品的辛烷值。对模型建立和应用过程中已知结构的化合物的辛烷值数据采用文献值。对只知碳数和类型而不能确定其化合物结构的组分,通过对文献数据进行统计计算,得到平均辛烷值与组分的碳数和类型的关系曲线,据此得到其平均辛烷值用于计算。与采用标准方法测定催化裂化汽油辛烷值的结果相比,该方法测定辛烷值的偏差约0.5个单位。该方法操作简单,用样量少,模型建立过程快速、简便,适合于微型反应器产物评价或炼厂稳定工艺装置的汽油辛烷值的监测。     

Fisher判别在气相色谱-质谱分析助燃剂燃烧残留物中的应用

为寻找一种用于火场助燃剂燃烧残留物鉴定的更为准确、有效的模式识别方法,对7种常见助燃剂在不同载体上的燃烧残留物样品及未知送检样品进行气相色谱-质谱（GC-MS）分析测试,通过特征组分分析鉴定出未知样品中含有汽油成分。同时运用Fisher判别及PCA（主成分分析）/Fisher判别联用两种判别方法对样本数据进行了分析处理,PCA/Fisher判别联用的结果表明送检样本中含有硝基油漆稀料成分,而仅使用Fisher判别的结果表明送检样本中含有93^#汽油。通过将两种分析方法所得结果与GC-MS特征组分分析的结果进行比对发现,Fisher判别能够对7种助燃剂燃烧残留物的样本实现更有效的分类,对未知样本的判别更为有效。该研究结果为火场助燃剂鉴定提供了新的数据分析手段。     

一种基于数据挖掘的计算机自动识别饱和脂肪烃的方法

通过对Nist107标准物质中饱和脂肪烃的质谱数据进行挖掘,发现了新的饱和脂肪烃的谱图特征。利用这些特征设置分步问答,使用MatLab编程实现并优化建立了自动识别饱和脂肪烃的方法。利用库内分子量小于450的标准质谱数据和本实验室测定的齿叶铁线莲(根)挥发油数据进行检验。库内检验结果表明:第一类错误(弃真错误)率为0.503%,第二类错误(存伪错误)率为0.008%。正构烷烃的误识率为0。实验数据验证结果表明:此法能准确识别饱和脂肪烃,用于识别正构烷烃时可克服因夹角余弦相似匹配中相似度接近而无法正确定性的缺点,为植物挥发成分测定、汽油样品的PONA分析和空气颗粒物中饱和脂肪烃的识别和总量的测定提供了一种新的快速的方法。     

气相色谱-原子发射光谱和溴加成法联合测定含烯烃汽油中的烯烃单体Ⅰ. 汽油中烯烃保留值的研究

通过对含烯烃汽油进行溴加成反应,对汽油中烯烃的保留值进行了研究。对171个溴化产物溴代烃进行了定性,并给出了溴代烃所对应的171个烯烃的定性结果,其中对C7前的溴代烃及烯烃组分给出了结构式。定性手段主要依据汽油中混合纯烯烃样品溴化后的气相色谱 质谱联用(GC MS)数据和气相色谱 原子发射光谱(GC AED)的元素比数据,确定溴代烃及其对应烯烃单体的分子式,然后利用已有的烯烃单体文献保留值数据、纯烯烃化合物的反应结果并结合碳数规律和沸点规律,确定了烯烃单体化合物的结构。该方法最大的优点是实现了选择性检测烯烃,从而排除了其他烃类化合物对烯烃测试的干扰,因此对于汽油中烯烃的分布有着更直观的反映。     

