11 Procedure
Safety precautions
Exercise normal laboratory safety precautions, wear gloves, impervious to mineral oil and light hydrocarbon solvents. Use only small quantities of flammable solvents on the workbench; handle larger volumes in a fume cupboard.
Ensure proper handling and disposal of PCB's and PCB contaminated equipment according to local regulations.
11.1 Sample treatment (clean-up)
11.1.1 Test portion
Weigh to the nearest 0,001 g, 0,9 g to 1,0 g of the test sample into a 10 ml volumetric flask. Add 1 ml, by pipette (5.8.3) of internal standard solution IS 2 (5.4.1) or IS 0,5 (5.4.2). Make up to the mark with solvent (5.1.1). Shake well to mix. If the sample is obviously wet, shown by the opacity of the solution, add anhydrous sodium sulphate and shake until a clear solution is obtained. This solution is designated solution A.
NOTE For samples of unknown PCB content, use IS 2 (5.4.1). For better accuracy, IS 0,5 is preferred for samples with expected PCB content less than 20 mg/kg.
11.1.2 Solid-phase column preparation
Attach the sulphonic acid column (5.9.1) to the top of the silica gel column (5.9.1) using an adapter (5.9.2).
Elute the column assembly three times with 2 ml of the solvent (5.1.1) in order to purify the stationary phase. Do not allow the adsorbent to dry at this stage.
11.1.3 Clean-up procedure
Transfer 500 μl +/- 5 μl of solution A (11.1.1) onto the top of the sulphonic acid upper column.
Add 0,5 ml solvent and apply a slight vacuum to distribute the sample evenly over the packing of the upper column. Wait at least 30 s before elution.
The elution shall be carried out at a maximum flow rate of 2 ml/min and the columns eluted each time until the solvent level is just above the top of the adsorbent (except for the final elution).
Elute the column combination twice with 1 ml aliquots of solvent (5.1.1) collecting the eluate in a 5 ml volumetric flask. Remove the sulphonic acid upper column and the adapter and elute the silica gel column twice with 0,5 ml of solvent, collecting the eluate in the same volumetric flask. Dilute to the volume mark with solvent (5.1.1); shake well to mix. This solution is designated solution B and used for GC analysis.
For some samples additional clean-up techniques may be necessary (see B.5).
11.1.4 Recovery
Dilute an Aroclor ® solution (5.2.1) in solvent (5.1.1) to obtain a concentration of 5 mg/l.
This solution is designated solution C.
Run a 500 μl aliquot of solution C through the clean-up procedure (11.1.3). The 5 ml eluate obtained is designated solution D.
Dilute 500 μl of solution C to 5 ml with solvent (5.1.1) to give a concentration of 0,5 mg/l Aroclor ®. This solution is designated solution E.
Add 50 4l internal standard solution IS2 (5.4.1) to solution D (from clean-up) and to solution E.
Inject solutions D and E into the GC and obtain a chromatogram and area table for each solution. Calculate the total amount of PCB for the Aroclor ® in each solution using the internal standard method (11.7.2.2) and calculate the recovery as follows:
The calculated recovery shall be greater than 95 %.
11.1.5 Commercial PCB standards
50 mg/kg commercial mixtures of PCBs in oil are used (5.2.2). If a standard of lower concentration is required, this is prepared by diluting (by weight) with unused PCB-free insulating liquid.
Commercial PCB standards are treated the same as the samples (11.1) and submitted to the clean-up procedure.
11.2 Background check
Run each new batch of solvent and a blank prepared with PCB-free insulating liquid (5.1.3) as per 11.1 through the GC to ensure that there are no spurious peaks.
A blank test portion should be run with every batch of samples and at least every 20 samples.
11.3 Determination
11.3.1 Sample and commercial PCB standard solutions, solution B (11.1.3) are injected into the GC. The GC is run under optimized conditions (clause 8).
11.3.2 Commercial PCB standards are run with every batch of samples, at least once every 10 samples.
Choose standards appropriate to the samples; typically run 10 mg/kg and 50 mg/kg standards of Aroclor ® 1260 in oil, (this is the most common commercial product) and other standards as required. Blank and control samples shall be subsequently used for quality assurance purposes.
NOTE Chromatograms of all commercial materials likely to be encountered may be run and retained for type recognition purposes.
11.3.3 Integrate the chromatogram to obtain a peak table listing peak number, retention time, peak area and/or peak height, as would be produced by any GC integrating system.
