926 PCBs AND ASSOCIATED AROMATICS
in the chromatogram is calculated from the individual
response factors of the detector. The total PCB content is
then obtained by summing the concentrations associated
with each peak.
An alternative, but more approximate, calculation,
noted by ASTM, can be made using the responses of the
larger, more cleanly separated peaks in both the standard
and the sample. Again, this makes the assumption that
the chromatograms of the standard and sample match one
another. ASTM notes that the PCB concentration calculated
in this way may be incorrect because the PCB content in an
individual peak may be reduced or relatively enhanced by
a non-standard isomer distribution or impurities. In order
to minimize the effect of variations between Aroclors,
ASTM suggests the use of a minimum of three peaks in the
sample being analyzed but adds that the simplified calcula-
tion should not be used in circumstances where maximum
accuracy is required. Indeed, EPA SW-846-8080 indicates
in $7.6.5.3 that PCB to residues should be quantitated by
comparing total area or height of residue peaks to total
area or height of peaks from appropriate Aroclor reference
materials.
The following calculations take the above considerations
into account on a different set of samples. Based on three
sets of calibration data for early, late and total areas of peaks,
chromatograms were recalculated to yield significantly dif-
ferent data as shown in Table 49.
The area of the chromatogram attributable to Aroclor
1016 in a standard was approximately 59%. The remaining
41% is attributable to a region from the start of timing to
2.16 min. Of this total, 39% is the response of the detector to
the solvent. This leaves only 2.5% which might be attribut-
able to Aroclor 1016. Thus, the error induced in the calcula-
tion by leaving this portion of the chromatogram out entirely
is very small relative to the error induced by including non-
PCB detector response in the same retention time range.
For example, for the sample marked as [‡] in Table 49, the
area of the total chromatogram which occurred prior to 2.19
min. was 74%. This should be compared with the 41% in an
Aroclor 1016 standard. In other words, in this case, about
70% of the total sample chromatogram area is attributable to
non PCB components.
§ 7.6.5.3 of the USEPA SW-846/8080 method states the
following: “Quantitate PCB residues by comparing total area
or height of residue peaks to total area or height of peaks
from appropriate Aroclor(s) reference materials. Measure
total area or height response from a common baseline under
all peaks. Use only those peaks from the sample that can
be attributed to chlorobiphenyls. These peaks must also be
present in a chromatogram of reference materials. Mixtures
of Aroclors may be required to provide a best match of GC
patterns of sample and reference.”
The gas chromatographic trace, in many cases, does
not conform with what would be expected from a stan-
dard Aroclor and, consequently, the quantitation of the
PCB components must be done by comparing total area
or height of residue peaks to total area or height of peaks
from appropriate Aroclor(s). These requirements have beenmet in Table 50 and are contrasted with other methods of
calculation.
It was not reasonable to attempt to quantitate the PCB
residue contained in these samples by peak height because
in part, the chromatogram derived from the instrumenta-
tion typically does not lend itself to that type of quantitation
method. In order to illustrate the point, PCB concentrations
have been calculated using the peak height method with
peaks appearing at different retention times and the results
listed in Table 51.
The peak height method is clearly not acceptable.
Inconsistencies in the peak height ratios have also been
calculated. The peak height ratios of peaks in a standard
Aroclor pattern should remain the same from sample to
sample. However, because the samples contained PCB resi-
dues and were not standard Aroclors the peak height ratios
vary. Peak height ratios are presented in Table 52. For this
fact alone, the method of peak height analysis is not valid
for these samples and the analyst must revert to the only other
alternative provided by the EPA SW-846 8080 method, that is,
the comparison of total areas.
Similar inconsistencies are shown by determining rela-
tive area ratios. Table 53 shows the result of calculating area
ratios relative to a peak at R t 7.75 min which is prominent
in the standard samples.
Clearly, the experience of the analyst is critical to the
analytical data derived. The precision and accuracy of an
analytical protocol, as measured by standard methods, may
indicate that the quantitation of ideal mixtures is within the
expected limits. Without such a measure of control there is a
much reduced likelihood of achieving reliable, scientifically
defensible, results.THE DISPOSAL OF POLYCHLORINATED
BIPHENYLS: CONCENTRATIONS OF CONCERNMethods for the dechlorination and destruction of polychlori-
nated biphenyls are reviewed in terms of the fundamental prin-
ciples involved and the reaction products formed. This section
is organized into three parts (1) the disposal of askarel liquids;
(2) the disposal of PCB contaminated liquids and (3) the reme-
diation of PCB contaminated soils. Figure 46 presents disposal
methods for high concentration PCBs.(1) PCBs are refractory organic compounds which
require a large energy input for their decompo-
sition. The methods in use for the destruction of
fluids which contain high percentages of PCBs are
therefore typically large energy sources. The most
widely used method for the disposal of askarel
fluids, and that preferred by the EPA is high tem-
perature incineration. However, other methods
discussed here include chemical methods, a DC
arc and the use of a microwave plasma.
(2) The EPA defines a PCB fluid as one which contains
more than 500 ppm of PCB. For this discussion,
liquids are considered to be contaminated withC016_003_r03.indd 926C016_003_r03.indd 926 11/18/2005 1:12:43 PM11/18/2005 1:12:43 PM