气相色谱-原子发射光谱和溴加成法联合测定含烯烃汽油中的烯烃单体Ⅰ. 汽油中烯烃保留值的研究

 通过对含烯烃汽油进行溴加成反应，对汽油中烯烃的保留值进行了研究。对171个溴化产物溴代烃进行了定性,并给出了溴代烃所对应的171个烯烃的定性结果,其中对C7前的溴代烃及烯烃组分给出了结构式。定性手段主要依据汽油中混合纯烯烃样品溴化后的气相色谱-质谱联用(GC-MS)数据和气相色谱-原子发射光谱(GC-AED)的元素比数据,确定溴代烃及其对应烯烃单体的分子式,然后利用已有的烯烃单体文献保留值数据、纯烯烃化合物的反应结果并结合碳数规律和沸点规律,确定了烯烃单体化合物的结构。该方法最大的优点是实现了选择性检测烯烃,从而排除了其他烃类化合物对烯烃测试的干扰,因此对于汽油中烯烃的分布有着更直观的反映。     

原位实时近红外光谱研究核壳乳液聚合过程

将苯乙烯(St)和丙烯酸丁酯(BA)单体以不同的聚合方式制备核壳乳液和共聚乳液, 并采用近红外光谱技术实现了对乳液反应过程的原位实时监测, 通过对近红外光谱的谱带归属和主成分分析, 为近红外光谱技术判别乳液聚合过程提供了科学依据, 也为判断反向核壳乳液核壳翻转的拐点提出了一种新的方法. 采用簇类独立软模式法(SIMCA)建立了定性判别模型, 得到了很好的判别结果, 为进一步研究近红外光谱技术用于核壳乳液聚合过程奠定了基础.     

光谱结合主成分分析和模糊聚类方法的样品聚类与识别

针对紫外光谱结合化学计量学方法快速测定渣油烃族组成模型适应性问题,对渣油光谱进行主成分分析,以主成分得分作为聚类的特征变量进行模糊聚类,建立了光谱结合主成分分析和模糊聚类方法的样品聚类与识别方法和识别,为光谱结合化学计量分析方法中构正模型的正确选择提供了依据。     

偏最小二乘法在红外光谱识别茶叶中的应用

采用漫反射傅立叶变换红外光谱(FTIR)法结合主成分分析(PCA)、偏最小二乘法(PLS)、簇类的独立软模式(SIMCA)识别法对十三种茶叶进行了分类判别研究。研究结果表明,通过多元散射校正(MSC)对原始光谱进行预处理,可以提高模式识别技术的分类判别效果。在此基础上,选取1 900~900 cm-1波长范围内的茶叶红外光谱建立识别模型,三种方法都得到了满意的分类判别效果。在对检验集中全部130个样本的判别中,PCA仅有两类样本无法判别,SIMCA的识别率和拒绝率都在90%以上,而PLS的识别效果最佳,全部样本都得到了正确的归类。这一研究结果表明傅立叶变换红外光谱法与化学计量学方法相结合可以实现茶叶品种的快速鉴别,这为茶叶的客观评审提供了一种新思路。     

毛细管气相色谱法同库仑法联合测定含烯汽油的烃族及芳烃组成

本文在使用填充柱色谱法与库仑法联合测定含烯汽油烃族组成的基础上,又使用涂以极性固定液的FFAP WCOT毛细柱分析了含烯汽油的饱和烃/烯烃及芳烃、芳烃单体烃的含量。通过芳烃单体烃的测定确定油品的碳数及准确扣除高沸点非芳烃组分在芳烃部分中的含量,使定量更加准确可靠。     

气相色谱/氢火焰离子化检测法分析烷基化物料中的C4氟代烃

以实验室合成的氟代叔丁烷、氟代仲丁烷、氟代正丁烷为参考,建立了烷基化物料中C4氟代烃的气相色谱/氢火焰离子化检测(GC/FID)分析方法。提出了利用气相色谱/原子发射光谱(GC/AED)按元素响应的特点求算C4氟代烃在GC/FID上相对校正因子的方法。方法采用OV-225(50 m×0.25 mm i.d.×0.25 μm)和SE-54(44 m×0.22 mm i.d.×0.25 μm)串联柱为分析柱,FID为检测器,校正归一化或间接外标方法进行定量,具有重复性好、应用移植便利、操作简单等特点,对氟代叔     

