This paper reports the first application of the silica based mesoporous material MCM-41 as a sorbent in solid phase extraction, to pre-concentrate pharmaceuticals of very different polarity (atenolol, nadolol, pindolol, timolol, bisoprolol, metoprolol, betaxolol, ketoprofen, naproxen, ibuprofen, diclofenac, tolfenamic acid, flufenamic acid and meclofenamic acid) in surface waters. The analytes were extracted from 100mL water samples at pH 2.0 (containing 10−3mol/L of sodium chloride) by passing the solution through a cartridge filled with 100mg of MCM-41. Following elution, the pharmaceuticals were determined by micro-liquid chromatography and triple quadrupole-mass spectrometry. Two selected reaction monitoring transitions were monitored per compound, the most intense one being used for quantification and the second one for confirmation. Matrix effect was found in real waters for most analytes and was overcome using the standard addition method, which compared favorably with the matrix matched calibration method. The detection limits in solvent (acetonitrile:water 10:90, v/v) ranged from 0.01 to 1.48μg/L and in real water extracts from 0.10 to 3.85μg/L (0.001–0.0385μg/L in the water samples). The quantitation limits in solvent were in the range 0.02–4.93μg/L, whereas in real water extracts were between 0.45 and 10.00μg/L (0.0045 and 0.1000μg/L in the water samples). When ultrapure water samples were spiked at two concentration levels of each pharmaceutical (0.1 and 0.2μg/L) and quantified using solvent based calibration graphs, recoveries were near 100%. However, recoveries for most pharmaceuticals were comparable or better than de described above, when river water samples (spiked at the same concentration levels) were quantified by the standard addition method and slightly worse using the matrix matched calibration method. Five real samples (two rivers, one dam and two fountain water samples) were analyzed by the developed method, atenolol, timolol, betaxolol, nadolol and diclofenac being found in some of them, at levels higher than their quantitation limits.
Read full abstract