蓝色圆珠笔油墨研究

发布时间：2018-10-31 18:09

【摘要】： 圆珠笔油墨字迹的鉴定已成为法庭科学的重要内容。建立快速可靠的分析检测方法,从而准确有效地鉴定圆珠笔的种类和书写形成时间,对解决刑事案件、经济和民事纠纷具有重要意义。圆珠笔油墨成分复杂,利用色谱技术和色谱-质谱联用技术可对其成分进行有效地分离和鉴定。由此产生的大量已知样品的光谱和色谱数据,可借助化学计量学方法建立数学模型,实现对未知样品的识别。这些为建立圆珠笔油墨库奠定了基础。本文研究的主要内容如下: 一、利用反相高效液相色谱和液相色谱-质谱联用技术对蓝色圆珠笔油墨字迹中难挥发性的染料成分进行研究。实验结果表明油墨中主要含有磺化铜酞菁、罗丹明B、碱性品蓝、甲基紫、结晶紫、碱性艳蓝B和碱性艳蓝BO等七种染料成分,据此用流程图将圆珠笔分成九大类。质谱强大的定性能力不仅确证了碱性品蓝和碱性艳蓝B脱甲基峰以及碱性艳蓝BO和罗丹明B脱乙基峰的存在而且发现了新的染料成分维多利亚4R和碱性品红及芳香胍类物质的存在,这样就解决了液相分析中一些未知的色谱峰的确认问题。二、运用化学计量学方法对蓝色圆珠笔油墨字迹的紫外可见光谱和液相色谱数据进行主成分分析和人工神经网络识别。首先利用主成分分析(PCA)分析蓝色圆珠笔油墨字迹的紫外可见光谱和液相色谱数据,通过对前三个主成分的二维、三维投影显示,对圆珠笔油墨进行了初步分类识别。其次将人工神经网络(主要是反向传播人工神经网络(BP-ANN)和径向基函数(RBF)神经网络)和主成分分析相结合来进行模式识别,通过对圆珠笔油墨样品的训练和预测,结果表明人工神经网络具有较高的分类准确率,可用于蓝色圆珠笔的分类识别。
[Abstract]:The identification of ink handwriting of ballpoint pen has become an important content of forensic science. It is of great significance to establish a rapid and reliable analysis and detection method to identify the type and writing time of ballpoint pen accurately and effectively, which is of great significance for solving criminal cases, economic disputes and civil disputes. The composition of ballpoint pen ink is complex and can be effectively separated and identified by chromatography and chromatography-mass spectrometry. The resulting spectral and chromatographic data of a large number of known samples can be identified by means of chemometrics to establish a mathematical model for the identification of unknown samples. These for the establishment of a ballpoint pen ink base laid the foundation. The main contents of this paper are as follows: 1. The volatile dyes in blue ballpoint pen ink were studied by reversed phase high performance liquid chromatography (RP-HPLC) and liquid chromatography-mass spectrometry (LC-MS). The experimental results showed that the ink mainly contained seven kinds of dyes, such as sulfonated copper phthalocyanine, rhodamine B, basic blue, methyl violet, crystal violet, alkaline brilliant blue B and alkaline brilliant blue BO, etc. According to the flow chart, the ballpoint pen was divided into nine categories. The strong qualitative ability of mass spectrometry not only confirmed the existence of demethylated peaks of basic blue and basic brilliant blue B, but also found new dye components Victoria 4R, basic fuchsin and aromatics, as well as the presence of basic brilliant blue BO and Rhodamine B deethylated peaks. The presence of vanguanidine, This solves the problem of confirming some unknown chromatographic peaks in liquid phase analysis. Secondly, principal component analysis (PCA) and artificial neural network (Ann) recognition of the UV spectrum and liquid chromatography data of blue ballpoint pen ink were carried out by means of chemometrics. At first, principal component analysis (PCA) was used to analyze the ultraviolet visible spectrum and liquid chromatography data of blue ballpoint pen ink. The first three principal components were shown by 2D and 3D projection, and the ink of ballpoint pen was preliminarily classified and identified. Secondly, the artificial neural network (BP-ANN) and the radial basis function (RBF) neural network are combined with the principal component analysis (PCA) to recognize the pattern and predict the ink sample of ballpoint pen. The results show that the artificial neural network has higher classification accuracy and can be used in the classification and recognition of blue ballpoint pen.
【学位授予单位】：首都师范大学
【学位级别】：硕士
【学位授予年份】：2008
【分类号】：D918.92

【参考文献】