色谱与色谱/质谱法相结合分析热裂解汽油C9馏分

采用毛细管气相色谱-氢火焰离子化检测器（CGC-FID）和气相色谱-质谱法(GC/MS)分析了热裂解汽油C9 馏分的组成。实验使用PONA毛细管气相色谱柱(100 m×0.25 mm i.d.×0.5 μm),根据烃类化合物在PONA柱上的保留规律,以正构烷烃标样保留值作为碳数分布依据,定量分析了裂解汽油C9 馏分中烃类化合物的碳数分布和单体烃含量;用GC/MS联用技术和CGC保留值定性法相结合对裂解汽油C9 馏分中相对含量大于0.2%的39种化合物进行了定性。     

气相色谱法测定车用汽油中含氧化合物和苯胺类化合物

利用中心切割技术和双毛细管色谱柱系统,采用两次进样的方式,建立了气相色谱测定车用汽油中含氧化合物和苯胺类化合物的分析方法。第一次进样分析,组分首先进入非极性DB-1色谱柱(30 m×0.32 mm×1.0μm),按沸点由低到高的顺序分离,通过电磁阀切换将沸点小于2-己酮的组分切割至强极性GS-OxyPLOT色谱柱(10 m×0.53 mm×10μm)或CP-Lowox色谱柱(10 m×0.53 mm×10μm),其余重烃组分通过阻尼柱进入FID检测器。在GS-OxyPLOT或CP-Lowox色谱柱上,烃类组分与含氧化合物分离并进入检测器检测,消除了大量的烃类组分对含氧化合物测定的影响。第二次进样分析,设定电磁阀切换时间为间-甲基苯胺从非极性色谱柱流出的时间,苯胺类化合物在GS-OxyPLOT或CP-Lowox色谱柱上与烃类和含氧化合物分离并进入检测器检测。以乙二醇二甲基醚为内标化合物进行内标法定量。实现了在一套系统上同时测定车用汽油中添加的甲基叔丁基醚(MTBE)、甲醇、甲缩醛、乙酸仲丁酯、乙酸乙酯、苯胺、邻/间/对-甲基苯胺和N-甲基苯胺的含量,各组分的检测范围为0.01%~10%(质量分数),回收率为86.0%~102.6%。该法可以为车用汽油的质量控制提供有效的检测手段。     

Prediction of the physicochemical properties of gasoline by comprehensive two-dimensional gas chromatography and multivariate data processing

The estimation of physicochemical parameters such as distillation points and relative densities still plays an important role in the quality control of gasoline and similar fuels. Their measurements according to standard ASTM procedures demands specific equipments and are time and work consuming. An alternative method to predict distillation points and relativity density by multivariate analysis of comprehensive two-dimensional gas chromatography with flame ionization detection (GC×GC-FID) data is presented here. Gasoline samples, previously tested according to standard methods, were used to build regression models, which were evaluated by external validation. The models for distillation points were built using variable selection methods, while the model for relativity density was built using the whole chromatograms. The root mean square prediction differences (RMSPD) obtained were 0.85%, 0.48%, 1.07% and 1.71% for 10, 50 and 90% v/v of distillation and for the final point of distillation, respectively. For relative density, the RMSPD was 0.24%. These results suggest that GC×GC-FID combined with multivariate analysis can be used to predict these physicochemical properties of gasoline.     

