Evaluation of automated clean-up for large scope pesticide multiresidue analysis by liquid chromatography coupled to mass spectrometry

•Automated µSPE workflow for the analysis of 243 pesticides in fruits and vegetables.•Optimization of the automated µSPE process for LC-MS/MS.•Comparison of automated µSPE and dSPE for linearity, recoveries, and matrix effects.•Application of automated µSPE workflow for the analysis of real samples by LC-MS/MS. Clean-up step is essential during the multiresidue sample preparation process to remove undesired matrix components that may cause analytical interferences or suppression effect. However, its application generally by specific sorbents entails time-consuming work producing low recoveries for some compounds. Moreover, it usually needs to be adapted to the different co-extractives from the matrix present in the samples by using different chemical sorbents increasing the number of validation procedures. Therefore, the development of a more efficient and automated and unified clean-up procedure means a significant time reduction and laboratory work with improved performance. In this study, extracts from different matrices (tomato, orange, rice, avocado and black tea) were purified by manual dispersive clean-up (different procedures according to the matrix group) in parallel with an automated µSPE clean-up workflow, in both cases based on QuEChERS extraction. The latter procedure employed clean-up cartridges containing a mixture of sorbent materials (anhydrous MgSO4/PSA/C18/CarbonX) suitable for multiple matrices. All the samples were analysed by liquid chromatography mass spectrometry and the results obtained from both procedures have been compared in terms of the extract cleanness, performance, interferences, and sample workflow. At the levels studied, similar recoveries were achieved by both techniques (manual and automated) except for reactive compounds when PSA was used as the sorbent material producing low recoveries. However, the µSPE recoveries were between 70–120%. Furthermore, closer calibration line slopes were provided when µSPE was applied to the different matrix groups studied. It is important to note that up to 30% more samples per day can be analysed using an automated µSPE compared to the manual method (which requires shaking, centrifuging, then taking the supernatant and adding formic acid in ACN); it also provides good repeatability - an RSD (%) < 10%. Consequently, this technique is a very useful option for routine analyses, greatly simplifying the work of muti-residue methods.