165 
solid phase extraction (SPE). The detection was done by GC-MS in El and NCI mode. 
This procedure was considered a suitable basis for the method used in the present study. 
Although the analysis of dicofol by GC techniques has been described in several 
publications, there are indications that this may be problematic. 
PCP and TCPy are too polar for a direct and sensitive GC analysis. Because of the high 
polarity of the PCP hydroxyl group, gaschromatographic separation of the underivatized 
substance is problematic (strong tailing; losses by adsorption in the injector), rendering 
a reproducible interpretation of the peaks, and thus quantification, impossible. 
Therefore, PCPy is generally analysed after derivatisation. However, this makes the 
analyses labour-intensive and more susceptible to errors. Alternatively, methods using 
HPLC techniques which do not require a derivatisation step generally have a lower 
sensitivity and selectivity. Both approaches were tested in this survey. 
6.3.2 Method development 
The limits of detection in published methods generally were much too high for this 
study. Considering the low concentrations at which organic pollutants are present in 
marine matrices, the planned limits of quantification (LOQs) were 30 pg/L for water, 
50 ng/kg dw for sediment, and 0.5 pg/kg ww for biota samples (project proposal). In the 
course of the project, these targets were lowered further because observed 
concentrations in the North Sea and Baltic Sea very often were below these limits. 
Chlorpyrifos (-ethyl, -methyl), endosulfan (I, II), and trifluralin were analysed by a 
commonly used procedure, as planned. 
Dicofol proved to be fairly unstable under various conditions and, therefore, had to be 
treated and analysed separately. 
Because of the common phenolic substructure of PCP and TCPy, the methodology for 
PCP and TCPy was developed in a common procedure but separate from the other 
substances.