zhpsЖурнал прикладной спектроскопииZhurnal Prikladnoii Spektroskopii0514-7506B. I. Stepanov Institute of Physics of the National Academy of Scienceszhps-787Research ArticleСтатьиОПРЕДЕЛЕНИЕ КОНЦЕНТРАЦИИ ПОЛИАРОМАТИЧЕСКИХ УГЛЕВОДОРОДОВ В ВОДЕ С ИСПОЛЬЗОВАНИЕМ ПОВЕРХНОСТНО-УСИЛЕННОЙ СПЕКТРОСКОПИИ КОМБИНАЦИОННОГО РАССЕЯНИЯ В СОЧЕТАНИИ С АНАЛИЗОМ ОСНОВНЫХ КОМПОНЕНТOB ЯДРА И РЕГРЕССИЕЙ ОПОРНЫХ ВЕКТОРОВDETERMINING THE CONCENTRATION OF POLYCYCLIC AROMATIC HYDROCARBONS IN WATER USING SURFACE ENHANCED RAMAN SPECTROSCOPY AND KERNEL PRINCIPAL COMPONENTS ANALYSIS COMBINED WITH SUPPORT VECTOR REGRESSIONJianC.JianC.

102600, Пекин

Beijing, 102600

caojian@ss.pku.edu.cnBoyanJ.BoyanJ.

102600, Пекин

Beijing, 102600

boyan1212@qq.comYingZh.YingZh.

102600, Пекин

Beijing, 102600

351687656@qq.comZhenyuW.ZhenyuW.

102600, Пекин

Beijing, 102600

zw@ss.pku.edu.cnШкола программного обеспечения и микроэлектроники Пекинского университетаSchool of Software and Microelectronics at Peking University202129012021881173(1)-173(8)Copyright © Jian C., Boyan J., Ying Z., Zhenyu W., 20212021Jian C., Boyan J., Ying Z., Zhenyu W.Jian C., Boyan J., Ying Z., Zhenyu W.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://zhps.ejournal.by/jour/article/view/787

Для определения концентрации полициклических ароматических углеводородов в воде предложен альтернативный аналитический метод, основанный на использовании поверхностно-усиленной КР-спектроскопии и анализа основных компонентов ядра в сочетании с методом регрессии опорных векторов. Составлен набор данных, содержащий 300 КР-спектров полициклических смесей ароматических углеводородов, содержащих нафталин (NAP), пирен (PYR) и фенантрен с концентрациями 0–1000 ppb. Для улучшения эффекта обнаружения применены методы предварительной обработки данных: нормализация, мультипликативная коррекция рассеяния, детрендирование, стандартное преобразование с нормальным отклонением и сглаживание Савицкого–Голея. Для сравнения использованы методы частичных наименьших квадратов и регрессии опорных векторов с полиномиальным ядром. Метод предварительной обработки с наилучшим прогнозирующим эффектом для трех веществ – преобразованиe с нормальным отклонением. Для NAP оптимальный коэффициент корреляции перекрестной проверки (Rcv), коэффициент корреляции предсказания (Rpred), RMSECV, RMSEP и RPD, соответственно, 0.90, 0.937, 138.9, 117.4 и 2.9 ppb, для PYR — Rcv, Rpred, RMSECV, RMSEP и RPD 0.881, 0.897, 152.3, 142.8 и 2.3 ppb соответственно, для PHE — Rcv, Rpred, RMSECV, RMSEP и RPD 0.980, 0.982, 64.5, 62.9 и 5.3 ppb соответственно.

Determining the concentration of polycyclic aromatic hydrocarbons (PAHs) in water is vital for reducing negative effects on human health, such as cancer and malformation. This study proposed an alternative analytical method based on surface enhanced Raman spectroscopy and kernel principal components analysis combined with support vector regression (SVR) for the determination of PAH concentration in water. For this, a dataset containing 300 Raman spectra of polycyclic aromatic hydrocarbon mixtures was made using naphthalene (NAP), pyrene (PYR), and phenanthrene (PHE) with concentrations ranging from 0 to 1000 ppb. In order to improve the effect of the model detection, different pre-processed methods were applied: normalization, multiplicative scatter correction, detrending, standart normal variate transformation, and Savitzky–Golay smoothing. For comparison, partial least squares (PLS) and SVR with the polynomialkernel were also used. The pre-processing method with the best prediction effect was SNV for all the three substances. For NAP, the optimal correlation coefficient of cross-validation (Rcv), correlation coefficient of prediction (Rpred), RMSECV, RMSEP, and RPD are 0.90, 0.937, 138.9, 117.4, and 2.9 ppb, respectively, while for PYR the optimal Rcv, Rpred, RMSECV, RMSEP, and RPD are respectively 0.881, 0.897, 152.3, 142.8, and 2.3 ppb. For PHE, the optimal Rcv, Rpred, RMSECV, RMSEP, and RPD are 0.980, 0.982, 64.5, 62.9, and 5.3 ppb, respectively. This study provides a new method with a better prediction effect for quantitative analysis of low concentrations of polycyclic aromatic hydrocarbons in water by using surface enhanced Raman spectroscopy.

