79 
As can be seen from Table 22, octachloronaphthalene (OCN) has a yield, which 
sometimes is more similar to the PC A compounds. Therefore, OCN was used as 
additional and alternative internal standard in case of a low recovery of 13 Cio-trans- 
chlordane. 
Chlordanes 
As already discussed before, frans-nonachlor cannot be collected completely in the PCA 
fraction. Tests of the two clean-up steps for biota gave good recoveries of 77-90 % (see 
Table 23). /ran.s-Nonachlor could be recovered to 75 % in the pre- and PCA-fraction. 
Therefore, both were quantified. 
However, cis- heptachlorepoxide was non-detectable, and the yield for trans-hepta- 
chlorepoxide was poor. The reason is degradation by the silica gel coated with sulphuric 
acid. Since this step is essential for a clean extract, these compounds could not be 
determined. 
Overall chlordane recoveries of the complete extraction and clean-up method were also 
tested with both sea-sand and air-dried sediment from the site UE 67 (Gauss 332, TOC 
0.12 %). Here, clean-up by silica gel coated with sulphuric acid could be omitted for 
chlordane analysis. Recoveries of the internal standards and the analytes were compa 
rable and in average >10%. 
Table 23: Relative recoveries [%] of selected chlordane compounds in biota after the two clean-up steps 
(column chromatography with silica gel/sulphuric acid and adsorption chromatography with Florisil ®). 
Samples of 10 g were spiked with 10 ng of each compound in 10 jal of cyclohexane. 
Compound 
Pre-fraction 
PCA-fraction 
ISTD 
13 Ci 0 -frani-chlordane 
4 
80 
e-hexachlorocyclohexane 
Chlordanes 
79 
<1 
frans-chlordane 
2 
77 
r/.v-chlordane 
2 
77 
r/.v-nonachlor 
1 
90 
frans-nonachlor 
29 
46 
Oxychlordane 
<1 
85 
rz.v-heptachlorepoxide 
<1 
24 
frans-heptachlorepoxide 
<1 
<1