Appendix D: Analytical Methods 



Retention behavior is temperature dependent and comparison of retention times obtained at two or 

 more temperatures may aid in determining a peak's identity. However, retention times are not 

 specific and despite the high resolution offered by capillary columns, two compounds of interest in 

 the same sample may have identical retention times. 



Strong additional evidence is obtained from comparison of retention times obtained on two 

 columns of different polarities, and from application of chemical reactions such as saponification, 

 treatment with Raney nickel, dechlorination reactions, treatment with basic ethanol or oxidation 

 with fuming sulphuric acid. The effects of the latter two methods on composition are summarized 

 in Table D.l. 



Confirmation of p,p'-DDT and p,p'-DDD involves dehydrochlorination with ethanolic 

 KOH solution. The p,p'-DDT is transformed to p,p'-DDE and p,p'-DDD is converted into p,p'- 

 DDMU. After GC/ECD analysis of the appropriate HPLC fraction, add 1 ml of ethanol and one 

 pellet of KOH to an aliquot of the extract in a ground glass concentrator tube equipped with a 

 Snyder column. Heat for 30 minutes. Allow the mixture to cool, remove the Snyder column, add 

 10 ml of clean distilled water and 1 ml iso-octane. Put a ground glass stopper on and extract by 

 shaking vigorously for 5 minutes. Centrifuge to separate the phases. Remove the iso-octane and 

 dry the extract by passing it through a Pasteur pipette plugged with glass wool which contains a 

 few grams of purified Na2SO"4. Concentrate the extracts to an appropriate volume for GC 

 analysis. The volume required to obtain sufficient response may be as low as 100 ill. The final 

 reduction in volume can be facilitated by using a gentle stream of high purity N2 over the solvent 

 (if this is not available, dry purified air may be substituted). 



P,p'-DDE is usually present at appreciably higher concentrations than the other members of 

 the DDT family. Its contribution can be distinguished qualitatively and quantitatively from that of 

 PCB components eluting at the same time in the HPLC fraction, by using fuming sulphuric acid as 

 an oxidising agent. This removes p,p'-DDE but does not affect the PCB's. After ECD/GC 

 analysis of this fraction, 1 ml of this oxidising agent added to an aliquot of the sample extract. The 

 two phases are shaken vigorously in a stoppered glass tube for 2 minutes. The phases are allowed 

 to separate and the upper layer is removed for injection into the GC. The decrease in the peak at 

 the appropriate retention time is a measure of the concentration of p,p'-DDE originally present in 

 the extract. 



Additional techniques, such as automated two-dimensional gas chromatography or negative 

 chemical ionisation GC/MS will be available for further confirmations as needed in the 

 confirmatory and reference laboratory of Professor Jan Duinker Insitut fur Meereskunde, Kiel, 

 Germany. 



99 