полициклические ароматические углеводородыповерхностно-усиленная КР-спектроскопияанализ основных компонентов ядрафункция ядрарегрессия опорных векторовpolycyclic aromatic hydrocarbonssurface enhanced Raman spectroscopykernel principal components analysiskernel functionsupport vector regressionThe research was funded by the Chinese Academy of Sciences Scientific Equipment Research Project (Grant No. YJKYYQ20170003).ReferencesI. Tongo, L. Ezemonye, K. A. Akpeh, Environ. Monit., 189, No. 6, 247 (2017).I. Tongo, L. Ezemonye, K. A. Akpeh, Environ. Monit., 189, No. 6, 247 (2017).N. Xiang, C. Jiang, T. Yang, P. Li, H. Wang, Y. Xie, et al. Chin. Ecotoxicol. Environ. Saf., 152, No. 8, 15 (2018).N. Xiang, C. Jiang, T. Yang, P. Li, H. Wang, Y. Xie, et al. Chin. Ecotoxicol. Environ. Saf., 152, No. 8, 15 (2018).M. L. Vestal, Science, 226, No. 4672, 275–281 (1984).M. L. Vestal, Science, 226, No. 4672, 275–281 (1984).B. Sharma, R. R. Frontiera, A.-I. Henry, E. Ringe, R. P. V. Duyne, SERS: Materials, Applications, and the Future. Mater. Today, 15, No. 1-2, 16–25 (2012).B. Sharma, R. R. Frontiera, A.-I. Henry, E. Ringe, R. P. V. Duyne, SERS: Materials, Applications, and the Future. Mater. Today, 15, No. 1-2, 16–25 (2012).J. Kneipp, H. Kneipp, K. Kneipp, Chem. Soc. Rev., 37, No. 5, 1052 (2008).J. Kneipp, H. Kneipp, K. Kneipp, Chem. Soc. Rev., 37, No. 5, 1052 (2008).H.-X. Gu, K. Hu, D.-W. Li, Y.-T. Long, Analyst, 141, No. 14, 4359 (2016).H.-X. Gu, K. Hu, D.-W. Li, Y.-T. Long, Analyst, 141, No. 14, 4359 (2016).X. Gu, S. Tian, Q. Zhou, J. Adkins, Z. Gu, X Li, et al. RSC Adv., 3, No. 48, 25989–25996 (2013).X. Gu, S. Tian, Q. Zhou, J. Adkins, Z. Gu, X Li, et al. RSC Adv., 3, No. 48, 25989–25996 (2013).D. W. Li, W. L. Zhai, Y. T. Li, Y. T. Long, Microchim. Acta, 181, No. 1-2, 23–43 (2014).D. W. Li, W. L. Zhai, Y. T. Li, Y. T. Long, Microchim. Acta, 181, No. 1-2, 23–43 (2014).R. Gautam, S. Vanga, F. Ariese, S. Umapathy, EPJ Tech. Instrum., 2, No. 1, 8 (2015).R. Gautam, S. Vanga, F. Ariese, S. Umapathy, EPJ Tech. Instrum., 2, No. 1, 8 (2015).S. Wold, Principal Component Anal., 2, No. 1, 37–52 (1987).S. Wold, Principal Component Anal., 2, No. 1, 37–52 (1987).A. C. Kak, A. M. Martínez, PCA versus LDA, 23, No. 3-4, 228–233 (2001).A. C. Kak, A. M. Martínez, PCA versus LDA, 23, No. 3-4, 228–233 (2001).P. Geladi, B. R. Kowalski, J. Anal. Chim. Acta, 185, No. 1, 1–17 (1986).P. Geladi, B. R. Kowalski, J. Anal. Chim. Acta, 185, No. 1, 1–17 (1986).A. J. Smola, B. Schölkopf, Stat. Comput., 14, No. 3, 199–222 (2004).A. J. Smola, B. Schölkopf, Stat. Comput., 14, No. 3, 199–222 (2004).L. A. Reisner, A. Cao, A. K. Pandya, Chemometr. Intell. Lab. Syst., 105, No. 1, 83–90 (2011).L. A. Reisner, A. Cao, A. K. Pandya, Chemometr. Intell. Lab. Syst., 105, No. 1, 83–90 (2011).T. Bocklitz, A. Walter, K. Hartmann, P. Rusch, J. Popp, Anal. Chim. Acta, 