Atmospheric pressure chemical ionisation (APCI) has previously been used as interface [17]. However, due to its poor sensitivity this ionsation technique was not applicable to the API 365 instrument used in this study. Thermal degradation of the analytes which may be attributed to the design of the ion source might be a reason for this observation. As a consequence, electrospray ionisation was chosen which is commonly known to be more influenced by coeluting matrix compounds compared to APCI. Because of its structural similarity to the analytes and similar chromatographic behaviour, 4-NBA semicarbazone was used as a suitable internal standard in order to compensate for variations of the MS response which may be caused by coeluting matrix compounds and day-to-day performance differences. A disadvantage of this internal standard protocol is, however, that 4-NBA semicarbazone had to be added after the derivatisation step with 2-NBA followed by the pH adjustment of the sample solution to pH 7.4 to avoid any side-reaction with the 2-NBA derivatives which might occur under acidic conditions, e.g. exchange of the 2-nitrophenyl group with the 4-nitrophenyl group due to equilibrium reactions. While this compromise does not compensate analyte loss due to incomplete extraction, the high recoveries and derivatisation yields of further sample preparation steps limit a further loss and ensure sufficient overall sensitivity of the method (see below). 4-NBA semicarbazone exhibits a distinctly lower sensitivity than its 2-NBA counterpart under the experimental conditions (voltage settings, MRM transition used) which made it necessary to add relatively large amounts equivalent to 2 μg g−1 tissue. The amount of internal standard was chosen so that similar signal areas were obtained for the I.S. and 100 μg kg−1 2-NBA-SC.