静电复印纸的鉴别是法庭科学物证检验中的一项重要工作。 建立显微共聚焦拉曼光谱技术结合化学计量学检验、 鉴别不同品牌、 型号静电复印纸的分析方法, 以实现对静电复印纸的无损检验和准确鉴别。 收集不同品牌、 不同型号的静电复印纸共计20种, 利用激光波长为785 nm的半导体激光器, 采集不同纸张样品的拉曼光谱数据, 分析每种纸张样品中的主要特征峰及对应的物质成分; 将光谱数据使用沃尔德系统聚类分析法进行分类, 并采用主成分分析法评价聚类分析的鉴别结果。 研究发现, 不同纸张样品的主要特征峰集中在900~1 700 cm^-1范围内, 分别位于714, 892, 1 092, 1 119, 1 143, 1 343, 1 385, 1 470, 1 510和1 600 cm^-1附近, 主要成分为纤维素、 木素和碳酸钙; 各纸张样品的光谱曲线虽然相互交叠, 但峰强度和峰面积存在一定差异, 可利用化学计量学中的聚类分析和主成分分析对纸张样品的光谱数据进行分类鉴别。 根据系统聚类分析的树状图和按计划表绘制的散点图可将20种不同品牌、 不同型号的静电复印纸样品分为四类, 其中第Ⅰ类中包含10份样品, 第Ⅱ类中包含3份样品, 第Ⅲ类中包含6份样品, 第Ⅳ类中仅包含1份样品。 再对纸张样品在900~1 700 cm^-1范围内的光谱数据进行主成分分析, 在17个主成分中前两个主成分累计贡献率已达到84%, 包含了绝大部分的光谱信息; 基于前两个主成分绘制纸张样品拉曼光谱数据的主成分得分图, 发现聚类分析的结果在主成分得分图中得到了很好的验证, 第Ⅰ~Ⅳ类所包含的各小类都能聚集在一块、 区分明显, 分类鉴别的结果准确、 合理。 该方法在使用时不会损坏纸张样品, 且操作过程简便, 鉴别效果较为理想, 可适用于法庭科学中对文件物证的检验和分析, 为物证溯源提供线索和依据。

The identification of electrostatic copy paper is an important work in forensic science physical examination. Establish the analysis method of microscopic confocal Raman spectroscopy combined with Chemometrics to examine different brands and models of copying paper, to achieve the non-destructive inspection and accurate identification of copy paper. The online shopping platform was used to collect 20 kinds of electrostatic copy paper of different brands and models. The Raman Spectra data of different paper samples were collected by using the laser wavelength of 785 nm semiconductor laser. The main characteristic peaks in each paper sample and their corresponding components were analyzed. The spectral data were classified by Wohlde hierarchical clustering analysis, and the discrimination results were evaluated by principal component analysis (PCA). It was found that the main characteristic peaks of different paper samples were concentrated in the range of 900~1 700 cm^-1, respectively around 714, 892, 1 092, 1 119, 1 143, 1 343, 1 385, 1 470, 1 510 and 1 600 cm^-1, and the main components were cellulose, lignin and calcium carbonate. Although the spectral curves of each paper sample overlap each other, there are some differences in peak intensity and peak area. The spectral data of paper samples can be classified and identified by cluster analysis and principal component analysis in Chemometrics. According to the tree diagram of the system cluster analysis and the scatter diagram drawn in light of the schedule Table, 20 kinds of copy paper samples in different brands and models can be divided into four categories. Among the four categories, 10 samples are included in ClassⅠ and 3 samples are included in ClassⅡ, ClassⅢ contains six samples and ClassⅣ contains only one sample. Then PCA of spectral data of paper samples in the range of 900~1 700 cm^-1, the contribution of the first two principal components in 17 principal components reached 84%, which contained most of the spectral information. Based on the first two principal components, the principal component scores of the Raman Spectrum data of paper samples were drawn. It was found that the results of cluster analysis were well verified in the principal component scores. All the subclasses contained in Class Ⅰ-Class Ⅳ can be grouped and distinguished clearly. The results of classification and identification are accurate and reasonable. This method can not damage the paper samples when used, the operation process is simple, and the effect of identification is ideal. It can be applied to the examination and analysis of documents material evidence in forensic science, and it can provide clues and a basis for tracing the source of material evidence.