704, No. 1, 47–56 (2011).T. Bocklitz, A. Walter, K. Hartmann, P. Rusch, J. Popp, Anal. Chim. Acta, 704, No. 1, 47–56 (2011).A. Rinnan, F. V. D. Berg, S. B. Engelsen, Trends Anal. Chem., 28, No. 10 1201–1222 (2009).A. Rinnan, F. V. D. Berg, S. B. Engelsen, Trends Anal. Chem., 28, No. 10 1201–1222 (2009).H. Chen, C. Tan, Z. Lin, T. Wu, Spectrochim. Acta A: Mol. Biomol. Spectrosc., 189, 183 (2017).H. Chen, C. Tan, Z. Lin, T. Wu, Spectrochim. Acta A: Mol. Biomol. Spectrosc., 189, 183 (2017).H. S. Tapp, M. Defernez, E. K. Kemsley, J. Agric. Food Chem., 51, No. 21, 6110–6115 (2003).H. S. Tapp, M. Defernez, E. K. Kemsley, J. Agric. Food Chem., 51, No. 21, 6110–6115 (2003).J. K. Holland, E. K. Kemsley, R. H. Wilson, J. Sci. Food Agric., 76, No. 2, 263–269 (2015).J. K. Holland, E. K. Kemsley, R. H. Wilson, J. Sci. Food Agric., 76, No. 2, 263–269 (2015).X. L. Yang, Y. F. Li, X. W. Zhang, S. Q. Hu, Appl. Mech. Mater., 494-495, 964–967 (2014).X. L. Yang, Y. F. Li, X. W. Zhang, S. Q. Hu, Appl. Mech. Mater., 494-495, 964–967 (2014).F. Kuang, W. Xu, S. Zhang, Appl. Soft Comput. J., 18C, 178–184 (2014).F. Kuang, W. Xu, S. Zhang, Appl. Soft Comput. J., 18C, 178–184 (2014).P. C. Lee, D. J. J. Meisel, J. Phys. Chem., 86, No. 17, 3391–3395 (1982).P. C. Lee, D. J. J. Meisel, J. Phys. Chem., 86, No. 17, 3391–3395 (1982).P. S. Sampaio, A. Soares, A. Castanho, A. S. Almeida, J. Oliveira, C. Brites, Food Chem., 242, 196–204 (2018).P. S. Sampaio, A. Soares, A. Castanho, A. S. Almeida, J. Oliveira, C. Brites, Food Chem., 242, 196–204 (2018).G. Krepper, F. Romeo, D. D. S. Fernandes, P. H. G. Diniz, M. C. U. de Araújo, M. S. Di Nezio, et al. Spectrochim. Acta A: Mol. Biomol. Spectrosc., 189, 300–306 (2017).G. Krepper, F. Romeo, D. D. S. Fernandes, P. H. G. Diniz, M. C. U. de Araújo, M. S. Di Nezio, et al. Spectrochim. Acta A: Mol. Biomol. Spectrosc., 189, 300–306 (2017).O. Frank, J. Jehlika, H. G. M. Edwards, Spectrochim. Acta A: Mol. Biomol. Spectrosc., 68, No. 4, 1065–1069 (2007).O. Frank, J. Jehlika, H. G. M. Edwards, Spectrochim. Acta A: Mol. Biomol. Spectrosc., 68, No. 4, 1065–1069 (2007).J. Neugebauer, E. J. Baerends, E. V. Efremov, F. Ariese, C. Gooijer, J. Phys. Chem. A, 109, No. 10, 2100–2106 (2005).J. Neugebauer, E. J. Baerends, E. V. Efremov, F. Ariese, C. Gooijer, J. Phys. Chem. A, 109, No. 10, 2100–2106 (2005).I. Bandyopadhyay, S. Manogaran, J. Mol. Struct. THEOCHEM, 496, No. 1, 107–119 (2000).I. Bandyopadhyay, S. Manogaran, J. Mol. Struct. THEOCHEM, 496, No. 1, 107–119 (2000).A. Bree, R. A. Kydd, T. N. Misra, V. V. B. Vilkos, Spectrochim. Acta A: Mol. Spectrosc., 27, No. 11, 2315–2332 (1971).A. Bree, R. A. Kydd, T. N. Misra, V. V. B. Vilkos, Spectrochim. Acta A: Mol. Spectrosc., 27, No. 11, 2315–2332 (1971).

The authors declare that there are no conflicts of interest present.