Quantitative Surface-enhanced Raman Spectroscopy Based On The Combination Of Microfluidics With An Advanced Chemometric Model

Due to its unique advantages, such as narrow spectroscopic bands with excellentmolecular specificity, reduced photo-bleaching, simple pretreatment, and ultra highsensitivity, surface-enhanced Raman spectroscopy (SERS) has attracted substantialresearch interests and has been successful applied to many areas. Although SERS hasabove-mentioned significant advantages, its limitations are also very straightforward.SERS effect relies heavily on the preparation of nano-rough metal enhancingsubstrates (such as nano-silver or nano-gold colloids). The absolute intensities ofSERS signals depend on not only the concentrations of the analytes of interest, butalso the degree of aggregation, the particle size and shape of the metal colloids, andlaser focusing position as well. Therefore, the heterogeneity of the enhancingsubstrates can cause significant variations in the absolute SERS intensities and hencesignificantly deteriorate the precision and accuracy of the quantitative results of SERSassays. Therefore, novel methods are highly desirable to effectively eliminate theconfounding effects caused by variations in the physical properties of enhancingsubstrates, the intensity and alignment/focusing of laser excitation source, and realizeaccurate quantitative SERS analysis. In this thesis, microfluidics and advancedcalibration model—Multiplicative Effects Model for Surface-Enhanced RamanSpectroscopy (MEMSERS) have been employed to eliminate the confounding effectscaused by variations other than the concentrations of chemical constituents, and henceto improve the accuracy and precision of quantitative SERS assays. The main contentsof this thesis are as follows:In Chapter2, microfluidics was combined with MEMSERSmodel to improve thequality of quantitative results in SERS assays. The utilization of microfluidics inSERS detection aims to obtain more reproducible SERS signals through morehomogeneous mixing, more consistent geometries and more efficient heat dissipationin the flow detection mode. The MEMSERSmodel is then adopted to further eliminatethe confounding effects on SERS signals caused by possible variations in the physicalproperties of nanocolloids as well as the intensity and alignment/focusing of laserexcitation source, and finally realize accurate quantitative SERS assays.In Chapter3, the combination of microfluidics with MEMSERSmodel was appliedto the quantification of malachite green in environmental water samples using SERSspectroscopy. Experimental results showed that the combination of microfluidics with MEMSERSmodel could effectively improve the quality of quantitative results of SERSspectroscopy, and realize accurate quantitative SERS analysis of malachite green inenvironmental water samplesIn Chapter4, SERS spectroscopy incorporating with MEMSERSmodel was usedto quantitative determine the concentrations of Cd2+and Hg2+in environmental watersamples. Experimental results demonstrated that due to the multiplicative effectparameter introduced in MEMSERSmodel and the “dual calibration” strategy employed,MEMSERSmodel could effectively eliminate the confounding effects on SERS signalscaused by possible variations in the physical properties of nanocolloids as well as theintensity and alignment/focusing of laser excitation source, hence realize accuratesimultaneous quantitative analysis of Cd2+and Hg2+in environmental water samples.