Advances in solid-phase microextraction as sample preparation method for food analysis.
Souza-Silva, Erica Aparecida
MetadataShow full item record
Within all steps involved in the analytical process, sample preparation is considered the most time-consuming step. Therefore, substantial efforts have focused on the search for automated sample preparation strategies that minimize sample handling and errors associated with human interference. Solid phase microextraction (SPME) addresses well the necessity for simple and automated sample preparation, with the integration of sampling, extraction, clean up and instrumental introduction into a single step. In SPME, selective extraction of compounds takes place based on the degree of distribution of the analyte between the SPME coating and the sample matrix. For this reason, the correct choice of SPME coating for a given application has great influence on the acquisition of reliable analytical data. In spite of its great potential, the implementation of SPME in the analysis of complex matrices, such as food, has been hindered by the lack of suitable SPME coatings that possess compatibility with complex matrices while maintaining sufficient sensitivity for trace applications. The main problem resides in the fact that the most matrix compatible coating, PDMS, has limited extraction efficiency towards less hydrophobic analytes, whereas the coating that exhibits best extraction efficiency towards pesticides, in general, is PDMS/DVB. PDMS/DVB as a solid coating suffers from the attachment of matrix components onto the coating surface, known as fouling. Fouling does not only considerably shorten coating reusability, but it also causes significant changes in extraction efficiency, skewing the reliability of the data obtained. Therefore, in this thesis, a new approach to fabricate a matrix-compatible SPME coating for GC-based analysis of food matrices is presented. The developed matrix-compatible coating was evaluated for its reusability in complex matrices, namely grape pulp and Concord grape juice, as well as for its extraction capabilities towards various analytes bearing different physicochemical properties. First, a method to impart matrix-compatibility to commercially available solid SPME coatings was developed. The method consists of applying a thin layer of PDMS onto the solid coating, in this case PDMS/DVB. The main premise behind this approach was to create a coating that presents the matrix compatibility of PDMS, while maintaining the sensitivity obtained with PDMS/DVB. The reusability of the obtained PDMS-modified coating was evaluated in grape pulp, and rewarding results were obtained since the coating could be reused for over 100 extractions. Moreover, the PDMS-modified coating presented a similar extraction efficacy to that presented by the original PDMS/DVB coating towards the triazole pesticides, used as model analytes. The developed PDMS-modified coating was then employed to develop a simple and fast DI-SPME-GC-ToFMS method for determination of ten triazole fungicides in grapes and strawberries. The method was successfully validated, and the figures of merit obtained with the SPME method were compared to those obtained with the QuEChERS method. The limits of quantitation reached by SPME were at least one order of magnitude lower than those achieved by the QuEChERS method, whereas precision and accuracy were comparable for both methods. Subsequently, given the vast option of commercial PDMS blends available, different types of PDMS were compared for their reusability in complex matrices, and parameters associated with the PDMS-overcoated fiber fabrication were investigated in regards to their effect on fiber longevity. Results showed that the long-term reusability of such coatings is a function of the coating’s fabrication process, such as achievement of smooth and uniform PDMS surface, and sealing of both fiber ends by PDMS layer. Regarding PDMS type, best results were obtained with Sylgard ® 184. Since one of the most important branches of food analysis involves the simultaneous analysis of pesticides with a wide range of polarities and from different classes, the PDMS-modified coating was evaluated for the extraction of analytes of different polarities (log P = 1.43 to 6) from water samples in order to understand the mass transfer of analytes within the PDMS outer layer during the mass uptake process. Results showed that for hydrophobic analytes, the kinetics of extraction of the PDMS-modified coating are quite similar to that of the original PDMS/DVB. However, for more polar analytes, the rate-limiting step is the diffusion through the coating; therefore, the PDMS layer affects the kinetic uptake. The main implication of these results is quite evident if a method aiming at simultaneous determinations of both polar and non-polar analytes is to be developed, such as is the case in multiclass pesticide analysis, since the sensitivity of the method at too short extraction times might not be enough for polar analytes. Finally, once the PDMS-overcoated fibers were proven to be robust and compatible for use in fruit pulp, the DI-SPME-ToFMS method for multiresidue pesticide determination in grapes was developed and SPME parameters that can affect extraction efficiency were optimized via multivariate methods. Despite a thorough investigation during optimization, the most polar pesticides, acephate and omethoate, could not be detected. Next, a careful evaluation of internal standards was presented and attentively discussed. The results showed that two pairs of internal standards, interchangeable amongst them (i.e. only two internal standards were needed) were sufficient to ensure reliable, precise and accurate analytical data. Interestingly, only two internal standards at the time were needed, and among the choices presented, the use of non-deuterated compounds presents an affordable, cost-effective solution for the method. Next, the method was fully validated for 40 pesticides in compliance to EU/SANCO requirements (R2 > 0.995, RSD < 20%, and 80% < accuracy < 120%). The validated method exhibited excellent performance for pesticides such as chlorothalonil, dicofol, and folpet, which are considered the weak link in QuEChERS-based multiresidue methods. Pyrethroid pesticides were not validated due to their non-specific adsorption onto the vial walls. For pyrethroids, a solvent pre-extraction step should be incorporated in order to avoid losses due to the interaction of these compounds with glassware. Overall, despite the challenges and limitations encountered, it is evident that the practical aspects of the PDMS-modified coating demonstrated in this thesis create new opportunities for SPME applied in food analysis.