11.4 Determination of experimental relative retention times (ERRT)
11.4.1 Run the test mixture (5.5) prepared as per 11.1, using the same GC conditions that are used for sample analyses. Identify all the peaks by comparison with the example in Figure A.1 and calculate the ERRT for each peak as follows:
where
x is the chosen peak;
30 is the congener 30 (reference);
209 is the congener 209 (reference and internal standard);
t is the retention time from injection.
Tabulate the results in a similar manner to the example as shown in Table A.1 and enter the results in the data files (clause 9).
NOTE 1 ERRTs shall be determined and entered into the data files for each individual GC system. The system should be recalibrated if there are any changes in GC conditions (e.g. temperature program, etc.).
NOTE 2 Congeners 30 and 209 are chosen as reference peaks for the determination of ERRT as they are at each end of the chromatogram (test mixture) isolated from congeners occurring in commercial mixtures and enable accurate repeatable values of ERRTs to be obtained.
11.4.2 Reference peaks for data-processing system
Use the ERRT values of the congeners listed in Table 2 below (determined in 11.4.1) to designate the reference peaks in the data-processing system.
11.5 Calculation of corrected relative response factors (CRRF)
11.5.1 The experimental relative response factors (ERRFs) of the congeners may vary from instrument to instrument depending on the injection mode and the conditions of the electron capture detector. They may also differ from those originating from [4].
11.5.2 The chromatogram is divided into nine segments (see Figure A.1) each represented by one of the congeners listed in Table 3. These congeners are normally present in commercial mixtures as major components (see Table A.1).
11.5.3 Using optimised chromatographic conditions as in clause 8, inject a suitable aliquot of the cleaned-up congener mix calibration solution (5.7).
Determine the relative response factors (to DCB) for the congeners listed in Table 3 using the following equation:
where
As is the peak area/height for internal standard (DCB);
Ai is the peak area/height for congener i;
Ms is the concentration (4g/ml) of internal standard (DCB);
Mi is the concentration (4g/ml) of congener i.
NOTE CAUTION: some software packages may produce the inverse of this ERRFi.
Take the average ERRF for a minimum of three determinations.
11.5.4 From the determined ERRF and the theoretical RRF value given in Table 3, calculate the correction factor K i for each congener:
Ki = ERRF/RRF
for example, for congener 180: segment 7
where
determined ERRF = 1,030
literature theoretical RRF = 1,137 (Table A.2)
11.5.5 Multiply the RRFs of all peaks in each segment of the chromatogram (Table A.1) by the correction factor calculated for the corresponding selected congener.
Example: segment 7: congener 180;
multiply each RRF by 0,906.
The resulting table of the corrected relative response factors (CRRF) for possibles and probables will be the one to use for data files (9.1).
11.6 Examination of the chromatograms
Chromatograms should be examined visually for any spurious peaks or chromatography problems and for interferences.
NOTE The commercial mixtures of "tetrachlorobenzyl toluene" may be encountered and can be mistaken for PCBs. These can be identified by their distinctive patterns in the PCB region of the chromatogram.
11.7 Calculation of results
11.7.1 Qualitative analysis
Compare the sample chromatograms with standard chromatograms to identify commercial materials such as Aroclor ® 1242, 1254 and 1260 (see Figure B.1).
11.7.2 Quantitative analysis
11.7.2.1 Calculation
The computer programme/software should produce a list, containing peak number with corresponding PCB congener numbers and the concentration of PCBs (mg/l) for each peak calculated from the concentration of internal standard DCB.
11.7.2.2 Mass of PCB in each peak
The mass of each congener or co-eluting congeners (for each peak i) is calculated by the internal standard method using the corrected relative response factors. Use the corrected response factors derived from the "all probable" RRF column in Table A.1 where there is no evidence for dechlorination or the "all possible" values for dechlorinated samples (see 9.1).
where:
i represents the peak due to a congener or co-eluting congeners;
mi is the mass of peak i in the test portion in milligrams;
ms is the mass of internal standard, in milligrams in the test portion 11.1.1 (nominally 0,002 or 0,0005);
Ai is the the area/height of peak i;
As is the area/height of internal standard peak;
RRFs is the relative response factor of internal standard (= 1,000);
CRRFi is the corrected relative response factor of peak i.
11.7.2.3 The mass of all the individual peaks are summed to give the total mass of PCBs (mg) in solution B.
Total PCB content of the sample (mg/kg) is calculated using the initial weight of sample in the test portion (11.1.1). These values may be entered into the software so that the actual result is the total PCB content of the sample in mg/kg.
