Blue Ballpoint Pen Ink Study

The identification of ballpoint pen ink handwritings was an important part for forensic science. And classifying and dating inks, which was of great significance to deal with crimes, economic cases and civil disputes, could be exactly and effectively identified by the exact, fast and reliable analysis mehods. The analytical techniques such as chromatography and chromatography coupled with mass spectromy had been explored as effective avenues for the separation and identification of components of writing inks, which were complex mixture. In order to deal with such a large number of spectrophotometry and chromatographic quantitative data of known ink samples, the chemometrics methods had been used to build mathematics models to recognize unkown samples. All above was the basis of building ballpoint pen ink databank.The main contents of this paper were as follows:1. The techniques of reverse phase high performance liquid chromatography(HPLC) and hyphenation of HPLC and mass spectromy(MS) had successfully been applied to investigating the nonvolatile dyestuff components of blue ballpoint pen ink handwritings. Experimental results showed that there were seven main dyestuffs of inks, such as Sulfonated Copper Phthalocyanine (SCP), Rhodamine B(RhB),Tetramethyl Para Rosaniline(TPR), Methyl Violet(MV), Crystal Violet(CV), Victoria Blue B(VBB), Victoria Blue BO(VBBO). According to these dye components all the ballpoint pen inks from different manufacturers could be classified into nine different groups in the flowchart form. Because of MS with powerful qualitative identification ability, it had been confirmed that TPR and VBB could give a daughter ion through demethylation, as well as VBBO and RhB could lose ethyl group. Furthermore, new dye component Victoria Blue 4R, basic fuchsin, and aryl guanidine were identified too. And thus some unknown chromatographic peaks of HPLC could be confirmed.2. The chemometrics methods such as principal component analysis(PCA) and artificial neural networks(ANN) had been used to analyze the UV–vis spectrophotometry and HPLC data of blue ballpoint pen inks. Firstly, PCA was applied to analyze the data of them for elementary discrimination on ink samples through two and three dimension projection display of the first three principal components. Secondly, , the methods of ANN(mainly back propagation artificial neural network(BP-ANN) and radial basis function(RBF) neural network) combined with PCA were adopted in the pattern recognition of UV–vis and HPLC data obtained from ink samples. By training and predicting the ink samples, the results indicated that ANN had such an accuracy rate of recognition that it could be used to discriminate the blue ballpoin pen inks.