Premium
High operationally stable sol–gel diglycidyloxycalix[4]arene fiber for solid‐phase microextraction of propranolol in human urine
Author(s) -
Li Xiujuan,
Zeng Zhaorui,
Hu Mingbai,
Mao Ming
Publication year - 2005
Publication title -
journal of separation science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.72
H-Index - 102
eISSN - 1615-9314
pISSN - 1615-9306
DOI - 10.1002/jssc.200500029
Subject(s) - solid phase microextraction , chromatography , fiber , chemistry , urine , materials science , gas chromatography–mass spectrometry , mass spectrometry , organic chemistry , biochemistry
A simple, sensitive, and accurate method for the determination of propranolol in human urine has been developed based on solid‐phase microextraction (SPME) followed by GC‐flame ionization detection (FID). The sol–gel 5,11,17,23‐tetra‐ tert ‐butyl‐25,27‐dihydroxy‐26,28‐diglycidyloxycalix[4]arene/hydroxy‐terminated silicone oil (diglycidyloxy‐C[4]/OH‐TSO) fiber was prepared to accommodate to the harsh extraction conditions. It possesses excellent alkali‐proof ability and retains its extraction characteristics intact even after treatment with highly alkaline (4 mol/L) NaOH solution. Direct chemical bonding of the coating to the fiber surface provides it with excellent solvent resistance and the introduction of calixarene enhances its thermal stability. The newly developed sol–gel calixarene coating was effectively used for the extraction of propranolol in human urine. No interference with the determination of propranolol was observed from the urine components. Standard curves were linear in the range 50–5000 μg/L for headspace‐SPME (HS‐SPME) and 25–25 000 μg/L for direct‐SPME (Dir‐SPME) with correlation coefficients better than 0.9999. The detection limit was 0.275 μg/L for HS‐SPME and 0.193 μg/L for Dir‐SPME. The method was validated using standard addition methodology and recovery values were between 91.4 and 117% for both the sampling modes with the RSDs less than 6% at different concentration levels in the linear ranges. The results obtained by both the sampling modes were feasible, and no significant differences between them regarding accuracy, precision, and detection limits were seen.