气相色谱/氢火焰检测器快速检测海洛因毒品及其来源推测

建立一种海洛因样本的快速检测和来源推测的方法。采用气相色谱/氢火焰检测器(GC/FID)和主成分分析(PCA)的方法对675份云南、新疆两地海洛因样本进行了检测,实现了云南、新疆两地缴获海洛因毒品的特征性划分,阳性率达到83.5%以上。采用GC/FID技术和主成分分析的方法可以实现海洛因样本地理来源区分。     

Towards gas chromatography-mass spectrometry coupling protocols for both identifying and quantification essential oils of Thymus capitatus Hoff et Link

Essential oil of Thymus capitatus Hoff et Link is analysed by using four techniques: GC/pyrolyse/MS, GC/FID, electronic impact GC/MS (quadripole), and GC/MS (ion trap). Both major and trace components are analysed. The GC/pyrolyse/MS coupling provides reference to the exact mass compositions without any need of the previously purified references, neither for major or trace components. The comparison between this reference analysis and GC/FID shows that the FID response coefficients may vary by a mean 7% from one component to another. As it was expected, quadripole or ion trap response coefficients vary to a much greater extent (a mean 25%), although the two MS techniques response coefficients are first order consistent. We conclude that GC/MS coupling could be used not only as it is usual for reliable identifications, but also for a complete quantitative routine analysis of essential oils. Expected precision could be very similar to GC/FID precision provided correcting species by species the MS analysis by a mean value of the response coefficient measured for the MS 70 eV electronic impact ionisation technologies. The GC/pyrolyse/MS coupling is proposed as a relevant tool for analysing reference samples containing trace natural species that could not be purified.     

Determination of Deoxynivalenol in Poultry Feed by Three Gas Chromatographic Detection Techniques

Three different gas chromatographic detection techniques were applied for the determination of deoxynivalenol (DON) present in poultry feed samples. Extraction and cleanup procedures were kept the same for GC–FID, GC–ECD and GC–MS methods. Although all three GC methods provided good and comparable results, but more attention was focused on GC–FID due to its lower cost and easy availability in many laboratories. Therefore, a short 15 m DB-1 short column was introduced for the determination of DON in poultry feed to reduce the time of analysis and initial cost of column. An inter-laboratory study for GC–FID was performed in two laboratories using four naturally DONS-contaminated feed samples and one spiked with standard. The relative standard deviations for repeatability (RSDr) and relative standard deviations for reproducibility (RSD_R) of naturally contaminated feed were in the range 5–23 and 11–24 %, respectively. The Horwitz Ratio (HORRAT) was less than 1.0 in each sample. From the spiking test, recovery, RSDr, RSD_R and value of HORRAT were 93, 5, 11 and 0.6 %, respectively. For GC–FID method, limit of quantification was found to be 6 μg kg^?1. Thus, GC–FID method using 15 m DB-1 capillary column is sensitive and validated analytical method for the determination of DON for poultry feed.     

Trace Analysis of Methyl Tert‐butyl Ether in Water Samples Using New Ultrasound‐assisted Dispersive Liquid‐liquid Microextraction and Gas Chromatography‐flame Ionization Detection

In this study, simple and efficient ultrasound‐assisted dispersive liquid‐liquid microextraction combined with gas chromatography (GC) was developed for the preconcentration and determination of methyl‐tert‐butyl ether (MTBE) in water samples. One hundred microliters of benzyl alcohol was injected slowly into 10 mL home‐designed centrifuge glass vial containing an aqueous sample with 30% (w/v) of NaCl that was located inside the ultrasonic water bath. The formed emulsion was centrifuged and 2 μL of separated benzyl alcohol was injected into a gas chromatographic system equipped with a flame ionization detector (GC‐FID) for analysis. Several factors influencing the extraction efficiency such as the nature and volume of organic solvent, extraction temperature, ionic strength and centrifugation times were investigated and optimized. Using optimum extraction conditions a detection limit of 0.05 μg L^‐1 and a good linearity (r² = 0.998) in a calibration range of 0.1‐500 μg L^‐1 were achieved. This proposed method was successfully applied to the analysis of MTBE in tap, well and a ground water sam ple contaminated by leaking gasoline from an underground storage tank (LUST) in a gasoline service station.     

烟用香精的气相色谱法质量控制研究

采用配有氢焰检测器或催燃检测器的气相色谱仪,对烟用香精进行了快速分析。方法操作简便,结果可靠,可作为烟用香精原料、产品质量监控的有效手段。