3. Results and discussion3.1. Background contamination controlAs mentioned in Section 2, due to a wide use of PFCs, alsobackground contamination originated from various sources in lab-oratory might represent a serious problem. Investigation of blanksperformed at the beginning of method development confirmedthese concerns.All blanks (even from ultrapure water) showed the presenceof peaks corresponding to some of the analytes of interest. Thisdenoted a generalized contamination. Moreover the source of thecontamination was hard to predict also because was variable frombatch to batch for both solvents and disposable plastic materi-als. Then we redistilled in glass the solvents and pretreated allthe materials used for analysis. As ultrapure water was obtainedby a water purification system, it might represent a possible con-tamination source. For this reason we used for washing stepsoligomineral water bottled in glass and ultrapure water was usedonly for HPLC mobile phase preparation. As a result, a cleaner blankchromatogram was obtained, but a contamination from PFUdAand (less intense) from perfluorooctanesulfonic acid still persisted.After excluding contamination by organic solvent and water, asmentioned above, and after systematic elimination of the otherpossible sources coming from extraction process (see Section 2.2),we concluded that the contamination came from the instrumen-tation, probably from chromatographic system tubes [5]. For thisreason we changed all possible HPLC tubing and replaced othertube in PEEK. At the end, the PFUdA contamination was determinedto originate from the degasser membrane [17]. For this reason, ashort C18column was placed between the mixer and the sampleloop for retaining the PFCs interferences coming from the mobilephase and HPLC apparatus [5]. In this way, retention times of PFCscoming from the system were higher than those of the injectedanalytes. Namely the short column delayed compounds retentiontimes coming from the LC system, and in this way were separatedfrom analytes contained in the sample. In Fig. 1A and B are shownthe chromatograms of a blank before and after, respectively, thetreatments described above.